TY - JOUR
AU - Addo, O, Yaw
AB - ABSTRACT Background Anemia is a moderate public health problem among adolescent girls in Ghana. Objectives We aimed to evaluate the barriers to and facilitators of program fidelity to a school-based anemia reduction program with weekly iron and folic acid (IFA) supplementation. Methods Authors analyzed directly observed weekly IFA consumption data collected longitudinally and cross-sectional data from a representative survey of 60 secondary schools and 1387 adolescent girls in the Northern and Volta regions of Ghana after 1 school year (2017–2018) of the intervention (30–36 wk). A bottleneck analysis was used to characterize the levels of IFA coverage and used adjusted generalized linear mixed-effects models to quantify the school and student drivers of IFA intake adherence. Results Of girls, 90% had ever consumed the tablet, whereas 56% had consumed ≥15 weekly tablets (mean: 16.4, range: 0–36), indicating average intake adherence was about half of the available tablets. Among ever consumers, 88% of girls liked the tablet, and 27% reported undesirable changes (primarily heavy menstrual flow). School-level factors represented 75% of the variance in IFA consumption over the school year. Total IFA tablets consumed was associated with the ability to make up missed IFA distributions (+1.4 tablets; 95% CI: +0.8, +2.0 tablets), junior compared with senior secondary school (+5.8; 95% CI: +0.1, +11.5), educators’ participating in a program-related training (+7.6; 95% CI: +2.9, 12.2), and educator perceptions that implementation was difficult (−6.9; 95% CI: −12.1, −1.7) and was an excessive time burden (−4.4; 95% CI: −8.4, −0.4). Conclusions Although the program reached Ghanaian schoolgirls, school-level factors were barriers to adherence. Modifications such as expanded training, formalized make-up IFA distributions, sensitization (awareness promotion), and additional support to senior high schools may improve adherence. Spreading the responsibility for IFA distribution to other teachers and streamlining monitoring may reduce the burden at the school level. Strengthening the health education component and improving knowledge of IFA among students may also be beneficial. adolescent girls, iron and folic acid, supplementation, anemia, national IFA program, secondary school Introduction The WHO recommends weekly intermittent supplementation with iron and folic acid (IFA) for the control of anemia among adolescent girls ages 15–19 y and school-aged children 5–12 y old in areas where the prevalence of anemia is ≥20% (1, 2). With a national anemia prevalence of 26.4% in 2017 among nonpregnant girls ages 15–19 y (no data available for those ages 10–14 y), anemia is a problem of moderate public health significance in Ghana (3, 4). To address this problem, the Girls’ Iron-Folic acid Tablet Supplementation (GIFTS) Program, an integrated health and nutrition education program with intermittent weekly IFA supplementation, was designed to reach adolescent girls ages 10–19 y simultaneously through 2 delivery platforms: schools and local health centers. The first phase of the program began in the Brong-Ahafo, Northern, Upper East, and Volta regions of Ghana beginning in October 2017. (In December 2018, Ghana's 10 regions were divided into 16. Brong-Ahafo became Bono, Bono East, and Ahafo. Northern became Savannah, Northern, and North East. Upper East maintained the same boundaries. Volta was divided into Volta and Oti.) School-based IFA supplementation has substantial benefit in reducing the prevalence of anemia under experimental conditions (5); however, there is limited information on how these interventions perform in real-world settings. Contextual factors may drive program performance and be responsible for the differences observed between efficacy and effectiveness studies. Guided by the Social-Ecological Model, potential factors affecting the success of school health programs can be identified at various levels: community, school, family, and individual factors (6). In other contexts, barriers and facilitators have been described in terms of resources, capacity, enabling environment, knowledge, attitudes, and perceptions (7, 8) and factors spanned from national policies to individual perceptions. Cross-sector collaborations have been identified by school-based programs as a key driver of program success in other countries because of the synergizing of resources and networks, but they may also create dysfunction (8–10). At the community level, caregivers and other community members influence programs through their support, resistance, or indifference, which may be influenced by cultural norms, media, and sensitization (awareness-promotion) activities (9). Several factors at the school level should also be considered such as procurement and stock outs, the system for administration of tablets, absenteeism, time commitment, and motivation of implementers (7, 8, 11). Training can improve the knowledge, attitudes, and skills of teachers which influence program fidelity (8). Teacher participation may also be affected by their workload and perceptions of anemia risk and IFA benefit (12). Knowledge, attitudes, and perceptions toward anemia risk and IFA may influence a student's willingness to participate, and after participating, any physical or cognitive changes attributed to the tablets may also affect adherence (13). In addition, students’ physical maturity and agency may influence their ability to swallow tablets and their household socioeconomic and food security status may dictate access to food and water for safe and appropriate consumption of IFA tablets (Figure 1). FIGURE 1 Open in new tabDownload slide Conceptual framework of the barriers to and facilitators of school-based IFA supplementation. This framework incorporates elements from Tanahashi's levels of health service coverage (1978) and the Innocenti Framework (UNICEF, 2018). IFA, iron and folic acid. FIGURE 1 Open in new tabDownload slide Conceptual framework of the barriers to and facilitators of school-based IFA supplementation. This framework incorporates elements from Tanahashi's levels of health service coverage (1978) and the Innocenti Framework (UNICEF, 2018). IFA, iron and folic acid. The school is a useful platform for intermittent IFA supplementation among adolescents because it has several advantages. One such advantage is regular access to a portion of the target population. In Ghana in 2018, nearly half of girls ages 12–14 y (49%) were enrolled in junior high schools (JHSs), and 30% of girls ages 15–17 y were enrolled in senior high schools (SHSs), up from 25% in 2016 (14). Enrollment of older girls (over-age enrollment) in secondary school was 16% in 2018, adding to the population reached by such an intervention. The gross enrollment ratio was higher (55%), but this represents the ratio of all girls enrolled in schools in Ghana regardless of age divided by the population of girls aged 15–17 y in Ghana (14). With free public SHS introduced in 2017, enrollment is expected to further increase. Schools also offer an existing infrastructure for administration of the program and an educational environment where sensitization of students can be easily integrated. With these advantages, addressing the barriers to school-based IFA supplementation and leveraging facilitators of adherence could lead to high coverage of the intervention. However, data evaluating the barriers to and facilitators of school-based IFA supplementation programs for adolescent girls are limited to a few studies in South and Southeast Asia and may not be broadly applicable to settings such as sub-Saharan Africa. The GIFTS program was rolled out in 3 phases: it began in 4 regions during the 2017–2018 school year and became a nationwide program in the 2019–2020 school year. An evaluation of 2 of the Phase I regions estimated a reduction in the prevalence of anemia from 25% to 19.5% over 1 school year, but adherence, defined as the number of tablets consumed each week of the school year after program launch (30–36 wk depending on the school), was suboptimal (approximately half consumed ≤15 tablets) (15). Improving program implementation fidelity and intake adherence could achieve a greater improvement in hemoglobin concentration and anemia control in the population. However, owing to very limited experience in adolescent IFA supplementation on the continent of Africa and none in Ghana, there is little evidence available for improving adherence to the intervention. Our aim was to identify program components limiting the coverage of IFA and to describe the key barriers and facilitators that may cause or mitigate against such bottlenecks during Phase I of Ghana's school-based adolescent IFA supplementation program. Methods Description of the program Training of staff from the education and health sector was conducted just before the rollout of the GIFTS program in each region. This was carried out in cascading trainings from the regional level to the schools and health centers. For the school-based portion of the intervention, the objective of the training was to deliver information regarding the program to the school focal person who would train other school staff. The training covered anemia causes, consequences, and prevention; benefits and potential side effects of IFA; program logistics; administration of IFA tablets; and monitoring. IFA tablets (60 mg elemental iron as ferrous fumarate and 400 µg folic acid) were distributed through the existing government health system to the district level, where they were provided in some cases to district education offices who distributed them to individual schools and in other cases to local health centers who distributed them to schools. The intervention provided IFA tablets to all girls aged 10–19 y in JHSs, SHSs, and technical/vocational schools, both public and private, on a set day of the week after a meal and with water. Some schools provided meals although most students brought their own. No tablets were given during holidays and breaks, leaving 30–36 distribution weeks depending on the school. Health and nutrition education were also addressed in the training; however, this component was carried out through the existing school health and nutrition education and promotion program. Malaria prevention was a key component of health education because Ghana is a malaria-endemic country. The intervention is fully described elsewhere (15). Sampling and data collection In September 2017, a baseline nutrition survey was conducted in JHSs, SHSs, and vocational/technical schools in the Northern and Volta regions of Ghana. These regions were purposively selected from the 4 Phase I regions for the GIFTS program by the Ministry of Health. Northern region was selected for its large size and generally low performance in other nutrition and health programs, whereas Volta, a smaller coastal region, had high performance in existing programs. It was expected that results from these 2 regions could be applied to the remaining 2 regions. Although a limited number were randomly selected for the evaluation, all schools and students within the 4 regions received the GIFTS program. A stratified 2-stage sampling frame was used. First, probability proportional to size sampling was used to select 60 schools—15 JHSs and 15 SHSs from each region—followed by simple random sampling of girls within each school. The sampling frame included all the public and private schools, treated as clusters, with school type (dichotomized as JHS or SHS) as strata. In Northern region, there was a denominator of 837 candidate schools (clusters) in the JHS-stratum and 68 in the SHS-stratum, and in Volta region, there were 1225 JHSs and 104 SHSs. Within each school, 29 girls were randomly selected by simple random sampling from forms 1 and 2 (equivalent to grades 7, 8, 10, and 11 in the North American system) to avoid early exam weeks required of students in form 3 (the graduating 12th graders). There were no other exclusion criteria. The derived sample size was powered to detect a 10% minimum change in the prevalence of anemia based on a background prevalence of 40% with 80% power and 95% significance level; anemia prevalence was 48% in 2014 (the latest data available at the time of study planning) (16). Calculations included 10% oversampling to account for nonparticipation and correcting for an intraclass correlation coefficient as high as 10% for repeated measures (17). A focal person (teacher) for the GIFTS program was surveyed at each school, for a total of 60 school-level interviews with a response rate of 100%. The baseline survey collected data from 1521 in-school adolescent girls ages 10–19 y with a response rate of 95%. Relevant to this analysis, the baseline survey collected demographic characteristics of girls. At the end of the school year, after ∼8 mo of the intervention, the participant schools (n = 60; 100% response rate) and girls (n = 1387; 86% response rate) were followed up for another round of data collection. Questionnaires were administered by trained enumerators in a private location. At follow-up, the school questionnaire (administered to a teacher) included information about school facilities, school health and nutrition activities, and specific questions related to the GIFTS program including perceptions of IFA and experiences with the distribution of IFA tablets. The student questionnaire asked girls about their knowledge of anemia and IFA, experiences with the program, and basic demographic information. The number of IFA tablets consumed by each survey participant in each school term was abstracted from classroom-level registers that collected individual IFA tablet consumption data for all girls in real time throughout the school year. These registers were completed weekly by the distributor of tablets after direct observation of tablet consumption. We used both self-reported and objective IFA consumption data for cross-validation and preservation of sample size. This study was approved by the ethical review board of the Ghana Health Service and exempted from review by the Emory University institutional review board. Procedures followed were in accordance with the ethical standards of these institutions. Written informed consent was obtained from parents or guardians of selected students and those in the age of majority (≥18 y). Verbal informed assent was obtained from each participant. Data were stored on password-protected devices before being transferred to encrypted cloud storage where they were anonymized before analysis. Statistical analysis Poststratification sampling weights were applied, and complex survey procedures in SAS version 9.4 (SAS Institute, Inc.) were used in all analyses to account for sampling strata and clusters. Statistical significance was defined by an α level of 0.05. Wealth ranking was calculated via principal components analysis of household possessions and durable goods (television, radio, bicycle, electricity, etc.) and divided into tertiles (18). We used census data to categorize schools as rural, peri-urban, or urban (19). Knowledge of anemia and IFA were calculated using a composite score of correct responses to 3 multiple-response questions about the causes, symptoms, and prevention of anemia and 2 multiple-response questions on awareness and benefits of IFA, respectively, and divided into tertiles. Indicators of IFA supplementation coverage were categorized based on Tanahashi's levels of health care coverage, which consist of availability, accessibility, acceptability, contact, and effectiveness (20). Often used in assessment of existing programs or service delivery, the Tanahashi bottleneck analyses model quantifies barriers (i.e., the bottleneck) to full coverage. A bottleneck is a barrier to 100% health care coverage, and usually quantified as 100 − % coverage (20). The total number of IFA tablets consumed, abstracted from registers, was categorized and compared over basic characteristics. Self-reported consumption of an IFA tablet during the school year (yes/no) was also compared by characteristics of individual students and schools. Rao–Scott chi-square tests were used for differences in proportions, and design-based t tests were used for continuous variables. To examine the associations between potential barriers and facilitators at the student and school levels and adherence to IFA supplementation, adjusted models of the continuous total number of IFA tablets consumed over the school year, abstracted from registers, were built using generalized linear mixed-effects models (GLMMs). Two adjusted GLMMs of the school- and student-level predictors of the total number of IFA tablets were constructed based on a conceptual model (Figure 1). The first model was restricted to students who had consumed ≥1 tablet to understand drivers of intake adherence (effectiveness coverage) among those who had ever consumed IFA tablets. The second model included all survey participants to understand the relation between these factors and intake adherence within the entire population of schoolgirls. Model diagnostics included multicollinearity and influential point checks (2 influential observations were excluded from the final mixed models). Additional regression diagnostics included a check of the residuals, Cook's D, Akaike Information Criterion, and Bayesian Information Criterion. A mixed model with clusters (schools) as a random intercept was used to estimate intraclass correlation. This enabled variance decomposition of the total number of IFA tablets consumed over the school year for quantifying the proportion of variance explained by student- or school-level factors. Results The final survey sample included 1387 girls ages 10–19 y and 60 schools. Classroom-level GIFTS program registers were missing for ≥1 term in 3 schools, reducing the sample with complete objective IFA consumption data to 1307 girls. There was 100% agreement between self-reported ever consuming IFA and objective register data of the same (data not shown). Figure 2 shows the levels of coverage and bottlenecks. Ninety percent (90% weighted; 1239 of 1307) of schoolgirls consumed ≥1 IFA tablet over the school year, implying a bottleneck of 10% in base-level effectiveness coverage, and the mean intake was 16.4 tablets (range: 0–36) over the year (data not shown). Only 56% consumed ≥15 tablets: approximately half of the tablets they were eligible to receive during the follow-up period (1 school year). In terms of recent exposure, 66% of the girls reported having consumed the tablet during the previous 2 wk, and 90% of schools had distributed IFA tablets in the week before the follow-up survey. Among those who had ever consumed an IFA tablet (n = 1314), nearly one-quarter (23%) of schoolgirls reported that they did not have another opportunity to make up a missed IFA distribution, 21% reported ever refusing to receive an IFA tablet, and 13% had received a tablet that they later threw away. From the school survey (n = 60), half of schools (48%) reported difficulties implementing the IFA program, but only 13% had difficulties with the program from community members at some point during the school year. Four schools (7%) had experienced a stock out of IFA tablets, of which only 2 schools had ever missed a distribution due to a stock out. FIGURE 2 Open in new tabDownload slide Levels of IFA supplementation coverage and bottlenecks among adolescent girls in Ghanaian schools in the Northern and Volta regions, 2017. Proportions are weighted at the student level to be representative of eligible girls in the school. an = 60 schools. Data from cross-sectional follow-up survey. bn = 1314 girls, only among those who self-reported that they had ever consumed a tablet. Data from cross-sectional follow-up survey. cn = 1307 girls, sample size reduced owing to missing IFA distribution registers. Data from IFA distribution registers. IFA, iron and folic acid. FIGURE 2 Open in new tabDownload slide Levels of IFA supplementation coverage and bottlenecks among adolescent girls in Ghanaian schools in the Northern and Volta regions, 2017. Proportions are weighted at the student level to be representative of eligible girls in the school. an = 60 schools. Data from cross-sectional follow-up survey. bn = 1314 girls, only among those who self-reported that they had ever consumed a tablet. Data from cross-sectional follow-up survey. cn = 1307 girls, sample size reduced owing to missing IFA distribution registers. Data from IFA distribution registers. IFA, iron and folic acid. Total IFA consumption differed by age, school level, and wealth. Girls 10–14 y of age had consumed 6.7 (95% CI: 2.2, 11.3) more tablets, on average, than the girls aged 15–19 y (22.6 compared with 15.9 tablets, P < 0.01; data not shown). IFA consumption also differed by school level (P < 0.01). The largest proportion of junior high schoolgirls (69%) had consumed 21–30 tablets over the school year, but among senior high and vocational schoolgirls 42% had consumed 11–20 tablets and only 23% of them had consumed 21–30 tablets (data not shown). IFA consumption differed significantly by wealth (P < 0.01): more girls from the low wealth tertile had consumed ≥11 tablets than their peers from higher wealth tertiles. No significant difference in IFA consumption by geographic location was observed (P = 0.24). Table 1 presents results of the bivariate analysis of student characteristics by self-reported consumption of IFA tablets. In crude comparisons, self-reported ever consuming an IFA tablet over the school year differed significantly by knowledge of IFA tablets and receiving health education sessions on malaria, deworming, and clean water/hand hygiene (each P < 0.05). There were no statistically significant differences in self-reported consumption of IFA tablets by age, menarche, wealth, knowledge of anemia, health education, IFA counseling, or receiving health education sessions on anemia, menstruation, eating iron-rich foods, or geophagy. Among consumers, 88% of girls liked the IFA tablets and 11% did not. Approximately 38% of IFA consumers experienced desirable changes such as improved health (22%), increased appetite (22%), increased strength (13%), increased activeness (8%), decreased sleepiness (5%), and improved concentration (3%) (data not shown). One-quarter of consumers (27%) noticed undesirable changes such as dizziness (4%), excessive hunger (3%), nausea, stomach pain, headaches, constipation, and dark/smelly stool (all ≤1%) (data not shown). Reported changes in menstruation were both desirable and undesirable: about one-third of those who noticed any changes reported heavier or longer menses, whereas an equal proportion said it made their menses more regular. Twenty-seven percent reported they had ever consumed an IFA tablet on an empty stomach. TABLE 1 Student characteristics by self-reported consumption of IFA tablets over 1 school year among adolescent girls in Ghanaian schools in the Northern and Volta regions, 20171 . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . Student characteristic . n . % (95% CI) . n . % (95% CI) . P value2 . n . % (95% CI) . Demographics  Age, y 0.079   10–14 8 2.7 (0.0, 6.6) 303 8.4 (4.1, 12.7) 311 7.6 (3.8, 11.5)   15–19 65 97.3 (93.4, 100) 1011 91.6 (87.3, 95.9) 1076 92.2 (88.2, 96.1)  Reached menarche 67 98.0 (95.5, 100) 1175 96.1 (94.3, 98.0) 0.272 1242 96.3 (94.6, 98.0)  Wealth tertile 0.229   High 31 48.5 (36.6, 60.4) 438 40.3 (32.2, 48.4) 469 41.2 (33.3, 49.0)   Middle 24 31.4 (26.8, 36.0) 439 31.9 (27.5, 36.3) 463 31.9 (28.1, 35.7)   Low 18 20.1 (8.1, 32.1) 437 27.7 (20.9, 34.6) 455 26.9 (20.1, 33.7) Knowledge, attitudes, and practices  Knowledge of anemia3 0.981   High 18 31.1 (5.8, 56.3) 369 32.1 (25.5, 38.7) 387 32.0 (25.0, 39.0)   Middle 25 33.6 (24.5, 42.6) 436 34.3 (29.5, 39.1) 461 34.2 (29.8, 38.7)   Low 30 35.4 (17.6, 53.1) 509 33.6 (28.3, 38.8) 539 33.7 (28.4, 39.1)  Knowledge of IFA4 0.036   High 12 23.2 (11.3, 35.2) 452 37.7 (32.0, 43.4) 464 36.6 (30.8, 42.3)   Middle 14 23.7 (12.3, 35.0) 324 23.5 (20.3, 26.8) 338 23.5 (20.2, 26.9)   Low 29 53.1 (39.3, 66.9) 528 38.8 (32.0, 45.5) 557 39.9 (33.2, 46.6) IFA program experiences  Took an IFA tablet on an empty stomach — — 361 26.9 (20.5, 33.4) — — —  Opinion of IFA tablets   Likes IFA tablets5 — — 1212 87.8 (83.3, 92.3) — — —   Does not like IFA tablets5 — — 75 11.3 (6.6, 15.9) — — —  Changes upon taking IFA tablets6   Desirable changes upon taking IFA tablets — — 594 37.9 (29.2, 46.7) — — —   Undesirable changes upon taking IFA tablets — — 322 26.9 (21.2, 32.6) — — —  Shared IFA tablet experiences with friends/family — — 714 52.5 (45.4, 59.6) — — —  Able to make up a missed IFA distribution — — 1111 76.9 (69.9, 83.8) — — —  Received IFA counseling 7 11.7 (2.2, 21.3) 298 18.4 (13.8, 23.0) 0.260 305 17.7 (13.4, 22.0)  Received health education 29 42.1 (25.4, 58.7) 702 46.4 (40.4, 52.4) 0.631 731 46.0 (40.4, 51.6)  Topics of health education sessions received in previous year   Anemia 27 34.1 (23.2, 45.0) 679 43.0 (37.3, 48.6) 0.187 706 42.1 (37.1, 47.0)   Malaria 31 36.7 (20.0, 53.4) 791 54.1 (49.0, 59.1) 0.035 822 52.3 (46.9, 57.7)   Deworming 14 15.2 (10.1, 20.3) 398 24.8 (19.8, 29.8) 0.009 412 23.8 (19.2, 28.3)   Clean water and/or handwashing 50 67.0 (56.6, 77.4) 1111 79.7 (74.6, 84.9) 0.022 1161 78.4 (73.7, 83.1)   Menstruation 52 69.6 (56.8, 82.4) 1055 74.8 (71.0, 78.5) 0.426 1107 74.2 (70.6, 77.8)   Eating iron-rich foods 31 39.7 (32.1, 47.3) 729 48.6 (42.2, 55.0) 0.132 760 47.7 (42.2, 57.8)   Avoiding eating soil or clay 21 19.1 (4.4, 33.8) 527 32.1 (26.0, 38.1) 0.130 548 30.7 (24.7, 36.7) . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . Student characteristic . n . % (95% CI) . n . % (95% CI) . P value2 . n . % (95% CI) . Demographics  Age, y 0.079   10–14 8 2.7 (0.0, 6.6) 303 8.4 (4.1, 12.7) 311 7.6 (3.8, 11.5)   15–19 65 97.3 (93.4, 100) 1011 91.6 (87.3, 95.9) 1076 92.2 (88.2, 96.1)  Reached menarche 67 98.0 (95.5, 100) 1175 96.1 (94.3, 98.0) 0.272 1242 96.3 (94.6, 98.0)  Wealth tertile 0.229   High 31 48.5 (36.6, 60.4) 438 40.3 (32.2, 48.4) 469 41.2 (33.3, 49.0)   Middle 24 31.4 (26.8, 36.0) 439 31.9 (27.5, 36.3) 463 31.9 (28.1, 35.7)   Low 18 20.1 (8.1, 32.1) 437 27.7 (20.9, 34.6) 455 26.9 (20.1, 33.7) Knowledge, attitudes, and practices  Knowledge of anemia3 0.981   High 18 31.1 (5.8, 56.3) 369 32.1 (25.5, 38.7) 387 32.0 (25.0, 39.0)   Middle 25 33.6 (24.5, 42.6) 436 34.3 (29.5, 39.1) 461 34.2 (29.8, 38.7)   Low 30 35.4 (17.6, 53.1) 509 33.6 (28.3, 38.8) 539 33.7 (28.4, 39.1)  Knowledge of IFA4 0.036   High 12 23.2 (11.3, 35.2) 452 37.7 (32.0, 43.4) 464 36.6 (30.8, 42.3)   Middle 14 23.7 (12.3, 35.0) 324 23.5 (20.3, 26.8) 338 23.5 (20.2, 26.9)   Low 29 53.1 (39.3, 66.9) 528 38.8 (32.0, 45.5) 557 39.9 (33.2, 46.6) IFA program experiences  Took an IFA tablet on an empty stomach — — 361 26.9 (20.5, 33.4) — — —  Opinion of IFA tablets   Likes IFA tablets5 — — 1212 87.8 (83.3, 92.3) — — —   Does not like IFA tablets5 — — 75 11.3 (6.6, 15.9) — — —  Changes upon taking IFA tablets6   Desirable changes upon taking IFA tablets — — 594 37.9 (29.2, 46.7) — — —   Undesirable changes upon taking IFA tablets — — 322 26.9 (21.2, 32.6) — — —  Shared IFA tablet experiences with friends/family — — 714 52.5 (45.4, 59.6) — — —  Able to make up a missed IFA distribution — — 1111 76.9 (69.9, 83.8) — — —  Received IFA counseling 7 11.7 (2.2, 21.3) 298 18.4 (13.8, 23.0) 0.260 305 17.7 (13.4, 22.0)  Received health education 29 42.1 (25.4, 58.7) 702 46.4 (40.4, 52.4) 0.631 731 46.0 (40.4, 51.6)  Topics of health education sessions received in previous year   Anemia 27 34.1 (23.2, 45.0) 679 43.0 (37.3, 48.6) 0.187 706 42.1 (37.1, 47.0)   Malaria 31 36.7 (20.0, 53.4) 791 54.1 (49.0, 59.1) 0.035 822 52.3 (46.9, 57.7)   Deworming 14 15.2 (10.1, 20.3) 398 24.8 (19.8, 29.8) 0.009 412 23.8 (19.2, 28.3)   Clean water and/or handwashing 50 67.0 (56.6, 77.4) 1111 79.7 (74.6, 84.9) 0.022 1161 78.4 (73.7, 83.1)   Menstruation 52 69.6 (56.8, 82.4) 1055 74.8 (71.0, 78.5) 0.426 1107 74.2 (70.6, 77.8)   Eating iron-rich foods 31 39.7 (32.1, 47.3) 729 48.6 (42.2, 55.0) 0.132 760 47.7 (42.2, 57.8)   Avoiding eating soil or clay 21 19.1 (4.4, 33.8) 527 32.1 (26.0, 38.1) 0.130 548 30.7 (24.7, 36.7) 1 Estimates are weighted at the student level to be representative of eligible girls in the school. IFA, iron-folic acid. 2 Rao–Scott chi-square testing the difference between never consumed and consumed ≥1 IFA tablet. 95% CIs and P values for student characteristics are based on Taylor series variance estimates to account for the complex sampling design. 3 Knowledge of anemia was calculated using a composite score of correct responses to questions about anemia (awareness, signs, causes, and effects) and divided into tertiles before weights were applied. 4 Knowledge of IFA was calculated using a composite score of correct responses to questions about IFA (awareness and effects) and divided into tertiles before weights were applied; 28 missing observations. 5 Remaining participants reported neutral feelings about IFA tablets. 6 Remaining participants noticed no changes. Open in new tab TABLE 1 Student characteristics by self-reported consumption of IFA tablets over 1 school year among adolescent girls in Ghanaian schools in the Northern and Volta regions, 20171 . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . Student characteristic . n . % (95% CI) . n . % (95% CI) . P value2 . n . % (95% CI) . Demographics  Age, y 0.079   10–14 8 2.7 (0.0, 6.6) 303 8.4 (4.1, 12.7) 311 7.6 (3.8, 11.5)   15–19 65 97.3 (93.4, 100) 1011 91.6 (87.3, 95.9) 1076 92.2 (88.2, 96.1)  Reached menarche 67 98.0 (95.5, 100) 1175 96.1 (94.3, 98.0) 0.272 1242 96.3 (94.6, 98.0)  Wealth tertile 0.229   High 31 48.5 (36.6, 60.4) 438 40.3 (32.2, 48.4) 469 41.2 (33.3, 49.0)   Middle 24 31.4 (26.8, 36.0) 439 31.9 (27.5, 36.3) 463 31.9 (28.1, 35.7)   Low 18 20.1 (8.1, 32.1) 437 27.7 (20.9, 34.6) 455 26.9 (20.1, 33.7) Knowledge, attitudes, and practices  Knowledge of anemia3 0.981   High 18 31.1 (5.8, 56.3) 369 32.1 (25.5, 38.7) 387 32.0 (25.0, 39.0)   Middle 25 33.6 (24.5, 42.6) 436 34.3 (29.5, 39.1) 461 34.2 (29.8, 38.7)   Low 30 35.4 (17.6, 53.1) 509 33.6 (28.3, 38.8) 539 33.7 (28.4, 39.1)  Knowledge of IFA4 0.036   High 12 23.2 (11.3, 35.2) 452 37.7 (32.0, 43.4) 464 36.6 (30.8, 42.3)   Middle 14 23.7 (12.3, 35.0) 324 23.5 (20.3, 26.8) 338 23.5 (20.2, 26.9)   Low 29 53.1 (39.3, 66.9) 528 38.8 (32.0, 45.5) 557 39.9 (33.2, 46.6) IFA program experiences  Took an IFA tablet on an empty stomach — — 361 26.9 (20.5, 33.4) — — —  Opinion of IFA tablets   Likes IFA tablets5 — — 1212 87.8 (83.3, 92.3) — — —   Does not like IFA tablets5 — — 75 11.3 (6.6, 15.9) — — —  Changes upon taking IFA tablets6   Desirable changes upon taking IFA tablets — — 594 37.9 (29.2, 46.7) — — —   Undesirable changes upon taking IFA tablets — — 322 26.9 (21.2, 32.6) — — —  Shared IFA tablet experiences with friends/family — — 714 52.5 (45.4, 59.6) — — —  Able to make up a missed IFA distribution — — 1111 76.9 (69.9, 83.8) — — —  Received IFA counseling 7 11.7 (2.2, 21.3) 298 18.4 (13.8, 23.0) 0.260 305 17.7 (13.4, 22.0)  Received health education 29 42.1 (25.4, 58.7) 702 46.4 (40.4, 52.4) 0.631 731 46.0 (40.4, 51.6)  Topics of health education sessions received in previous year   Anemia 27 34.1 (23.2, 45.0) 679 43.0 (37.3, 48.6) 0.187 706 42.1 (37.1, 47.0)   Malaria 31 36.7 (20.0, 53.4) 791 54.1 (49.0, 59.1) 0.035 822 52.3 (46.9, 57.7)   Deworming 14 15.2 (10.1, 20.3) 398 24.8 (19.8, 29.8) 0.009 412 23.8 (19.2, 28.3)   Clean water and/or handwashing 50 67.0 (56.6, 77.4) 1111 79.7 (74.6, 84.9) 0.022 1161 78.4 (73.7, 83.1)   Menstruation 52 69.6 (56.8, 82.4) 1055 74.8 (71.0, 78.5) 0.426 1107 74.2 (70.6, 77.8)   Eating iron-rich foods 31 39.7 (32.1, 47.3) 729 48.6 (42.2, 55.0) 0.132 760 47.7 (42.2, 57.8)   Avoiding eating soil or clay 21 19.1 (4.4, 33.8) 527 32.1 (26.0, 38.1) 0.130 548 30.7 (24.7, 36.7) . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . Student characteristic . n . % (95% CI) . n . % (95% CI) . P value2 . n . % (95% CI) . Demographics  Age, y 0.079   10–14 8 2.7 (0.0, 6.6) 303 8.4 (4.1, 12.7) 311 7.6 (3.8, 11.5)   15–19 65 97.3 (93.4, 100) 1011 91.6 (87.3, 95.9) 1076 92.2 (88.2, 96.1)  Reached menarche 67 98.0 (95.5, 100) 1175 96.1 (94.3, 98.0) 0.272 1242 96.3 (94.6, 98.0)  Wealth tertile 0.229   High 31 48.5 (36.6, 60.4) 438 40.3 (32.2, 48.4) 469 41.2 (33.3, 49.0)   Middle 24 31.4 (26.8, 36.0) 439 31.9 (27.5, 36.3) 463 31.9 (28.1, 35.7)   Low 18 20.1 (8.1, 32.1) 437 27.7 (20.9, 34.6) 455 26.9 (20.1, 33.7) Knowledge, attitudes, and practices  Knowledge of anemia3 0.981   High 18 31.1 (5.8, 56.3) 369 32.1 (25.5, 38.7) 387 32.0 (25.0, 39.0)   Middle 25 33.6 (24.5, 42.6) 436 34.3 (29.5, 39.1) 461 34.2 (29.8, 38.7)   Low 30 35.4 (17.6, 53.1) 509 33.6 (28.3, 38.8) 539 33.7 (28.4, 39.1)  Knowledge of IFA4 0.036   High 12 23.2 (11.3, 35.2) 452 37.7 (32.0, 43.4) 464 36.6 (30.8, 42.3)   Middle 14 23.7 (12.3, 35.0) 324 23.5 (20.3, 26.8) 338 23.5 (20.2, 26.9)   Low 29 53.1 (39.3, 66.9) 528 38.8 (32.0, 45.5) 557 39.9 (33.2, 46.6) IFA program experiences  Took an IFA tablet on an empty stomach — — 361 26.9 (20.5, 33.4) — — —  Opinion of IFA tablets   Likes IFA tablets5 — — 1212 87.8 (83.3, 92.3) — — —   Does not like IFA tablets5 — — 75 11.3 (6.6, 15.9) — — —  Changes upon taking IFA tablets6   Desirable changes upon taking IFA tablets — — 594 37.9 (29.2, 46.7) — — —   Undesirable changes upon taking IFA tablets — — 322 26.9 (21.2, 32.6) — — —  Shared IFA tablet experiences with friends/family — — 714 52.5 (45.4, 59.6) — — —  Able to make up a missed IFA distribution — — 1111 76.9 (69.9, 83.8) — — —  Received IFA counseling 7 11.7 (2.2, 21.3) 298 18.4 (13.8, 23.0) 0.260 305 17.7 (13.4, 22.0)  Received health education 29 42.1 (25.4, 58.7) 702 46.4 (40.4, 52.4) 0.631 731 46.0 (40.4, 51.6)  Topics of health education sessions received in previous year   Anemia 27 34.1 (23.2, 45.0) 679 43.0 (37.3, 48.6) 0.187 706 42.1 (37.1, 47.0)   Malaria 31 36.7 (20.0, 53.4) 791 54.1 (49.0, 59.1) 0.035 822 52.3 (46.9, 57.7)   Deworming 14 15.2 (10.1, 20.3) 398 24.8 (19.8, 29.8) 0.009 412 23.8 (19.2, 28.3)   Clean water and/or handwashing 50 67.0 (56.6, 77.4) 1111 79.7 (74.6, 84.9) 0.022 1161 78.4 (73.7, 83.1)   Menstruation 52 69.6 (56.8, 82.4) 1055 74.8 (71.0, 78.5) 0.426 1107 74.2 (70.6, 77.8)   Eating iron-rich foods 31 39.7 (32.1, 47.3) 729 48.6 (42.2, 55.0) 0.132 760 47.7 (42.2, 57.8)   Avoiding eating soil or clay 21 19.1 (4.4, 33.8) 527 32.1 (26.0, 38.1) 0.130 548 30.7 (24.7, 36.7) 1 Estimates are weighted at the student level to be representative of eligible girls in the school. IFA, iron-folic acid. 2 Rao–Scott chi-square testing the difference between never consumed and consumed ≥1 IFA tablet. 95% CIs and P values for student characteristics are based on Taylor series variance estimates to account for the complex sampling design. 3 Knowledge of anemia was calculated using a composite score of correct responses to questions about anemia (awareness, signs, causes, and effects) and divided into tertiles before weights were applied. 4 Knowledge of IFA was calculated using a composite score of correct responses to questions about IFA (awareness and effects) and divided into tertiles before weights were applied; 28 missing observations. 5 Remaining participants reported neutral feelings about IFA tablets. 6 Remaining participants noticed no changes. Open in new tab Table 2 presents results of the bivariate analysis of school-level characteristics by self-reported consumption of IFA tablets. Among girls who had ever consumed an IFA tablet, 96% attended a school that kept a written record of their supply of IFA tablets, whereas only 70% of girls who reported never taking an IFA tablet attended a school that recorded their IFA supply appropriately (P < 0.01). Educator training on the GIFTS program was associated with consuming ≥1 IFA tablet compared with never consuming (67% compared with 22%, P = 0.002). Most (98.5%) of the never consumers were in SHSs (P < 0.01). School size was also a factor in the proportion of girls who never consumed an IFA tablet. For example, the largest schools (>500 girls) had 84% of the never consumers and only 61% of the consumers (P < 0.05). Nearly 70% of consumers attended schools where the educator respondent reported issues with girls refusing IFA tablets, whereas >99% of never consumers attended schools reporting the same (P < 0.05). Fifty-five percent of consumers attended schools where the educator respondent thought the program was too time-consuming, whereas only 2% of never consumers attended a school where this perception was reported (P < 0.01). Fewer than half (48.3%) of girls who never consumed an IFA tablet attended a school whose educator respondent thought the IFA program was important for the health of girls, whereas 82% of consumers attended a school whose educator respondent thought it was important (P < 0.01). TABLE 2 School-level characteristics by self-reported consumption of IFA tablets over 1 school year among adolescent girls in Ghanaian schools in the Northern and Volta regions, 20171 . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . School-level characteristic2 . n . % (95% CI) . n . % (95% CI) . P value . n . % (95% CI) . Government systems and supply of tablets  Experienced stock outs of IFA tablets 0 — 89 1.8 (0.0, 4.1) — 89 1.6 (0.0, 3.6)  Missed a distribution due to stock out 0 — 34 0.4 (0.0, 1.1) — 34 0.4 (0.0, 1.0)  Storage of IFA tablets   At the school 73 100 1207 83.2 (64.1, 100) 1280 84.9 (67.4, 100)   At someone's home 0 — 107 16.8 (0.0, 35.9) — 107 15.1 (0.0, 32.6)  Records IFA tablet supply appropriately 45 70.3 (24.7, 100) 1189 96.1 (92.3, 99.9) 0.011 1234 93.4 (86.5, 100)  Educator received program training 22 21.9 (0.0, 51.8) 954 67.0 (48.3, 85.6) 0.002 976 62.3 (42.8, 81.7) School characteristics  School level <0.001   Junior high 8 1.5 (0.0, 3.6) 700 15.1 (6.9, 23.3) 708 13.7 (6.3, 21.1)   Senior high/vocational 65 98.5 (96.4, 100) 614 84.9 (76.7, 93.1) 679 86.3 (78.9, 93.7)  Enrollment, girls 0.030   <100 6 0.8 (0.0, 2.2) 498 7.3 (3.0, 11.5) 504 6.6 (2.7, 10.4)   100–299 2 0.7 (0.0, 2.0) 443 20.6 (9.1, 32.0) 445 18.5 (8.1, 28.9)   300–500 18 15.0 (0.0, 39.1) 111 11.6 (1.0, 22.2) 129 11.9 (1.3, 22.5)   >500 47 83.5 (58.9, 100) 262 60.6 (43.1, 78.1) 309 63.0 (46.5, 79.5)  Location 0.346   Rural 14 8.5 (0.0, 20.6) 732 30.8 (14.2, 47.4) 746 28.5 (12.9, 44.1)   Peri-urban 42 55.0 (8.5, 100) 352 45.7 (25.0, 66.4) 394 46.7 (26.5, 66.8)   Urban 17 36.6 (0.0, 83.8) 230 23.5 (6.7, 40.4) 247 24.9 (6.9, 42.8) Resources and capacity  Functioning toilets/latrines 66 86.4 (59.7, 100) 1088 90.3 (81.7, 98.8) 0.668 1154 89.8 (80.1, 99.6)  Functioning handwashing stations 29 42.7 (0.0, 89.2) 937 69.0 (49.7, 88.3) 0.247 966 66.2 (47.2, 85.3)  Active health clubs 25 42.3 (0.0, 89.0) 613 56.7 (37.4, 76.0) 0.530 638 55.2 (36.0, 74.4)  Previous year's activities (school perspective)   Nutrition counseling 49 48.4 (2.0, 94.7) 599 39.2 (19.0, 59.4) 0.692 648 40.1 (20.5, 59.8)   Deworming 29 39.0 (0.0, 84.5) 372 45.9 (25.0, 66.5) 0.770 401 45.0 (24.8, 65.3)   Anemia screening 33 52.8 (6.5, 99.1) 568 29.9 (12.6, 47.3) 0.224 601 32.3 (13.8, 50.9)   Malaria control counseling 51 77.6 (49.0, 100) 706 55.4 (35.6, 75.2) 0.176 757 57.7 (38.6, 76.9)   Counseling on anemia 61 67.4 (19.6, 100) 684 42.0 (22.5, 61.5) 0.311 745 44.7 (25.2, 64.1)   Counseling on IFA tablets 61 67.4 (19.6, 100) 890 50.4 (30.1, 70.7) 0.495 951 52.2 (32.0, 72.3)  Frequency of health sessions —   More than once per month 7 1.1 (0.0, 2.9) 215 5.0 (0.7, 9.3) 222 4.6 (0.7, 8.5)   Once per month 15 35.8 (0.0, 83.2) 445 23.0 (8.1, 37.8) 460 24.3 (7.7, 40.9)   Once or twice per term 51 63.1 (15.9, 100) 602 70.0 (53.7, 85.9) 653 69.1 (51.9, 86.3)   Once or twice per year 0 — 52 2.3 (0.0, 5.9) 52 2.0 (0.0, 5.3)  Average time spent giving health education sessions 0.045    <30 min 1 0.4 (0.0, 1.2) 319 30.6 (10.0, 51.2) 320 27.5 (8.1, 46.8)    30–60 min 60 67.0 (19.4, 100) 863 52.5 (32.0, 73.0) 923 54.0 (33.7, 74.3)    >1 h 12 32.6 (0.0, 80.4) 132 16.9 (1.6, 32.1) 144 18.5 (1.5, 35.5) Perceptions and experiences  Experienced difficulties implementingIFA program 29 34.3 (0.0, 72.5) 637 65.7 (46.4, 85.0) 0.080 637 65.7 (46.4, 85.0)  Experienced difficulties with IFAprogram from community members 15 10.1 (0.0, 24.3) 180 11.1 (0.0, 22.5) 0.858 195 11.0 (0.0, 22.2)  Girls had concerns about taking IFA 51 71.6 (25.7, 100) 905 68.3 (47.6, 89.0) 0.894 956 68.6 (49.0, 88.2)  Girls refused to consume IFA tablets 72 99.1 (97.0, 100) 880 69.7 (49.0, 90.4) 0.024 952 72.7 (53.3, 92.2)  Thoughts on IFA program   Difficult to implement 15 33.6 (0.0, 81.2) 200 36.5 (14.8, 58.2) 0.903 215 36.2 (14.9, 57.5)   Too time consuming 2 2.1 (0.0, 6.1) 546 54.9 (35.3, 74.5) 0.004 548 49.4 (29.4, 69.4)   Important for the health of girls 40 48.3 (1.9, 94.7) 1048 82.3 (69.2, 95.4) 0.010 1088 78.8 (62.6, 95.0)  Educator knowledge of anemia3 0.641   High 19 46.3 (0.0, 93.2) 445 31.1 (12.4, 49.8) 464 32.7 (13.3, 52.0)   Middle 14 11.7 (0.0, 34.4) 392 24.9 (9.6, 40.2) 406 23.5 (9.1, 38.0)   Low 40 42.0 (0.0, 87.6) 477 44.0 (23.4, 64.6) 517 43.8 (23.8, 63.8) . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . School-level characteristic2 . n . % (95% CI) . n . % (95% CI) . P value . n . % (95% CI) . Government systems and supply of tablets  Experienced stock outs of IFA tablets 0 — 89 1.8 (0.0, 4.1) — 89 1.6 (0.0, 3.6)  Missed a distribution due to stock out 0 — 34 0.4 (0.0, 1.1) — 34 0.4 (0.0, 1.0)  Storage of IFA tablets   At the school 73 100 1207 83.2 (64.1, 100) 1280 84.9 (67.4, 100)   At someone's home 0 — 107 16.8 (0.0, 35.9) — 107 15.1 (0.0, 32.6)  Records IFA tablet supply appropriately 45 70.3 (24.7, 100) 1189 96.1 (92.3, 99.9) 0.011 1234 93.4 (86.5, 100)  Educator received program training 22 21.9 (0.0, 51.8) 954 67.0 (48.3, 85.6) 0.002 976 62.3 (42.8, 81.7) School characteristics  School level <0.001   Junior high 8 1.5 (0.0, 3.6) 700 15.1 (6.9, 23.3) 708 13.7 (6.3, 21.1)   Senior high/vocational 65 98.5 (96.4, 100) 614 84.9 (76.7, 93.1) 679 86.3 (78.9, 93.7)  Enrollment, girls 0.030   <100 6 0.8 (0.0, 2.2) 498 7.3 (3.0, 11.5) 504 6.6 (2.7, 10.4)   100–299 2 0.7 (0.0, 2.0) 443 20.6 (9.1, 32.0) 445 18.5 (8.1, 28.9)   300–500 18 15.0 (0.0, 39.1) 111 11.6 (1.0, 22.2) 129 11.9 (1.3, 22.5)   >500 47 83.5 (58.9, 100) 262 60.6 (43.1, 78.1) 309 63.0 (46.5, 79.5)  Location 0.346   Rural 14 8.5 (0.0, 20.6) 732 30.8 (14.2, 47.4) 746 28.5 (12.9, 44.1)   Peri-urban 42 55.0 (8.5, 100) 352 45.7 (25.0, 66.4) 394 46.7 (26.5, 66.8)   Urban 17 36.6 (0.0, 83.8) 230 23.5 (6.7, 40.4) 247 24.9 (6.9, 42.8) Resources and capacity  Functioning toilets/latrines 66 86.4 (59.7, 100) 1088 90.3 (81.7, 98.8) 0.668 1154 89.8 (80.1, 99.6)  Functioning handwashing stations 29 42.7 (0.0, 89.2) 937 69.0 (49.7, 88.3) 0.247 966 66.2 (47.2, 85.3)  Active health clubs 25 42.3 (0.0, 89.0) 613 56.7 (37.4, 76.0) 0.530 638 55.2 (36.0, 74.4)  Previous year's activities (school perspective)   Nutrition counseling 49 48.4 (2.0, 94.7) 599 39.2 (19.0, 59.4) 0.692 648 40.1 (20.5, 59.8)   Deworming 29 39.0 (0.0, 84.5) 372 45.9 (25.0, 66.5) 0.770 401 45.0 (24.8, 65.3)   Anemia screening 33 52.8 (6.5, 99.1) 568 29.9 (12.6, 47.3) 0.224 601 32.3 (13.8, 50.9)   Malaria control counseling 51 77.6 (49.0, 100) 706 55.4 (35.6, 75.2) 0.176 757 57.7 (38.6, 76.9)   Counseling on anemia 61 67.4 (19.6, 100) 684 42.0 (22.5, 61.5) 0.311 745 44.7 (25.2, 64.1)   Counseling on IFA tablets 61 67.4 (19.6, 100) 890 50.4 (30.1, 70.7) 0.495 951 52.2 (32.0, 72.3)  Frequency of health sessions —   More than once per month 7 1.1 (0.0, 2.9) 215 5.0 (0.7, 9.3) 222 4.6 (0.7, 8.5)   Once per month 15 35.8 (0.0, 83.2) 445 23.0 (8.1, 37.8) 460 24.3 (7.7, 40.9)   Once or twice per term 51 63.1 (15.9, 100) 602 70.0 (53.7, 85.9) 653 69.1 (51.9, 86.3)   Once or twice per year 0 — 52 2.3 (0.0, 5.9) 52 2.0 (0.0, 5.3)  Average time spent giving health education sessions 0.045    <30 min 1 0.4 (0.0, 1.2) 319 30.6 (10.0, 51.2) 320 27.5 (8.1, 46.8)    30–60 min 60 67.0 (19.4, 100) 863 52.5 (32.0, 73.0) 923 54.0 (33.7, 74.3)    >1 h 12 32.6 (0.0, 80.4) 132 16.9 (1.6, 32.1) 144 18.5 (1.5, 35.5) Perceptions and experiences  Experienced difficulties implementingIFA program 29 34.3 (0.0, 72.5) 637 65.7 (46.4, 85.0) 0.080 637 65.7 (46.4, 85.0)  Experienced difficulties with IFAprogram from community members 15 10.1 (0.0, 24.3) 180 11.1 (0.0, 22.5) 0.858 195 11.0 (0.0, 22.2)  Girls had concerns about taking IFA 51 71.6 (25.7, 100) 905 68.3 (47.6, 89.0) 0.894 956 68.6 (49.0, 88.2)  Girls refused to consume IFA tablets 72 99.1 (97.0, 100) 880 69.7 (49.0, 90.4) 0.024 952 72.7 (53.3, 92.2)  Thoughts on IFA program   Difficult to implement 15 33.6 (0.0, 81.2) 200 36.5 (14.8, 58.2) 0.903 215 36.2 (14.9, 57.5)   Too time consuming 2 2.1 (0.0, 6.1) 546 54.9 (35.3, 74.5) 0.004 548 49.4 (29.4, 69.4)   Important for the health of girls 40 48.3 (1.9, 94.7) 1048 82.3 (69.2, 95.4) 0.010 1088 78.8 (62.6, 95.0)  Educator knowledge of anemia3 0.641   High 19 46.3 (0.0, 93.2) 445 31.1 (12.4, 49.8) 464 32.7 (13.3, 52.0)   Middle 14 11.7 (0.0, 34.4) 392 24.9 (9.6, 40.2) 406 23.5 (9.1, 38.0)   Low 40 42.0 (0.0, 87.6) 477 44.0 (23.4, 64.6) 517 43.8 (23.8, 63.8) 1 Estimates are weighted at the student level to be representative of eligible girls in the school. IFA, iron and folic acid. 2 Rao–Scott chi-square testing the difference between never consumed and consumed ≥1 IFA tablet. 3 Educator knowledge of anemia was calculated using a composite score of correct responses to questions about anemia (awareness, signs, causes, and effects) and divided into tertiles before weights were applied. Open in new tab TABLE 2 School-level characteristics by self-reported consumption of IFA tablets over 1 school year among adolescent girls in Ghanaian schools in the Northern and Volta regions, 20171 . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . School-level characteristic2 . n . % (95% CI) . n . % (95% CI) . P value . n . % (95% CI) . Government systems and supply of tablets  Experienced stock outs of IFA tablets 0 — 89 1.8 (0.0, 4.1) — 89 1.6 (0.0, 3.6)  Missed a distribution due to stock out 0 — 34 0.4 (0.0, 1.1) — 34 0.4 (0.0, 1.0)  Storage of IFA tablets   At the school 73 100 1207 83.2 (64.1, 100) 1280 84.9 (67.4, 100)   At someone's home 0 — 107 16.8 (0.0, 35.9) — 107 15.1 (0.0, 32.6)  Records IFA tablet supply appropriately 45 70.3 (24.7, 100) 1189 96.1 (92.3, 99.9) 0.011 1234 93.4 (86.5, 100)  Educator received program training 22 21.9 (0.0, 51.8) 954 67.0 (48.3, 85.6) 0.002 976 62.3 (42.8, 81.7) School characteristics  School level <0.001   Junior high 8 1.5 (0.0, 3.6) 700 15.1 (6.9, 23.3) 708 13.7 (6.3, 21.1)   Senior high/vocational 65 98.5 (96.4, 100) 614 84.9 (76.7, 93.1) 679 86.3 (78.9, 93.7)  Enrollment, girls 0.030   <100 6 0.8 (0.0, 2.2) 498 7.3 (3.0, 11.5) 504 6.6 (2.7, 10.4)   100–299 2 0.7 (0.0, 2.0) 443 20.6 (9.1, 32.0) 445 18.5 (8.1, 28.9)   300–500 18 15.0 (0.0, 39.1) 111 11.6 (1.0, 22.2) 129 11.9 (1.3, 22.5)   >500 47 83.5 (58.9, 100) 262 60.6 (43.1, 78.1) 309 63.0 (46.5, 79.5)  Location 0.346   Rural 14 8.5 (0.0, 20.6) 732 30.8 (14.2, 47.4) 746 28.5 (12.9, 44.1)   Peri-urban 42 55.0 (8.5, 100) 352 45.7 (25.0, 66.4) 394 46.7 (26.5, 66.8)   Urban 17 36.6 (0.0, 83.8) 230 23.5 (6.7, 40.4) 247 24.9 (6.9, 42.8) Resources and capacity  Functioning toilets/latrines 66 86.4 (59.7, 100) 1088 90.3 (81.7, 98.8) 0.668 1154 89.8 (80.1, 99.6)  Functioning handwashing stations 29 42.7 (0.0, 89.2) 937 69.0 (49.7, 88.3) 0.247 966 66.2 (47.2, 85.3)  Active health clubs 25 42.3 (0.0, 89.0) 613 56.7 (37.4, 76.0) 0.530 638 55.2 (36.0, 74.4)  Previous year's activities (school perspective)   Nutrition counseling 49 48.4 (2.0, 94.7) 599 39.2 (19.0, 59.4) 0.692 648 40.1 (20.5, 59.8)   Deworming 29 39.0 (0.0, 84.5) 372 45.9 (25.0, 66.5) 0.770 401 45.0 (24.8, 65.3)   Anemia screening 33 52.8 (6.5, 99.1) 568 29.9 (12.6, 47.3) 0.224 601 32.3 (13.8, 50.9)   Malaria control counseling 51 77.6 (49.0, 100) 706 55.4 (35.6, 75.2) 0.176 757 57.7 (38.6, 76.9)   Counseling on anemia 61 67.4 (19.6, 100) 684 42.0 (22.5, 61.5) 0.311 745 44.7 (25.2, 64.1)   Counseling on IFA tablets 61 67.4 (19.6, 100) 890 50.4 (30.1, 70.7) 0.495 951 52.2 (32.0, 72.3)  Frequency of health sessions —   More than once per month 7 1.1 (0.0, 2.9) 215 5.0 (0.7, 9.3) 222 4.6 (0.7, 8.5)   Once per month 15 35.8 (0.0, 83.2) 445 23.0 (8.1, 37.8) 460 24.3 (7.7, 40.9)   Once or twice per term 51 63.1 (15.9, 100) 602 70.0 (53.7, 85.9) 653 69.1 (51.9, 86.3)   Once or twice per year 0 — 52 2.3 (0.0, 5.9) 52 2.0 (0.0, 5.3)  Average time spent giving health education sessions 0.045    <30 min 1 0.4 (0.0, 1.2) 319 30.6 (10.0, 51.2) 320 27.5 (8.1, 46.8)    30–60 min 60 67.0 (19.4, 100) 863 52.5 (32.0, 73.0) 923 54.0 (33.7, 74.3)    >1 h 12 32.6 (0.0, 80.4) 132 16.9 (1.6, 32.1) 144 18.5 (1.5, 35.5) Perceptions and experiences  Experienced difficulties implementingIFA program 29 34.3 (0.0, 72.5) 637 65.7 (46.4, 85.0) 0.080 637 65.7 (46.4, 85.0)  Experienced difficulties with IFAprogram from community members 15 10.1 (0.0, 24.3) 180 11.1 (0.0, 22.5) 0.858 195 11.0 (0.0, 22.2)  Girls had concerns about taking IFA 51 71.6 (25.7, 100) 905 68.3 (47.6, 89.0) 0.894 956 68.6 (49.0, 88.2)  Girls refused to consume IFA tablets 72 99.1 (97.0, 100) 880 69.7 (49.0, 90.4) 0.024 952 72.7 (53.3, 92.2)  Thoughts on IFA program   Difficult to implement 15 33.6 (0.0, 81.2) 200 36.5 (14.8, 58.2) 0.903 215 36.2 (14.9, 57.5)   Too time consuming 2 2.1 (0.0, 6.1) 546 54.9 (35.3, 74.5) 0.004 548 49.4 (29.4, 69.4)   Important for the health of girls 40 48.3 (1.9, 94.7) 1048 82.3 (69.2, 95.4) 0.010 1088 78.8 (62.6, 95.0)  Educator knowledge of anemia3 0.641   High 19 46.3 (0.0, 93.2) 445 31.1 (12.4, 49.8) 464 32.7 (13.3, 52.0)   Middle 14 11.7 (0.0, 34.4) 392 24.9 (9.6, 40.2) 406 23.5 (9.1, 38.0)   Low 40 42.0 (0.0, 87.6) 477 44.0 (23.4, 64.6) 517 43.8 (23.8, 63.8) . Never consumed IFA (n = 73) . Consumed ≥1 IFA tablet (n = 1314) . Overall (n = 1387) . School-level characteristic2 . n . % (95% CI) . n . % (95% CI) . P value . n . % (95% CI) . Government systems and supply of tablets  Experienced stock outs of IFA tablets 0 — 89 1.8 (0.0, 4.1) — 89 1.6 (0.0, 3.6)  Missed a distribution due to stock out 0 — 34 0.4 (0.0, 1.1) — 34 0.4 (0.0, 1.0)  Storage of IFA tablets   At the school 73 100 1207 83.2 (64.1, 100) 1280 84.9 (67.4, 100)   At someone's home 0 — 107 16.8 (0.0, 35.9) — 107 15.1 (0.0, 32.6)  Records IFA tablet supply appropriately 45 70.3 (24.7, 100) 1189 96.1 (92.3, 99.9) 0.011 1234 93.4 (86.5, 100)  Educator received program training 22 21.9 (0.0, 51.8) 954 67.0 (48.3, 85.6) 0.002 976 62.3 (42.8, 81.7) School characteristics  School level <0.001   Junior high 8 1.5 (0.0, 3.6) 700 15.1 (6.9, 23.3) 708 13.7 (6.3, 21.1)   Senior high/vocational 65 98.5 (96.4, 100) 614 84.9 (76.7, 93.1) 679 86.3 (78.9, 93.7)  Enrollment, girls 0.030   <100 6 0.8 (0.0, 2.2) 498 7.3 (3.0, 11.5) 504 6.6 (2.7, 10.4)   100–299 2 0.7 (0.0, 2.0) 443 20.6 (9.1, 32.0) 445 18.5 (8.1, 28.9)   300–500 18 15.0 (0.0, 39.1) 111 11.6 (1.0, 22.2) 129 11.9 (1.3, 22.5)   >500 47 83.5 (58.9, 100) 262 60.6 (43.1, 78.1) 309 63.0 (46.5, 79.5)  Location 0.346   Rural 14 8.5 (0.0, 20.6) 732 30.8 (14.2, 47.4) 746 28.5 (12.9, 44.1)   Peri-urban 42 55.0 (8.5, 100) 352 45.7 (25.0, 66.4) 394 46.7 (26.5, 66.8)   Urban 17 36.6 (0.0, 83.8) 230 23.5 (6.7, 40.4) 247 24.9 (6.9, 42.8) Resources and capacity  Functioning toilets/latrines 66 86.4 (59.7, 100) 1088 90.3 (81.7, 98.8) 0.668 1154 89.8 (80.1, 99.6)  Functioning handwashing stations 29 42.7 (0.0, 89.2) 937 69.0 (49.7, 88.3) 0.247 966 66.2 (47.2, 85.3)  Active health clubs 25 42.3 (0.0, 89.0) 613 56.7 (37.4, 76.0) 0.530 638 55.2 (36.0, 74.4)  Previous year's activities (school perspective)   Nutrition counseling 49 48.4 (2.0, 94.7) 599 39.2 (19.0, 59.4) 0.692 648 40.1 (20.5, 59.8)   Deworming 29 39.0 (0.0, 84.5) 372 45.9 (25.0, 66.5) 0.770 401 45.0 (24.8, 65.3)   Anemia screening 33 52.8 (6.5, 99.1) 568 29.9 (12.6, 47.3) 0.224 601 32.3 (13.8, 50.9)   Malaria control counseling 51 77.6 (49.0, 100) 706 55.4 (35.6, 75.2) 0.176 757 57.7 (38.6, 76.9)   Counseling on anemia 61 67.4 (19.6, 100) 684 42.0 (22.5, 61.5) 0.311 745 44.7 (25.2, 64.1)   Counseling on IFA tablets 61 67.4 (19.6, 100) 890 50.4 (30.1, 70.7) 0.495 951 52.2 (32.0, 72.3)  Frequency of health sessions —   More than once per month 7 1.1 (0.0, 2.9) 215 5.0 (0.7, 9.3) 222 4.6 (0.7, 8.5)   Once per month 15 35.8 (0.0, 83.2) 445 23.0 (8.1, 37.8) 460 24.3 (7.7, 40.9)   Once or twice per term 51 63.1 (15.9, 100) 602 70.0 (53.7, 85.9) 653 69.1 (51.9, 86.3)   Once or twice per year 0 — 52 2.3 (0.0, 5.9) 52 2.0 (0.0, 5.3)  Average time spent giving health education sessions 0.045    <30 min 1 0.4 (0.0, 1.2) 319 30.6 (10.0, 51.2) 320 27.5 (8.1, 46.8)    30–60 min 60 67.0 (19.4, 100) 863 52.5 (32.0, 73.0) 923 54.0 (33.7, 74.3)    >1 h 12 32.6 (0.0, 80.4) 132 16.9 (1.6, 32.1) 144 18.5 (1.5, 35.5) Perceptions and experiences  Experienced difficulties implementingIFA program 29 34.3 (0.0, 72.5) 637 65.7 (46.4, 85.0) 0.080 637 65.7 (46.4, 85.0)  Experienced difficulties with IFAprogram from community members 15 10.1 (0.0, 24.3) 180 11.1 (0.0, 22.5) 0.858 195 11.0 (0.0, 22.2)  Girls had concerns about taking IFA 51 71.6 (25.7, 100) 905 68.3 (47.6, 89.0) 0.894 956 68.6 (49.0, 88.2)  Girls refused to consume IFA tablets 72 99.1 (97.0, 100) 880 69.7 (49.0, 90.4) 0.024 952 72.7 (53.3, 92.2)  Thoughts on IFA program   Difficult to implement 15 33.6 (0.0, 81.2) 200 36.5 (14.8, 58.2) 0.903 215 36.2 (14.9, 57.5)   Too time consuming 2 2.1 (0.0, 6.1) 546 54.9 (35.3, 74.5) 0.004 548 49.4 (29.4, 69.4)   Important for the health of girls 40 48.3 (1.9, 94.7) 1048 82.3 (69.2, 95.4) 0.010 1088 78.8 (62.6, 95.0)  Educator knowledge of anemia3 0.641   High 19 46.3 (0.0, 93.2) 445 31.1 (12.4, 49.8) 464 32.7 (13.3, 52.0)   Middle 14 11.7 (0.0, 34.4) 392 24.9 (9.6, 40.2) 406 23.5 (9.1, 38.0)   Low 40 42.0 (0.0, 87.6) 477 44.0 (23.4, 64.6) 517 43.8 (23.8, 63.8) 1 Estimates are weighted at the student level to be representative of eligible girls in the school. IFA, iron and folic acid. 2 Rao–Scott chi-square testing the difference between never consumed and consumed ≥1 IFA tablet. 3 Educator knowledge of anemia was calculated using a composite score of correct responses to questions about anemia (awareness, signs, causes, and effects) and divided into tertiles before weights were applied. Open in new tab Table 3 shows the results of the adjusted linear mixed model among consumers of IFA tablets adjusted for student- and school-level factors. The ability to make up a missed IFA distribution was the only statistically significant student-level predictor. JHS compared with SHS, educator training on the GIFTS program, and educator experiences and perceptions of the program were significant school-level predictors. Girls who reported the ability to make up missed IFA distributions consumed a mean of 1.44 (P < 0.01) more tablets over 1 school year than those who did not have that ability. Girls from schools with an educator who received GIFTS program implementation training consumed an adjusted mean of 7.57 (P < 0.01) more tablets. Girls in JHSs consumed an adjusted mean of 5.76 (P < 0.05) more tablets than girls in SHSs. In schools reporting issues with refusals to consume IFA tablets, girls consumed an adjusted mean of 4.02 (P < 0.05) fewer tablets. In schools where the school educator respondent thought the program was difficult to implement and too time-consuming, girls consumed an adjusted mean of 6.92 (P < 0.01) and 4.41 (P < 0.05) fewer tablets over 1 school year, respectively. A model including both consumers and nonconsumers of IFA tablets had identical significant school-level predictors of total IFA consumption and no significant individual-level predictors (data not shown). Over 75% of the variance around the total number of IFA tablets consumed over 1 school year was explained at the school level, with ∼25% coming from factors at the student level. TABLE 3 Predictors of the total number of IFA tablets consumed over 1 school year, among consumers of IFA tablets among adolescent girls in Ghanaian schools in the Northern and Volta regions, 20171 Variable . Adjusted estimate (95% CI) . Student  Demographics   Age, standardized 0.00 (−0.40, 0.39)   Wealth tertile    High −0.40 (−1.10, 0.30)    Middle 0.51 (−0.17, 1.18)    Low Ref.  Knowledge, attitudes, and practices   Knowledge of anemia, standardized2 −0.19 (−0.48, 0.09)   Knowledge of IFA tablets, standardized2 0.03 (−0.26, 0.32)   Has ever consumed an IFA tablet on an empty stomach −0.11 (−0.69, 0.47)   Likes IFA 0.50 (−0.35, 1.34)   Changes noticed upon taking IFA    No changes Ref.    Desirable changes 0.44 (−0.13, 1.01)    Undesirable changes −0.36 (−0.94, 0.23)   Shared IFA tablet experiences with friends/family 0.04 (−0.50, 0.57)   Received health education 0.32 (−0.29, 0.94)   Received IFA counseling −0.14 (−0.95, 0.66)   Able to make up a missed IFA distribution 1.44 (0.83, 2.01)*** School  Government systems   Missed a distribution due to stock out −0.30 (−11.37, 10.77)   Storage of IFA tablets    At the school 2.06 (−6.19, 10.31)    At someone's home Ref.   Records IFA tablet supply appropriately 1.51 (−4.13, 7.14)   Educator respondent received training on the IFA program 7.57 (2.89, 12.24)**  Resources and capacity   School level    Junior high 5.76 (0.07, 11.45)*    Senior high/vocational Ref.   Enrollment, standardized −1.16 (−4.24, 1.92)   Location    Rural 1.85 (−3.13, 6.83)    Peri-urban 1.35 (−3.33, 6.03)    Urban Ref.   Active health clubs 0.73 (−3.95, 5.41)   Previous year's activities    Nutrition counseling −0.08 (−3.73, 3.57)    Deworming 1.17 (−3.56, 5.90)    Anemia screening −0.84 (−5.76, 4.08)    Malaria control counseling 1.73 (−2.32, 5.78)    Counseling on anemia −6.26 (−14.33, 1.81)    Counseling on IFA tablets 0.16 (−5.87, 6.19)   Frequency of health sessions    More than once per month 3.24 (−3.87, 10.34)    Once per month Ref.    Once or twice per term 1.53 (−2.61, 5.68)    Once or twice per year −2.09 (−10.96, 6.77)   Average time spent giving health sessions    <30 min Ref.    30–60 min −1.47 (−5.72, 2.78)    >1 h 4.95 (−0.94, 10.85)  Perceptions and experiences   Experienced difficulties implementing IFA program 1.38 (−2.76, 5.52)   Experienced difficulties with IFA program from communitymembers −0.99 (−5.94, 3.96)   Girls had concerns about taking IFA tablets −0.55 (−5.53, 4.43)   Girls refused to consume IFA tablets −4.02 (−8.04, −0.01)*   Thoughts on IFA program    Difficult to implement −6.92 (−12.14, −1.70)**    Too time consuming −4.41 (−8.44, −0.37)*    Important for the health of girls 3.34 (−1.31, 7.98)   Educator knowledge of anemia, standardized3 −0.31 (−3.21, 2.59) Variable . Adjusted estimate (95% CI) . Student  Demographics   Age, standardized 0.00 (−0.40, 0.39)   Wealth tertile    High −0.40 (−1.10, 0.30)    Middle 0.51 (−0.17, 1.18)    Low Ref.  Knowledge, attitudes, and practices   Knowledge of anemia, standardized2 −0.19 (−0.48, 0.09)   Knowledge of IFA tablets, standardized2 0.03 (−0.26, 0.32)   Has ever consumed an IFA tablet on an empty stomach −0.11 (−0.69, 0.47)   Likes IFA 0.50 (−0.35, 1.34)   Changes noticed upon taking IFA    No changes Ref.    Desirable changes 0.44 (−0.13, 1.01)    Undesirable changes −0.36 (−0.94, 0.23)   Shared IFA tablet experiences with friends/family 0.04 (−0.50, 0.57)   Received health education 0.32 (−0.29, 0.94)   Received IFA counseling −0.14 (−0.95, 0.66)   Able to make up a missed IFA distribution 1.44 (0.83, 2.01)*** School  Government systems   Missed a distribution due to stock out −0.30 (−11.37, 10.77)   Storage of IFA tablets    At the school 2.06 (−6.19, 10.31)    At someone's home Ref.   Records IFA tablet supply appropriately 1.51 (−4.13, 7.14)   Educator respondent received training on the IFA program 7.57 (2.89, 12.24)**  Resources and capacity   School level    Junior high 5.76 (0.07, 11.45)*    Senior high/vocational Ref.   Enrollment, standardized −1.16 (−4.24, 1.92)   Location    Rural 1.85 (−3.13, 6.83)    Peri-urban 1.35 (−3.33, 6.03)    Urban Ref.   Active health clubs 0.73 (−3.95, 5.41)   Previous year's activities    Nutrition counseling −0.08 (−3.73, 3.57)    Deworming 1.17 (−3.56, 5.90)    Anemia screening −0.84 (−5.76, 4.08)    Malaria control counseling 1.73 (−2.32, 5.78)    Counseling on anemia −6.26 (−14.33, 1.81)    Counseling on IFA tablets 0.16 (−5.87, 6.19)   Frequency of health sessions    More than once per month 3.24 (−3.87, 10.34)    Once per month Ref.    Once or twice per term 1.53 (−2.61, 5.68)    Once or twice per year −2.09 (−10.96, 6.77)   Average time spent giving health sessions    <30 min Ref.    30–60 min −1.47 (−5.72, 2.78)    >1 h 4.95 (−0.94, 10.85)  Perceptions and experiences   Experienced difficulties implementing IFA program 1.38 (−2.76, 5.52)   Experienced difficulties with IFA program from communitymembers −0.99 (−5.94, 3.96)   Girls had concerns about taking IFA tablets −0.55 (−5.53, 4.43)   Girls refused to consume IFA tablets −4.02 (−8.04, −0.01)*   Thoughts on IFA program    Difficult to implement −6.92 (−12.14, −1.70)**    Too time consuming −4.41 (−8.44, −0.37)*    Important for the health of girls 3.34 (−1.31, 7.98)   Educator knowledge of anemia, standardized3 −0.31 (−3.21, 2.59) 1 n = 1231. Estimates are weighted at the student level to be representative of eligible girls in the school. Data from program registers. Sample size reduced owing to missing data. *P < 0.05; **P < 0.01; ***P < 0.001. A total of 1239 girls had consumed ≥1 IFA tablet as recorded on program registers: 1 outlier was excluded; there were 7 missing observations for ≥1 covariates including “Likes IFA,” “Received health education,” and “Received IFA counseling.” IFA, iron and folic acid. 2 Knowledge of anemia and IFA were calculated using composite scores of correct responses to questions about anemia and IFA, respectively. 3 Educator knowledge of anemia was calculated using a composite score of correct responses to questions about anemia (awareness, signs, causes, and effects). Open in new tab TABLE 3 Predictors of the total number of IFA tablets consumed over 1 school year, among consumers of IFA tablets among adolescent girls in Ghanaian schools in the Northern and Volta regions, 20171 Variable . Adjusted estimate (95% CI) . Student  Demographics   Age, standardized 0.00 (−0.40, 0.39)   Wealth tertile    High −0.40 (−1.10, 0.30)    Middle 0.51 (−0.17, 1.18)    Low Ref.  Knowledge, attitudes, and practices   Knowledge of anemia, standardized2 −0.19 (−0.48, 0.09)   Knowledge of IFA tablets, standardized2 0.03 (−0.26, 0.32)   Has ever consumed an IFA tablet on an empty stomach −0.11 (−0.69, 0.47)   Likes IFA 0.50 (−0.35, 1.34)   Changes noticed upon taking IFA    No changes Ref.    Desirable changes 0.44 (−0.13, 1.01)    Undesirable changes −0.36 (−0.94, 0.23)   Shared IFA tablet experiences with friends/family 0.04 (−0.50, 0.57)   Received health education 0.32 (−0.29, 0.94)   Received IFA counseling −0.14 (−0.95, 0.66)   Able to make up a missed IFA distribution 1.44 (0.83, 2.01)*** School  Government systems   Missed a distribution due to stock out −0.30 (−11.37, 10.77)   Storage of IFA tablets    At the school 2.06 (−6.19, 10.31)    At someone's home Ref.   Records IFA tablet supply appropriately 1.51 (−4.13, 7.14)   Educator respondent received training on the IFA program 7.57 (2.89, 12.24)**  Resources and capacity   School level    Junior high 5.76 (0.07, 11.45)*    Senior high/vocational Ref.   Enrollment, standardized −1.16 (−4.24, 1.92)   Location    Rural 1.85 (−3.13, 6.83)    Peri-urban 1.35 (−3.33, 6.03)    Urban Ref.   Active health clubs 0.73 (−3.95, 5.41)   Previous year's activities    Nutrition counseling −0.08 (−3.73, 3.57)    Deworming 1.17 (−3.56, 5.90)    Anemia screening −0.84 (−5.76, 4.08)    Malaria control counseling 1.73 (−2.32, 5.78)    Counseling on anemia −6.26 (−14.33, 1.81)    Counseling on IFA tablets 0.16 (−5.87, 6.19)   Frequency of health sessions    More than once per month 3.24 (−3.87, 10.34)    Once per month Ref.    Once or twice per term 1.53 (−2.61, 5.68)    Once or twice per year −2.09 (−10.96, 6.77)   Average time spent giving health sessions    <30 min Ref.    30–60 min −1.47 (−5.72, 2.78)    >1 h 4.95 (−0.94, 10.85)  Perceptions and experiences   Experienced difficulties implementing IFA program 1.38 (−2.76, 5.52)   Experienced difficulties with IFA program from communitymembers −0.99 (−5.94, 3.96)   Girls had concerns about taking IFA tablets −0.55 (−5.53, 4.43)   Girls refused to consume IFA tablets −4.02 (−8.04, −0.01)*   Thoughts on IFA program    Difficult to implement −6.92 (−12.14, −1.70)**    Too time consuming −4.41 (−8.44, −0.37)*    Important for the health of girls 3.34 (−1.31, 7.98)   Educator knowledge of anemia, standardized3 −0.31 (−3.21, 2.59) Variable . Adjusted estimate (95% CI) . Student  Demographics   Age, standardized 0.00 (−0.40, 0.39)   Wealth tertile    High −0.40 (−1.10, 0.30)    Middle 0.51 (−0.17, 1.18)    Low Ref.  Knowledge, attitudes, and practices   Knowledge of anemia, standardized2 −0.19 (−0.48, 0.09)   Knowledge of IFA tablets, standardized2 0.03 (−0.26, 0.32)   Has ever consumed an IFA tablet on an empty stomach −0.11 (−0.69, 0.47)   Likes IFA 0.50 (−0.35, 1.34)   Changes noticed upon taking IFA    No changes Ref.    Desirable changes 0.44 (−0.13, 1.01)    Undesirable changes −0.36 (−0.94, 0.23)   Shared IFA tablet experiences with friends/family 0.04 (−0.50, 0.57)   Received health education 0.32 (−0.29, 0.94)   Received IFA counseling −0.14 (−0.95, 0.66)   Able to make up a missed IFA distribution 1.44 (0.83, 2.01)*** School  Government systems   Missed a distribution due to stock out −0.30 (−11.37, 10.77)   Storage of IFA tablets    At the school 2.06 (−6.19, 10.31)    At someone's home Ref.   Records IFA tablet supply appropriately 1.51 (−4.13, 7.14)   Educator respondent received training on the IFA program 7.57 (2.89, 12.24)**  Resources and capacity   School level    Junior high 5.76 (0.07, 11.45)*    Senior high/vocational Ref.   Enrollment, standardized −1.16 (−4.24, 1.92)   Location    Rural 1.85 (−3.13, 6.83)    Peri-urban 1.35 (−3.33, 6.03)    Urban Ref.   Active health clubs 0.73 (−3.95, 5.41)   Previous year's activities    Nutrition counseling −0.08 (−3.73, 3.57)    Deworming 1.17 (−3.56, 5.90)    Anemia screening −0.84 (−5.76, 4.08)    Malaria control counseling 1.73 (−2.32, 5.78)    Counseling on anemia −6.26 (−14.33, 1.81)    Counseling on IFA tablets 0.16 (−5.87, 6.19)   Frequency of health sessions    More than once per month 3.24 (−3.87, 10.34)    Once per month Ref.    Once or twice per term 1.53 (−2.61, 5.68)    Once or twice per year −2.09 (−10.96, 6.77)   Average time spent giving health sessions    <30 min Ref.    30–60 min −1.47 (−5.72, 2.78)    >1 h 4.95 (−0.94, 10.85)  Perceptions and experiences   Experienced difficulties implementing IFA program 1.38 (−2.76, 5.52)   Experienced difficulties with IFA program from communitymembers −0.99 (−5.94, 3.96)   Girls had concerns about taking IFA tablets −0.55 (−5.53, 4.43)   Girls refused to consume IFA tablets −4.02 (−8.04, −0.01)*   Thoughts on IFA program    Difficult to implement −6.92 (−12.14, −1.70)**    Too time consuming −4.41 (−8.44, −0.37)*    Important for the health of girls 3.34 (−1.31, 7.98)   Educator knowledge of anemia, standardized3 −0.31 (−3.21, 2.59) 1 n = 1231. Estimates are weighted at the student level to be representative of eligible girls in the school. Data from program registers. Sample size reduced owing to missing data. *P < 0.05; **P < 0.01; ***P < 0.001. A total of 1239 girls had consumed ≥1 IFA tablet as recorded on program registers: 1 outlier was excluded; there were 7 missing observations for ≥1 covariates including “Likes IFA,” “Received health education,” and “Received IFA counseling.” IFA, iron and folic acid. 2 Knowledge of anemia and IFA were calculated using composite scores of correct responses to questions about anemia and IFA, respectively. 3 Educator knowledge of anemia was calculated using a composite score of correct responses to questions about anemia (awareness, signs, causes, and effects). Open in new tab Discussion In this study, we found that 90% of adolescent girls attending schools in 2 regions of Ghana had ever consumed an IFA tablet during the school year, and 56% consumed ≥15 weekly tablets during the year. The school was largely responsible for the variability in the total number of IFA tablets consumed, indicating that addressing school-level factors would be effective in improving adherence to the intervention. Other evaluations of similar interventions have highlighted contextual factors as predictors of adherence, but also pointed to student-level predictors such as sensitization, perceptions, and side effects (11, 21, 22). Importantly, previous studies have not quantified the relative contributions of factors to consumption of IFA tablets. Key strengths of our study are the representative nature of the study sample, longitudinally collected data for IFA consumption, and use of variance decomposition methods. The study is representative of 2234 schools and ∼400,000 schoolgirls in 2 regions of Ghana. The longitudinal collection of the number of IFA tablets consumed through direct observation may reduce recall bias in the outcome, and the multilevel model accounts for student- and school-level factors in the hierarchical structure in which they exist and enables interpretation of cluster effects. This analysis has several limitations. We were unable to examine ecological variables such as political and economic factors; perceptions, experiences, and motivations of other educators; and community attitudes and beliefs. Schools were majority co-ed with only 2 girls-only schools. Hence, social factors that affect program adherence identified in this work might be reflective of what one would encounter among girls from a typical co-ed school in Ghana but not necessarily representative of girls-only schools. We were also unable to collect attendance or school performance data that may have added to our understanding. There may also be social desirability and recall bias in the self-reported predictors at the student and school levels. There is excellent agreement between the self-reported and monitoring consumption data, suggesting that consumption may be accurately captured. However, 13% reported receiving an IFA tablet and throwing it away at some point during the school year, a source of error in the abstracted consumption data. The educator respondent for the school questionnaire, although in most cases the focal person for the program, did vary in some cases. Finally, there may be residual confounding for which the models were unable to account. School-level factors were prominent in the GLMM analysis, whereas student-level factors had little relation with IFA consumption. In adjusted models, the ability to make up a missed IFA distribution, educator training, and the educators’ experiences and perceptions were associated with the number of IFA tablets consumed. Having another opportunity to make up a missed IFA distribution may mitigate the effects of absenteeism on program adherence. Making up a missed distribution reflects both student-level agency in seeking a missed IFA tablet and school-level structures that tracked students and allowed for make-up distributions. Educator training on the program remained a key predictor of the number of IFA tablets consumed, even in adjusted models. The educator's experiences and perceptions of the program may influence their motivation and ultimately the performance of the program in their school. This is supported by research showing that knowledge of risks and benefits improves adherence to micronutrient interventions (23, 24). However, educator training was not associated with perceptions of the program on the school survey (data not shown). Negative perceptions of the difficulty and time burden suggest the need for other modifications at the school level that may also improve adherence, such as spreading the responsibility for distribution to other teachers or student leaders and streamlining monitoring. Adherence to IFA consumption was somewhat lower in the GIFTS program than in weekly IFA programs in India where complete adherence (1 tablet every week throughout the school year) was >80% in most regions, but ranged from 53% to 99% (7, 22). Bottlenecks at the levels of acceptability, accessibility, and contact may have been responsible for suboptimal adherence to the program (Figure 2). This study investigated the student- and school-level facilitators and barriers that may be responsible for creating or eliminating these bottlenecks. In bivariate analyses, the number of IFA tablets consumed differed by girls’ age, household wealth, and level of school (JHS compared with SHS). Adolescents experience barriers to swallowing tablets that are often overcome with age and practice (25). However, the inverse relation observed between age and IFA tablet adherence may be related to increased decision-making power with age that is increasingly influenced by factors outside of parents and teachers (26), or it may be a proxy for level of school. Students at JHSs consumed more tablets than did students at SHSs. These differences may be related to the smaller ratio of students to focal persons at JHSs and other structural differences, which allow for better management of distribution and sensitization activities. Household wealth was expected to have a positive association with IFA consumption because it may be a proxy for school attendance, increasing opportunities to consume tablets (27, 28). However, the inverse was observed in crude analyses and significance was attenuated with no clear direction of the association in adjusted models. Girls from higher-income households may have a greater sense of agency to reject IFA tablets. Although only a small number of schools reported issues coming from parents and community members, routine monitoring reported pervasive early misconceptions about the IFA tablets and their supposed relation to reproductive health. However, such misconceptions were mostly limited to the start of the program, which may explain why educators did not report this issue at follow-up and why it was not associated with IFA consumption. School-level factors may also be responsible for the observed crude associations because girls from the high wealth tertile were more represented in SHSs (46.2%) than in JHSs (32.1%). The evaluation took place during the first year of free SHS, which is expected to increase enrollment in SHS and may change the wealth profiles therein. Addressing structural barriers in this evolving context may present additional challenges. Consuming ≥1 tablet over the school year may not be biologically meaningful; however, it is an important indication of program coverage and uptake at the population level. In crude analyses, a girl's level of knowledge about IFA tablets and receipt of certain health education sessions were associated with whether she ever consumed an IFA tablet. Students with lower knowledge of IFA most commonly thought the tablets were related to reproductive health, such as being a way to regulate menstruation, a contraceptive, or a fertility drug. It is unclear why health sessions related to malaria, deworming, and clean water/hand hygiene were related to whether a student consumed IFA tablets. Health teachers may highlight the linkage of these topics with anemia, increasing awareness of the condition; however, sessions specifically covering anemia and IFA tablets were not associated with whether students consumed an IFA tablet. These relations were no longer present in adjusted models, suggesting the presence of confounding. Level of school, enrollment, educator training, appropriate accounting for IFA tablets, and the educators’ experiences and perceptions were also associated with whether girls ever consumed an IFA tablet. As previously discussed, level and enrollment may both be elements of the school's capacity for effective programming. Educator training emerged as a strong predictor of IFA consumption. It is notable that training only occurred 1 time in these regions, highlighting the impact of a single training. Annual training of school focal persons and expanding the list of trainees to include other key stakeholders such as head teachers may improve program fidelity and educator perceptions of the program and their motivation. Appropriate accounting for the IFA tablets in the school may improve the availability of IFA tablets, because this would improve the school's ability to maintain knowledge of the location, quantity, and condition of its stock. Challenges such as improper storage of IFA tablets can be addressed through adequate training at all levels (9, 29, 30), frequent interaction between health staff and teachers, and thorough monitoring (11). Our findings suggest that school-level factors are most responsible for the coverage of and adherence to IFA supplementation. Additional support to large schools and SHSs may improve coverage where the program is a greater burden because of the number of students and organizational structure of the school. Differences in these schools such as boarding, multiple tracks, and rotation of students through classrooms present challenges for distribution of tablets, tracking adherence, and organizing education sessions. However, these challenges also present opportunities for the intervention to increase the number of contacts and integrate the intervention into existing activities. Ensuring annual training of school staff, including school leadership, may also be beneficial. Training for school staff and students should include guidance on making up missed distributions. Strategies to sensitize and motivate school staff may also benefit the program, because attitudes about the program predicted coverage and adherence to it. Strengthening the health education component and improving knowledge of IFA among students may also produce benefit after school-level factors are addressed. These factors arising from the first-year evaluation may be useful areas of focus for improving the GIFTS program as it matures into a national program. ACKNOWLEDGEMENTS The authors’ responsibilities were as follows—OYA, KT, EFA, LS, LG, MEJ, RM, and UR: designed the research; ABM, EFA, and LS: conducted the research; OYA, LG, AJS, and KT: provided technical assistance and training for the implementation of the study; LG: analyzed the data; LG, OYA, MEJ, and AJS: wrote the paper; LG: had primary responsibility for the final content; and all authors: read and approved the final manuscript. Notes LG is supported by the Oakridge Institute for Science and Education, Department of Energy and by the Laney Graduate School at Emory University. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Authors were employed by their respective organizations during the research. Author disclosures: LG received a contract from UNICEF-Ghana to conduct a portion of this evaluation. All other authors report no conflicts of interest. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC. Abbreviations used: GIFTS, Girls’ Iron-Folic acid Tablet Supplementation; GLMM, generalized linear mixed-effects model; IFA, iron and folic acid; JHS, junior high school; SHS, senior high school. References 1. World Health Organization (WHO) . Guideline: intermittent iron and folic acid supplementation in menstruating women . [Internet] . Geneva : WHO ; 2011 . [cited 2019 Oct 7]. Available from: https://www.who.int/nutrition/publications/micronutrients/guidelines/guideline_iron_folicacid_suppl_women/en/ . PubMed OpenURL Placeholder Text Google Scholar Google Preview WorldCat COPAC 2. World Health Organization (WHO) . Guideline: intermittent iron supplementation in preschool and school-age children . [Internet] . Geneva : WHO ; 2011 . [cited 2019 Oct 7]. Available from: https://www.who.int/nutrition/publications/micronutrients/guidelines/guideline_iron_supplementation_children/en/ . PubMed OpenURL Placeholder Text Google Scholar Google Preview WorldCat COPAC 3. University of Ghana, GroundWork, University of Wisconsin-Madison, KEMRI-Wellcome Trust, UNICEF . Ghana micronutrient survey 2017 . [Internet] . Accra (Ghana) : UNICEF ; 2017 . [cited 2019 Oct 7]. Available from: https://www.unicef.org/ghana/reports/ghana-micro-nutrient-survey . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 4. World Health Organization (WHO) . Iron deficiency anaemia assessment, prevention, and control: a guide for programme managers . [Internet] . Geneva : WHO, UNICEF, UNU ; 2001 . [cited 2019 Oct 7] . Available from: https://www.who.int/nutrition/publications/micronutrients/anaemia_iron_deficiency/WHO_NHD_01.3/en/ . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 5. Salam RA , Hooda M, Das JK, Arshad A, Lassi ZS, Middleton P, Bhutta ZA. Interventions to improve adolescent nutrition: a systematic review and meta-analysis . J Adolesc Health . 2016 ; 59 ( 4, Supplement ): S29 – 39 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Lohrmann DK . A complementary ecological model of the coordinated school health program . Public Health Rep . 2008 ; 123 ( 6 ): 695 – 703 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Aguayo VM , Paintal K, Singh G. The Adolescent Girls’ Anaemia Control Programme: a decade of programming experience to break the inter-generational cycle of malnutrition in India . Public Health Nutr . 2013 ; 16 ( 9 ): 1667 – 76 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Roche ML , Bury L, Yusadiredja IN, Asri EK, Purwanti TS, Kusyuniati S, Bhardwaj A, Izwardy D. Adolescent girls’ nutrition and prevention of anaemia: a school based multisectoral collaboration in Indonesia . BMJ . 2018 ; 363 : k4541 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Malhotra S , Yadav K, Kusuma Y, Sinha S, Yadav V, Pandav CS. Challenges in scaling up successful public health interventions: lessons learnt from resistance to a nationwide roll-out of the weekly iron-folic acid supplementation programme for adolescents in India . Natl Med J India . 2015 ; 28 ( 2 ): 81 – 5 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 10. Cavalli-Sforza T , Berger J, Smitasiri S, Viteri F. Weekly iron-folic acid supplementation of women of reproductive age: impact overview, lessons learned, expansion plans, and contributions toward achievement of the millennium development goals . Nutr Rev . 2005 ; 63 ( 12 Pt 2 ): S152 – 8 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 11. Risonar M , Tengco L, Rayco-Solon P, Solon F. The effect of a school-based weekly iron supplementation delivery system among anemic schoolchildren in the Philippines . Eur J Clin Nutr . 2008 ; 62 ( 8 ): 991 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat 12. D'Agostino A , Ssebiryo F, Murphy H, Cristello A, Nakiwala R, Otim K, Sarkar D, Ngalombi S, Schott W, Katuntu D et al. Facility- and community-based delivery of micronutrient powders in Uganda: opening the black box of implementation using mixed methods . Matern Child Nutr . 2019 ; 15 ( S5 ): e12798 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Priya SH , Datta SS, Bahurupi YA, Narayan KA, Nishanthini N, Ramya MR. Factors influencing weekly iron folic acid supplementation programme among school children: where to focus our attention? . Saudi J Health Sci . 2016 ; 5 ( 1 ): 28 – 33 . Google Scholar Crossref Search ADS WorldCat 14. United Nations Educational, Scientific and Cultural Organization (UNESCO). Enrolment ratios . [Internet] . Paris : UNESCO ; 2019 ; [cited 2020 Feb 4]. Available from: http://data.uis.unesco.org . 15. Ghana Health Service, Ghana Education Service, UNICEF-Ghana, Emory University Global Health Institute, CDC . Impact evaluation of a school-based integrated adolescent nutrition and health programme with iron and folic-acid supplementation intervention among adolescent girls in Ghana . Cantonments, Accra (Ghana) : UNICEF-Ghana ; 2019 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 16. Ghana Statistical Service (GSS), Ghana Health Service (GHS), ICF International . Ghana Demographic and Health Survey 2014 . [Internet] . Rockville (MD) : GSS, GHS, and ICF International ; 2015 . [cited 2019 Nov 4]. Available from: http://dhsprogram.com/pubs/pdf/FR307/FR307.pdf . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 17. Murray DM . Design and analysis of group-randomized trials . New York : Oxford University Press ; 1998 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 18. Vyas S , Kumaranayake L. Constructing socio-economic status indices: how to use principal components analysis . Health Policy Plan . 2006 ; 21 ( 6 ): 459 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Ghana Statistical Service . Population and Housing Census . Accra (Ghana) : Ghana Statistical Service ; 2010 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 20. Tanahashi T . Health service coverage and its evaluation . Bull World Health Organ . 1978 ; 56 ( 2 ): 295 – 303 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 21. Kheirouri S , Alizadeh M. Process evaluation of a national school-based iron supplementation program for adolescent girls in Iran . BMC Public Health . 2014 ; 14 ( 1 ): 959 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Vir SC , Singh N, Nigam AK, Jain R. Weekly iron and folic acid supplementation with counseling reduces anemia in adolescent girls: a large-scale effectiveness study in Uttar Pradesh, India . Food Nutr Bull . 2008 ; 29 ( 3 ): 186 – 94 . Google Scholar Crossref Search ADS PubMed WorldCat 23. Tumilowicz A , Habicht J-P, Mbuya MNN, Beal T, Ntozini R, Rohner F, Pelto GH, Fisseha T, Haidar J, Assefa N et al. Bottlenecks and predictors of coverage and adherence outcomes for a micronutrient powder program in Ethiopia . Matern Child Nutr . 2019 ; 15 ( S5 ): e12807 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 24. Ford ND , Ruth LJ, Ngalombi S, Lubowa A, Halati S, Ahimbisibwe M, Mapango C, Whitehead RD Jr, Jefferds ME. Predictors of micronutrient powder sachet coverage and recent intake among children 12–23 months in Eastern Uganda . Matern Child Nutr . 2019 ; 15 ( S5 ): e12792 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Hansen DL , Tulinius D, Hansen EH. Adolescents’ struggles with swallowing tablets: barriers, strategies and learning . Pharm World Sci . 2007 ; 30 ( 1 ): 65 . Google Scholar Crossref Search ADS PubMed WorldCat 26. Patton GC , Sawyer SM, Santelli JS, Ross DA, Afifi R, Allen NB, Arora M, Azzopardi P, Baldwin W, Bonell C et al. Our future: a Lancet commission on adolescent health and wellbeing . Lancet . 2016 ; 387 ( 10036 ): 2423 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Saasa SK . Education among Zambian children: linking head of household characteristics to school attendance . Vulnerable Child Youth Stud . 2018 ; 13 ( 3 ): 239 – 46 . Google Scholar Crossref Search ADS WorldCat 28. Shehu HK . Factors influencing primary school non-attendance among children in north west Nigeria . Lit Inf Comput Educ J . 2018 ; 9 ( 2 ): 2916 – 22 . OpenURL Placeholder Text WorldCat 29. Jefferds MED , Mirkovic KR, Subedi GR, Mebrahtu S, Dahal P, Perrine CG. Predictors of micronutrient powder sachet coverage in Nepal . Matern Child Nutr . 2015 ; 11 ( S4 ): 77 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Reerink I , Namaste SM, Poonawala A, Nyhus Dhillon C, Aburto N, Chaudhery D, Kroeun H, Griffiths M, Haque MR, Bonvecchio A et al. Experiences and lessons learned for delivery of micronutrient powders interventions . Matern Child Nutr . 2017 ; 13 ( S1 ): e12495 . Google Scholar Crossref Search ADS WorldCat Published by Oxford University Press on behalf of the American Society for Nutrition 2020. This work is written by (a) US Government employee(s) and is in the public domain in the US.
TI - Barriers to and Facilitators of Iron and Folic Acid Supplementation within a School-Based Integrated Nutrition and Health Promotion Program among Ghanaian Adolescent Girls
JF - Current Developments in Nutrition
DO - 10.1093/cdn/nzaa135
DA - 2020-09-01
UR - https://www.deepdyve.com/lp/oxford-university-press/barriers-to-and-facilitators-of-iron-and-folic-acid-supplementation-70tibcMAf2
VL - 4
IS - 9
DP - DeepDyve
ER -