Parent-targeted home-based interventions for increasing fruit and vegetable intake in children: a systematic review and meta-analysis

Parent-targeted home-based interventions for increasing fruit and vegetable intake in children: a... Abstract Context Parent interventions delivered in the home represent a valuable approach to improving children’s diets. Objective This review aims to examine the effectiveness of parent-targeted in-home interventions in increasing fruit and vegetable intake in children. Data Sources Five electronic databases were searched: MEDLINE, Embase, PubMed, CINAHL, and PsycINFO. Study Selection Randomized and nonrandomized trials conducted in children aged 2 to 12 years and published in English from 2000 to 2016 were eligible. Data Extraction Eighteen publications were reviewed, and 12 randomized trials were analyzed. Studies were pooled on the basis of outcome measure and type of intervention, resulting in 3 separate meta-analyses. Results Nutrition education interventions resulted in a small but significant increase in fruit intake (Hedges’ g = 0.112; P = 0.028). Taste exposure interventions led to a significant increase in vegetable intake, with a moderate effect (Hedges’ g = 0.438; P < 0.001). Interventions involving daily or weekly sessions reported positive outcomes more frequently than those using monthly sessions. Conclusions Future interventions should incorporate regular taste exposure to maximize increases in vegetable intake in children. This is particularly important because fewer children meet national recommendations for vegetable intake than for fruit intake. child, fruit, home, parent, vegetables INTRODUCTION Eating fruit and vegetables (FVs) can protect against obesity1 and reduce the risk of mortality from cardiovascular disease.2 Despite this, children across much of the developed world do not meet the recommendations for daily intake of FVs. In the United States and the United Kingdom, children’s intake of FVs remains below national recommendations for age and sex.3,4 In Europe, a high proportion (76.5%) of children do not meet the World Health Organization’s recommendation of 400 g of FV per day.5 According to the most recent Australian 2014–2015 National Health Survey, only 1 in 20 children aged 2 to 18 years meet the Australian guidelines for recommended daily servings of vegetables or FVs combined.6 Longitudinal analyses of children’s eating behaviors suggest that food preferences are developed early and are likely to persist into adulthood.7,8 This has led to increasing recognition of the need for early interventions that promote FV intake among children.9 School-based interventions are among the most commonly used strategies for increasing children’s intake of FVs, despite having minimal impact on children’s vegetable intake.10 However, systematic reviews suggest that school-based interventions that directly target parents, for example, via parent attendance at intervention sessions are more likely to report positive or mixed dietary outcomes (eg, increased FV intake and reduced fat intake) than are interventions that indirectly involve parents, for example, via a parent newsletter.11 It is well documented that parents exert substantial influence over their children’s FV intake.5,12 For example, studies have identified a number of parental factors associated with increases in children’s intake of FVs, including parent intake of FVs, parent providing of FVs to children, structured family mealtimes, set rules around eating, and the availability and accessibility of FVs in the home.5,13,14 Targeting parents in nutrition interventions may therefore represent a particularly effective strategy for increasing children’s FV intake. Several systematic reviews on interventions aimed at increasing FV intake in children are published in the literature.9,11,15 However, few have evaluated parent-targeted interventions (< 50%)9,11,15 and, where parental involvement was included, most did not require direct parental involvement (< 25%).11 Parent-targeted interventions have the potential to be more acceptable to parents when compared with community-based or multicomponent school-based interventions that involve a parent or family component. This is because parent-targeted interventions are more accessible, with the majority of interventions being delivered in the home environment via home visits,16 telephone calls,17 and online delivery methods, eliminating the need for parents to travel.17,18 Reducing barriers of participant access to interventions is also imperative for policymakers and health professionals interested in delivering parent-targeted strategies in rural and remote areas, where intervention is often limited.19,20 An increasing number of parent-targeted, home-based FV interventions have become available. For example, 2 recently published randomized controlled trials (RCTs) reported significant increases in children’s FV intake up to 12 months after the completion of an online21 or telephone-delivered22 parent-targeted FV intervention. Overall, however, the effects of parent-targeted home-based FV interventions are mixed, with some home-visiting parent-targeted FV interventions reporting no significant increases in children’s FV intake.19,23 It is therefore difficult to ascertain which parent-targeted home-based interventions may be most effective at promoting FV intake among children. It is possible that differences in the type of parent-targeted intervention being evaluated may have contributed to the inconsistent findings to date. Parent-targeted, home-based interventions have tended to focus on 2 strategies: repeated taste exposure, which involves repeatedly exposing children to a target vegetable at home (with or without a reward), and nutrition- and skills-based education. Taste exposure interventions are based on the theory of learned safety in which repeated exposure to vegetables, without negative consequences (eg, gastrointestinal upset), is hypothesized to support greater vegetable acceptance among children.24 Nutrition education interventions, however, have largely targeted evidence-based parent and home environment factors associated with increases in children’s FV intake. These include parent education and training in the national dietary guidelines, along with strategies to encourage parent modeling of FV intake, provide FVs to children daily, and increase the availability of ready-to-eat FVs to children in the home (eg, clean and chopped FVs).16 Although taste exposure and nutrition education are potentially valuable approaches for improving children’s FV intake, previous systematic reviews have not specifically examined the effect of parent-targeted, home-based FV interventions according to the type of intervention strategy delivered. Therefore, a systematic review of the acceptability and feasibility of taste exposure and nutrition education interventions and their effectiveness on children’s FV intake is needed. The findings will provide evidence-based information for parents, health professionals, and policymakers seeking effective strategies to increase FV intake in children. METHODS The aims of this study were to systematically identify and review the quality, feasibility, and acceptability of parent-targeted, home-based interventions aimed at promoting FVs to children and to undertake a meta-analysis to analyze the effects of such interventions on FV intake in children. Using the PICOS (Participants, Intervention, Comparators, Outcomes, Study Design) model for framing research questions, 4 key questions were identified: (1) Are parent-targeted home-based interventions aimed at increasing FV intake in children aged 2 to 12 years feasible and acceptable to parents? (2) Are parent-targeted home-based taste exposure and nutrition education interventions effective at promoting significant increases in children’s FV intake? (3) Do parent-targeted home-based taste exposure and nutrition education interventions, compared with no intervention, result in significantly greater FV intake in children aged 2 to 12 years? (4) What characteristics are common among parent-targeted home-based interventions that achieve significant increases in FV intake in children aged 2 to 12 years? The PICOS criteria are shown in Table 1. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (see Appendix S1 in the Supporting Information online), considered the gold standard for reporting the evaluation of interventions,25 were used when designing this review. This required the following: (1) a specific set of inclusion and exclusion criteria (a priori); (2) an extensive literature search across multiple databases; (3) full documentation of the electronic search strategy to enable replication; (4) use of a standardized process to screen and select studies; and (5) use of a standardized process to collect data and assess the risk of bias of individual studies. The protocol for this review has not been published and is not registered. Search strategy Five electronic databases (PubMed, PsycINFO, CINAHL, Embase, and Ovid MEDLINE) were searched for human studies published in English between January 2000 and February 2016. The following search terms were used in each database: (parent* OR home*) AND (child* OR preschool*) AND (fruit* OR vegetable*) AND (intervention*). This search was updated in August 2016. Selection criteria Inclusion and exclusion criteria are described in Appendix S2 in the Supporting Information online. Studies that evaluated a parent-targeted home-based intervention aimed at increasing children’s fruit and/or vegetable intake and were published in a peer-reviewed journal were included. Additional criteria were included, as follows. (1) The intervention was published between January 2000 and August 2016. (2) The intervention required a parent-only component (eg, parent telephone calls, parent website/newsletter, and parent-led exposure activity) and was delivered to the parent at home, for example, via the internet, telephone, or mail. Interventions that included a parent-only component but were delivered in a school, community, or research setting were excluded. (3) Interventions were aimed at increasing children’s fruit and/or vegetable intake. A child was defined as a person between the ages of 2 and 12 years. This broad definition was designed to capture preschool- and primary school-aged children, both of whom demonstrate poor intake of FVs.3,6,26 This is also the age range during which parents have the strongest influence on children’s food preferences.27 (4) Interventions were specifically designed to effect change in at least 1 measure of child fruit and/or vegetable intake. A range of outcomes was included to account for the variability of outcome measures used across studies.28 These included intake of a target vegetable in grams or daily servings of a child’s fruit and/or vegetable intake via parent self-report (eg, food frequency questionnaires, 24-hour recalls, or survey items). (5) Studies were designed as RCTs, nonrandomized controlled trials, or pre–post interventions. Study selection Figure 1 summarizes the process of identifying, screening, and including studies. Two searches yielded a total of 1867 abstracts, which were screened by 3 authors (L.M.T., M.S., and A.M.G.) using the established inclusion and exclusion criteria. To determine eligibility, full-text articles were retrieved for all abstracts that were judged to be eligible by at least 1 of these authors or that did not yield sufficient information in the abstract (n = 35). Initial disagreements were resolved through a comprehensive discussion between the authors in which the inclusion and exclusion criteria were compared against the full text. When an agreement could not be reached, consensus was achieved by employing a fourth author (J.C.). An extended search of the reference lists of included studies and table of contents of key journals did not yield any additional studies. Data extraction One author (L.M.T.) extracted the data from the included studies using a standardized table in Microsoft Excel.29 Extracted data items are shown in Table 2.16,17,19,21–24,30–42 Quality assessment The methodological quality of the included studies was assessed by consensus between 2 authors (L.M.T. and M.S.) using an adapted version of the Downs and Black checklist for assessing the quality of health care intervention studies.43 A third author (A.M.G.) was included when full agreement was not reached for all criteria (Table 316,17,19,21–24,30–42). Constructs of study quality included the following: (1) clear reporting of study hypotheses or aims, participant characteristics, and main findings; (2) clear reporting of attrition rate and methods of randomization, if applicable; (3) blinding of study participants to the intervention they received (RCT only); (4) blinding of those measuring the main outcomes of the intervention (RCTs only); (5) use of valid and reliable outcome measures; and (6) clear reporting of a sample power calculation.43,44 Studies that reported high attrition rates (> 30%) or reported outcome data only for participants who completed the intervention were evaluated as having lower quality due to potential outcome bias. Each construct of study quality was scored using the criteria “yes,” “no,” or “unclear” against the adapted Downs and Black criteria (Table 3).44 Table 3 Appraisal of methodological quality of the 18 publications included in the systematic review, based on the parameters proposed by Downs and Black43 Reference  Aim(s) clearly described  Participant characteristics clearly described  Main findings clearly described  Attrition rate acceptable (< 30%)  Randomization method clearly described  Participants blinded  Assessors blinded  Outcome measures valid and reliable  Risk of residual confounding  Sample power calculation clearly reported  Potential outcome bias  Corsini et al. (2013)30  Yes  Yes  Unclear  Yes  Unclear  No  No  Yes  Unclear  Yes  No  Cravener et al. (2015)31  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Dulin Keita et al. (2014)32  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Fildes et al. (2014)33  Yes  Yes  Yes  No  Unclear  No  Unclear  Yes  Unclear  No  Yes  Haire-Joshu et al. (2008)16  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Holley et al. (2015)24  Yes  Yes  Yes  Yes  No  Unclear  No  Yes  Low  Yes  No  Horton et al. (2013)23  No  Yes  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Knowlden et al. (2015)34  Yes  Yes  Yes  Yes  Yes  Yes  No  Yes  Low  Yes  No  McGowan et al. (2013)35  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Remington et al. (2012)36  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Schwinn et al. (2014)37  Yes  No  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Spurrier et al. (2016)19  Yes  Yes  Yes  No  N/A  N/A  N/A  Yes  N/A  No  Yes  Tabak et al. (2012)38  Yes  Yes  Yes  Yes  Unclear  No  No  Yes  Low  Yes  No  Thompson et al. (2015)39  No  Yes  Yes  Yes  Unclear  No  Unclear  Yes  Low  Yes  No  Tomayako et al. (2016)40  Yes  Yes  Yes  No  Unclear  Unclear  No  Yes  Low  No  No  Wardle et al. (2003)41  Yes  No  Yes  Yes  Unclear  No  No  Yes  Unclear  No  No  Wyse et al. (2012)17  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Wyse et al. (2011)42  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Percent “yes”  89  89  83  83  40  7  0  100  20  67  11  Reference  Aim(s) clearly described  Participant characteristics clearly described  Main findings clearly described  Attrition rate acceptable (< 30%)  Randomization method clearly described  Participants blinded  Assessors blinded  Outcome measures valid and reliable  Risk of residual confounding  Sample power calculation clearly reported  Potential outcome bias  Corsini et al. (2013)30  Yes  Yes  Unclear  Yes  Unclear  No  No  Yes  Unclear  Yes  No  Cravener et al. (2015)31  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Dulin Keita et al. (2014)32  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Fildes et al. (2014)33  Yes  Yes  Yes  No  Unclear  No  Unclear  Yes  Unclear  No  Yes  Haire-Joshu et al. (2008)16  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Holley et al. (2015)24  Yes  Yes  Yes  Yes  No  Unclear  No  Yes  Low  Yes  No  Horton et al. (2013)23  No  Yes  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Knowlden et al. (2015)34  Yes  Yes  Yes  Yes  Yes  Yes  No  Yes  Low  Yes  No  McGowan et al. (2013)35  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Remington et al. (2012)36  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Schwinn et al. (2014)37  Yes  No  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Spurrier et al. (2016)19  Yes  Yes  Yes  No  N/A  N/A  N/A  Yes  N/A  No  Yes  Tabak et al. (2012)38  Yes  Yes  Yes  Yes  Unclear  No  No  Yes  Low  Yes  No  Thompson et al. (2015)39  No  Yes  Yes  Yes  Unclear  No  Unclear  Yes  Low  Yes  No  Tomayako et al. (2016)40  Yes  Yes  Yes  No  Unclear  Unclear  No  Yes  Low  No  No  Wardle et al. (2003)41  Yes  No  Yes  Yes  Unclear  No  No  Yes  Unclear  No  No  Wyse et al. (2012)17  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Wyse et al. (2011)42  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Percent “yes”  89  89  83  83  40  7  0  100  20  67  11  Abbreviation: N/A; not applicable. A child’s FV intake may be influenced by factors other than age and sex, such as socioeconomic status, parent FV intake, and parent body mass index.12,45–47 Therefore, each RCT (n = 14) and nonrandomized controlled trial (n = 1) was further assessed for risk of residual confounders. Randomized control trials were rated as having adequately controlled for potential confounders (low risk of bias) when authors provided evidence of a comparison between the intervention group and the control group on important confounders at baseline and when no between-group differences were observed (eg, demographics table) or when differences were identified but were adequately controlled for in the final analysis (eg, by the inclusion of confounder[s] as covariate[s] in the analysis). Meta-analysis Of the 20 studies included in the systematic review,16,17,19,21–24,30–42 8 were excluded from the meta-analysis,17,19,21,22,32,34,40,42 leaving 12 interventions for analysis. Reasons for exclusion included the publication of long-term follow-up data of a previously included study,21,22 the absence of a control (n = 3),19,32,42 and the lack of an appropriate control group (n = 1) (eg, control group received a mail-based version of the intervention).40 Because of the differences between fruit intake and vegetable intake, data for FVs were extracted separately. The authors of studies that reported an outcome measure only for fruit intake and vegetable intake combined were contacted to request information that would enable a separate analysis. One author was unable to provide separate scores for fruit intake and vegetable intake, and 1 author could not be reached, resulting in the exclusion of an additional 2 studies.17,34 When studies reported data for more than 1 pre- or postmeasurement time point, data closest to the beginning (but before delivery of intervention materials) and the end of the intervention were selected. Interventions differed according to 2 key theoretical and design-based criteria: the type of intervention (taste exposure vs nutrition education) and the principal summary measures targeted, such as type of outcome (fruit intake vs vegetable intake vs combined FV intake). To reduce exclusion criteria, conserve data points, and minimize the estimation of variance, outcome data were pooled separately for taste exposure and nutrition education interventions instead of being pooled into a single meta-regression on the basis of effect sizes. For each study outcome (fruit intake or vegetable intake), the standardized mean difference (Hedges’ g) was calculated as the mean between-group difference in pre–post intervention change scores divided by the standard deviation of the change scores pooled over the groups. Comprehensive meta-analysis software (BioStat, version 3) was used to calculate effect sizes and meta-analyze these effect sizes. In the results, a random-effects model was reported because of heterogeneity in some outcomes, but no outcome differed when a fixed-effects model was used. Heterogeneity of effect sizes was examined using the I2 statistic.48 Data extraction was checked by having a second author (V.Q.) re-extract 40% of the data. Wherever possible, means and standard deviations of pre–post change scores were extracted for taste exposure and nutrition education interventions. When means and standard deviations for pre–post interventions were available only separately, the scores for standard deviation of change, assuming a pre–post correlation of 0.6, were estimated. For interventions with more than 1 intervention condition of the same type of treatment, Hedges’ g was calculated between each intervention arm and control arm and then these were combined into 1 summary effect, assuming a between-subgroup correlation of 0.5.49 Sensitivity analyses were conducted for both imputed correlations for a range of plausible values (for pre–post correlation, 0.5, 0.6, and 0.7; and for subgroup variation, 0.4, 0.5, and 0.6) to ensure that no assumption altered the significance of a meta-analysis. A separate sensitivity analysis was conducted to examine whether removing studies identified as having potential bias (n = 2) (Table 3) altered the results of the meta-analysis. Studies were grouped according to whether the intervention primarily used a taste exposure or a nutrition education intervention (intervention type), and whether interventions measured fruit intake or vegetable intake (outcome type). As all studies with fruit-only outcomes were education interventions, this method produced 3 meta-analyses: the effect of nutrition education intervention vs no intervention on children’s fruit intake, the effect of nutrition education intervention vs no intervention on children’s vegetable intake, and the effect of taste exposure intervention vs no intervention on children’s vegetable intake. Consequently, only 1 effect size per study was included in any 1 meta-analysis. RESULTS A total of 622 studies were identified, and 9 RCTs were included in the meta-analysis (Figure 1). Figure 1 View largeDownload slide Flow diagram of the literature search process. A total of 1209 studies were identified across 2 searches. Eighteen studies were included in the systematic review and 12 in the meta-analysis. Figure 1 View largeDownload slide Flow diagram of the literature search process. A total of 1209 studies were identified across 2 searches. Eighteen studies were included in the systematic review and 12 in the meta-analysis. Study selection After the initial screening, 2 searches identified 1209 studies, 40 of which were considered potentially eligible and were retrieved for full-text review (35 from the initial search and 5 from the second search). A further 20 studies were excluded because the intervention did not target FV intake in children or was not an intervention study (n = 9), because the intervention was not home based (eg, it was community- or school-based) (n = 9), or because the article did not report original research (eg, protocol or dissertation) (n = 2). Although this resulted in a total of 20 eligible studies, 2 studies reported the long-term follow-up data of a previously identified and eligible intervention.21,22 These studies were presented with the original intervention rather than as a separate study (Table 2) (n = 2). The systematic review therefore included 18 intervention studies reporting 3759 children with a mean age of 5.3 years (range, 2–11 y; SD = 2.7). Table 1 summarizes the characteristics of the eligible studies. Two intervention studies were assessed as having potential outcome bias.19,33 To address this issue, a sensitivity analysis was conducted, which resulted in the exclusion of these studies from the meta-analysis. Table 1 PICOS criteria for inclusion and exclusion of studies Parameter  Inclusion criteria  Exclusion criteria  Population  Children aged 2–12 y and their parents  Children aged < 2 y or > 12 y and their parents  Intervention  Fruit and/or vegetable intake in children before and after a parent-targeted, home-based intervention  Interventions delivered in a school, community, or research setting; interventions without a parent-only component  Comparison interventions  Intervention vs no intervention (inactive controls); intervention vs active controls    Outcomes  Fruit and/or vegetable intake in children (including daily intake of a target vegetable in grams or daily intake of servings of fruit and/or vegetables via parent self-report)  Parent fruit and/or vegetable intake  Study types  Randomized controlled trial, nonrandomized controlled trial, and pre- and postintervention studies  Conference abstracts, systematic reviews, meta-analyses, book chapters, dissertations, case–control studies, cross-sectional studies  Parameter  Inclusion criteria  Exclusion criteria  Population  Children aged 2–12 y and their parents  Children aged < 2 y or > 12 y and their parents  Intervention  Fruit and/or vegetable intake in children before and after a parent-targeted, home-based intervention  Interventions delivered in a school, community, or research setting; interventions without a parent-only component  Comparison interventions  Intervention vs no intervention (inactive controls); intervention vs active controls    Outcomes  Fruit and/or vegetable intake in children (including daily intake of a target vegetable in grams or daily intake of servings of fruit and/or vegetables via parent self-report)  Parent fruit and/or vegetable intake  Study types  Randomized controlled trial, nonrandomized controlled trial, and pre- and postintervention studies  Conference abstracts, systematic reviews, meta-analyses, book chapters, dissertations, case–control studies, cross-sectional studies  Table 2 Characteristics of parent-targeted home-based interventions to increase fruit and vegetable intake in children Reference; country  Sample size; age range of children  Inclusion criteria  Characteristics of intervention (mode of intervention delivery; duration of intervention; frequency and length of intervention sessions; length of data collection from baseline to final point of data collection; type/training of interventionist)  Outcome measures (reported outcome measures in relation to child fruit and or vegetable intake; measure of acceptability and/or feasibility identified; potential confounders balanced or controlled for in RCT)  Randomized control trials  Corsini et al. (2013)30; Australia  N = 185; 4–6 y  Child’s age (4–6.99 y) Parent commitment to undertake a short activity daily for 2 wk Parent willingness to have 4 fieldworker home visits Parent ability to communicate in English  Mail-based (intervention materials), home-based (parent-led exposure), and home visits (parent training); 14 d of taste exposure only (EO) vs exposure and sticker reward (E + R) vs control (no intervention) and 4 home visits; length of intervention sessions unclear; data were collected 4 wk and 3 mo from baseline; trained fieldworkers  Intake of the target vegetable (in grams), usual vegetable intake measured using the Children’s Dietary Questionnaire, vegetable intake frequency via parent report (0–4 times), parents used a checklist of 23 vegetables to indicate how many vegetables children consumed in the past week and children’s liking of the target vegetable using a 3-point visual facial scale; no measure(s) of feasibility and/or acceptability reported; child age, sex, baseline FV intake in children  Cravener et al. (2015)31; USA  N = 24; 3–5 y  Child without pre-existing medical conditions (including food allergies) Child intake of < 2 servings of vegetables per day Child at risk for obesity on basis of family history, defined as having at least 1 parent with a BMI ≥ 25 kg/m2  Home visits and printed materials (food packages with parent-instructions and child-targeted education sessions); 4 wk; length of intervention sessions unclear; data were collected 4 wk from baseline; training of interventionists unclear  Vegetable intake as the difference between pre- and post-weights (in grams) of the foods provided, child-rated liking of 6 vegetables; qualitative postintervention feedback; child sex, age, SES, vegetable intake at baseline, BMI z-score, SES (paternal and maternal education level), maternal BMI, paternal BMI, and ethnicity  Fildes et al. (2014)33; England and Wales  N = 98; 3–4 y  Child’s age (3–4 yr) Child enrolled in GEMINI cohort  Mail-based and web-based; 14 d of taste exposure followed by a sticker reward vs control (no intervention); length of intervention sessions unclear; data were collected 14 d from baseline; N/A (instructions delivered via mail and online)  Number of pieces eaten measured the child’s intake of the target vegetable; children’s liking of the target variable measured using parent report on 9-point scale (“dislikes a lot” to “likes a lot”); quantitative postintervention evaluation questionnaire; child sex, age, SES (maternal education), and maternal BMI  Horton et al. (2013)23; Mexico  N = 361; 7–13 y  Child’s age (7–13 y) Child lives at home Child not on a medically prescribed diet Mother’s age (≥ 18 y) Mother married or living with partner Mother Spanish speaking Family living in Imperial County with no plans to move  Home visits (family intervention sessions) and telephone-based (support calls); 14 wk; home visits delivered weekly for 8 wk followed by 6 wk of weekly alternating home visits and telephone support calls (length of sessions unclear), ie, total intervention length was 16.5 h or 990 min; data were collected 14 wk from baseline; trained community members  Daily FV intake using 2 questions from the National Cancer Institute’s Food Attitudes and Behavior survey and child self-reports of FV variety; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, percentage of parents married, SES (maternal education level, median household size, percentage on food assistance, percentage that own their home), and maternal race  Knowlden et al. (2015)34; USAa Knowlden & Sharma (2016)21; USA (12-mo FU)a  N = 57; 4–6 y N = 44; 4–6 y  Child’s age (4–6 y) Parent English speaking Family internet and telephone access Mother not pregnant Child without disability Child without a medical condition associated with weight gain or prescribed weight management medication Child not enrolled in a weight-management program  Web-based; 4 wk; sessions of 20–30 min/wk; 5 educational sessions plus a booster; 1–15-min audio-visual presentation, interactive worksheet, and a discussion board post; data were collected 8 wk and 1 y from baseline; N/A (intervention delivered via a website)  Child’s FV consumption (dietary recall, measured in cups), and fruit availability; process evaluation data collected via telephone counseling and postintervention evaluation surveys (intervention fidelity, dose delivered, dose received, reach, recruitment, and potential cross-contamination between the groups); child sex, age, FV intake at baseline, race, SES (maternal marital status, maternal employment status); and maternal race  McGowan et al. (2013)35; UK  N = 126; 2–6 y  Child’s age (2–6 y) Child without known medical or psychological condition affecting diet Parent English speaking  Home visits and printed materials; 8-wk; sessions of 1 h/wk; researcher worked through an intervention booklet with the parent; data were collected 8 wk from baseline; researchers received training prior to the intervention  Child’s daily FV intake via parent self-report (“How many servings of fruit [vegetables] does your child typically eat?”; 7-point scale from “less than 1 per day” to “5 per day”); postintervention interview covering intervention acceptability; child sex, age, FV intake at baseline, ethnicity, parent age, and SES (parent education and parent living status)  Remington et al. (2012)36; UK  N = 173; 3–4 y  Child’s age (3–4 y) Child attending a selected nursery school  Home visits; 12 d of daily taste exposure to a target vegetable followed by praise vs a sticker for tasting vs control group; length of intervention sessions unclear; data were collected 4 wk and 12 wk from baseline; trained researchers  Child’s liking of the target vegetable via parent report using a faces scale and intake of the target vegetable (in grams) using a digital scale; postintervention qualitative feedback; child age, sex, vegetable intake, or vegetable liking at baseline, parent age, parent ethnicity, and SES (parent home ownership, parent education)  Schwinn et al. (2014)37; USA  N = 67; girls, 10–12 y  Girls Child’s age (10–12 y) Girls and mothers living in publicly subsidized housing  Web-based (website); 3 wk; sessions of 25 min/wk; data were collected 3 wk and 5 mo from baseline; N/A (intervention delivered via an online platform)  Child’s score on the Youth and Adolescent FFQ via parent-self-report; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, ethnicity, maternal age, maternal FV intake at baseline, and SES (maternal employment, whether child qualifies for reduced-price or free school lunch)  Tabak et al. (2012)38; USA  N = 43; 2–5 y  Child’s age (2–5 y) Family living in current residence for next 6 mo  Mail- and telephone-based; 4 mo: 4 tailored newsletters and 2 motivational phone calls delivered every 4 wk; mean length of telephone sessions was 34 min; data were collected 4 mo from baseline; registered dietitian trained in motivational interviewing techniques  Child’s score on the Block Kids FFQ via parent self-report; quantitative postintervention evaluation questionnaire, postintervention qualitative feedback and process evaluation data (telephone session duration in minutes); child age, sex, parent age, parent sex, parent BMI, ethnicity, and SES (parent income)  Thompson et al. (2015)39; USA  N = 387; 9–11 y  Child in 4th or 5th grade (age 9–11 y Family English speaking Computer and high-speed internet access Parent willing to participate in telephone sessions  Online (video game) and web-based (website and electronic newsletters); 10 sessions delivered over 3 mo (length of sessions unclear) across 4 groups—action, coping, action + coping, and control; data were collected 3 mo and 6 mo from baseline; trained staff conducted dietary recall  Child’s FV intake via 3 unannounced 24-h dietary recalls (2 d during week, 1 d on weekend) conducted over the telephone by trained staff and 3 d of FV intake at each data collection period were averaged; process evaluation data (participation rates); parent satisfaction scale; child sex, FV intake at baseline, ethnicity, parent age, parent ethnicity, and SES (parent education level)  Tomayako et al. (2016)40; USA  N = 150; 2–5 y  Family of American Indian background; child’s age (2–5 y) Child lived with at least 1 primary caregiver (eg, mother, father, grandmother, aunt) in a home setting Child free of any major physical or behavioral disorders  Family-based randomized trial of a healthy lifestyle toolkit delivered via 2 formats: in-home mentoring via 12 monthly home visits or by mail. Each lesson addressed 1 of 4 target areas: (1) eat more FV, (2) consume less soda and added sugar, (3) become more active, and (4) watch less TV; 12-mo intervention period; home visits of 60 min duration; data were collected 12 mo from baseline; interventionists were tribal members or individuals who had longstanding employment within the community and were trained to administer the intervention  Daily servings of FV, sugar-sweetened drinks, and candy/junk food using Nutrition Data System for Research software 2010; focus group testing covering intervention acceptability; child age, sex, FV intake at baseline, BMI percentile, BMI z-score, ethnicity, caregiver age, caregiver sex, caregiver ethnicity, adult BMI, caregiver FV intake, and SES (caregiver education level)  Wardle et al. (2003)41; Australia  N = 156; 2–6 y  Child’s age (2–6 y)  Home visits; 2 wk of daily taste exposure to target vegetable vs nutrition education vs no intervention vs a postintervention taste test; session length unclear; data were collected 2 wk from baseline; training of interventionists unclear  Child’s liking of the 6 test vegetables measured using a 3-point faces scale, child’s consumption of a target vegetable measured (in grams) using a digital scale; semistructured postintervention interview covering intervention acceptability; child sex, age, and vegetable intake at baseline  Wyse et al. (2012)17; Australiaa Wolfenden et al. (2014)22; Australia (12- and 18-mo FU)a  N = 394; 3–5 y N = 164; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk; parent must have some responsibility for providing meals and snacks to child; parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2–6-mo, 12 mo, and 18 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention acceptability questionnaire; child age, sex, SES (decile of disadvantage classification associated with child’s preschool postal code), and children’s FV intake at baseline Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (number of intervention calls completed, delivery of key topics, delivery of the intervention as per protocol); child age, sex, ethnicity, FV intake at baseline, parent age, parent sex, SES (household income, university education), and parent FV intake)  Haire-Joshu et al. (2008)16; USA  N = 1306; 2–5 y  Child’s age (2–5 y) Parent’s age (20–59 y) Child and parent living in rural, southeast Missouri  Mail-based (1 tailored newsletter); 4 home visits of 60 min duration (frequency unclear) and 4 sing-a-long story books and audio cassettes (delivered at each home visit); mean length of data were collected 7 mo from baseline; parent educators received 4 h of PAT training on nutrition and material content  Intake frequency measured using the Saint Louis University for Kids FFQ, child-feeding practices, parent modeling of FV intake, nutrition knowledge, FV availability in the home; quantitative postintervention evaluation questionnaire; child sex, age, FV intake at baseline, parent age, parent sex, and SES (parent education level)  Nonrandomized control trials  Holley et al. (2015)24; UK  N = 115; 2–4 y  Child’s age (2–4 y)  Home-based (parent-led exposure) and community-based (assessments held at preschool); 14-d intervention across 5 groups: repeated exposure (RE), modeling plus repeated exposure (M + RE), rewards plus repeated exposure (R + RE), modeling, rewards, and RE (condition 4) vs a no-treatment control group; session length unclear; data were collected 2 wk from baseline; unclear if interventionists received training  Intake (in grams) and liking (measured using a 3-point smiley-face scale) of the target vegetable; no measure(s) of feasibility and/or acceptability reported; child age, sex, vegetable intake at baseline, BMI z-score, parent age, and parent sex  Pre–post studies  Dulin Keita et al. (2014)32; USA  N = 39 (data completed at baseline and at FU; 2–5 y  Child’s age (3–5 y) Child’s age/sex-specific BMI is ≥ 50th percentile Parent’s age (≥ 18 y) Parent lives with child at least 75% of the time Parent can speak and read English Parent is knowledgeable about child’s diet and physical activity  Mail- and telephone-based; 4 mo; 4 tailored mailouts, 3 motivational telephone calls, activity video (session length unclear); data were collected 4 mo from baseline; lay counselors received 12 h of motivational interviewing training from MINT-qualified trainer  Child’s score on the National Cancer Institute’s FV all-day screener tool,; process evaluation data (number of counselor reported calls completed), quantitative postintervention evaluation questionnaire; N/A  Spurrier et al. (2016)19; Australia  N = 24 (22 families); 4–12 y  Child’s age (4–12 y) Child’s age/sex-specific BMI indicative of overweight or obesity according to IOTF definitions Child living in metropolitan Adelaide, South Australia Child not diagnosed with a medical condition affecting weight or growth or not enrolled in a weight-management program  Home-based (parent-led) education program; 3 home visits and 2 FU telephone calls were offered to each family (session length unclear); data were collected approximately 6 mo from baseline (21–44 wk); researchers had backgrounds in nutrition, occupational therapy, or human movement and received 3 h of training and education prior to intervention  Score on the FV subscale of the Children’s Dietary Questionnaire (no measure[s] of feasibility and/or acceptability reported); N/A  Wyse et al. (2011)42; Australia  N = 34; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Parent must have some responsibility for providing meals and snacks to child Parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2 mo and 6 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Score on the FV Subscale of the Children’s Dietary Questionnaire and Household Food Expenditure Survey, noncore subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention evaluation questionnaire; N/A  Reference; country  Sample size; age range of children  Inclusion criteria  Characteristics of intervention (mode of intervention delivery; duration of intervention; frequency and length of intervention sessions; length of data collection from baseline to final point of data collection; type/training of interventionist)  Outcome measures (reported outcome measures in relation to child fruit and or vegetable intake; measure of acceptability and/or feasibility identified; potential confounders balanced or controlled for in RCT)  Randomized control trials  Corsini et al. (2013)30; Australia  N = 185; 4–6 y  Child’s age (4–6.99 y) Parent commitment to undertake a short activity daily for 2 wk Parent willingness to have 4 fieldworker home visits Parent ability to communicate in English  Mail-based (intervention materials), home-based (parent-led exposure), and home visits (parent training); 14 d of taste exposure only (EO) vs exposure and sticker reward (E + R) vs control (no intervention) and 4 home visits; length of intervention sessions unclear; data were collected 4 wk and 3 mo from baseline; trained fieldworkers  Intake of the target vegetable (in grams), usual vegetable intake measured using the Children’s Dietary Questionnaire, vegetable intake frequency via parent report (0–4 times), parents used a checklist of 23 vegetables to indicate how many vegetables children consumed in the past week and children’s liking of the target vegetable using a 3-point visual facial scale; no measure(s) of feasibility and/or acceptability reported; child age, sex, baseline FV intake in children  Cravener et al. (2015)31; USA  N = 24; 3–5 y  Child without pre-existing medical conditions (including food allergies) Child intake of < 2 servings of vegetables per day Child at risk for obesity on basis of family history, defined as having at least 1 parent with a BMI ≥ 25 kg/m2  Home visits and printed materials (food packages with parent-instructions and child-targeted education sessions); 4 wk; length of intervention sessions unclear; data were collected 4 wk from baseline; training of interventionists unclear  Vegetable intake as the difference between pre- and post-weights (in grams) of the foods provided, child-rated liking of 6 vegetables; qualitative postintervention feedback; child sex, age, SES, vegetable intake at baseline, BMI z-score, SES (paternal and maternal education level), maternal BMI, paternal BMI, and ethnicity  Fildes et al. (2014)33; England and Wales  N = 98; 3–4 y  Child’s age (3–4 yr) Child enrolled in GEMINI cohort  Mail-based and web-based; 14 d of taste exposure followed by a sticker reward vs control (no intervention); length of intervention sessions unclear; data were collected 14 d from baseline; N/A (instructions delivered via mail and online)  Number of pieces eaten measured the child’s intake of the target vegetable; children’s liking of the target variable measured using parent report on 9-point scale (“dislikes a lot” to “likes a lot”); quantitative postintervention evaluation questionnaire; child sex, age, SES (maternal education), and maternal BMI  Horton et al. (2013)23; Mexico  N = 361; 7–13 y  Child’s age (7–13 y) Child lives at home Child not on a medically prescribed diet Mother’s age (≥ 18 y) Mother married or living with partner Mother Spanish speaking Family living in Imperial County with no plans to move  Home visits (family intervention sessions) and telephone-based (support calls); 14 wk; home visits delivered weekly for 8 wk followed by 6 wk of weekly alternating home visits and telephone support calls (length of sessions unclear), ie, total intervention length was 16.5 h or 990 min; data were collected 14 wk from baseline; trained community members  Daily FV intake using 2 questions from the National Cancer Institute’s Food Attitudes and Behavior survey and child self-reports of FV variety; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, percentage of parents married, SES (maternal education level, median household size, percentage on food assistance, percentage that own their home), and maternal race  Knowlden et al. (2015)34; USAa Knowlden & Sharma (2016)21; USA (12-mo FU)a  N = 57; 4–6 y N = 44; 4–6 y  Child’s age (4–6 y) Parent English speaking Family internet and telephone access Mother not pregnant Child without disability Child without a medical condition associated with weight gain or prescribed weight management medication Child not enrolled in a weight-management program  Web-based; 4 wk; sessions of 20–30 min/wk; 5 educational sessions plus a booster; 1–15-min audio-visual presentation, interactive worksheet, and a discussion board post; data were collected 8 wk and 1 y from baseline; N/A (intervention delivered via a website)  Child’s FV consumption (dietary recall, measured in cups), and fruit availability; process evaluation data collected via telephone counseling and postintervention evaluation surveys (intervention fidelity, dose delivered, dose received, reach, recruitment, and potential cross-contamination between the groups); child sex, age, FV intake at baseline, race, SES (maternal marital status, maternal employment status); and maternal race  McGowan et al. (2013)35; UK  N = 126; 2–6 y  Child’s age (2–6 y) Child without known medical or psychological condition affecting diet Parent English speaking  Home visits and printed materials; 8-wk; sessions of 1 h/wk; researcher worked through an intervention booklet with the parent; data were collected 8 wk from baseline; researchers received training prior to the intervention  Child’s daily FV intake via parent self-report (“How many servings of fruit [vegetables] does your child typically eat?”; 7-point scale from “less than 1 per day” to “5 per day”); postintervention interview covering intervention acceptability; child sex, age, FV intake at baseline, ethnicity, parent age, and SES (parent education and parent living status)  Remington et al. (2012)36; UK  N = 173; 3–4 y  Child’s age (3–4 y) Child attending a selected nursery school  Home visits; 12 d of daily taste exposure to a target vegetable followed by praise vs a sticker for tasting vs control group; length of intervention sessions unclear; data were collected 4 wk and 12 wk from baseline; trained researchers  Child’s liking of the target vegetable via parent report using a faces scale and intake of the target vegetable (in grams) using a digital scale; postintervention qualitative feedback; child age, sex, vegetable intake, or vegetable liking at baseline, parent age, parent ethnicity, and SES (parent home ownership, parent education)  Schwinn et al. (2014)37; USA  N = 67; girls, 10–12 y  Girls Child’s age (10–12 y) Girls and mothers living in publicly subsidized housing  Web-based (website); 3 wk; sessions of 25 min/wk; data were collected 3 wk and 5 mo from baseline; N/A (intervention delivered via an online platform)  Child’s score on the Youth and Adolescent FFQ via parent-self-report; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, ethnicity, maternal age, maternal FV intake at baseline, and SES (maternal employment, whether child qualifies for reduced-price or free school lunch)  Tabak et al. (2012)38; USA  N = 43; 2–5 y  Child’s age (2–5 y) Family living in current residence for next 6 mo  Mail- and telephone-based; 4 mo: 4 tailored newsletters and 2 motivational phone calls delivered every 4 wk; mean length of telephone sessions was 34 min; data were collected 4 mo from baseline; registered dietitian trained in motivational interviewing techniques  Child’s score on the Block Kids FFQ via parent self-report; quantitative postintervention evaluation questionnaire, postintervention qualitative feedback and process evaluation data (telephone session duration in minutes); child age, sex, parent age, parent sex, parent BMI, ethnicity, and SES (parent income)  Thompson et al. (2015)39; USA  N = 387; 9–11 y  Child in 4th or 5th grade (age 9–11 y Family English speaking Computer and high-speed internet access Parent willing to participate in telephone sessions  Online (video game) and web-based (website and electronic newsletters); 10 sessions delivered over 3 mo (length of sessions unclear) across 4 groups—action, coping, action + coping, and control; data were collected 3 mo and 6 mo from baseline; trained staff conducted dietary recall  Child’s FV intake via 3 unannounced 24-h dietary recalls (2 d during week, 1 d on weekend) conducted over the telephone by trained staff and 3 d of FV intake at each data collection period were averaged; process evaluation data (participation rates); parent satisfaction scale; child sex, FV intake at baseline, ethnicity, parent age, parent ethnicity, and SES (parent education level)  Tomayako et al. (2016)40; USA  N = 150; 2–5 y  Family of American Indian background; child’s age (2–5 y) Child lived with at least 1 primary caregiver (eg, mother, father, grandmother, aunt) in a home setting Child free of any major physical or behavioral disorders  Family-based randomized trial of a healthy lifestyle toolkit delivered via 2 formats: in-home mentoring via 12 monthly home visits or by mail. Each lesson addressed 1 of 4 target areas: (1) eat more FV, (2) consume less soda and added sugar, (3) become more active, and (4) watch less TV; 12-mo intervention period; home visits of 60 min duration; data were collected 12 mo from baseline; interventionists were tribal members or individuals who had longstanding employment within the community and were trained to administer the intervention  Daily servings of FV, sugar-sweetened drinks, and candy/junk food using Nutrition Data System for Research software 2010; focus group testing covering intervention acceptability; child age, sex, FV intake at baseline, BMI percentile, BMI z-score, ethnicity, caregiver age, caregiver sex, caregiver ethnicity, adult BMI, caregiver FV intake, and SES (caregiver education level)  Wardle et al. (2003)41; Australia  N = 156; 2–6 y  Child’s age (2–6 y)  Home visits; 2 wk of daily taste exposure to target vegetable vs nutrition education vs no intervention vs a postintervention taste test; session length unclear; data were collected 2 wk from baseline; training of interventionists unclear  Child’s liking of the 6 test vegetables measured using a 3-point faces scale, child’s consumption of a target vegetable measured (in grams) using a digital scale; semistructured postintervention interview covering intervention acceptability; child sex, age, and vegetable intake at baseline  Wyse et al. (2012)17; Australiaa Wolfenden et al. (2014)22; Australia (12- and 18-mo FU)a  N = 394; 3–5 y N = 164; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk; parent must have some responsibility for providing meals and snacks to child; parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2–6-mo, 12 mo, and 18 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention acceptability questionnaire; child age, sex, SES (decile of disadvantage classification associated with child’s preschool postal code), and children’s FV intake at baseline Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (number of intervention calls completed, delivery of key topics, delivery of the intervention as per protocol); child age, sex, ethnicity, FV intake at baseline, parent age, parent sex, SES (household income, university education), and parent FV intake)  Haire-Joshu et al. (2008)16; USA  N = 1306; 2–5 y  Child’s age (2–5 y) Parent’s age (20–59 y) Child and parent living in rural, southeast Missouri  Mail-based (1 tailored newsletter); 4 home visits of 60 min duration (frequency unclear) and 4 sing-a-long story books and audio cassettes (delivered at each home visit); mean length of data were collected 7 mo from baseline; parent educators received 4 h of PAT training on nutrition and material content  Intake frequency measured using the Saint Louis University for Kids FFQ, child-feeding practices, parent modeling of FV intake, nutrition knowledge, FV availability in the home; quantitative postintervention evaluation questionnaire; child sex, age, FV intake at baseline, parent age, parent sex, and SES (parent education level)  Nonrandomized control trials  Holley et al. (2015)24; UK  N = 115; 2–4 y  Child’s age (2–4 y)  Home-based (parent-led exposure) and community-based (assessments held at preschool); 14-d intervention across 5 groups: repeated exposure (RE), modeling plus repeated exposure (M + RE), rewards plus repeated exposure (R + RE), modeling, rewards, and RE (condition 4) vs a no-treatment control group; session length unclear; data were collected 2 wk from baseline; unclear if interventionists received training  Intake (in grams) and liking (measured using a 3-point smiley-face scale) of the target vegetable; no measure(s) of feasibility and/or acceptability reported; child age, sex, vegetable intake at baseline, BMI z-score, parent age, and parent sex  Pre–post studies  Dulin Keita et al. (2014)32; USA  N = 39 (data completed at baseline and at FU; 2–5 y  Child’s age (3–5 y) Child’s age/sex-specific BMI is ≥ 50th percentile Parent’s age (≥ 18 y) Parent lives with child at least 75% of the time Parent can speak and read English Parent is knowledgeable about child’s diet and physical activity  Mail- and telephone-based; 4 mo; 4 tailored mailouts, 3 motivational telephone calls, activity video (session length unclear); data were collected 4 mo from baseline; lay counselors received 12 h of motivational interviewing training from MINT-qualified trainer  Child’s score on the National Cancer Institute’s FV all-day screener tool,; process evaluation data (number of counselor reported calls completed), quantitative postintervention evaluation questionnaire; N/A  Spurrier et al. (2016)19; Australia  N = 24 (22 families); 4–12 y  Child’s age (4–12 y) Child’s age/sex-specific BMI indicative of overweight or obesity according to IOTF definitions Child living in metropolitan Adelaide, South Australia Child not diagnosed with a medical condition affecting weight or growth or not enrolled in a weight-management program  Home-based (parent-led) education program; 3 home visits and 2 FU telephone calls were offered to each family (session length unclear); data were collected approximately 6 mo from baseline (21–44 wk); researchers had backgrounds in nutrition, occupational therapy, or human movement and received 3 h of training and education prior to intervention  Score on the FV subscale of the Children’s Dietary Questionnaire (no measure[s] of feasibility and/or acceptability reported); N/A  Wyse et al. (2011)42; Australia  N = 34; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Parent must have some responsibility for providing meals and snacks to child Parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2 mo and 6 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Score on the FV Subscale of the Children’s Dietary Questionnaire and Household Food Expenditure Survey, noncore subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention evaluation questionnaire; N/A  Abbreviations: BMI, body mass index; CATI, computer-assisted telephone interviewing; FFQ, food frequency questionnaire; FV, fruit(s) and vegetable(s); FU, follow-up (time from baseline to final point of data collection); IOTF, International Obesity Task Force; MINT, Motivational Interviewing Network of Trainers; N/A, not available; PAT, Parents as Teachers; RCT, randomized control trial; SES, socioeconomic status. a Study had a separate follow-up publication. Follow-up data was treated as an extension of a single paper in this systematic review. Study characteristics Table 2 presents the study characteristics of both exposure and education interventions, respectively. Fourteen interventions (78%) were RCTs, 1 (6%) was a nonrandomized control trial (in which groups were systematically assigned by the primary investigator), and 3 (17%) were pre–post studies. Interventions were most commonly conducted in the United States, followed by the United Kingdom, Australia, and Mexico. The mean sample size was 229 participants (range, 24–1306; SD = 305). The mean age of children targeted by interventions was 5.3 years (SD = 2.7). Parent age was reported by 70% of studies, with a mean of 36.3 years (SD = 4.7). The percentage of female caregivers was consistently higher across studies (≥ 85%) compared with the percentage of male caregivers (≤ 1%). Interventions included home visits (10 of 18; 56%), telephone sessions (6 of 18; 33%), written materials (12 of 18; 67%), and online delivery methods (eg, website, electronic newsletters) (5 of 18; 28%) (Table 4).16,17,19,21–24,30–42 Nutrition education programs were the most common type of intervention (12/18, 67%). These interventions typically focused on increasing knowledge about the nutritional guidelines (eg, the recommended daily servings of FVs) and behavioral strategies for achieving the guidelines via a healthy home food environment (eg, parent providing of FVs, role modeling of FV intake, and adequate FV availability) (Table 3). Six studies (33%) used a repeated taste exposure paradigm to examine whether exposure, with or without positive reinforcement (eg, sticker reward or verbal praise), increased children’s intake of a target vegetable.24,30,31,33,36,41 Although only 2 studies met all of the applicable quality criteria,32,42 the majority of studies met most of the quality requirements (Table 3), which included a clear presentation of information related to study aims, hypotheses, and participant characteristics (16/18, 89%), a clear description of the main study findings (15/18, 83%),43,44 and reporting of an acceptable attrition rate (≤ 30%) (15/18, 83%) (Table 3).44 All studies (100%, 18/18) used valid and reliable outcome measures. Studies were less likely to meet the quality requirements for reporting of sample power calculations, with calculations not reported in 12 of the 18 studies (67%). The descriptions of randomization methods, blinding of participants, and blinding of assessors were also unclear or unavailable in 60% (9/15), 93% (14/15), and 100% (15/15) of applicable studies, respectively. In the majority of RCTs, the distribution of important confounders appeared balanced after randomization, with 3 studies (20%) providing insufficient detail to adequately determine the risk of residual confounding (Table 3). Two studies met the criteria for potential outcome bias.19,33 To address this issue, a sensitivity analysis that excluded biased studies in the meta-analysis was conducted. This procedure did not alter the significance of any meta-analytic outcome, suggesting that the inclusion of these studies in the systematic review was appropriate. Inter-rater reliability There was high inter-rater reliability between the 2 authors for both searches (2 disagreements out of 1145 abstracts for the initial search [κ = 0.94] and 2 unresolved disagreements out of 69 abstracts for the updated search [κ = 0.74], both of which required consultation with a third author). Research question 1: Are parent-targeted home-based interventions aimed at increasing FV intake in children feasible and acceptable to parents? Thirteen studies (72%) reported process evaluation data and/or quantitative or qualitative data on intervention acceptability and participant satisfaction (Table 2). Of these, 100% of studies reported evidence of intervention feasibility, including the completion of sessions within the recommended time frame,17,34,38,42 the completion of all intervention sessions by over 70% of participants,17,32,42 or a low attrition rate (< 30%, Table 3). Several feasibility issues were identified for both types of interventions. One nutrition education intervention found that the rate of participants who received telephone calls (16%) was lower than the rate of those who received mailed intervention materials (42%).32 In an online nutrition education intervention, lower participant engagement was reported for parents and caregivers (28% of parents reported reading more than 60% of intervention session materials) than for children (91%).39 For exposure interventions, 1 home-visiting study found that 41% of participants did not complete the minimum number of exposure sessions.41 Another online and mail-based exposure intervention reported lower rates of participant engagement with an online instruction video (17%) than with a written instruction leaflet (94%).33 Two taste exposure studies reported quantitative data on the overall acceptability of interventions. Among these, 80% of participants agreed that the taste exposure procedure was helpful (eg, increased children’s willingness to try vegetables),33 85% felt that the advice given was useful,41 and 70% agreed that the intervention had a long-lasting effect on their child’s liking of the target vegetable.41 In addition, 85% of participants agreed that they would use the taste exposure procedure again,33 and 65% reported that they had already used the strategy for a different vegetable.41 Nutrition education interventions were also rated favorably, with 83% to 97% of participants agreeing that participation was helpful (eg, they could set reasonable limits for their children after participating in the program)16 or worthwhile,42 92% to 100% rating the intervention materials as easy to read32 or relevant,42 and 87% reporting that they were still using the intervention materials provided.32 For both taste exposure and nutrition education interventions, participants identified time (eg, the time needed to undertake the intervention) as the most common barrier to acceptability.31,38,41 Research question 2: Are parent-targeted home-based taste exposure and nutrition education interventions effective at increasing FV intake in children? Overall, positive outcomes were more frequently reported with taste exposure interventions than with nutrition education interventions, with all 6 taste exposure interventions demonstrating a significant increase in children’s vegetable intake (Table 3). In addition, 4 of the 6 taste exposure interventions reported an increase in child-reported liking of a target vegetable (Table 3). Of the 12 nutrition education interventions reviewed, 3 (25%) yielded no effect on children’s FV intake.19,23,38 However, 2 of the 3 nutrition education interventions that failed to find an intervention effect had a relatively small sample size (< 45),19,38 suggesting these interventions may not have had adequate power to detect a significant intervention effect. Research question 3: Do parent-targeted home-based taste exposure and nutrition education interventions result in significantly greater FV intake in children? A meta-analysis was performed to statistically evaluate the effect of taste exposure and nutrition education interventions compared with no intervention on FV intake in children. Nutrition education interventions (n = 4) had a significant but small positive effect on child fruit intake (Hedges’ g = 0.112; SE = 0.051; 95%CI, 0.012–0.212; P = 0.028), with nonsignificant heterogeneity (I2= 5.945; P = 0.363) (Figure 2A16,35,37,39). There was no significant effect of education interventions on child vegetable intake (n = 6) (Figure 2B16,23,35,37–39) (Hedges’ g = 0.125; SE = 0.082; 95%CI, −0.035 to 0.285; P = 0.125), but there was significant heterogeneity (I2= 55.84; P = 0.045). Figure 2 View largeDownload slide (A) Comparison of the effect of a parent-targeted home-based nutrition education intervention vs no intervention on (A) fruit intake (no. of daily servings) and (B) vegetable intake (no. of daily servings) in children. Analysis A was performed using 4 randomized controlled studies.16,35,37,39 Analysis B was performed using 6 randomized controlled studies.16,23,35,37–39 Figure 2 View largeDownload slide (A) Comparison of the effect of a parent-targeted home-based nutrition education intervention vs no intervention on (A) fruit intake (no. of daily servings) and (B) vegetable intake (no. of daily servings) in children. Analysis A was performed using 4 randomized controlled studies.16,35,37,39 Analysis B was performed using 6 randomized controlled studies.16,23,35,37–39 Figure 3 View largeDownload slide Comparison of the effect of a parent-targeted home-based taste exposure intervention vs no intervention on vegetable intake (in grams) in children. This analysis was performed using 6 randomized controlled studies.24,30,31,33,36,41 Figure 3 View largeDownload slide Comparison of the effect of a parent-targeted home-based taste exposure intervention vs no intervention on vegetable intake (in grams) in children. This analysis was performed using 6 randomized controlled studies.24,30,31,33,36,41 For taste exposure interventions (Figure 324,30,31,33,36,41) (n = 6), there was a moderately large and significant effect of intervention on child vegetable intake (Hedges’ g = 0.438; I2 =0.000; SE = 0.064; 95%CI, 0.312–0.564; P < 0.001). This result remained significant when the study identified as having potential outcome bias was removed from the analysis33 (Hedges’ g = 0.372; I2 =0.000; SE = 0.086; 95%CI, 0.204–0.540; P < 0.001). Sensitivity analyses revealed that using the stipulated range of plausible correlations for pre–post correlation or subgroup variation imputations did not alter the significance of any meta-analytic outcome. Research question 4: What characteristics are common among parent-targeted home-based interventions that achieve significant increases in FV intake in children? Overall, online-delivered interventions appeared to be most effective, with 100% of online-delivered interventions reporting a significant increase in FV intake in children compared with 80% of home-visiting interventions and 50% of telephone-delivered interventions. Interventions providing daily sessions had the highest percentage of reported significant intervention effects (100%) compared with interventions delivered weekly (83%) or monthly (50%). The number of intervention sessions did not appear to strongly influence the effectiveness of interventions. Studies that included more than 14 sessions had the highest percentage of reported significant intervention effects, but this statistic was based on 1 intervention (n = 1 of 1). The percentage of reported significant intervention effects was slightly lower among studies that provided 11 to 14 sessions (83%, n = 5 of 6), 6 to 10 sessions (75%, n = 3 of 4), and 1 to 5 sessions (86%, n = 6 of 7) (Table 4). Table 4 Summary of outcomes from parent-targeted home-based interventions to increase fruit and/or vegetable intake in children Characteristics of intervention  Study outcomes   Increased FV intakea  Increased FV liking  No effect on FV intake  Type of intervention   Taste exposure24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies   Nutrition education16,17,19,23,32,34,35,37–40,42  9 studies16,17,32,34,35,37,39,40,42  No studies  3 studies19,23,38  Recruitment strategy   Media or advertising16,23,30,31,34,35,37,39  7 studies16,30,31,34,35,37,39  No studies  1 study23   Written information, mailed out16,17,23,33,36,38,39,42  5 studies17,33,36,39,42  No studies  2 studies23,38   Researcher contact (face-to-face or telephone)16,17,19,24,32,35,41,42  7 studies16,17,24,32,35,41,42  No studies  1 study19   Online, eg, social media31,37–39  3 studies31,37,39  No studies  1 study38  Type of incentives   Financial16,30,32,34,35,37,38  6 studies16,30,32,34,35,37  No studies  1 study38   None17,19,23,24,31,33,36,39–42  9 studies17,24,31,33,36,39–42  4 studies24,33,36,41  2 studies19,23  Mode of delivery   Written information or resources16,17,23,24,30–33,35,38,41,42  10 studies16,17,24,30–33,35,41,42  3 studies24,33,41  2 studies23,38   Telephone calls17,19,23,32,38,42  3 studies17,32,42  No studies  3 studies19,23,38   Home visits by researchers16,19,23,30–32,35,36,40,41  8 studies16,30–32,35,36,40,41  2 studies36,41  2 studies19,23   Online16,33,34,37,39  5 studies16,33,34,37,39  1 study33  No studies  Frequency of sessions   Daily24,30,33,36,41  4 studies24,30,36,41  2 studies36,41  No studies   Weekly17,23,31,34,37,42  5 studies17,31,34,37,42  No studies  1 study23   Monthly19,32,38,40  2 studies32,40  No studies  2 studies19,38   Variable/unclear16,35,39  3 studies16,35,39  No studies  No studies  No. of sessions   1–5 sessions17,19,31,34,35,37,42  6 studies17,31,34,35,37,42  No studies  1 study19   6–10 sessions16,32,38,39  3 studies16,32,39  No studies  1study38   11–14 sessions23,24,30,33,36,41  5 studies24,30,33,36,41  4 studies24,33,36,41  1study 23   > 14 sessions40  1 study40  No studies  No studies  Components of intervention   Goal setting17,19,23,32,34,35,37,38,42  6 studies17,32,34,35,37,42  No studies  3 studies19,23,38   Meal planning17,23,34,35,37–39,42  6 studies17,34,35,37,39,42  No studies  2 studies23,38   Nutrition education16,17,19,23,24,31,34,35,37–42  10 studies16,17,31,34,35,37,39–42  1 study24  3 studies19,23,38   Written materials16,17,19,30–35,37–42  13studies16,17,30–35,37,39–42  2 studies33,41  2 studies19,38   Rewards24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies  Provision of tailored information/feedback   Yes16,17,19,32,42  4 studies16,17,32,42  No studies  1 study19   No23,24,30,31,33–41  11 studies21,24,30,31,33,35–37,39–41  1 study33  2 studies23,38  Characteristics of intervention  Study outcomes   Increased FV intakea  Increased FV liking  No effect on FV intake  Type of intervention   Taste exposure24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies   Nutrition education16,17,19,23,32,34,35,37–40,42  9 studies16,17,32,34,35,37,39,40,42  No studies  3 studies19,23,38  Recruitment strategy   Media or advertising16,23,30,31,34,35,37,39  7 studies16,30,31,34,35,37,39  No studies  1 study23   Written information, mailed out16,17,23,33,36,38,39,42  5 studies17,33,36,39,42  No studies  2 studies23,38   Researcher contact (face-to-face or telephone)16,17,19,24,32,35,41,42  7 studies16,17,24,32,35,41,42  No studies  1 study19   Online, eg, social media31,37–39  3 studies31,37,39  No studies  1 study38  Type of incentives   Financial16,30,32,34,35,37,38  6 studies16,30,32,34,35,37  No studies  1 study38   None17,19,23,24,31,33,36,39–42  9 studies17,24,31,33,36,39–42  4 studies24,33,36,41  2 studies19,23  Mode of delivery   Written information or resources16,17,23,24,30–33,35,38,41,42  10 studies16,17,24,30–33,35,41,42  3 studies24,33,41  2 studies23,38   Telephone calls17,19,23,32,38,42  3 studies17,32,42  No studies  3 studies19,23,38   Home visits by researchers16,19,23,30–32,35,36,40,41  8 studies16,30–32,35,36,40,41  2 studies36,41  2 studies19,23   Online16,33,34,37,39  5 studies16,33,34,37,39  1 study33  No studies  Frequency of sessions   Daily24,30,33,36,41  4 studies24,30,36,41  2 studies36,41  No studies   Weekly17,23,31,34,37,42  5 studies17,31,34,37,42  No studies  1 study23   Monthly19,32,38,40  2 studies32,40  No studies  2 studies19,38   Variable/unclear16,35,39  3 studies16,35,39  No studies  No studies  No. of sessions   1–5 sessions17,19,31,34,35,37,42  6 studies17,31,34,35,37,42  No studies  1 study19   6–10 sessions16,32,38,39  3 studies16,32,39  No studies  1study38   11–14 sessions23,24,30,33,36,41  5 studies24,30,33,36,41  4 studies24,33,36,41  1study 23   > 14 sessions40  1 study40  No studies  No studies  Components of intervention   Goal setting17,19,23,32,34,35,37,38,42  6 studies17,32,34,35,37,42  No studies  3 studies19,23,38   Meal planning17,23,34,35,37–39,42  6 studies17,34,35,37,39,42  No studies  2 studies23,38   Nutrition education16,17,19,23,24,31,34,35,37–42  10 studies16,17,31,34,35,37,39–42  1 study24  3 studies19,23,38   Written materials16,17,19,30–35,37–42  13studies16,17,30–35,37,39–42  2 studies33,41  2 studies19,38   Rewards24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies  Provision of tailored information/feedback   Yes16,17,19,32,42  4 studies16,17,32,42  No studies  1 study19   No23,24,30,31,33–41  11 studies21,24,30,31,33,35–37,39–41  1 study33  2 studies23,38  a A significant increase in FV intake was set as P < 0.05. Five of 6 taste exposure interventions (83%) included an exposure plus a reward condition. Of these, 80% found that children who received exposure along with a sticker reward ate significantly more of the target vegetable from baseline to the end of the intervention compared with children who did not receive a sticker reward.24,30,31,33,36 When the type of reward was examined, a tangible reward such as a sticker resulted in a greater increase in target vegetable intake compared with a social reward, such as parent verbal praise (P = 0.001).36 Five studies (29%) provided participants with personalized feedback.16,17,19,32,42 This included written materials with the name of the parent and their child,19 the selection of intervention topics to suit the target child behaviors,16,32 tailored feedback about FV intake specific to the parent and/or child,16,17,42 and tailored goals.17,42 Four of the 5 studies that used personalized feedback (80%) found a significant intervention affect (Table 3). Thirteen of the 14 studies without personalized feedback also reported similar outcomes (Table 4). Whether studies provided a financial incentive for participation did not appear to influence the outcome of interventions: 86% of studies using incentives reported a significant increase in FV intake in children compared with 82% of studies that did not use incentives. DISCUSSION The majority of published parent-targeted home-based interventions have reported an improvement in children’s FV intake. However, this review and meta-analysis demonstrates that interventions may not be equally effective at increasing the intake of both fruit and vegetables. Specifically, taste exposure interventions, but not nutrition education interventions, resulted in a significant increase in children’s vegetable intake. Despite having no effect on vegetable intake, nutrition education interventions did increase children’s fruit intake significantly. Online and home-visiting interventions more frequently reported significant increases in children’s FV intake compared with telephone-based interventions. Furthermore, not all interventions were equally acceptable to parents, with parents identifying time as a major limitation to completing taste exposure activities and telephone-based nutrition education sessions. The current meta-analysis found a moderate to large increase in children’s vegetable intake in parent-targeted, home-based taste exposure interventions. This is an important finding, given that fewer children meet the national guidelines for recommended daily servings of vegetables than for fruit6 and most children do not consume the recommended variety of vegetables.50 Although previous studies have identified taste exposure as an effective intervention strategy for increasing children’s FV intake,51 a previous meta-analysis found no effect of a parent-delivered taste exposure intervention on children’s vegetable intake when compared with no intervention.9 This discrepancy may be explained by the fact that the previous meta-analysis included only 2 interventions, while the current analysis included 6, meaning the former may have lacked sufficient power to detect an intervention effect.9 The effects of nutrition education interventions on FV intake in children were mixed. The meta-analysis of 2290 participants from interventions comparing nutrition education with no intervention revealed a significant intervention effect for fruit intake but not vegetable intake. It is unclear why taste exposure interventions, but not nutrition education interventions, resulted in a significant increase in vegetable intake. Recent population data indicates that children aged 2 to 18 years are more likely to meet the recommended daily intake for fruit (68%) than for vegetables (5%).50 This may suggest that vegetable intake is more difficult to increase than fruit intake. It is well known that children have a biological predisposition to reject bitter-tasting foods, such as vegetables, to protect themselves against possible toxins.52 Children may also become reluctant to taste unfamiliar vegetables as a result of food neophobia or the fear of eating new foods,53 which is common between the ages of 2 and 6.54 Promoting substantial changes in children’s vegetable intake may therefore necessitate more intensive strategies than parent-targeted nutrition education, particularly for bitter-tasting and unfamiliar vegetables. Overall, the pattern of results supports the findings obtained with other commonly used FV interventions, including school- and family-based interventions. The school environment is one of the most common settings for implementing FV interventions in children because of its capacity to target large numbers of children.11 A meta-analysis of 27 school-based interventions involving 26 361 children found that children’s intake of fruit increased by a useful amount (0.24 servings), whereas intake of vegetables did not (0.07 servings).10 The majority of these interventions, however, involved the delivery of nutrition education or the provision of free or subsidized FVs and did not include taste exposure. Given the findings of the current review, it is plausible that the absence of taste exposure may have contributed to the negligible effect of school-based interventions on children’s vegetable intake.10 Supporting this hypothesis are the findings of Skouteris et al.,55 who reported that children whose parents received taste exposure training as part of a family-based nutrition education intervention ate significantly more vegetables at the end of the intervention compared with controls. Taken together, these findings suggest that, irrespective of the context (eg, school, community, or home), parent involvement and taste exposure are likely to be important strategies to include in any intervention aimed at promoting meaningful increases in children’s FV intake. Feasibility and acceptability are important issues when considering the efficacy of an intervention. The literature review showed that parent-targeted interventions conducted in the home setting appeared feasible to deliver, with the majority of studies reporting an attrition rate of less than 30% (Table 3). However, participant burden is commonly identified as a key factor for nonparticipation in research studies.56 This was partly confirmed by the finding that parents identified time as an important limitation for both taste exposure and telephone-delivered nutrition education interventions, including the time to undertake taste exposure activities and to complete daily record forms,31,41 multiple (n = 3) assessment sessions,33 and telephone sessions.38 To enhance acceptability, future interventions may benefit from using alternative strategies to daily written record forms and structured taste exposure procedures. For example, studies have successfully trialed the use of mobile applications in the treatment of clinical anxiety to enhance participant self-monitoring and recording of treatment tasks.57 New evidence also suggests that modified versions of the taste exposure method may be effective at increasing children’s FV intake. A recent RCT of a nutrition education intervention involving 394 parents found that daily parent providing of FVs to children, regardless of the outcome, mediated the long-term positive effect (up to 12 months) of the intervention on children’s FV intake.12,22 In this approach, parents are broadly encouraged to increase the number of occasions when vegetables are offered to children each day. This is different from traditional taste exposure methods in which parents are instructed to offer a specific target vegetable on a set number of days, allowing for less flexibility. Given that time was identified as a common barrier to completing interventions, identifying effective but flexible parent-targeted intervention strategies is an important goal for future parent-targeted FV interventions. Contrary to both theory58 and earlier literature,59 evidence that longer interventions (11–14 sessions), compared with briefer interventions (1–5 sessions), were more likely to report significant increases in children’s FV intake at end of treatment was not strong. Although these findings suggest that both longer and shorter interventions may be equally effective, the majority of short-term intervention studies included in this review did not include long-term follow-up data.21,22 Where short-term interventions (eg, 4–6 intervention sessions) did include long-term follow-up data, increases in children’s FV intake were observed only up to 12 months of follow-up.21,22 Therefore, ongoing contact may still be important in maintaining increases in children’s FV intake over the long term.21,58 This is consistent with theories of health behavior change that recommend multiple points of contact to maintain behavior change over time.58 Moreover, a systematic review of behavioral therapy trials found that booster sessions, used in combination with standard treatment, significantly aided behavior change in over 50% of studies.59 This may be attributable to the reinforcement of skills learned during treatment, a reduction in participant concerns about treatment termination, and the ability to maintain participant accountability in implementing skills acquired during treatment.59 Although preliminary, the findings of this review have important implications for health professionals interested in delivering parent-targeted home-based interventions as well as for parents seeking effective strategies to increase children’s FV intake. Similarly to previous systematic reviews, the present findings suggest that taste exposure is more effective at increasing children’s FV intake when combined with a tangible reward, at least in the short term.9,60 However, the usefulness of rewarding children for eating FVs remains controversial and should be considered. Taste exposure interventions may be susceptible to the overjustification effect, which argues that repeatedly offering a reward for eating a target vegetable diminishes a child’s liking and acceptance of that vegetable over time.61,62 In this context, rewards may act as a signal to children that tasting a target or unfamiliar vegetable is unpleasant because it requires immediate compensation.61 However, not all rewards may negatively affect children’s acceptance of a target FV. For example, some research has shown that unexpected rewards may actually boost motivation.63 Future research is therefore needed to better ascertain whether using a short-term tangible reward (after approximately 14 taste occasions) increases or decreases children’s long-term (> 12 months) willingness to taste a target vegetable and to identify the most effective method for delivering a reward (eg, expected or unexpected).63 LIMITATIONS AND FUTURE DIRECTIONS Although the current meta-analysis enabled a rigorous examination of published interventions, the methodology of meta-analysis is vulnerable to bias. Specifically, interventions with nonsignificant findings may be less frequently published in the literature,64 resulting in a nonrandom sample of included studies. A quantitative risk assessment could not be performed in the current meta-analysis because of the small number of included studies. There was also some unexplained heterogeneity in effect size. Meta-regression is a technique that can use study-level variables to explain heterogeneity within a random-effects model. However, the lack of consistency in potential explanatory variables across studies in the current meta-analysis precluded a meta-regression from being conducted. For example, 1 study had both an exposure group and a nutrition education group,41 numerous studies only assessed vegetable and not fruit intake, and multiple studies had several intervention conditions. In addition, only 2 of the 18 interventions included in this review targeted non-white, non-English-speaking families.23,40 This is an important limitation, given that children’s dietary behaviors are influenced in part by socioeconomic status and culture.65 It is therefore unclear whether the intervention effects observed in this systematic review can be generalized across different cultures, ethnicities, and economic backgrounds. Few studies included a long-term follow-up assessment, making it difficult to ascertain whether interventions resulted in sustained increases in children’s FV intake. The risk of interventions reporting false positive or false negative results was identified as low, with most studies adequately balancing or controlling for children’s age, sex, and baseline FV intake (Table 3). However, parent FV intake, parent body mass index, and parent educational status were less frequently reported, despite being associated with children’s FV intake,12,45–47 and indices of socioeconomic status varied across studies (Table 2). Therefore future intervention studies should also control for these factors to ensure rigorous examination of the effect of parent-targeted home-based interventions on children’s FV intake. There is currently no gold standard for the assessment of children’s dietary intake, with methods varying considerably across interventions.66 In the studies included in the present review, assessment methods ranged from a dietary intake questionnaire to a repeated 24-hour dietary recall delivered by a trained health professional. Different dietary intake assessment methods are associated with different forms of bias, including the over- or underestimation of child dietary intake.66 Therefore, future parent-targeted home-based nutrition interventions need to include the same assessment methods to facilitate the comparison of FV intake data across studies.66 In all of the interventions evaluated here, data on children’s FV intake were collected using parent self-report. Although this method is demonstrated to be reliable and valid in children under the age of 8, the evidence is unclear in older children,66 with parents often over-reporting children's school and out-of-home food intake.67 Including children over the age of 10 in the assessment of children’s FV intake may help to increase the validity and reliability of outcome data from future dietary interventions.66 Given that parent-targeted, home-based interventions have the potential to significantly increase children’s FV intake, it is important that future research identify ways to increase the number of parents being offered support, for example, by reducing delivery costs.68 Unfortunately, the equivocal findings of this review regarding the effect of using financial incentives to increase participation in FV interventions in children provide little direction for health professionals interested in delivering parent-targeted, home-based interventions. Additional RCTs are therefore needed to examine the impact of parent-targeted home-based interventions, with and without financial incentives, on FV intake in children. In addition, none of the telephone-based intervention studies included in this review utilized mobile phone applications (apps) or text messaging. Given the increasing use of phone-based technologies in interventions designed to effect change in health behaviors,69–71 incorporation of mobile phone apps in future research is likely to contribute to the development of more affordable parent-targeted interventions to increase children’s FV intake on a mass scale. CONCLUSION This review demonstrates that taste exposure is an effective parent-targeted, home-based intervention strategy for promoting short-term increases (up to 12 months) in children’s vegetable intake, which remains below the national recommendations across much of the developed world. Interventions delivered in the home setting, particularly via online modalities or home visits, also have strong merit, resulting in a significant increase in children’s fruit intake (lasting up to 12 months). To best inform parents and health professionals, an evaluation of the long-term impact (> 12 months) of these interventions is needed to identify effective strategies for achieving sustained increases in children’s FV intake. Despite this limitation, the majority of the parent-targeted home-based FV interventions reviewed here appeared feasible to deliver, as evidenced by the low attrition rates. Taken together, these findings suggest that such interventions are likely to have substantial public health appeal, offering an effective, practical, and geographically far-reaching approach for promoting increases in children’s FV intake. This is particularly important for parents and health professionals in rural and remote areas, for whom access to effective health promotion interventions is often very limited. Acknowledgments Author contributions. L.M.T., C.E.W., and J.C. designed the research; L.M.T., M.S., and A.M.G. contributed to the systematic search; L.M.T., V.F.Q., and D.C. performed the statistical analysis; L.M.T. and A.M.G. were responsible for preparing the manuscript for publication; L.M.T. had primary responsibility for the final content; and C.E.W., J.C., and R.J.C. critically reviewed the intellectual content of the draft manuscript prior to submission. All authors read and approved the final manuscript. Funding/support This project was supported by the Kids Cancer Alliance, which is supported by a Cancer Institute NSW grant (no. 11/TRC/1–03) awarded to C.E.W., J.C., and R.J.C. The Behavioural Sciences Unit at the Kids Cancer Center, Sydney Children’s Hospital, Sydney, Australia, is proudly supported by the Kids with Cancer Foundation. This research was also funded by the Cancer Council NSW (program grant no. PG16–02) with the support of the Estate of the Late Harry McPaul. The funders had no role in the design of the study; in the collection, management, analysis, or interpretation of the data; in the preparation and revision of the manuscript; or in the publication decisions related to this manuscript. Authors received complete access to the data pertaining to this publication. Declaration of interest The authors have no relevant interests to declare. Supporting Information Appendix S1Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement Appendix S2Search strategies References 1 Vioque J, Weinbrenner T, Castelló A et al.   Intake of fruits and vegetables in relation to 10‐year weight gain among Spanish adults. Obesity . 2008; 16: 664– 670. Google Scholar CrossRef Search ADS PubMed  2 Kaikkonen JE, Mikkilä V, Raitakari OT. Role of childhood food patterns on adult cardiovascular disease risk. Curr Atheroscler Rep.  2014; 16: 1– 15. Google Scholar CrossRef Search ADS   3 Ruxton CHS, Derbyshire E. Diet adequacy in UK schoolchildren. Nutr Food Sci . 2011; 41: 20– 33. http://dx.doi.org/10.1108/00346651111102865 Google Scholar CrossRef Search ADS   4 Fox MK, Gearan E, Cannon J, et al.   Usual food intakes of 2- and 3-year old US children are not consistent with dietary guidelines. BMC Nutr.  2016; 2: 67. doi:10.1186/s40795-016-0106-2. Google Scholar CrossRef Search ADS   5 Wyse R, Campbell E, Nathan N, et al.   Associations between characteristics of the home food environment and fruit and vegetable intake in preschool children: a cross-sectional study. BMC Public Health . 2011; 11: 938. 10.1186/1471-2458-11-938 Google Scholar CrossRef Search ADS PubMed  6 Australian Bureau of Statistics. National Health Survey: first results, 2014-15. http://www.abs.gov.au/ausstats/abs@.nsf/mf/4364.0.55.001. Published August 12, 2015. Accessed January 5, 2017. 7 Nicklaus S, Boggio V, Chabanet C, et al.   A prospective study of food preferences in childhood. Food Qual Prefer . 2004; 15: 805– 818. http://dx.doi.org/10.1016/j.foodqual.2004.02.010 Google Scholar CrossRef Search ADS   8 Mikkilä V, Räsänen L, Raitakari OT, et al.   Consistent dietary patterns identified from childhood to adulthood: the Cardiovascular Risk in Young Finns Study. Br J Nutr.  2005; 93: 923– 931. Google Scholar CrossRef Search ADS PubMed  9 Wolfenden L, Wyse RJ, Britton BI, et al.   Interventions for increasing fruit and vegetable consumption in children aged 5 years and under. Cochrane Database Syst Rev . 2012;(11): CD008552. doi:10.1002/14651858.CD008552.pub2 10 Evans CE, Christian MS, Cleghorn CL, et al.   Systematic review and meta-analysis of school-based interventions to improve daily fruit and vegetable intake in children aged 5 to 12 y. Am J Clin Nutr . 2012; 96: 889– 901. http://dx.doi.org/10.3945/ajcn.111.030270 Google Scholar CrossRef Search ADS PubMed  11 Hingle MD, O'Connor TM, Dave JM, et al.   Parental involvement in interventions to improve child dietary intake: a systematic review. Prev Med.  2010; 51: 103– 111. http://dx.doi.org/10.1016/j.ypmed.2010.04.014 Google Scholar CrossRef Search ADS PubMed  12 Wyse R, Wolfenden L, Bisquera A. Characteristics of the home food environment that mediate immediate and sustained increases in child fruit and vegetable consumption: mediation analysis from the Healthy Habits cluster randomised controlled trial. Int J Behav Nutr Phys Act.  2015; 12: 118. doi:10.1186/s12966-015-0281-6. Google Scholar CrossRef Search ADS PubMed  13 Benton D. Role of parents in the determination of the food preferences of children and the development of obesity. Int J Obes Relat Metab Disord.  2004; 28: 858– 869. http://dx.doi.org/10.1038/sj.ijo.0802532 Google Scholar CrossRef Search ADS PubMed  14 Pearson N, Biddle SJ, Gorely T. Family correlates of fruit and vegetable consumption in children and adolescents: a systematic review. Public Health Nutr.  2009; 12: 267– 283. http://dx.doi.org/10.1017/S1368980008002589 Google Scholar CrossRef Search ADS PubMed  15 Knai C, Pomerleau J, Lock K, et al.   Getting children to eat more fruit and vegetables: a systematic review. Prev Med.  2006; 42: 85– 95. http://dx.doi.org/10.1016/j.ypmed.2005.11.012 Google Scholar CrossRef Search ADS PubMed  16 Haire-Joshu D, Elliott MB, Caito NM, et al.   High 5 for Kids: the impact of a home visiting program on fruit and vegetable intake of parents and their preschool children. Prev Med . 2008; 47: 77– 82. http://dx.doi.org/10.1016/j.ypmed.2008.03.016 Google Scholar CrossRef Search ADS PubMed  17 Wyse R, Wolfenden L, Campbell E, et al.   A cluster randomized controlled trial of a telephone-based parent intervention to increase preschoolers’ fruit and vegetable consumption. Am J Clin Nutr.  2012; 96: 102– 110. Google Scholar CrossRef Search ADS PubMed  18 Paineau DL, Beaufils F, Boulier A, et al.   Family dietary coaching to improve nutritional intakes and body weight control: a randomized controlled trial. Arch Pediatr Adolesc Med.  2008; 162: 34– 43. http://dx.doi.org/10.1001/archpediatrics.2007.2 Google Scholar CrossRef Search ADS PubMed  19 Spurrier NJ, Bell L, Wilson A, et al.   Minimal change in children's lifestyle behaviours and adiposity following a home-based obesity intervention: results from a pilot study. BMC Res Notes.  2016; 9: 26. doi: 10.1186/s13104-015-1796-2. Google Scholar CrossRef Search ADS PubMed  20 Cohen JF, Kraak VI, Choumenkovitch SF, et al.   The CHANGE study: a healthy-lifestyles intervention to improve rural children's diet quality. J Acad Nutr Diet . 2014; 114: 48– 53. http://dx.doi.org/10.1016/j.jand.2013.08.014 Google Scholar CrossRef Search ADS PubMed  21 Knowlden A, Sharma M. One-year efficacy testing of Enabling Mothers to Prevent Pediatric Obesity through Web-Based Education and Reciprocal Determinism (EMPOWER) randomized control trial. Health Educ Behav.  2016; 43: 94– 106. http://dx.doi.org/10.1177/1090198115596737 Google Scholar CrossRef Search ADS PubMed  22 Wolfenden L, Wyse R, Campbell E, et al.   Randomized controlled trial of a telephone-based intervention for child fruit and vegetable intake: long-term follow-up. Am J Clin Nutr.  2014; 99: 543– 550. http://dx.doi.org/10.3945/ajcn.113.071738 Google Scholar CrossRef Search ADS PubMed  23 Horton LA, Parada H, Slymen DJ, et al.   Targeting children's dietary behaviors in a family intervention:'Entre familia: reflejos de salud'. Salud Pública Méx . 2013; 55(suppl 3): 397– 405. Google Scholar PubMed  24 Holley CE, Haycraft E, Farrow C. ‘Why don’t you try it again?’A comparison of parent led, home based interventions aimed at increasing children's consumption of a disliked vegetable. Appetite . 2015; 87: 215– 222. Google Scholar CrossRef Search ADS PubMed  25 Moher D, Liberati A, Tetzlaff J, et al.   Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med.  2009; 151: 264– 269. http://dx.doi.org/10.7326/0003-4819-151-4-200908180-00135 Google Scholar CrossRef Search ADS PubMed  26 Kim SA, Moore LV, Galuska D, et al.   Vital signs: fruit and vegetable intake among children—United States, 2003–2010. MMWR Morb Mortal Wkly Rep . 2014; 63: 671– 676. Google Scholar PubMed  27 Skinner JD, Carruth BR, Bounds W, et al.   Children's food preferences: a longitudinal analysis. J Am Diet Assoc.  2002; 102: 1638– 1647. http://dx.doi.org/10.1016/S0002-8223(02)90349-4 Google Scholar CrossRef Search ADS PubMed  28 Delgado-Noguera M, Tort S, Martínez-Zapata MJ, et al.   Primary school interventions to promote fruit and vegetable consumption: a systematic review and meta-analysis. Prev Med.  2011; 53: 3– 9. Google Scholar CrossRef Search ADS PubMed  29 Treweek S, Lockhart P, Pitkethly M, et al.   Methods to improve recruitment to randomised controlled trials: Cochrane systematic review and meta-analysis. BMJ Open . 2013; 3. 10.1136/bmjopen-2012-002360 30 Corsini N, Slater A, Harrison A, et al.   Rewards can be used effectively with repeated exposure to increase liking of vegetables in 4–6-year-old children. Public Health Nutr.  2013; 16: 942– 951. http://dx.doi.org/10.1017/S1368980011002035 Google Scholar CrossRef Search ADS PubMed  31 Cravener TL, Schlechter H, Loeb KL, et al.   Feeding strategies derived from behavioral economics and psychology can increase vegetable intake in children as part of a home-based intervention: results of a pilot study. J Acad Nutr Diet.  2015; 115: 1798– 1807. http://dx.doi.org/10.1016/j.jand.2015.03.024 Google Scholar CrossRef Search ADS PubMed  32 Dulin Keita A, Risica PM, Drenner KL, et al.   Feasibility and acceptability of an early childhood obesity prevention intervention: results from the healthy homes, healthy families pilot study. J Obes . 2014; 2014:378501. doi:10.1155/2014/378501 33 Fildes A, van Jaarsveld CH, Wardle J, et al.   Parent-administered exposure to increase children's vegetable acceptance: a randomized controlled trial. J Acad Nutr Diet . 2014; 114: 881– 888. http://dx.doi.org/10.1016/j.jand.2013.07.040 Google Scholar CrossRef Search ADS PubMed  34 Knowlden AP, Sharma M, Cottrell RR, et al.   Impact evaluation of Enabling Mothers to Prevent Pediatric Obesity through Web-Based Education and Reciprocal Determinism (EMPOWER) randomized control trial. Health Educ Behav.  2015; 42: 171– 184. http://dx.doi.org/10.1177/1090198114547816 Google Scholar CrossRef Search ADS PubMed  35 McGowan L, Cooke LJ, Gardner B, et al.   Healthy feeding habits: efficacy results from a cluster-randomized, controlled exploratory trial of a novel, habit-based intervention with parents. Am J Clin Nutr.  2013; 98: 769– 777. http://dx.doi.org/10.3945/ajcn.112.052159 Google Scholar CrossRef Search ADS PubMed  36 Remington A, An E, Croker H, et al.   Increasing food acceptance in the home setting: a randomized controlled trial of parent-administered taste exposure with incentives. Am J Clin Nutr . 2012; 95: 72– 77. http://dx.doi.org/10.3945/ajcn.111.024596 Google Scholar CrossRef Search ADS PubMed  37 Schwinn TM, Schinke S, Fang L, et al.   A web-based, health promotion program for adolescent girls and their mothers who reside in public housing. Addict Behav . 2014; 39: 757– 760. http://dx.doi.org/10.1016/j.addbeh.2013.11.029 Google Scholar CrossRef Search ADS PubMed  38 Tabak RG, Tate DF, Stevens J, et al.   Family ties to health program: a randomized intervention to improve vegetable intake in children. J Nutr Educ Behav . 2012; 44: 166– 171. http://dx.doi.org/10.1016/j.jneb.2011.06.009 Google Scholar CrossRef Search ADS PubMed  39 Thompson D, Bhatt R, Vazquez I, et al.   Creating action plans in a serious video game increases and maintains child fruit-vegetable intake: a randomized controlled trial. Int J Behav Nutr Phys Act.  2015; 12: 39. doi: 10.1186/s12966-015-0199-z Google Scholar CrossRef Search ADS PubMed  40 Tomayko EJ, Prince RJ, Cronin KA, et al.   The Healthy Children, Strong Families intervention promotes improvements in nutrition, activity and body weight in American Indian families with young children. Public Health Nutr . 2016;19: 1– 10. 41 Wardle J, Cooke LJ, Gibson EL, et al.   Increasing children's acceptance of vegetables; a randomized trial of parent-led exposure. Appetite . 2003; 40: 155– 162. http://dx.doi.org/10.1016/S0195-6663(02)00135-6 Google Scholar CrossRef Search ADS PubMed  42 Wyse R, Wolfenden L, Campbell E, et al.   A pilot study of a telephone-based parental intervention to increase fruit and vegetable consumption in 3–5-year-old children. Public Health Nutr.  2011; 14: 2245– 2253. http://dx.doi.org/10.1017/S1368980011001170 Google Scholar CrossRef Search ADS PubMed  43 Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health . 1998; 52: 377– 384. http://dx.doi.org/10.1136/jech.52.6.377 Google Scholar CrossRef Search ADS PubMed  44 Luz F, Hay P, Gibson A, et al.   Does severe dietary energy restriction increase binge eating in overweight or obese individuals? A systematic review. Obes Rev.  2015; 16: 652– 665. http://dx.doi.org/10.1111/obr.12295 Google Scholar CrossRef Search ADS PubMed  45 Golan M, Crow S. Parents are key players in the prevention and treatment of weight-related problems. Nutr Rev.  2004; 62: 39– 50. http://dx.doi.org/10.1111/j.1753-4887.2004.tb00005.x Google Scholar CrossRef Search ADS PubMed  46 Jones LR, Steer CD, Rogers IS, et al.   Influences on child fruit and vegetable intake: sociodemographic, parental and child factors in a longitudinal cohort study. Public Health Nutr.  2010; 13: 1122– 1130. http://dx.doi.org/10.1017/S1368980010000133 Google Scholar CrossRef Search ADS PubMed  47 Rasmussen M, Krolner R, Klepp KI, et al.   Determinants of fruit and vegetable consumption among children and adolescents: a review of the literature. Part I: quantitative studies. Int J Behav Nutr Phys Act.  2006; 3: 22. doi: 10.1186/1479-5868-3-22 Google Scholar CrossRef Search ADS PubMed  48 Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med.  2002; 21: 1539– 1558. http://dx.doi.org/10.1002/sim.1186 Google Scholar CrossRef Search ADS PubMed  49 Borenstein M, Hedges LV, Higgins JPT, et al.   Introduction to Meta-Analysis . Chicester, West Sussex, UK: John Wiley & Sons Ltd; 2009. Google Scholar CrossRef Search ADS   50 Guenther PM, Dodd KW, Reedy J, et al.   Most Americans eat much less than recommended amounts of fruits and vegetables. J Am Diet Assoc . 2006; 106:1371–1379. http://dx.doi.org/10.1016/j.jada.2006.06.002 51 Blanchette L, Brug J. Determinants of fruit and vegetable consumption among 6–12‐year‐old children and effective interventions to increase consumption. J Hum Nutr Diet.  2005; 18: 431– 443. Google Scholar CrossRef Search ADS PubMed  52 Wardle J, Cooke L. Genetic and environmental determinants of children's food preferences. Br J Nutr.  2008; 99(suppl 1): S15– S21. Google Scholar PubMed  53 Cooke L, Carnell S, Wardle J. Food neophobia and mealtime food consumption in 4–5 year old children. Int J Behav Nutr Phys Act.  2006; 3: 14. doi: 10.1186/1479-5868-3-14 Google Scholar CrossRef Search ADS PubMed  54 Cooke L, Wardle J, Gibson E, et al.   Demographic, familial and trait predictors of fruit and vegetable consumption by pre-school children. Public Health Nutr . 2004; 7: 295– 302. Google Scholar CrossRef Search ADS PubMed  55 Skouteris H, Hill B, McCabe M, et al.   A parent-based intervention to promote healthy eating and active behaviours in pre-school children: evaluation of the MEND 2–4 randomized controlled trial. Pediatr Obes . 2016; 11: 4– 10. doi:10.1111/ijpo.12011 Google Scholar CrossRef Search ADS PubMed  56 Wakefield CE, Fardell JE, Doolan EL, et al.   Participation in psychosocial oncology and quality-of-life research: a systematic review. Lancet Oncol.  2017; 18: e153– e165. Google Scholar CrossRef Search ADS PubMed  57 Price M, Yuen EK, Goetter EM, et al.   mHealth: a mechanism to deliver more accessible, more effective mental health care. Clin Psychol Psychother.  2014; 21: 427– 436. http://dx.doi.org/10.1002/cpp.1855 Google Scholar CrossRef Search ADS PubMed  58 Prochaska JO., Velicer WF . The transtheoretical model of health behavior change. Am J Health Promot.  1997; 12: 38– 48. http://dx.doi.org/10.4278/0890-1171-12.1.38 Google Scholar CrossRef Search ADS PubMed  59 Whisman MA. The efficacy of booster maintenance sessions in behavior therapy: review and methodological critique. Clin Psychol Rev . 1990; 10: 155– 170. http://dx.doi.org/10.1016/0272-7358(90)90055-F Google Scholar CrossRef Search ADS   60 Kessler RC. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry . 2005; 62:617–627. http://dx.doi.org/10.1001/archpsyc.62.6.617 61 Lepper MR, Greene D, Nisbett RE. Undermining children's intrinsic interest with extrinsic reward: a test of the “overjustification” hypothesis. J Pers Soc Psychol . 1973; 28: 129–137. Google Scholar CrossRef Search ADS   62 Wardle J, Herrera ML, Cooke L, et al.   Modifying children's food preferences: the effects of exposure and reward on acceptance of an unfamiliar vegetable. Eur J Clin Nutr.  2003; 57: 341– 348. http://dx.doi.org/10.1038/sj.ejcn.1601541 Google Scholar CrossRef Search ADS PubMed  63 Hidi S. Revisiting the role of rewards in motivation and learning: implications of neuroscientific research. Educ Psychol Rev.  2016; 28: 61– 93. http://dx.doi.org/10.1007/s10648-015-9307-5 Google Scholar CrossRef Search ADS   64 Dickersin K, Min Y-I, Meinert CL. Factors influencing publication of research results: follow-up of applications submitted to two institutional review boards. JAMA . 1992; 267: 374– 378. http://dx.doi.org/10.1001/jama.1992.03480030052036 Google Scholar CrossRef Search ADS PubMed  65 Dixon B, Peña M-M, Taveras EM. Lifecourse approach to racial/ethnic disparities in childhood obesity. Adv Nutr.  2012; 3: 73– 82. Google Scholar CrossRef Search ADS PubMed  66 Burrows TL, Martin RJ, Collins CE. A systematic review of the validity of dietary assessment methods in children when compared with the method of doubly labeled water. J Am Diet Assoc . 2010; 110: 1501– 1510. http://dx.doi.org/10.1016/j.jada.2010.07.008 Google Scholar CrossRef Search ADS PubMed  67 Livingstone MB, Prentice AM, Coward WA, et al.   Validation of estimates of energy intake by weighed dietary record and diet history in children and adolescents. Am J Clin Nutr . 1992; 56: 29– 35. Google Scholar CrossRef Search ADS PubMed  68 Johns B, Baltussen R, Hutubessy R. Programme costs in the economic evaluation of health interventions. Cost Eff Resour Alloc.  2003; 1: 1. doi: 10.1186/1478-7547-1-1 Google Scholar CrossRef Search ADS PubMed  69 Head KJ, Noar SM, Iannarino NT, et al.   Efficacy of text messaging-based interventions for health promotion: a meta-analysis. Soc Sci Med.  2013; 97: 41– 48. http://dx.doi.org/10.1016/j.socscimed.2013.08.003 Google Scholar CrossRef Search ADS PubMed  70 Schoeppe S, Alley S, Van Lippevelde W, et al.   Efficacy of interventions that use apps to improve diet, physical activity and sedentary behaviour: a systematic review. Int J Behav Nutr Phys Act.  2016; 13: 127. 10.1186/s12966-016-0454-y Google Scholar CrossRef Search ADS PubMed  71 Zhao J, Freeman B, Li M. Can mobile phone apps influence people's health behavior change? An evidence review. J Med Internet Res.  2016; 18: e287. doi: 10.2196/jmir.5692 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nutrition Reviews Oxford University Press

Parent-targeted home-based interventions for increasing fruit and vegetable intake in children: a systematic review and meta-analysis

Loading next page...
 
/lp/ou_press/parent-targeted-home-based-interventions-for-increasing-fruit-and-hN00QMKNcb
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.
ISSN
0029-6643
eISSN
1753-4887
D.O.I.
10.1093/nutrit/nux066
Publisher site
See Article on Publisher Site

Abstract

Abstract Context Parent interventions delivered in the home represent a valuable approach to improving children’s diets. Objective This review aims to examine the effectiveness of parent-targeted in-home interventions in increasing fruit and vegetable intake in children. Data Sources Five electronic databases were searched: MEDLINE, Embase, PubMed, CINAHL, and PsycINFO. Study Selection Randomized and nonrandomized trials conducted in children aged 2 to 12 years and published in English from 2000 to 2016 were eligible. Data Extraction Eighteen publications were reviewed, and 12 randomized trials were analyzed. Studies were pooled on the basis of outcome measure and type of intervention, resulting in 3 separate meta-analyses. Results Nutrition education interventions resulted in a small but significant increase in fruit intake (Hedges’ g = 0.112; P = 0.028). Taste exposure interventions led to a significant increase in vegetable intake, with a moderate effect (Hedges’ g = 0.438; P < 0.001). Interventions involving daily or weekly sessions reported positive outcomes more frequently than those using monthly sessions. Conclusions Future interventions should incorporate regular taste exposure to maximize increases in vegetable intake in children. This is particularly important because fewer children meet national recommendations for vegetable intake than for fruit intake. child, fruit, home, parent, vegetables INTRODUCTION Eating fruit and vegetables (FVs) can protect against obesity1 and reduce the risk of mortality from cardiovascular disease.2 Despite this, children across much of the developed world do not meet the recommendations for daily intake of FVs. In the United States and the United Kingdom, children’s intake of FVs remains below national recommendations for age and sex.3,4 In Europe, a high proportion (76.5%) of children do not meet the World Health Organization’s recommendation of 400 g of FV per day.5 According to the most recent Australian 2014–2015 National Health Survey, only 1 in 20 children aged 2 to 18 years meet the Australian guidelines for recommended daily servings of vegetables or FVs combined.6 Longitudinal analyses of children’s eating behaviors suggest that food preferences are developed early and are likely to persist into adulthood.7,8 This has led to increasing recognition of the need for early interventions that promote FV intake among children.9 School-based interventions are among the most commonly used strategies for increasing children’s intake of FVs, despite having minimal impact on children’s vegetable intake.10 However, systematic reviews suggest that school-based interventions that directly target parents, for example, via parent attendance at intervention sessions are more likely to report positive or mixed dietary outcomes (eg, increased FV intake and reduced fat intake) than are interventions that indirectly involve parents, for example, via a parent newsletter.11 It is well documented that parents exert substantial influence over their children’s FV intake.5,12 For example, studies have identified a number of parental factors associated with increases in children’s intake of FVs, including parent intake of FVs, parent providing of FVs to children, structured family mealtimes, set rules around eating, and the availability and accessibility of FVs in the home.5,13,14 Targeting parents in nutrition interventions may therefore represent a particularly effective strategy for increasing children’s FV intake. Several systematic reviews on interventions aimed at increasing FV intake in children are published in the literature.9,11,15 However, few have evaluated parent-targeted interventions (< 50%)9,11,15 and, where parental involvement was included, most did not require direct parental involvement (< 25%).11 Parent-targeted interventions have the potential to be more acceptable to parents when compared with community-based or multicomponent school-based interventions that involve a parent or family component. This is because parent-targeted interventions are more accessible, with the majority of interventions being delivered in the home environment via home visits,16 telephone calls,17 and online delivery methods, eliminating the need for parents to travel.17,18 Reducing barriers of participant access to interventions is also imperative for policymakers and health professionals interested in delivering parent-targeted strategies in rural and remote areas, where intervention is often limited.19,20 An increasing number of parent-targeted, home-based FV interventions have become available. For example, 2 recently published randomized controlled trials (RCTs) reported significant increases in children’s FV intake up to 12 months after the completion of an online21 or telephone-delivered22 parent-targeted FV intervention. Overall, however, the effects of parent-targeted home-based FV interventions are mixed, with some home-visiting parent-targeted FV interventions reporting no significant increases in children’s FV intake.19,23 It is therefore difficult to ascertain which parent-targeted home-based interventions may be most effective at promoting FV intake among children. It is possible that differences in the type of parent-targeted intervention being evaluated may have contributed to the inconsistent findings to date. Parent-targeted, home-based interventions have tended to focus on 2 strategies: repeated taste exposure, which involves repeatedly exposing children to a target vegetable at home (with or without a reward), and nutrition- and skills-based education. Taste exposure interventions are based on the theory of learned safety in which repeated exposure to vegetables, without negative consequences (eg, gastrointestinal upset), is hypothesized to support greater vegetable acceptance among children.24 Nutrition education interventions, however, have largely targeted evidence-based parent and home environment factors associated with increases in children’s FV intake. These include parent education and training in the national dietary guidelines, along with strategies to encourage parent modeling of FV intake, provide FVs to children daily, and increase the availability of ready-to-eat FVs to children in the home (eg, clean and chopped FVs).16 Although taste exposure and nutrition education are potentially valuable approaches for improving children’s FV intake, previous systematic reviews have not specifically examined the effect of parent-targeted, home-based FV interventions according to the type of intervention strategy delivered. Therefore, a systematic review of the acceptability and feasibility of taste exposure and nutrition education interventions and their effectiveness on children’s FV intake is needed. The findings will provide evidence-based information for parents, health professionals, and policymakers seeking effective strategies to increase FV intake in children. METHODS The aims of this study were to systematically identify and review the quality, feasibility, and acceptability of parent-targeted, home-based interventions aimed at promoting FVs to children and to undertake a meta-analysis to analyze the effects of such interventions on FV intake in children. Using the PICOS (Participants, Intervention, Comparators, Outcomes, Study Design) model for framing research questions, 4 key questions were identified: (1) Are parent-targeted home-based interventions aimed at increasing FV intake in children aged 2 to 12 years feasible and acceptable to parents? (2) Are parent-targeted home-based taste exposure and nutrition education interventions effective at promoting significant increases in children’s FV intake? (3) Do parent-targeted home-based taste exposure and nutrition education interventions, compared with no intervention, result in significantly greater FV intake in children aged 2 to 12 years? (4) What characteristics are common among parent-targeted home-based interventions that achieve significant increases in FV intake in children aged 2 to 12 years? The PICOS criteria are shown in Table 1. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (see Appendix S1 in the Supporting Information online), considered the gold standard for reporting the evaluation of interventions,25 were used when designing this review. This required the following: (1) a specific set of inclusion and exclusion criteria (a priori); (2) an extensive literature search across multiple databases; (3) full documentation of the electronic search strategy to enable replication; (4) use of a standardized process to screen and select studies; and (5) use of a standardized process to collect data and assess the risk of bias of individual studies. The protocol for this review has not been published and is not registered. Search strategy Five electronic databases (PubMed, PsycINFO, CINAHL, Embase, and Ovid MEDLINE) were searched for human studies published in English between January 2000 and February 2016. The following search terms were used in each database: (parent* OR home*) AND (child* OR preschool*) AND (fruit* OR vegetable*) AND (intervention*). This search was updated in August 2016. Selection criteria Inclusion and exclusion criteria are described in Appendix S2 in the Supporting Information online. Studies that evaluated a parent-targeted home-based intervention aimed at increasing children’s fruit and/or vegetable intake and were published in a peer-reviewed journal were included. Additional criteria were included, as follows. (1) The intervention was published between January 2000 and August 2016. (2) The intervention required a parent-only component (eg, parent telephone calls, parent website/newsletter, and parent-led exposure activity) and was delivered to the parent at home, for example, via the internet, telephone, or mail. Interventions that included a parent-only component but were delivered in a school, community, or research setting were excluded. (3) Interventions were aimed at increasing children’s fruit and/or vegetable intake. A child was defined as a person between the ages of 2 and 12 years. This broad definition was designed to capture preschool- and primary school-aged children, both of whom demonstrate poor intake of FVs.3,6,26 This is also the age range during which parents have the strongest influence on children’s food preferences.27 (4) Interventions were specifically designed to effect change in at least 1 measure of child fruit and/or vegetable intake. A range of outcomes was included to account for the variability of outcome measures used across studies.28 These included intake of a target vegetable in grams or daily servings of a child’s fruit and/or vegetable intake via parent self-report (eg, food frequency questionnaires, 24-hour recalls, or survey items). (5) Studies were designed as RCTs, nonrandomized controlled trials, or pre–post interventions. Study selection Figure 1 summarizes the process of identifying, screening, and including studies. Two searches yielded a total of 1867 abstracts, which were screened by 3 authors (L.M.T., M.S., and A.M.G.) using the established inclusion and exclusion criteria. To determine eligibility, full-text articles were retrieved for all abstracts that were judged to be eligible by at least 1 of these authors or that did not yield sufficient information in the abstract (n = 35). Initial disagreements were resolved through a comprehensive discussion between the authors in which the inclusion and exclusion criteria were compared against the full text. When an agreement could not be reached, consensus was achieved by employing a fourth author (J.C.). An extended search of the reference lists of included studies and table of contents of key journals did not yield any additional studies. Data extraction One author (L.M.T.) extracted the data from the included studies using a standardized table in Microsoft Excel.29 Extracted data items are shown in Table 2.16,17,19,21–24,30–42 Quality assessment The methodological quality of the included studies was assessed by consensus between 2 authors (L.M.T. and M.S.) using an adapted version of the Downs and Black checklist for assessing the quality of health care intervention studies.43 A third author (A.M.G.) was included when full agreement was not reached for all criteria (Table 316,17,19,21–24,30–42). Constructs of study quality included the following: (1) clear reporting of study hypotheses or aims, participant characteristics, and main findings; (2) clear reporting of attrition rate and methods of randomization, if applicable; (3) blinding of study participants to the intervention they received (RCT only); (4) blinding of those measuring the main outcomes of the intervention (RCTs only); (5) use of valid and reliable outcome measures; and (6) clear reporting of a sample power calculation.43,44 Studies that reported high attrition rates (> 30%) or reported outcome data only for participants who completed the intervention were evaluated as having lower quality due to potential outcome bias. Each construct of study quality was scored using the criteria “yes,” “no,” or “unclear” against the adapted Downs and Black criteria (Table 3).44 Table 3 Appraisal of methodological quality of the 18 publications included in the systematic review, based on the parameters proposed by Downs and Black43 Reference  Aim(s) clearly described  Participant characteristics clearly described  Main findings clearly described  Attrition rate acceptable (< 30%)  Randomization method clearly described  Participants blinded  Assessors blinded  Outcome measures valid and reliable  Risk of residual confounding  Sample power calculation clearly reported  Potential outcome bias  Corsini et al. (2013)30  Yes  Yes  Unclear  Yes  Unclear  No  No  Yes  Unclear  Yes  No  Cravener et al. (2015)31  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Dulin Keita et al. (2014)32  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Fildes et al. (2014)33  Yes  Yes  Yes  No  Unclear  No  Unclear  Yes  Unclear  No  Yes  Haire-Joshu et al. (2008)16  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Holley et al. (2015)24  Yes  Yes  Yes  Yes  No  Unclear  No  Yes  Low  Yes  No  Horton et al. (2013)23  No  Yes  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Knowlden et al. (2015)34  Yes  Yes  Yes  Yes  Yes  Yes  No  Yes  Low  Yes  No  McGowan et al. (2013)35  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Remington et al. (2012)36  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Schwinn et al. (2014)37  Yes  No  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Spurrier et al. (2016)19  Yes  Yes  Yes  No  N/A  N/A  N/A  Yes  N/A  No  Yes  Tabak et al. (2012)38  Yes  Yes  Yes  Yes  Unclear  No  No  Yes  Low  Yes  No  Thompson et al. (2015)39  No  Yes  Yes  Yes  Unclear  No  Unclear  Yes  Low  Yes  No  Tomayako et al. (2016)40  Yes  Yes  Yes  No  Unclear  Unclear  No  Yes  Low  No  No  Wardle et al. (2003)41  Yes  No  Yes  Yes  Unclear  No  No  Yes  Unclear  No  No  Wyse et al. (2012)17  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Wyse et al. (2011)42  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Percent “yes”  89  89  83  83  40  7  0  100  20  67  11  Reference  Aim(s) clearly described  Participant characteristics clearly described  Main findings clearly described  Attrition rate acceptable (< 30%)  Randomization method clearly described  Participants blinded  Assessors blinded  Outcome measures valid and reliable  Risk of residual confounding  Sample power calculation clearly reported  Potential outcome bias  Corsini et al. (2013)30  Yes  Yes  Unclear  Yes  Unclear  No  No  Yes  Unclear  Yes  No  Cravener et al. (2015)31  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Dulin Keita et al. (2014)32  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Fildes et al. (2014)33  Yes  Yes  Yes  No  Unclear  No  Unclear  Yes  Unclear  No  Yes  Haire-Joshu et al. (2008)16  Yes  Yes  No  Yes  Yes  No  No  Yes  Low  Yes  No  Holley et al. (2015)24  Yes  Yes  Yes  Yes  No  Unclear  No  Yes  Low  Yes  No  Horton et al. (2013)23  No  Yes  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Knowlden et al. (2015)34  Yes  Yes  Yes  Yes  Yes  Yes  No  Yes  Low  Yes  No  McGowan et al. (2013)35  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Remington et al. (2012)36  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Schwinn et al. (2014)37  Yes  No  Yes  Yes  Unclear  No  No  Yes  Low  No  No  Spurrier et al. (2016)19  Yes  Yes  Yes  No  N/A  N/A  N/A  Yes  N/A  No  Yes  Tabak et al. (2012)38  Yes  Yes  Yes  Yes  Unclear  No  No  Yes  Low  Yes  No  Thompson et al. (2015)39  No  Yes  Yes  Yes  Unclear  No  Unclear  Yes  Low  Yes  No  Tomayako et al. (2016)40  Yes  Yes  Yes  No  Unclear  Unclear  No  Yes  Low  No  No  Wardle et al. (2003)41  Yes  No  Yes  Yes  Unclear  No  No  Yes  Unclear  No  No  Wyse et al. (2012)17  Yes  Yes  Yes  Yes  Yes  No  No  Yes  Low  Yes  No  Wyse et al. (2011)42  Yes  Yes  Yes  Yes  N/A  N/A  N/A  Yes  N/A  Yes  No  Percent “yes”  89  89  83  83  40  7  0  100  20  67  11  Abbreviation: N/A; not applicable. A child’s FV intake may be influenced by factors other than age and sex, such as socioeconomic status, parent FV intake, and parent body mass index.12,45–47 Therefore, each RCT (n = 14) and nonrandomized controlled trial (n = 1) was further assessed for risk of residual confounders. Randomized control trials were rated as having adequately controlled for potential confounders (low risk of bias) when authors provided evidence of a comparison between the intervention group and the control group on important confounders at baseline and when no between-group differences were observed (eg, demographics table) or when differences were identified but were adequately controlled for in the final analysis (eg, by the inclusion of confounder[s] as covariate[s] in the analysis). Meta-analysis Of the 20 studies included in the systematic review,16,17,19,21–24,30–42 8 were excluded from the meta-analysis,17,19,21,22,32,34,40,42 leaving 12 interventions for analysis. Reasons for exclusion included the publication of long-term follow-up data of a previously included study,21,22 the absence of a control (n = 3),19,32,42 and the lack of an appropriate control group (n = 1) (eg, control group received a mail-based version of the intervention).40 Because of the differences between fruit intake and vegetable intake, data for FVs were extracted separately. The authors of studies that reported an outcome measure only for fruit intake and vegetable intake combined were contacted to request information that would enable a separate analysis. One author was unable to provide separate scores for fruit intake and vegetable intake, and 1 author could not be reached, resulting in the exclusion of an additional 2 studies.17,34 When studies reported data for more than 1 pre- or postmeasurement time point, data closest to the beginning (but before delivery of intervention materials) and the end of the intervention were selected. Interventions differed according to 2 key theoretical and design-based criteria: the type of intervention (taste exposure vs nutrition education) and the principal summary measures targeted, such as type of outcome (fruit intake vs vegetable intake vs combined FV intake). To reduce exclusion criteria, conserve data points, and minimize the estimation of variance, outcome data were pooled separately for taste exposure and nutrition education interventions instead of being pooled into a single meta-regression on the basis of effect sizes. For each study outcome (fruit intake or vegetable intake), the standardized mean difference (Hedges’ g) was calculated as the mean between-group difference in pre–post intervention change scores divided by the standard deviation of the change scores pooled over the groups. Comprehensive meta-analysis software (BioStat, version 3) was used to calculate effect sizes and meta-analyze these effect sizes. In the results, a random-effects model was reported because of heterogeneity in some outcomes, but no outcome differed when a fixed-effects model was used. Heterogeneity of effect sizes was examined using the I2 statistic.48 Data extraction was checked by having a second author (V.Q.) re-extract 40% of the data. Wherever possible, means and standard deviations of pre–post change scores were extracted for taste exposure and nutrition education interventions. When means and standard deviations for pre–post interventions were available only separately, the scores for standard deviation of change, assuming a pre–post correlation of 0.6, were estimated. For interventions with more than 1 intervention condition of the same type of treatment, Hedges’ g was calculated between each intervention arm and control arm and then these were combined into 1 summary effect, assuming a between-subgroup correlation of 0.5.49 Sensitivity analyses were conducted for both imputed correlations for a range of plausible values (for pre–post correlation, 0.5, 0.6, and 0.7; and for subgroup variation, 0.4, 0.5, and 0.6) to ensure that no assumption altered the significance of a meta-analysis. A separate sensitivity analysis was conducted to examine whether removing studies identified as having potential bias (n = 2) (Table 3) altered the results of the meta-analysis. Studies were grouped according to whether the intervention primarily used a taste exposure or a nutrition education intervention (intervention type), and whether interventions measured fruit intake or vegetable intake (outcome type). As all studies with fruit-only outcomes were education interventions, this method produced 3 meta-analyses: the effect of nutrition education intervention vs no intervention on children’s fruit intake, the effect of nutrition education intervention vs no intervention on children’s vegetable intake, and the effect of taste exposure intervention vs no intervention on children’s vegetable intake. Consequently, only 1 effect size per study was included in any 1 meta-analysis. RESULTS A total of 622 studies were identified, and 9 RCTs were included in the meta-analysis (Figure 1). Figure 1 View largeDownload slide Flow diagram of the literature search process. A total of 1209 studies were identified across 2 searches. Eighteen studies were included in the systematic review and 12 in the meta-analysis. Figure 1 View largeDownload slide Flow diagram of the literature search process. A total of 1209 studies were identified across 2 searches. Eighteen studies were included in the systematic review and 12 in the meta-analysis. Study selection After the initial screening, 2 searches identified 1209 studies, 40 of which were considered potentially eligible and were retrieved for full-text review (35 from the initial search and 5 from the second search). A further 20 studies were excluded because the intervention did not target FV intake in children or was not an intervention study (n = 9), because the intervention was not home based (eg, it was community- or school-based) (n = 9), or because the article did not report original research (eg, protocol or dissertation) (n = 2). Although this resulted in a total of 20 eligible studies, 2 studies reported the long-term follow-up data of a previously identified and eligible intervention.21,22 These studies were presented with the original intervention rather than as a separate study (Table 2) (n = 2). The systematic review therefore included 18 intervention studies reporting 3759 children with a mean age of 5.3 years (range, 2–11 y; SD = 2.7). Table 1 summarizes the characteristics of the eligible studies. Two intervention studies were assessed as having potential outcome bias.19,33 To address this issue, a sensitivity analysis was conducted, which resulted in the exclusion of these studies from the meta-analysis. Table 1 PICOS criteria for inclusion and exclusion of studies Parameter  Inclusion criteria  Exclusion criteria  Population  Children aged 2–12 y and their parents  Children aged < 2 y or > 12 y and their parents  Intervention  Fruit and/or vegetable intake in children before and after a parent-targeted, home-based intervention  Interventions delivered in a school, community, or research setting; interventions without a parent-only component  Comparison interventions  Intervention vs no intervention (inactive controls); intervention vs active controls    Outcomes  Fruit and/or vegetable intake in children (including daily intake of a target vegetable in grams or daily intake of servings of fruit and/or vegetables via parent self-report)  Parent fruit and/or vegetable intake  Study types  Randomized controlled trial, nonrandomized controlled trial, and pre- and postintervention studies  Conference abstracts, systematic reviews, meta-analyses, book chapters, dissertations, case–control studies, cross-sectional studies  Parameter  Inclusion criteria  Exclusion criteria  Population  Children aged 2–12 y and their parents  Children aged < 2 y or > 12 y and their parents  Intervention  Fruit and/or vegetable intake in children before and after a parent-targeted, home-based intervention  Interventions delivered in a school, community, or research setting; interventions without a parent-only component  Comparison interventions  Intervention vs no intervention (inactive controls); intervention vs active controls    Outcomes  Fruit and/or vegetable intake in children (including daily intake of a target vegetable in grams or daily intake of servings of fruit and/or vegetables via parent self-report)  Parent fruit and/or vegetable intake  Study types  Randomized controlled trial, nonrandomized controlled trial, and pre- and postintervention studies  Conference abstracts, systematic reviews, meta-analyses, book chapters, dissertations, case–control studies, cross-sectional studies  Table 2 Characteristics of parent-targeted home-based interventions to increase fruit and vegetable intake in children Reference; country  Sample size; age range of children  Inclusion criteria  Characteristics of intervention (mode of intervention delivery; duration of intervention; frequency and length of intervention sessions; length of data collection from baseline to final point of data collection; type/training of interventionist)  Outcome measures (reported outcome measures in relation to child fruit and or vegetable intake; measure of acceptability and/or feasibility identified; potential confounders balanced or controlled for in RCT)  Randomized control trials  Corsini et al. (2013)30; Australia  N = 185; 4–6 y  Child’s age (4–6.99 y) Parent commitment to undertake a short activity daily for 2 wk Parent willingness to have 4 fieldworker home visits Parent ability to communicate in English  Mail-based (intervention materials), home-based (parent-led exposure), and home visits (parent training); 14 d of taste exposure only (EO) vs exposure and sticker reward (E + R) vs control (no intervention) and 4 home visits; length of intervention sessions unclear; data were collected 4 wk and 3 mo from baseline; trained fieldworkers  Intake of the target vegetable (in grams), usual vegetable intake measured using the Children’s Dietary Questionnaire, vegetable intake frequency via parent report (0–4 times), parents used a checklist of 23 vegetables to indicate how many vegetables children consumed in the past week and children’s liking of the target vegetable using a 3-point visual facial scale; no measure(s) of feasibility and/or acceptability reported; child age, sex, baseline FV intake in children  Cravener et al. (2015)31; USA  N = 24; 3–5 y  Child without pre-existing medical conditions (including food allergies) Child intake of < 2 servings of vegetables per day Child at risk for obesity on basis of family history, defined as having at least 1 parent with a BMI ≥ 25 kg/m2  Home visits and printed materials (food packages with parent-instructions and child-targeted education sessions); 4 wk; length of intervention sessions unclear; data were collected 4 wk from baseline; training of interventionists unclear  Vegetable intake as the difference between pre- and post-weights (in grams) of the foods provided, child-rated liking of 6 vegetables; qualitative postintervention feedback; child sex, age, SES, vegetable intake at baseline, BMI z-score, SES (paternal and maternal education level), maternal BMI, paternal BMI, and ethnicity  Fildes et al. (2014)33; England and Wales  N = 98; 3–4 y  Child’s age (3–4 yr) Child enrolled in GEMINI cohort  Mail-based and web-based; 14 d of taste exposure followed by a sticker reward vs control (no intervention); length of intervention sessions unclear; data were collected 14 d from baseline; N/A (instructions delivered via mail and online)  Number of pieces eaten measured the child’s intake of the target vegetable; children’s liking of the target variable measured using parent report on 9-point scale (“dislikes a lot” to “likes a lot”); quantitative postintervention evaluation questionnaire; child sex, age, SES (maternal education), and maternal BMI  Horton et al. (2013)23; Mexico  N = 361; 7–13 y  Child’s age (7–13 y) Child lives at home Child not on a medically prescribed diet Mother’s age (≥ 18 y) Mother married or living with partner Mother Spanish speaking Family living in Imperial County with no plans to move  Home visits (family intervention sessions) and telephone-based (support calls); 14 wk; home visits delivered weekly for 8 wk followed by 6 wk of weekly alternating home visits and telephone support calls (length of sessions unclear), ie, total intervention length was 16.5 h or 990 min; data were collected 14 wk from baseline; trained community members  Daily FV intake using 2 questions from the National Cancer Institute’s Food Attitudes and Behavior survey and child self-reports of FV variety; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, percentage of parents married, SES (maternal education level, median household size, percentage on food assistance, percentage that own their home), and maternal race  Knowlden et al. (2015)34; USAa Knowlden & Sharma (2016)21; USA (12-mo FU)a  N = 57; 4–6 y N = 44; 4–6 y  Child’s age (4–6 y) Parent English speaking Family internet and telephone access Mother not pregnant Child without disability Child without a medical condition associated with weight gain or prescribed weight management medication Child not enrolled in a weight-management program  Web-based; 4 wk; sessions of 20–30 min/wk; 5 educational sessions plus a booster; 1–15-min audio-visual presentation, interactive worksheet, and a discussion board post; data were collected 8 wk and 1 y from baseline; N/A (intervention delivered via a website)  Child’s FV consumption (dietary recall, measured in cups), and fruit availability; process evaluation data collected via telephone counseling and postintervention evaluation surveys (intervention fidelity, dose delivered, dose received, reach, recruitment, and potential cross-contamination between the groups); child sex, age, FV intake at baseline, race, SES (maternal marital status, maternal employment status); and maternal race  McGowan et al. (2013)35; UK  N = 126; 2–6 y  Child’s age (2–6 y) Child without known medical or psychological condition affecting diet Parent English speaking  Home visits and printed materials; 8-wk; sessions of 1 h/wk; researcher worked through an intervention booklet with the parent; data were collected 8 wk from baseline; researchers received training prior to the intervention  Child’s daily FV intake via parent self-report (“How many servings of fruit [vegetables] does your child typically eat?”; 7-point scale from “less than 1 per day” to “5 per day”); postintervention interview covering intervention acceptability; child sex, age, FV intake at baseline, ethnicity, parent age, and SES (parent education and parent living status)  Remington et al. (2012)36; UK  N = 173; 3–4 y  Child’s age (3–4 y) Child attending a selected nursery school  Home visits; 12 d of daily taste exposure to a target vegetable followed by praise vs a sticker for tasting vs control group; length of intervention sessions unclear; data were collected 4 wk and 12 wk from baseline; trained researchers  Child’s liking of the target vegetable via parent report using a faces scale and intake of the target vegetable (in grams) using a digital scale; postintervention qualitative feedback; child age, sex, vegetable intake, or vegetable liking at baseline, parent age, parent ethnicity, and SES (parent home ownership, parent education)  Schwinn et al. (2014)37; USA  N = 67; girls, 10–12 y  Girls Child’s age (10–12 y) Girls and mothers living in publicly subsidized housing  Web-based (website); 3 wk; sessions of 25 min/wk; data were collected 3 wk and 5 mo from baseline; N/A (intervention delivered via an online platform)  Child’s score on the Youth and Adolescent FFQ via parent-self-report; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, ethnicity, maternal age, maternal FV intake at baseline, and SES (maternal employment, whether child qualifies for reduced-price or free school lunch)  Tabak et al. (2012)38; USA  N = 43; 2–5 y  Child’s age (2–5 y) Family living in current residence for next 6 mo  Mail- and telephone-based; 4 mo: 4 tailored newsletters and 2 motivational phone calls delivered every 4 wk; mean length of telephone sessions was 34 min; data were collected 4 mo from baseline; registered dietitian trained in motivational interviewing techniques  Child’s score on the Block Kids FFQ via parent self-report; quantitative postintervention evaluation questionnaire, postintervention qualitative feedback and process evaluation data (telephone session duration in minutes); child age, sex, parent age, parent sex, parent BMI, ethnicity, and SES (parent income)  Thompson et al. (2015)39; USA  N = 387; 9–11 y  Child in 4th or 5th grade (age 9–11 y Family English speaking Computer and high-speed internet access Parent willing to participate in telephone sessions  Online (video game) and web-based (website and electronic newsletters); 10 sessions delivered over 3 mo (length of sessions unclear) across 4 groups—action, coping, action + coping, and control; data were collected 3 mo and 6 mo from baseline; trained staff conducted dietary recall  Child’s FV intake via 3 unannounced 24-h dietary recalls (2 d during week, 1 d on weekend) conducted over the telephone by trained staff and 3 d of FV intake at each data collection period were averaged; process evaluation data (participation rates); parent satisfaction scale; child sex, FV intake at baseline, ethnicity, parent age, parent ethnicity, and SES (parent education level)  Tomayako et al. (2016)40; USA  N = 150; 2–5 y  Family of American Indian background; child’s age (2–5 y) Child lived with at least 1 primary caregiver (eg, mother, father, grandmother, aunt) in a home setting Child free of any major physical or behavioral disorders  Family-based randomized trial of a healthy lifestyle toolkit delivered via 2 formats: in-home mentoring via 12 monthly home visits or by mail. Each lesson addressed 1 of 4 target areas: (1) eat more FV, (2) consume less soda and added sugar, (3) become more active, and (4) watch less TV; 12-mo intervention period; home visits of 60 min duration; data were collected 12 mo from baseline; interventionists were tribal members or individuals who had longstanding employment within the community and were trained to administer the intervention  Daily servings of FV, sugar-sweetened drinks, and candy/junk food using Nutrition Data System for Research software 2010; focus group testing covering intervention acceptability; child age, sex, FV intake at baseline, BMI percentile, BMI z-score, ethnicity, caregiver age, caregiver sex, caregiver ethnicity, adult BMI, caregiver FV intake, and SES (caregiver education level)  Wardle et al. (2003)41; Australia  N = 156; 2–6 y  Child’s age (2–6 y)  Home visits; 2 wk of daily taste exposure to target vegetable vs nutrition education vs no intervention vs a postintervention taste test; session length unclear; data were collected 2 wk from baseline; training of interventionists unclear  Child’s liking of the 6 test vegetables measured using a 3-point faces scale, child’s consumption of a target vegetable measured (in grams) using a digital scale; semistructured postintervention interview covering intervention acceptability; child sex, age, and vegetable intake at baseline  Wyse et al. (2012)17; Australiaa Wolfenden et al. (2014)22; Australia (12- and 18-mo FU)a  N = 394; 3–5 y N = 164; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk; parent must have some responsibility for providing meals and snacks to child; parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2–6-mo, 12 mo, and 18 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention acceptability questionnaire; child age, sex, SES (decile of disadvantage classification associated with child’s preschool postal code), and children’s FV intake at baseline Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (number of intervention calls completed, delivery of key topics, delivery of the intervention as per protocol); child age, sex, ethnicity, FV intake at baseline, parent age, parent sex, SES (household income, university education), and parent FV intake)  Haire-Joshu et al. (2008)16; USA  N = 1306; 2–5 y  Child’s age (2–5 y) Parent’s age (20–59 y) Child and parent living in rural, southeast Missouri  Mail-based (1 tailored newsletter); 4 home visits of 60 min duration (frequency unclear) and 4 sing-a-long story books and audio cassettes (delivered at each home visit); mean length of data were collected 7 mo from baseline; parent educators received 4 h of PAT training on nutrition and material content  Intake frequency measured using the Saint Louis University for Kids FFQ, child-feeding practices, parent modeling of FV intake, nutrition knowledge, FV availability in the home; quantitative postintervention evaluation questionnaire; child sex, age, FV intake at baseline, parent age, parent sex, and SES (parent education level)  Nonrandomized control trials  Holley et al. (2015)24; UK  N = 115; 2–4 y  Child’s age (2–4 y)  Home-based (parent-led exposure) and community-based (assessments held at preschool); 14-d intervention across 5 groups: repeated exposure (RE), modeling plus repeated exposure (M + RE), rewards plus repeated exposure (R + RE), modeling, rewards, and RE (condition 4) vs a no-treatment control group; session length unclear; data were collected 2 wk from baseline; unclear if interventionists received training  Intake (in grams) and liking (measured using a 3-point smiley-face scale) of the target vegetable; no measure(s) of feasibility and/or acceptability reported; child age, sex, vegetable intake at baseline, BMI z-score, parent age, and parent sex  Pre–post studies  Dulin Keita et al. (2014)32; USA  N = 39 (data completed at baseline and at FU; 2–5 y  Child’s age (3–5 y) Child’s age/sex-specific BMI is ≥ 50th percentile Parent’s age (≥ 18 y) Parent lives with child at least 75% of the time Parent can speak and read English Parent is knowledgeable about child’s diet and physical activity  Mail- and telephone-based; 4 mo; 4 tailored mailouts, 3 motivational telephone calls, activity video (session length unclear); data were collected 4 mo from baseline; lay counselors received 12 h of motivational interviewing training from MINT-qualified trainer  Child’s score on the National Cancer Institute’s FV all-day screener tool,; process evaluation data (number of counselor reported calls completed), quantitative postintervention evaluation questionnaire; N/A  Spurrier et al. (2016)19; Australia  N = 24 (22 families); 4–12 y  Child’s age (4–12 y) Child’s age/sex-specific BMI indicative of overweight or obesity according to IOTF definitions Child living in metropolitan Adelaide, South Australia Child not diagnosed with a medical condition affecting weight or growth or not enrolled in a weight-management program  Home-based (parent-led) education program; 3 home visits and 2 FU telephone calls were offered to each family (session length unclear); data were collected approximately 6 mo from baseline (21–44 wk); researchers had backgrounds in nutrition, occupational therapy, or human movement and received 3 h of training and education prior to intervention  Score on the FV subscale of the Children’s Dietary Questionnaire (no measure[s] of feasibility and/or acceptability reported); N/A  Wyse et al. (2011)42; Australia  N = 34; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Parent must have some responsibility for providing meals and snacks to child Parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2 mo and 6 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Score on the FV Subscale of the Children’s Dietary Questionnaire and Household Food Expenditure Survey, noncore subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention evaluation questionnaire; N/A  Reference; country  Sample size; age range of children  Inclusion criteria  Characteristics of intervention (mode of intervention delivery; duration of intervention; frequency and length of intervention sessions; length of data collection from baseline to final point of data collection; type/training of interventionist)  Outcome measures (reported outcome measures in relation to child fruit and or vegetable intake; measure of acceptability and/or feasibility identified; potential confounders balanced or controlled for in RCT)  Randomized control trials  Corsini et al. (2013)30; Australia  N = 185; 4–6 y  Child’s age (4–6.99 y) Parent commitment to undertake a short activity daily for 2 wk Parent willingness to have 4 fieldworker home visits Parent ability to communicate in English  Mail-based (intervention materials), home-based (parent-led exposure), and home visits (parent training); 14 d of taste exposure only (EO) vs exposure and sticker reward (E + R) vs control (no intervention) and 4 home visits; length of intervention sessions unclear; data were collected 4 wk and 3 mo from baseline; trained fieldworkers  Intake of the target vegetable (in grams), usual vegetable intake measured using the Children’s Dietary Questionnaire, vegetable intake frequency via parent report (0–4 times), parents used a checklist of 23 vegetables to indicate how many vegetables children consumed in the past week and children’s liking of the target vegetable using a 3-point visual facial scale; no measure(s) of feasibility and/or acceptability reported; child age, sex, baseline FV intake in children  Cravener et al. (2015)31; USA  N = 24; 3–5 y  Child without pre-existing medical conditions (including food allergies) Child intake of < 2 servings of vegetables per day Child at risk for obesity on basis of family history, defined as having at least 1 parent with a BMI ≥ 25 kg/m2  Home visits and printed materials (food packages with parent-instructions and child-targeted education sessions); 4 wk; length of intervention sessions unclear; data were collected 4 wk from baseline; training of interventionists unclear  Vegetable intake as the difference between pre- and post-weights (in grams) of the foods provided, child-rated liking of 6 vegetables; qualitative postintervention feedback; child sex, age, SES, vegetable intake at baseline, BMI z-score, SES (paternal and maternal education level), maternal BMI, paternal BMI, and ethnicity  Fildes et al. (2014)33; England and Wales  N = 98; 3–4 y  Child’s age (3–4 yr) Child enrolled in GEMINI cohort  Mail-based and web-based; 14 d of taste exposure followed by a sticker reward vs control (no intervention); length of intervention sessions unclear; data were collected 14 d from baseline; N/A (instructions delivered via mail and online)  Number of pieces eaten measured the child’s intake of the target vegetable; children’s liking of the target variable measured using parent report on 9-point scale (“dislikes a lot” to “likes a lot”); quantitative postintervention evaluation questionnaire; child sex, age, SES (maternal education), and maternal BMI  Horton et al. (2013)23; Mexico  N = 361; 7–13 y  Child’s age (7–13 y) Child lives at home Child not on a medically prescribed diet Mother’s age (≥ 18 y) Mother married or living with partner Mother Spanish speaking Family living in Imperial County with no plans to move  Home visits (family intervention sessions) and telephone-based (support calls); 14 wk; home visits delivered weekly for 8 wk followed by 6 wk of weekly alternating home visits and telephone support calls (length of sessions unclear), ie, total intervention length was 16.5 h or 990 min; data were collected 14 wk from baseline; trained community members  Daily FV intake using 2 questions from the National Cancer Institute’s Food Attitudes and Behavior survey and child self-reports of FV variety; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, percentage of parents married, SES (maternal education level, median household size, percentage on food assistance, percentage that own their home), and maternal race  Knowlden et al. (2015)34; USAa Knowlden & Sharma (2016)21; USA (12-mo FU)a  N = 57; 4–6 y N = 44; 4–6 y  Child’s age (4–6 y) Parent English speaking Family internet and telephone access Mother not pregnant Child without disability Child without a medical condition associated with weight gain or prescribed weight management medication Child not enrolled in a weight-management program  Web-based; 4 wk; sessions of 20–30 min/wk; 5 educational sessions plus a booster; 1–15-min audio-visual presentation, interactive worksheet, and a discussion board post; data were collected 8 wk and 1 y from baseline; N/A (intervention delivered via a website)  Child’s FV consumption (dietary recall, measured in cups), and fruit availability; process evaluation data collected via telephone counseling and postintervention evaluation surveys (intervention fidelity, dose delivered, dose received, reach, recruitment, and potential cross-contamination between the groups); child sex, age, FV intake at baseline, race, SES (maternal marital status, maternal employment status); and maternal race  McGowan et al. (2013)35; UK  N = 126; 2–6 y  Child’s age (2–6 y) Child without known medical or psychological condition affecting diet Parent English speaking  Home visits and printed materials; 8-wk; sessions of 1 h/wk; researcher worked through an intervention booklet with the parent; data were collected 8 wk from baseline; researchers received training prior to the intervention  Child’s daily FV intake via parent self-report (“How many servings of fruit [vegetables] does your child typically eat?”; 7-point scale from “less than 1 per day” to “5 per day”); postintervention interview covering intervention acceptability; child sex, age, FV intake at baseline, ethnicity, parent age, and SES (parent education and parent living status)  Remington et al. (2012)36; UK  N = 173; 3–4 y  Child’s age (3–4 y) Child attending a selected nursery school  Home visits; 12 d of daily taste exposure to a target vegetable followed by praise vs a sticker for tasting vs control group; length of intervention sessions unclear; data were collected 4 wk and 12 wk from baseline; trained researchers  Child’s liking of the target vegetable via parent report using a faces scale and intake of the target vegetable (in grams) using a digital scale; postintervention qualitative feedback; child age, sex, vegetable intake, or vegetable liking at baseline, parent age, parent ethnicity, and SES (parent home ownership, parent education)  Schwinn et al. (2014)37; USA  N = 67; girls, 10–12 y  Girls Child’s age (10–12 y) Girls and mothers living in publicly subsidized housing  Web-based (website); 3 wk; sessions of 25 min/wk; data were collected 3 wk and 5 mo from baseline; N/A (intervention delivered via an online platform)  Child’s score on the Youth and Adolescent FFQ via parent-self-report; no measure(s) of feasibility and/or acceptability reported; child age, sex, FV intake, ethnicity, maternal age, maternal FV intake at baseline, and SES (maternal employment, whether child qualifies for reduced-price or free school lunch)  Tabak et al. (2012)38; USA  N = 43; 2–5 y  Child’s age (2–5 y) Family living in current residence for next 6 mo  Mail- and telephone-based; 4 mo: 4 tailored newsletters and 2 motivational phone calls delivered every 4 wk; mean length of telephone sessions was 34 min; data were collected 4 mo from baseline; registered dietitian trained in motivational interviewing techniques  Child’s score on the Block Kids FFQ via parent self-report; quantitative postintervention evaluation questionnaire, postintervention qualitative feedback and process evaluation data (telephone session duration in minutes); child age, sex, parent age, parent sex, parent BMI, ethnicity, and SES (parent income)  Thompson et al. (2015)39; USA  N = 387; 9–11 y  Child in 4th or 5th grade (age 9–11 y Family English speaking Computer and high-speed internet access Parent willing to participate in telephone sessions  Online (video game) and web-based (website and electronic newsletters); 10 sessions delivered over 3 mo (length of sessions unclear) across 4 groups—action, coping, action + coping, and control; data were collected 3 mo and 6 mo from baseline; trained staff conducted dietary recall  Child’s FV intake via 3 unannounced 24-h dietary recalls (2 d during week, 1 d on weekend) conducted over the telephone by trained staff and 3 d of FV intake at each data collection period were averaged; process evaluation data (participation rates); parent satisfaction scale; child sex, FV intake at baseline, ethnicity, parent age, parent ethnicity, and SES (parent education level)  Tomayako et al. (2016)40; USA  N = 150; 2–5 y  Family of American Indian background; child’s age (2–5 y) Child lived with at least 1 primary caregiver (eg, mother, father, grandmother, aunt) in a home setting Child free of any major physical or behavioral disorders  Family-based randomized trial of a healthy lifestyle toolkit delivered via 2 formats: in-home mentoring via 12 monthly home visits or by mail. Each lesson addressed 1 of 4 target areas: (1) eat more FV, (2) consume less soda and added sugar, (3) become more active, and (4) watch less TV; 12-mo intervention period; home visits of 60 min duration; data were collected 12 mo from baseline; interventionists were tribal members or individuals who had longstanding employment within the community and were trained to administer the intervention  Daily servings of FV, sugar-sweetened drinks, and candy/junk food using Nutrition Data System for Research software 2010; focus group testing covering intervention acceptability; child age, sex, FV intake at baseline, BMI percentile, BMI z-score, ethnicity, caregiver age, caregiver sex, caregiver ethnicity, adult BMI, caregiver FV intake, and SES (caregiver education level)  Wardle et al. (2003)41; Australia  N = 156; 2–6 y  Child’s age (2–6 y)  Home visits; 2 wk of daily taste exposure to target vegetable vs nutrition education vs no intervention vs a postintervention taste test; session length unclear; data were collected 2 wk from baseline; training of interventionists unclear  Child’s liking of the 6 test vegetables measured using a 3-point faces scale, child’s consumption of a target vegetable measured (in grams) using a digital scale; semistructured postintervention interview covering intervention acceptability; child sex, age, and vegetable intake at baseline  Wyse et al. (2012)17; Australiaa Wolfenden et al. (2014)22; Australia (12- and 18-mo FU)a  N = 394; 3–5 y N = 164; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk; parent must have some responsibility for providing meals and snacks to child; parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2–6-mo, 12 mo, and 18 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention acceptability questionnaire; child age, sex, SES (decile of disadvantage classification associated with child’s preschool postal code), and children’s FV intake at baseline Child’s score on the FV Subscale of the Children’s Dietary Questionnaire; process evaluation data (number of intervention calls completed, delivery of key topics, delivery of the intervention as per protocol); child age, sex, ethnicity, FV intake at baseline, parent age, parent sex, SES (household income, university education), and parent FV intake)  Haire-Joshu et al. (2008)16; USA  N = 1306; 2–5 y  Child’s age (2–5 y) Parent’s age (20–59 y) Child and parent living in rural, southeast Missouri  Mail-based (1 tailored newsletter); 4 home visits of 60 min duration (frequency unclear) and 4 sing-a-long story books and audio cassettes (delivered at each home visit); mean length of data were collected 7 mo from baseline; parent educators received 4 h of PAT training on nutrition and material content  Intake frequency measured using the Saint Louis University for Kids FFQ, child-feeding practices, parent modeling of FV intake, nutrition knowledge, FV availability in the home; quantitative postintervention evaluation questionnaire; child sex, age, FV intake at baseline, parent age, parent sex, and SES (parent education level)  Nonrandomized control trials  Holley et al. (2015)24; UK  N = 115; 2–4 y  Child’s age (2–4 y)  Home-based (parent-led exposure) and community-based (assessments held at preschool); 14-d intervention across 5 groups: repeated exposure (RE), modeling plus repeated exposure (M + RE), rewards plus repeated exposure (R + RE), modeling, rewards, and RE (condition 4) vs a no-treatment control group; session length unclear; data were collected 2 wk from baseline; unclear if interventionists received training  Intake (in grams) and liking (measured using a 3-point smiley-face scale) of the target vegetable; no measure(s) of feasibility and/or acceptability reported; child age, sex, vegetable intake at baseline, BMI z-score, parent age, and parent sex  Pre–post studies  Dulin Keita et al. (2014)32; USA  N = 39 (data completed at baseline and at FU; 2–5 y  Child’s age (3–5 y) Child’s age/sex-specific BMI is ≥ 50th percentile Parent’s age (≥ 18 y) Parent lives with child at least 75% of the time Parent can speak and read English Parent is knowledgeable about child’s diet and physical activity  Mail- and telephone-based; 4 mo; 4 tailored mailouts, 3 motivational telephone calls, activity video (session length unclear); data were collected 4 mo from baseline; lay counselors received 12 h of motivational interviewing training from MINT-qualified trainer  Child’s score on the National Cancer Institute’s FV all-day screener tool,; process evaluation data (number of counselor reported calls completed), quantitative postintervention evaluation questionnaire; N/A  Spurrier et al. (2016)19; Australia  N = 24 (22 families); 4–12 y  Child’s age (4–12 y) Child’s age/sex-specific BMI indicative of overweight or obesity according to IOTF definitions Child living in metropolitan Adelaide, South Australia Child not diagnosed with a medical condition affecting weight or growth or not enrolled in a weight-management program  Home-based (parent-led) education program; 3 home visits and 2 FU telephone calls were offered to each family (session length unclear); data were collected approximately 6 mo from baseline (21–44 wk); researchers had backgrounds in nutrition, occupational therapy, or human movement and received 3 h of training and education prior to intervention  Score on the FV subscale of the Children’s Dietary Questionnaire (no measure[s] of feasibility and/or acceptability reported); N/A  Wyse et al. (2011)42; Australia  N = 34; 3–5 y  Child’s age (3–5 y) Child attending a participating preschool Child did not have a condition requiring specialized dietary information or advice Parent resided with child for ≥ 4 d/wk Parent must have some responsibility for providing meals and snacks to child Parent must be able to understand spoken and written English  Telephone-based (CATI); 4 wk; sessions of 30 min/wk; data were collected 2 mo and 6 mo from baseline; 2 d of training in script delivery, nutrition, and parenting by dietitian, psychologist, or health-promotion practitioners plus 10 h of delivery practice  Score on the FV Subscale of the Children’s Dietary Questionnaire and Household Food Expenditure Survey, noncore subscale of the Children’s Dietary Questionnaire; process evaluation data (participation rate, average call duration, average days elapsed between calls, and average number of call attempts) and quantitative postintervention evaluation questionnaire; N/A  Abbreviations: BMI, body mass index; CATI, computer-assisted telephone interviewing; FFQ, food frequency questionnaire; FV, fruit(s) and vegetable(s); FU, follow-up (time from baseline to final point of data collection); IOTF, International Obesity Task Force; MINT, Motivational Interviewing Network of Trainers; N/A, not available; PAT, Parents as Teachers; RCT, randomized control trial; SES, socioeconomic status. a Study had a separate follow-up publication. Follow-up data was treated as an extension of a single paper in this systematic review. Study characteristics Table 2 presents the study characteristics of both exposure and education interventions, respectively. Fourteen interventions (78%) were RCTs, 1 (6%) was a nonrandomized control trial (in which groups were systematically assigned by the primary investigator), and 3 (17%) were pre–post studies. Interventions were most commonly conducted in the United States, followed by the United Kingdom, Australia, and Mexico. The mean sample size was 229 participants (range, 24–1306; SD = 305). The mean age of children targeted by interventions was 5.3 years (SD = 2.7). Parent age was reported by 70% of studies, with a mean of 36.3 years (SD = 4.7). The percentage of female caregivers was consistently higher across studies (≥ 85%) compared with the percentage of male caregivers (≤ 1%). Interventions included home visits (10 of 18; 56%), telephone sessions (6 of 18; 33%), written materials (12 of 18; 67%), and online delivery methods (eg, website, electronic newsletters) (5 of 18; 28%) (Table 4).16,17,19,21–24,30–42 Nutrition education programs were the most common type of intervention (12/18, 67%). These interventions typically focused on increasing knowledge about the nutritional guidelines (eg, the recommended daily servings of FVs) and behavioral strategies for achieving the guidelines via a healthy home food environment (eg, parent providing of FVs, role modeling of FV intake, and adequate FV availability) (Table 3). Six studies (33%) used a repeated taste exposure paradigm to examine whether exposure, with or without positive reinforcement (eg, sticker reward or verbal praise), increased children’s intake of a target vegetable.24,30,31,33,36,41 Although only 2 studies met all of the applicable quality criteria,32,42 the majority of studies met most of the quality requirements (Table 3), which included a clear presentation of information related to study aims, hypotheses, and participant characteristics (16/18, 89%), a clear description of the main study findings (15/18, 83%),43,44 and reporting of an acceptable attrition rate (≤ 30%) (15/18, 83%) (Table 3).44 All studies (100%, 18/18) used valid and reliable outcome measures. Studies were less likely to meet the quality requirements for reporting of sample power calculations, with calculations not reported in 12 of the 18 studies (67%). The descriptions of randomization methods, blinding of participants, and blinding of assessors were also unclear or unavailable in 60% (9/15), 93% (14/15), and 100% (15/15) of applicable studies, respectively. In the majority of RCTs, the distribution of important confounders appeared balanced after randomization, with 3 studies (20%) providing insufficient detail to adequately determine the risk of residual confounding (Table 3). Two studies met the criteria for potential outcome bias.19,33 To address this issue, a sensitivity analysis that excluded biased studies in the meta-analysis was conducted. This procedure did not alter the significance of any meta-analytic outcome, suggesting that the inclusion of these studies in the systematic review was appropriate. Inter-rater reliability There was high inter-rater reliability between the 2 authors for both searches (2 disagreements out of 1145 abstracts for the initial search [κ = 0.94] and 2 unresolved disagreements out of 69 abstracts for the updated search [κ = 0.74], both of which required consultation with a third author). Research question 1: Are parent-targeted home-based interventions aimed at increasing FV intake in children feasible and acceptable to parents? Thirteen studies (72%) reported process evaluation data and/or quantitative or qualitative data on intervention acceptability and participant satisfaction (Table 2). Of these, 100% of studies reported evidence of intervention feasibility, including the completion of sessions within the recommended time frame,17,34,38,42 the completion of all intervention sessions by over 70% of participants,17,32,42 or a low attrition rate (< 30%, Table 3). Several feasibility issues were identified for both types of interventions. One nutrition education intervention found that the rate of participants who received telephone calls (16%) was lower than the rate of those who received mailed intervention materials (42%).32 In an online nutrition education intervention, lower participant engagement was reported for parents and caregivers (28% of parents reported reading more than 60% of intervention session materials) than for children (91%).39 For exposure interventions, 1 home-visiting study found that 41% of participants did not complete the minimum number of exposure sessions.41 Another online and mail-based exposure intervention reported lower rates of participant engagement with an online instruction video (17%) than with a written instruction leaflet (94%).33 Two taste exposure studies reported quantitative data on the overall acceptability of interventions. Among these, 80% of participants agreed that the taste exposure procedure was helpful (eg, increased children’s willingness to try vegetables),33 85% felt that the advice given was useful,41 and 70% agreed that the intervention had a long-lasting effect on their child’s liking of the target vegetable.41 In addition, 85% of participants agreed that they would use the taste exposure procedure again,33 and 65% reported that they had already used the strategy for a different vegetable.41 Nutrition education interventions were also rated favorably, with 83% to 97% of participants agreeing that participation was helpful (eg, they could set reasonable limits for their children after participating in the program)16 or worthwhile,42 92% to 100% rating the intervention materials as easy to read32 or relevant,42 and 87% reporting that they were still using the intervention materials provided.32 For both taste exposure and nutrition education interventions, participants identified time (eg, the time needed to undertake the intervention) as the most common barrier to acceptability.31,38,41 Research question 2: Are parent-targeted home-based taste exposure and nutrition education interventions effective at increasing FV intake in children? Overall, positive outcomes were more frequently reported with taste exposure interventions than with nutrition education interventions, with all 6 taste exposure interventions demonstrating a significant increase in children’s vegetable intake (Table 3). In addition, 4 of the 6 taste exposure interventions reported an increase in child-reported liking of a target vegetable (Table 3). Of the 12 nutrition education interventions reviewed, 3 (25%) yielded no effect on children’s FV intake.19,23,38 However, 2 of the 3 nutrition education interventions that failed to find an intervention effect had a relatively small sample size (< 45),19,38 suggesting these interventions may not have had adequate power to detect a significant intervention effect. Research question 3: Do parent-targeted home-based taste exposure and nutrition education interventions result in significantly greater FV intake in children? A meta-analysis was performed to statistically evaluate the effect of taste exposure and nutrition education interventions compared with no intervention on FV intake in children. Nutrition education interventions (n = 4) had a significant but small positive effect on child fruit intake (Hedges’ g = 0.112; SE = 0.051; 95%CI, 0.012–0.212; P = 0.028), with nonsignificant heterogeneity (I2= 5.945; P = 0.363) (Figure 2A16,35,37,39). There was no significant effect of education interventions on child vegetable intake (n = 6) (Figure 2B16,23,35,37–39) (Hedges’ g = 0.125; SE = 0.082; 95%CI, −0.035 to 0.285; P = 0.125), but there was significant heterogeneity (I2= 55.84; P = 0.045). Figure 2 View largeDownload slide (A) Comparison of the effect of a parent-targeted home-based nutrition education intervention vs no intervention on (A) fruit intake (no. of daily servings) and (B) vegetable intake (no. of daily servings) in children. Analysis A was performed using 4 randomized controlled studies.16,35,37,39 Analysis B was performed using 6 randomized controlled studies.16,23,35,37–39 Figure 2 View largeDownload slide (A) Comparison of the effect of a parent-targeted home-based nutrition education intervention vs no intervention on (A) fruit intake (no. of daily servings) and (B) vegetable intake (no. of daily servings) in children. Analysis A was performed using 4 randomized controlled studies.16,35,37,39 Analysis B was performed using 6 randomized controlled studies.16,23,35,37–39 Figure 3 View largeDownload slide Comparison of the effect of a parent-targeted home-based taste exposure intervention vs no intervention on vegetable intake (in grams) in children. This analysis was performed using 6 randomized controlled studies.24,30,31,33,36,41 Figure 3 View largeDownload slide Comparison of the effect of a parent-targeted home-based taste exposure intervention vs no intervention on vegetable intake (in grams) in children. This analysis was performed using 6 randomized controlled studies.24,30,31,33,36,41 For taste exposure interventions (Figure 324,30,31,33,36,41) (n = 6), there was a moderately large and significant effect of intervention on child vegetable intake (Hedges’ g = 0.438; I2 =0.000; SE = 0.064; 95%CI, 0.312–0.564; P < 0.001). This result remained significant when the study identified as having potential outcome bias was removed from the analysis33 (Hedges’ g = 0.372; I2 =0.000; SE = 0.086; 95%CI, 0.204–0.540; P < 0.001). Sensitivity analyses revealed that using the stipulated range of plausible correlations for pre–post correlation or subgroup variation imputations did not alter the significance of any meta-analytic outcome. Research question 4: What characteristics are common among parent-targeted home-based interventions that achieve significant increases in FV intake in children? Overall, online-delivered interventions appeared to be most effective, with 100% of online-delivered interventions reporting a significant increase in FV intake in children compared with 80% of home-visiting interventions and 50% of telephone-delivered interventions. Interventions providing daily sessions had the highest percentage of reported significant intervention effects (100%) compared with interventions delivered weekly (83%) or monthly (50%). The number of intervention sessions did not appear to strongly influence the effectiveness of interventions. Studies that included more than 14 sessions had the highest percentage of reported significant intervention effects, but this statistic was based on 1 intervention (n = 1 of 1). The percentage of reported significant intervention effects was slightly lower among studies that provided 11 to 14 sessions (83%, n = 5 of 6), 6 to 10 sessions (75%, n = 3 of 4), and 1 to 5 sessions (86%, n = 6 of 7) (Table 4). Table 4 Summary of outcomes from parent-targeted home-based interventions to increase fruit and/or vegetable intake in children Characteristics of intervention  Study outcomes   Increased FV intakea  Increased FV liking  No effect on FV intake  Type of intervention   Taste exposure24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies   Nutrition education16,17,19,23,32,34,35,37–40,42  9 studies16,17,32,34,35,37,39,40,42  No studies  3 studies19,23,38  Recruitment strategy   Media or advertising16,23,30,31,34,35,37,39  7 studies16,30,31,34,35,37,39  No studies  1 study23   Written information, mailed out16,17,23,33,36,38,39,42  5 studies17,33,36,39,42  No studies  2 studies23,38   Researcher contact (face-to-face or telephone)16,17,19,24,32,35,41,42  7 studies16,17,24,32,35,41,42  No studies  1 study19   Online, eg, social media31,37–39  3 studies31,37,39  No studies  1 study38  Type of incentives   Financial16,30,32,34,35,37,38  6 studies16,30,32,34,35,37  No studies  1 study38   None17,19,23,24,31,33,36,39–42  9 studies17,24,31,33,36,39–42  4 studies24,33,36,41  2 studies19,23  Mode of delivery   Written information or resources16,17,23,24,30–33,35,38,41,42  10 studies16,17,24,30–33,35,41,42  3 studies24,33,41  2 studies23,38   Telephone calls17,19,23,32,38,42  3 studies17,32,42  No studies  3 studies19,23,38   Home visits by researchers16,19,23,30–32,35,36,40,41  8 studies16,30–32,35,36,40,41  2 studies36,41  2 studies19,23   Online16,33,34,37,39  5 studies16,33,34,37,39  1 study33  No studies  Frequency of sessions   Daily24,30,33,36,41  4 studies24,30,36,41  2 studies36,41  No studies   Weekly17,23,31,34,37,42  5 studies17,31,34,37,42  No studies  1 study23   Monthly19,32,38,40  2 studies32,40  No studies  2 studies19,38   Variable/unclear16,35,39  3 studies16,35,39  No studies  No studies  No. of sessions   1–5 sessions17,19,31,34,35,37,42  6 studies17,31,34,35,37,42  No studies  1 study19   6–10 sessions16,32,38,39  3 studies16,32,39  No studies  1study38   11–14 sessions23,24,30,33,36,41  5 studies24,30,33,36,41  4 studies24,33,36,41  1study 23   > 14 sessions40  1 study40  No studies  No studies  Components of intervention   Goal setting17,19,23,32,34,35,37,38,42  6 studies17,32,34,35,37,42  No studies  3 studies19,23,38   Meal planning17,23,34,35,37–39,42  6 studies17,34,35,37,39,42  No studies  2 studies23,38   Nutrition education16,17,19,23,24,31,34,35,37–42  10 studies16,17,31,34,35,37,39–42  1 study24  3 studies19,23,38   Written materials16,17,19,30–35,37–42  13studies16,17,30–35,37,39–42  2 studies33,41  2 studies19,38   Rewards24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies  Provision of tailored information/feedback   Yes16,17,19,32,42  4 studies16,17,32,42  No studies  1 study19   No23,24,30,31,33–41  11 studies21,24,30,31,33,35–37,39–41  1 study33  2 studies23,38  Characteristics of intervention  Study outcomes   Increased FV intakea  Increased FV liking  No effect on FV intake  Type of intervention   Taste exposure24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies   Nutrition education16,17,19,23,32,34,35,37–40,42  9 studies16,17,32,34,35,37,39,40,42  No studies  3 studies19,23,38  Recruitment strategy   Media or advertising16,23,30,31,34,35,37,39  7 studies16,30,31,34,35,37,39  No studies  1 study23   Written information, mailed out16,17,23,33,36,38,39,42  5 studies17,33,36,39,42  No studies  2 studies23,38   Researcher contact (face-to-face or telephone)16,17,19,24,32,35,41,42  7 studies16,17,24,32,35,41,42  No studies  1 study19   Online, eg, social media31,37–39  3 studies31,37,39  No studies  1 study38  Type of incentives   Financial16,30,32,34,35,37,38  6 studies16,30,32,34,35,37  No studies  1 study38   None17,19,23,24,31,33,36,39–42  9 studies17,24,31,33,36,39–42  4 studies24,33,36,41  2 studies19,23  Mode of delivery   Written information or resources16,17,23,24,30–33,35,38,41,42  10 studies16,17,24,30–33,35,41,42  3 studies24,33,41  2 studies23,38   Telephone calls17,19,23,32,38,42  3 studies17,32,42  No studies  3 studies19,23,38   Home visits by researchers16,19,23,30–32,35,36,40,41  8 studies16,30–32,35,36,40,41  2 studies36,41  2 studies19,23   Online16,33,34,37,39  5 studies16,33,34,37,39  1 study33  No studies  Frequency of sessions   Daily24,30,33,36,41  4 studies24,30,36,41  2 studies36,41  No studies   Weekly17,23,31,34,37,42  5 studies17,31,34,37,42  No studies  1 study23   Monthly19,32,38,40  2 studies32,40  No studies  2 studies19,38   Variable/unclear16,35,39  3 studies16,35,39  No studies  No studies  No. of sessions   1–5 sessions17,19,31,34,35,37,42  6 studies17,31,34,35,37,42  No studies  1 study19   6–10 sessions16,32,38,39  3 studies16,32,39  No studies  1study38   11–14 sessions23,24,30,33,36,41  5 studies24,30,33,36,41  4 studies24,33,36,41  1study 23   > 14 sessions40  1 study40  No studies  No studies  Components of intervention   Goal setting17,19,23,32,34,35,37,38,42  6 studies17,32,34,35,37,42  No studies  3 studies19,23,38   Meal planning17,23,34,35,37–39,42  6 studies17,34,35,37,39,42  No studies  2 studies23,38   Nutrition education16,17,19,23,24,31,34,35,37–42  10 studies16,17,31,34,35,37,39–42  1 study24  3 studies19,23,38   Written materials16,17,19,30–35,37–42  13studies16,17,30–35,37,39–42  2 studies33,41  2 studies19,38   Rewards24,30,31,33,36,41  6 studies24,30,31,33,36,41  4 studies24,33,36,41  No studies  Provision of tailored information/feedback   Yes16,17,19,32,42  4 studies16,17,32,42  No studies  1 study19   No23,24,30,31,33–41  11 studies21,24,30,31,33,35–37,39–41  1 study33  2 studies23,38  a A significant increase in FV intake was set as P < 0.05. Five of 6 taste exposure interventions (83%) included an exposure plus a reward condition. Of these, 80% found that children who received exposure along with a sticker reward ate significantly more of the target vegetable from baseline to the end of the intervention compared with children who did not receive a sticker reward.24,30,31,33,36 When the type of reward was examined, a tangible reward such as a sticker resulted in a greater increase in target vegetable intake compared with a social reward, such as parent verbal praise (P = 0.001).36 Five studies (29%) provided participants with personalized feedback.16,17,19,32,42 This included written materials with the name of the parent and their child,19 the selection of intervention topics to suit the target child behaviors,16,32 tailored feedback about FV intake specific to the parent and/or child,16,17,42 and tailored goals.17,42 Four of the 5 studies that used personalized feedback (80%) found a significant intervention affect (Table 3). Thirteen of the 14 studies without personalized feedback also reported similar outcomes (Table 4). Whether studies provided a financial incentive for participation did not appear to influence the outcome of interventions: 86% of studies using incentives reported a significant increase in FV intake in children compared with 82% of studies that did not use incentives. DISCUSSION The majority of published parent-targeted home-based interventions have reported an improvement in children’s FV intake. However, this review and meta-analysis demonstrates that interventions may not be equally effective at increasing the intake of both fruit and vegetables. Specifically, taste exposure interventions, but not nutrition education interventions, resulted in a significant increase in children’s vegetable intake. Despite having no effect on vegetable intake, nutrition education interventions did increase children’s fruit intake significantly. Online and home-visiting interventions more frequently reported significant increases in children’s FV intake compared with telephone-based interventions. Furthermore, not all interventions were equally acceptable to parents, with parents identifying time as a major limitation to completing taste exposure activities and telephone-based nutrition education sessions. The current meta-analysis found a moderate to large increase in children’s vegetable intake in parent-targeted, home-based taste exposure interventions. This is an important finding, given that fewer children meet the national guidelines for recommended daily servings of vegetables than for fruit6 and most children do not consume the recommended variety of vegetables.50 Although previous studies have identified taste exposure as an effective intervention strategy for increasing children’s FV intake,51 a previous meta-analysis found no effect of a parent-delivered taste exposure intervention on children’s vegetable intake when compared with no intervention.9 This discrepancy may be explained by the fact that the previous meta-analysis included only 2 interventions, while the current analysis included 6, meaning the former may have lacked sufficient power to detect an intervention effect.9 The effects of nutrition education interventions on FV intake in children were mixed. The meta-analysis of 2290 participants from interventions comparing nutrition education with no intervention revealed a significant intervention effect for fruit intake but not vegetable intake. It is unclear why taste exposure interventions, but not nutrition education interventions, resulted in a significant increase in vegetable intake. Recent population data indicates that children aged 2 to 18 years are more likely to meet the recommended daily intake for fruit (68%) than for vegetables (5%).50 This may suggest that vegetable intake is more difficult to increase than fruit intake. It is well known that children have a biological predisposition to reject bitter-tasting foods, such as vegetables, to protect themselves against possible toxins.52 Children may also become reluctant to taste unfamiliar vegetables as a result of food neophobia or the fear of eating new foods,53 which is common between the ages of 2 and 6.54 Promoting substantial changes in children’s vegetable intake may therefore necessitate more intensive strategies than parent-targeted nutrition education, particularly for bitter-tasting and unfamiliar vegetables. Overall, the pattern of results supports the findings obtained with other commonly used FV interventions, including school- and family-based interventions. The school environment is one of the most common settings for implementing FV interventions in children because of its capacity to target large numbers of children.11 A meta-analysis of 27 school-based interventions involving 26 361 children found that children’s intake of fruit increased by a useful amount (0.24 servings), whereas intake of vegetables did not (0.07 servings).10 The majority of these interventions, however, involved the delivery of nutrition education or the provision of free or subsidized FVs and did not include taste exposure. Given the findings of the current review, it is plausible that the absence of taste exposure may have contributed to the negligible effect of school-based interventions on children’s vegetable intake.10 Supporting this hypothesis are the findings of Skouteris et al.,55 who reported that children whose parents received taste exposure training as part of a family-based nutrition education intervention ate significantly more vegetables at the end of the intervention compared with controls. Taken together, these findings suggest that, irrespective of the context (eg, school, community, or home), parent involvement and taste exposure are likely to be important strategies to include in any intervention aimed at promoting meaningful increases in children’s FV intake. Feasibility and acceptability are important issues when considering the efficacy of an intervention. The literature review showed that parent-targeted interventions conducted in the home setting appeared feasible to deliver, with the majority of studies reporting an attrition rate of less than 30% (Table 3). However, participant burden is commonly identified as a key factor for nonparticipation in research studies.56 This was partly confirmed by the finding that parents identified time as an important limitation for both taste exposure and telephone-delivered nutrition education interventions, including the time to undertake taste exposure activities and to complete daily record forms,31,41 multiple (n = 3) assessment sessions,33 and telephone sessions.38 To enhance acceptability, future interventions may benefit from using alternative strategies to daily written record forms and structured taste exposure procedures. For example, studies have successfully trialed the use of mobile applications in the treatment of clinical anxiety to enhance participant self-monitoring and recording of treatment tasks.57 New evidence also suggests that modified versions of the taste exposure method may be effective at increasing children’s FV intake. A recent RCT of a nutrition education intervention involving 394 parents found that daily parent providing of FVs to children, regardless of the outcome, mediated the long-term positive effect (up to 12 months) of the intervention on children’s FV intake.12,22 In this approach, parents are broadly encouraged to increase the number of occasions when vegetables are offered to children each day. This is different from traditional taste exposure methods in which parents are instructed to offer a specific target vegetable on a set number of days, allowing for less flexibility. Given that time was identified as a common barrier to completing interventions, identifying effective but flexible parent-targeted intervention strategies is an important goal for future parent-targeted FV interventions. Contrary to both theory58 and earlier literature,59 evidence that longer interventions (11–14 sessions), compared with briefer interventions (1–5 sessions), were more likely to report significant increases in children’s FV intake at end of treatment was not strong. Although these findings suggest that both longer and shorter interventions may be equally effective, the majority of short-term intervention studies included in this review did not include long-term follow-up data.21,22 Where short-term interventions (eg, 4–6 intervention sessions) did include long-term follow-up data, increases in children’s FV intake were observed only up to 12 months of follow-up.21,22 Therefore, ongoing contact may still be important in maintaining increases in children’s FV intake over the long term.21,58 This is consistent with theories of health behavior change that recommend multiple points of contact to maintain behavior change over time.58 Moreover, a systematic review of behavioral therapy trials found that booster sessions, used in combination with standard treatment, significantly aided behavior change in over 50% of studies.59 This may be attributable to the reinforcement of skills learned during treatment, a reduction in participant concerns about treatment termination, and the ability to maintain participant accountability in implementing skills acquired during treatment.59 Although preliminary, the findings of this review have important implications for health professionals interested in delivering parent-targeted home-based interventions as well as for parents seeking effective strategies to increase children’s FV intake. Similarly to previous systematic reviews, the present findings suggest that taste exposure is more effective at increasing children’s FV intake when combined with a tangible reward, at least in the short term.9,60 However, the usefulness of rewarding children for eating FVs remains controversial and should be considered. Taste exposure interventions may be susceptible to the overjustification effect, which argues that repeatedly offering a reward for eating a target vegetable diminishes a child’s liking and acceptance of that vegetable over time.61,62 In this context, rewards may act as a signal to children that tasting a target or unfamiliar vegetable is unpleasant because it requires immediate compensation.61 However, not all rewards may negatively affect children’s acceptance of a target FV. For example, some research has shown that unexpected rewards may actually boost motivation.63 Future research is therefore needed to better ascertain whether using a short-term tangible reward (after approximately 14 taste occasions) increases or decreases children’s long-term (> 12 months) willingness to taste a target vegetable and to identify the most effective method for delivering a reward (eg, expected or unexpected).63 LIMITATIONS AND FUTURE DIRECTIONS Although the current meta-analysis enabled a rigorous examination of published interventions, the methodology of meta-analysis is vulnerable to bias. Specifically, interventions with nonsignificant findings may be less frequently published in the literature,64 resulting in a nonrandom sample of included studies. A quantitative risk assessment could not be performed in the current meta-analysis because of the small number of included studies. There was also some unexplained heterogeneity in effect size. Meta-regression is a technique that can use study-level variables to explain heterogeneity within a random-effects model. However, the lack of consistency in potential explanatory variables across studies in the current meta-analysis precluded a meta-regression from being conducted. For example, 1 study had both an exposure group and a nutrition education group,41 numerous studies only assessed vegetable and not fruit intake, and multiple studies had several intervention conditions. In addition, only 2 of the 18 interventions included in this review targeted non-white, non-English-speaking families.23,40 This is an important limitation, given that children’s dietary behaviors are influenced in part by socioeconomic status and culture.65 It is therefore unclear whether the intervention effects observed in this systematic review can be generalized across different cultures, ethnicities, and economic backgrounds. Few studies included a long-term follow-up assessment, making it difficult to ascertain whether interventions resulted in sustained increases in children’s FV intake. The risk of interventions reporting false positive or false negative results was identified as low, with most studies adequately balancing or controlling for children’s age, sex, and baseline FV intake (Table 3). However, parent FV intake, parent body mass index, and parent educational status were less frequently reported, despite being associated with children’s FV intake,12,45–47 and indices of socioeconomic status varied across studies (Table 2). Therefore future intervention studies should also control for these factors to ensure rigorous examination of the effect of parent-targeted home-based interventions on children’s FV intake. There is currently no gold standard for the assessment of children’s dietary intake, with methods varying considerably across interventions.66 In the studies included in the present review, assessment methods ranged from a dietary intake questionnaire to a repeated 24-hour dietary recall delivered by a trained health professional. Different dietary intake assessment methods are associated with different forms of bias, including the over- or underestimation of child dietary intake.66 Therefore, future parent-targeted home-based nutrition interventions need to include the same assessment methods to facilitate the comparison of FV intake data across studies.66 In all of the interventions evaluated here, data on children’s FV intake were collected using parent self-report. Although this method is demonstrated to be reliable and valid in children under the age of 8, the evidence is unclear in older children,66 with parents often over-reporting children's school and out-of-home food intake.67 Including children over the age of 10 in the assessment of children’s FV intake may help to increase the validity and reliability of outcome data from future dietary interventions.66 Given that parent-targeted, home-based interventions have the potential to significantly increase children’s FV intake, it is important that future research identify ways to increase the number of parents being offered support, for example, by reducing delivery costs.68 Unfortunately, the equivocal findings of this review regarding the effect of using financial incentives to increase participation in FV interventions in children provide little direction for health professionals interested in delivering parent-targeted, home-based interventions. Additional RCTs are therefore needed to examine the impact of parent-targeted home-based interventions, with and without financial incentives, on FV intake in children. In addition, none of the telephone-based intervention studies included in this review utilized mobile phone applications (apps) or text messaging. Given the increasing use of phone-based technologies in interventions designed to effect change in health behaviors,69–71 incorporation of mobile phone apps in future research is likely to contribute to the development of more affordable parent-targeted interventions to increase children’s FV intake on a mass scale. CONCLUSION This review demonstrates that taste exposure is an effective parent-targeted, home-based intervention strategy for promoting short-term increases (up to 12 months) in children’s vegetable intake, which remains below the national recommendations across much of the developed world. Interventions delivered in the home setting, particularly via online modalities or home visits, also have strong merit, resulting in a significant increase in children’s fruit intake (lasting up to 12 months). To best inform parents and health professionals, an evaluation of the long-term impact (> 12 months) of these interventions is needed to identify effective strategies for achieving sustained increases in children’s FV intake. Despite this limitation, the majority of the parent-targeted home-based FV interventions reviewed here appeared feasible to deliver, as evidenced by the low attrition rates. Taken together, these findings suggest that such interventions are likely to have substantial public health appeal, offering an effective, practical, and geographically far-reaching approach for promoting increases in children’s FV intake. This is particularly important for parents and health professionals in rural and remote areas, for whom access to effective health promotion interventions is often very limited. Acknowledgments Author contributions. L.M.T., C.E.W., and J.C. designed the research; L.M.T., M.S., and A.M.G. contributed to the systematic search; L.M.T., V.F.Q., and D.C. performed the statistical analysis; L.M.T. and A.M.G. were responsible for preparing the manuscript for publication; L.M.T. had primary responsibility for the final content; and C.E.W., J.C., and R.J.C. critically reviewed the intellectual content of the draft manuscript prior to submission. All authors read and approved the final manuscript. Funding/support This project was supported by the Kids Cancer Alliance, which is supported by a Cancer Institute NSW grant (no. 11/TRC/1–03) awarded to C.E.W., J.C., and R.J.C. The Behavioural Sciences Unit at the Kids Cancer Center, Sydney Children’s Hospital, Sydney, Australia, is proudly supported by the Kids with Cancer Foundation. This research was also funded by the Cancer Council NSW (program grant no. PG16–02) with the support of the Estate of the Late Harry McPaul. The funders had no role in the design of the study; in the collection, management, analysis, or interpretation of the data; in the preparation and revision of the manuscript; or in the publication decisions related to this manuscript. Authors received complete access to the data pertaining to this publication. Declaration of interest The authors have no relevant interests to declare. Supporting Information Appendix S1Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement Appendix S2Search strategies References 1 Vioque J, Weinbrenner T, Castelló A et al.   Intake of fruits and vegetables in relation to 10‐year weight gain among Spanish adults. Obesity . 2008; 16: 664– 670. Google Scholar CrossRef Search ADS PubMed  2 Kaikkonen JE, Mikkilä V, Raitakari OT. Role of childhood food patterns on adult cardiovascular disease risk. Curr Atheroscler Rep.  2014; 16: 1– 15. Google Scholar CrossRef Search ADS   3 Ruxton CHS, Derbyshire E. Diet adequacy in UK schoolchildren. Nutr Food Sci . 2011; 41: 20– 33. http://dx.doi.org/10.1108/00346651111102865 Google Scholar CrossRef Search ADS   4 Fox MK, Gearan E, Cannon J, et al.   Usual food intakes of 2- and 3-year old US children are not consistent with dietary guidelines. BMC Nutr.  2016; 2: 67. doi:10.1186/s40795-016-0106-2. Google Scholar CrossRef Search ADS   5 Wyse R, Campbell E, Nathan N, et al.   Associations between characteristics of the home food environment and fruit and vegetable intake in preschool children: a cross-sectional study. BMC Public Health . 2011; 11: 938. 10.1186/1471-2458-11-938 Google Scholar CrossRef Search ADS PubMed  6 Australian Bureau of Statistics. National Health Survey: first results, 2014-15. http://www.abs.gov.au/ausstats/abs@.nsf/mf/4364.0.55.001. Published August 12, 2015. Accessed January 5, 2017. 7 Nicklaus S, Boggio V, Chabanet C, et al.   A prospective study of food preferences in childhood. Food Qual Prefer . 2004; 15: 805– 818. http://dx.doi.org/10.1016/j.foodqual.2004.02.010 Google Scholar CrossRef Search ADS   8 Mikkilä V, Räsänen L, Raitakari OT, et al.   Consistent dietary patterns identified from childhood to adulthood: the Cardiovascular Risk in Young Finns Study. Br J Nutr.  2005; 93: 923– 931. Google Scholar CrossRef Search ADS PubMed  9 Wolfenden L, Wyse RJ, Britton BI, et al.   Interventions for increasing fruit and vegetable consumption in children aged 5 years and under. Cochrane Database Syst Rev . 2012;(11): CD008552. doi:10.1002/14651858.CD008552.pub2 10 Evans CE, Christian MS, Cleghorn CL, et al.   Systematic review and meta-analysis of school-based interventions to improve daily fruit and vegetable intake in children aged 5 to 12 y. Am J Clin Nutr . 2012; 96: 889– 901. http://dx.doi.org/10.3945/ajcn.111.030270 Google Scholar CrossRef Search ADS PubMed  11 Hingle MD, O'Connor TM, Dave JM, et al.   Parental involvement in interventions to improve child dietary intake: a systematic review. Prev Med.  2010; 51: 103– 111. http://dx.doi.org/10.1016/j.ypmed.2010.04.014 Google Scholar CrossRef Search ADS PubMed  12 Wyse R, Wolfenden L, Bisquera A. Characteristics of the home food environment that mediate immediate and sustained increases in child fruit and vegetable consumption: mediation analysis from the Healthy Habits cluster randomised controlled trial. Int J Behav Nutr Phys Act.  2015; 12: 118. doi:10.1186/s12966-015-0281-6. Google Scholar CrossRef Search ADS PubMed  13 Benton D. Role of parents in the determination of the food preferences of children and the development of obesity. Int J Obes Relat Metab Disord.  2004; 28: 858– 869. http://dx.doi.org/10.1038/sj.ijo.0802532 Google Scholar CrossRef Search ADS PubMed  14 Pearson N, Biddle SJ, Gorely T. Family correlates of fruit and vegetable consumption in children and adolescents: a systematic review. Public Health Nutr.  2009; 12: 267– 283. http://dx.doi.org/10.1017/S1368980008002589 Google Scholar CrossRef Search ADS PubMed  15 Knai C, Pomerleau J, Lock K, et al.   Getting children to eat more fruit and vegetables: a systematic review. Prev Med.  2006; 42: 85– 95. http://dx.doi.org/10.1016/j.ypmed.2005.11.012 Google Scholar CrossRef Search ADS PubMed  16 Haire-Joshu D, Elliott MB, Caito NM, et al.   High 5 for Kids: the impact of a home visiting program on fruit and vegetable intake of parents and their preschool children. Prev Med . 2008; 47: 77– 82. http://dx.doi.org/10.1016/j.ypmed.2008.03.016 Google Scholar CrossRef Search ADS PubMed  17 Wyse R, Wolfenden L, Campbell E, et al.   A cluster randomized controlled trial of a telephone-based parent intervention to increase preschoolers’ fruit and vegetable consumption. Am J Clin Nutr.  2012; 96: 102– 110. Google Scholar CrossRef Search ADS PubMed  18 Paineau DL, Beaufils F, Boulier A, et al.   Family dietary coaching to improve nutritional intakes and body weight control: a randomized controlled trial. Arch Pediatr Adolesc Med.  2008; 162: 34– 43. http://dx.doi.org/10.1001/archpediatrics.2007.2 Google Scholar CrossRef Search ADS PubMed  19 Spurrier NJ, Bell L, Wilson A, et al.   Minimal change in children's lifestyle behaviours and adiposity following a home-based obesity intervention: results from a pilot study. BMC Res Notes.  2016; 9: 26. doi: 10.1186/s13104-015-1796-2. Google Scholar CrossRef Search ADS PubMed  20 Cohen JF, Kraak VI, Choumenkovitch SF, et al.   The CHANGE study: a healthy-lifestyles intervention to improve rural children's diet quality. J Acad Nutr Diet . 2014; 114: 48– 53. http://dx.doi.org/10.1016/j.jand.2013.08.014 Google Scholar CrossRef Search ADS PubMed  21 Knowlden A, Sharma M. One-year efficacy testing of Enabling Mothers to Prevent Pediatric Obesity through Web-Based Education and Reciprocal Determinism (EMPOWER) randomized control trial. Health Educ Behav.  2016; 43: 94– 106. http://dx.doi.org/10.1177/1090198115596737 Google Scholar CrossRef Search ADS PubMed  22 Wolfenden L, Wyse R, Campbell E, et al.   Randomized controlled trial of a telephone-based intervention for child fruit and vegetable intake: long-term follow-up. Am J Clin Nutr.  2014; 99: 543– 550. http://dx.doi.org/10.3945/ajcn.113.071738 Google Scholar CrossRef Search ADS PubMed  23 Horton LA, Parada H, Slymen DJ, et al.   Targeting children's dietary behaviors in a family intervention:'Entre familia: reflejos de salud'. Salud Pública Méx . 2013; 55(suppl 3): 397– 405. Google Scholar PubMed  24 Holley CE, Haycraft E, Farrow C. ‘Why don’t you try it again?’A comparison of parent led, home based interventions aimed at increasing children's consumption of a disliked vegetable. Appetite . 2015; 87: 215– 222. Google Scholar CrossRef Search ADS PubMed  25 Moher D, Liberati A, Tetzlaff J, et al.   Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med.  2009; 151: 264– 269. http://dx.doi.org/10.7326/0003-4819-151-4-200908180-00135 Google Scholar CrossRef Search ADS PubMed  26 Kim SA, Moore LV, Galuska D, et al.   Vital signs: fruit and vegetable intake among children—United States, 2003–2010. MMWR Morb Mortal Wkly Rep . 2014; 63: 671– 676. Google Scholar PubMed  27 Skinner JD, Carruth BR, Bounds W, et al.   Children's food preferences: a longitudinal analysis. J Am Diet Assoc.  2002; 102: 1638– 1647. http://dx.doi.org/10.1016/S0002-8223(02)90349-4 Google Scholar CrossRef Search ADS PubMed  28 Delgado-Noguera M, Tort S, Martínez-Zapata MJ, et al.   Primary school interventions to promote fruit and vegetable consumption: a systematic review and meta-analysis. Prev Med.  2011; 53: 3– 9. Google Scholar CrossRef Search ADS PubMed  29 Treweek S, Lockhart P, Pitkethly M, et al.   Methods to improve recruitment to randomised controlled trials: Cochrane systematic review and meta-analysis. BMJ Open . 2013; 3. 10.1136/bmjopen-2012-002360 30 Corsini N, Slater A, Harrison A, et al.   Rewards can be used effectively with repeated exposure to increase liking of vegetables in 4–6-year-old children. Public Health Nutr.  2013; 16: 942– 951. http://dx.doi.org/10.1017/S1368980011002035 Google Scholar CrossRef Search ADS PubMed  31 Cravener TL, Schlechter H, Loeb KL, et al.   Feeding strategies derived from behavioral economics and psychology can increase vegetable intake in children as part of a home-based intervention: results of a pilot study. J Acad Nutr Diet.  2015; 115: 1798– 1807. http://dx.doi.org/10.1016/j.jand.2015.03.024 Google Scholar CrossRef Search ADS PubMed  32 Dulin Keita A, Risica PM, Drenner KL, et al.   Feasibility and acceptability of an early childhood obesity prevention intervention: results from the healthy homes, healthy families pilot study. J Obes . 2014; 2014:378501. doi:10.1155/2014/378501 33 Fildes A, van Jaarsveld CH, Wardle J, et al.   Parent-administered exposure to increase children's vegetable acceptance: a randomized controlled trial. J Acad Nutr Diet . 2014; 114: 881– 888. http://dx.doi.org/10.1016/j.jand.2013.07.040 Google Scholar CrossRef Search ADS PubMed  34 Knowlden AP, Sharma M, Cottrell RR, et al.   Impact evaluation of Enabling Mothers to Prevent Pediatric Obesity through Web-Based Education and Reciprocal Determinism (EMPOWER) randomized control trial. Health Educ Behav.  2015; 42: 171– 184. http://dx.doi.org/10.1177/1090198114547816 Google Scholar CrossRef Search ADS PubMed  35 McGowan L, Cooke LJ, Gardner B, et al.   Healthy feeding habits: efficacy results from a cluster-randomized, controlled exploratory trial of a novel, habit-based intervention with parents. Am J Clin Nutr.  2013; 98: 769– 777. http://dx.doi.org/10.3945/ajcn.112.052159 Google Scholar CrossRef Search ADS PubMed  36 Remington A, An E, Croker H, et al.   Increasing food acceptance in the home setting: a randomized controlled trial of parent-administered taste exposure with incentives. Am J Clin Nutr . 2012; 95: 72– 77. http://dx.doi.org/10.3945/ajcn.111.024596 Google Scholar CrossRef Search ADS PubMed  37 Schwinn TM, Schinke S, Fang L, et al.   A web-based, health promotion program for adolescent girls and their mothers who reside in public housing. Addict Behav . 2014; 39: 757– 760. http://dx.doi.org/10.1016/j.addbeh.2013.11.029 Google Scholar CrossRef Search ADS PubMed  38 Tabak RG, Tate DF, Stevens J, et al.   Family ties to health program: a randomized intervention to improve vegetable intake in children. J Nutr Educ Behav . 2012; 44: 166– 171. http://dx.doi.org/10.1016/j.jneb.2011.06.009 Google Scholar CrossRef Search ADS PubMed  39 Thompson D, Bhatt R, Vazquez I, et al.   Creating action plans in a serious video game increases and maintains child fruit-vegetable intake: a randomized controlled trial. Int J Behav Nutr Phys Act.  2015; 12: 39. doi: 10.1186/s12966-015-0199-z Google Scholar CrossRef Search ADS PubMed  40 Tomayko EJ, Prince RJ, Cronin KA, et al.   The Healthy Children, Strong Families intervention promotes improvements in nutrition, activity and body weight in American Indian families with young children. Public Health Nutr . 2016;19: 1– 10. 41 Wardle J, Cooke LJ, Gibson EL, et al.   Increasing children's acceptance of vegetables; a randomized trial of parent-led exposure. Appetite . 2003; 40: 155– 162. http://dx.doi.org/10.1016/S0195-6663(02)00135-6 Google Scholar CrossRef Search ADS PubMed  42 Wyse R, Wolfenden L, Campbell E, et al.   A pilot study of a telephone-based parental intervention to increase fruit and vegetable consumption in 3–5-year-old children. Public Health Nutr.  2011; 14: 2245– 2253. http://dx.doi.org/10.1017/S1368980011001170 Google Scholar CrossRef Search ADS PubMed  43 Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health . 1998; 52: 377– 384. http://dx.doi.org/10.1136/jech.52.6.377 Google Scholar CrossRef Search ADS PubMed  44 Luz F, Hay P, Gibson A, et al.   Does severe dietary energy restriction increase binge eating in overweight or obese individuals? A systematic review. Obes Rev.  2015; 16: 652– 665. http://dx.doi.org/10.1111/obr.12295 Google Scholar CrossRef Search ADS PubMed  45 Golan M, Crow S. Parents are key players in the prevention and treatment of weight-related problems. Nutr Rev.  2004; 62: 39– 50. http://dx.doi.org/10.1111/j.1753-4887.2004.tb00005.x Google Scholar CrossRef Search ADS PubMed  46 Jones LR, Steer CD, Rogers IS, et al.   Influences on child fruit and vegetable intake: sociodemographic, parental and child factors in a longitudinal cohort study. Public Health Nutr.  2010; 13: 1122– 1130. http://dx.doi.org/10.1017/S1368980010000133 Google Scholar CrossRef Search ADS PubMed  47 Rasmussen M, Krolner R, Klepp KI, et al.   Determinants of fruit and vegetable consumption among children and adolescents: a review of the literature. Part I: quantitative studies. Int J Behav Nutr Phys Act.  2006; 3: 22. doi: 10.1186/1479-5868-3-22 Google Scholar CrossRef Search ADS PubMed  48 Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med.  2002; 21: 1539– 1558. http://dx.doi.org/10.1002/sim.1186 Google Scholar CrossRef Search ADS PubMed  49 Borenstein M, Hedges LV, Higgins JPT, et al.   Introduction to Meta-Analysis . Chicester, West Sussex, UK: John Wiley & Sons Ltd; 2009. Google Scholar CrossRef Search ADS   50 Guenther PM, Dodd KW, Reedy J, et al.   Most Americans eat much less than recommended amounts of fruits and vegetables. J Am Diet Assoc . 2006; 106:1371–1379. http://dx.doi.org/10.1016/j.jada.2006.06.002 51 Blanchette L, Brug J. Determinants of fruit and vegetable consumption among 6–12‐year‐old children and effective interventions to increase consumption. J Hum Nutr Diet.  2005; 18: 431– 443. Google Scholar CrossRef Search ADS PubMed  52 Wardle J, Cooke L. Genetic and environmental determinants of children's food preferences. Br J Nutr.  2008; 99(suppl 1): S15– S21. Google Scholar PubMed  53 Cooke L, Carnell S, Wardle J. Food neophobia and mealtime food consumption in 4–5 year old children. Int J Behav Nutr Phys Act.  2006; 3: 14. doi: 10.1186/1479-5868-3-14 Google Scholar CrossRef Search ADS PubMed  54 Cooke L, Wardle J, Gibson E, et al.   Demographic, familial and trait predictors of fruit and vegetable consumption by pre-school children. Public Health Nutr . 2004; 7: 295– 302. Google Scholar CrossRef Search ADS PubMed  55 Skouteris H, Hill B, McCabe M, et al.   A parent-based intervention to promote healthy eating and active behaviours in pre-school children: evaluation of the MEND 2–4 randomized controlled trial. Pediatr Obes . 2016; 11: 4– 10. doi:10.1111/ijpo.12011 Google Scholar CrossRef Search ADS PubMed  56 Wakefield CE, Fardell JE, Doolan EL, et al.   Participation in psychosocial oncology and quality-of-life research: a systematic review. Lancet Oncol.  2017; 18: e153– e165. Google Scholar CrossRef Search ADS PubMed  57 Price M, Yuen EK, Goetter EM, et al.   mHealth: a mechanism to deliver more accessible, more effective mental health care. Clin Psychol Psychother.  2014; 21: 427– 436. http://dx.doi.org/10.1002/cpp.1855 Google Scholar CrossRef Search ADS PubMed  58 Prochaska JO., Velicer WF . The transtheoretical model of health behavior change. Am J Health Promot.  1997; 12: 38– 48. http://dx.doi.org/10.4278/0890-1171-12.1.38 Google Scholar CrossRef Search ADS PubMed  59 Whisman MA. The efficacy of booster maintenance sessions in behavior therapy: review and methodological critique. Clin Psychol Rev . 1990; 10: 155– 170. http://dx.doi.org/10.1016/0272-7358(90)90055-F Google Scholar CrossRef Search ADS   60 Kessler RC. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry . 2005; 62:617–627. http://dx.doi.org/10.1001/archpsyc.62.6.617 61 Lepper MR, Greene D, Nisbett RE. Undermining children's intrinsic interest with extrinsic reward: a test of the “overjustification” hypothesis. J Pers Soc Psychol . 1973; 28: 129–137. Google Scholar CrossRef Search ADS   62 Wardle J, Herrera ML, Cooke L, et al.   Modifying children's food preferences: the effects of exposure and reward on acceptance of an unfamiliar vegetable. Eur J Clin Nutr.  2003; 57: 341– 348. http://dx.doi.org/10.1038/sj.ejcn.1601541 Google Scholar CrossRef Search ADS PubMed  63 Hidi S. Revisiting the role of rewards in motivation and learning: implications of neuroscientific research. Educ Psychol Rev.  2016; 28: 61– 93. http://dx.doi.org/10.1007/s10648-015-9307-5 Google Scholar CrossRef Search ADS   64 Dickersin K, Min Y-I, Meinert CL. Factors influencing publication of research results: follow-up of applications submitted to two institutional review boards. JAMA . 1992; 267: 374– 378. http://dx.doi.org/10.1001/jama.1992.03480030052036 Google Scholar CrossRef Search ADS PubMed  65 Dixon B, Peña M-M, Taveras EM. Lifecourse approach to racial/ethnic disparities in childhood obesity. Adv Nutr.  2012; 3: 73– 82. Google Scholar CrossRef Search ADS PubMed  66 Burrows TL, Martin RJ, Collins CE. A systematic review of the validity of dietary assessment methods in children when compared with the method of doubly labeled water. J Am Diet Assoc . 2010; 110: 1501– 1510. http://dx.doi.org/10.1016/j.jada.2010.07.008 Google Scholar CrossRef Search ADS PubMed  67 Livingstone MB, Prentice AM, Coward WA, et al.   Validation of estimates of energy intake by weighed dietary record and diet history in children and adolescents. Am J Clin Nutr . 1992; 56: 29– 35. Google Scholar CrossRef Search ADS PubMed  68 Johns B, Baltussen R, Hutubessy R. Programme costs in the economic evaluation of health interventions. Cost Eff Resour Alloc.  2003; 1: 1. doi: 10.1186/1478-7547-1-1 Google Scholar CrossRef Search ADS PubMed  69 Head KJ, Noar SM, Iannarino NT, et al.   Efficacy of text messaging-based interventions for health promotion: a meta-analysis. Soc Sci Med.  2013; 97: 41– 48. http://dx.doi.org/10.1016/j.socscimed.2013.08.003 Google Scholar CrossRef Search ADS PubMed  70 Schoeppe S, Alley S, Van Lippevelde W, et al.   Efficacy of interventions that use apps to improve diet, physical activity and sedentary behaviour: a systematic review. Int J Behav Nutr Phys Act.  2016; 13: 127. 10.1186/s12966-016-0454-y Google Scholar CrossRef Search ADS PubMed  71 Zhao J, Freeman B, Li M. Can mobile phone apps influence people's health behavior change? An evidence review. J Med Internet Res.  2016; 18: e287. doi: 10.2196/jmir.5692 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Journal

Nutrition ReviewsOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from
Google Scholar,
PubMed
Create lists to
organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off