A tailored intervention to promote uptake of retinal screening among young adults with type 2 diabetes - an intervention mapping approach

A tailored intervention to promote uptake of retinal screening among young adults with type 2... Background: Young adults (18–39 years) with type 2 diabetes are at risk of early development and rapid progression of diabetic retinopathy, a leading cause of vision loss and blindness in working-age adults. Retinal screening is key to the early detection of diabetic retinopathy, with risk of vision loss significantly reduced by timely treatment thereafter. Despite this, retinal screening rates are low among this at-risk group. The objective of this study was to develop a theoretically-grounded, evidence-based retinal screening promotion leaflet, tailored to young adults with type 2diabetes. Methods: Utilising the six steps of Intervention Mapping, our multidisciplinary planning team conducted a mixed-methods needs assessment (Step 1); identified modifiable behavioural determinants of screening behaviour and constructed a matrix of change objectives (Step 2); designed, reviewed and debriefed leaflet content with stakeholders (Steps 3 and 4); and developed program implementation and evaluation plans (Steps 5 and 6). Results: Step 1 included in-depth qualitative interviews (N = 10) and an online survey that recruited a nationally- representative sample (N = 227), both informed by literature review. The needs assessment highlighted the crucial roles of knowledge (about diabetic retinopathy and screening), perception of personal risk, awareness of the approval of significant others and engagement with healthcare team, on retinal screening intentions and uptake. In Step 2, we selected five modifiable behavioural determinants to be targeted: knowledge, attitudes, normative beliefs, intention, and behavioural skills. In Steps 3 and 4, the “Who is looking after your eyes?” leaflet was developed, containing persuasive messages targeting each determinant and utilising engaging, cohort-appropriate imagery. In Steps 5 and 6, we planned Statewide implementation and designed a randomised controlled trial to evaluate the leaflet. Conclusions: This research provides an example of a systematic, evidence-based approach to the development of a simple health intervention designed to promote uptake of screening in accordance with national guidelines. The methods and findings illustrate how Intervention Mapping can be employed to develop tailored retinal screening promotion materials for specific priority populations. This paper has implications for future program planners and is intended to assist those wishing to use Intervention Mapping to create similar theoretically-driven, tailored resources. Keywords: Type 2 diabetes, Young adults, Diabetic retinopathy, Intervention mapping, Needs assessment, Retinal screening, Young-onset, Health behaviour change * Correspondence: alake@acbrd.org.au School of Psychology, Deakin University, Geelong, Australia The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria, Melbourne, Australia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lake et al. BMC Health Services Research (2018) 18:396 Page 2 of 18 Background been shown to be effective both in identifying determi- Worldwide increase in the prevalence of type 2 diabetes nants and increasing uptake for a range of disease pre- (T2D) in young adults (< 40 years), with its associated vention interventions [32]. Utilising IM, the aim of the considerable morbidity and mortality, is a burgeoning current study was to identify determinants of screening public health concern [1–5]. Adverse phenotype [6], behaviour for young adults with T2D, and develop an sub-optimal glycemic (blood glucose) control and long engaging psycho-educational resource to target these diabetes duration expose young adults with T2D to a factors, designed to promote screening uptake. high lifetime risk of diabetes-related complications [7, 8]. One of the most common is diabetic retinopathy (DR), Method and results which is a leading cause of vision loss and blindness in In this section, IM steps 1–4 are presented in detail, working age adults [9, 10]. followed by summaries of Steps 5 and 6. Method and re- Early detection of DR via retinal screening (‘screening’), sults are reported separately for each step, including il- followed by timely treatment, are crucial factors in pre- lustrative examples of key IM activities (with full detail venting vision loss [11]. Australian national Guidelines for provided in Additional files). Table 1 provides an over- the Management of Diabetic Retinopathy recommend view of each IM step as it was applied to this project. screening uptake at diabetes diagnosis, repeated at least every two years thereafter [12], an interval less frequent Logic model of the problem than that prescribed for adults with T2D in the United Establish and work with a planning group States (US) and United Kingdom (UK) [13, 14]. Unfortu- A six-person multidisciplinary planning team was con- nately however, young adults (aged 18–39 years) are the vened comprising representatives from The Australian least likely to initiate retinal screening in accordance with Centre for Behavioural Research in Diabetes (AJL, JLB, national guidelines and have lower overall screening rates JS); Centre for Eye Research Australia (GR); Diabetes than older adults (aged ≥40 years) or young adults with Victoria (CH); and Vision 2020 Australia (DT). Combined, type 1 diabetes [15–17]. In addition to their low engage- the planning team provided expertise in psychosocial and ment with existing diabetes services [18], additional com- clinical aspects of diabetes and vision loss, health munication challenges exist due to the lack of dedicated programs, hubs or services for young adults with T2D. Table 1 Overview of IM steps and activities applied to the current leaflet development program Thus, there is need for the development of tailored, evidence-based health promotion resources, using an IM steps IM activities application appropriate to the culture and context, in Step 1: Logic model of � Establish and work with a planning group the problem order to encourage screening uptake among this � Conduct mixed-methods needs assessment priority population [19–23]. � Create logic model of the problem Best-practice development of health promotion re- � Describe context of the intervention and sources targets modifiable behavioural determinants for state program goals a clearly specified health behaviour. The UK Medical Step 2: Program outcomes � State expected behavioural outcomes Research Council (MRC) framework for the design and and objectives; logic model and Performance Objectives (PO) evaluation of complex interventions recommends use of of change � Create logic model of change good quality evidence from a range of sources, strong � Create matrix of Change Objectives theoretical underpinnings, causal modelling and a well- Step 3: Program design � Generate program themes, components, designed evaluation [24]. Intervention mapping (IM) is a scope and sequence six-step protocol encompassing MRC elements, which � Choose theory and evidence-based change provides an effective and useful framework for this pur- methods pose [25]. Key activities are: 1) detailed needs assess- Step 4: Program production � Draft persuasive message content and ment, developing causal logic model of the problem, 2) leaflet stating program outcomes and performance objectives, � Pre-test, refine and produce leaflet developing logic model of change, 3) utilising theory and Step 5: Program � Identify program implementers, adopters evidence-based change methods, designing program to implementation and maintainers target identified behavioural determinants, 4) producing, � Design implementation and liaise with pre-testing and refining program with broad stakeholder program implementers input, 5) planning for program implementation, and 6) Step 6: Program evaluation � Write effect and process evaluation planning for evaluation [26]. Intervention mapping has questions been widely used by intervention planners to guide the � Develop measures for assessment development of effective health promotion materials in a � Specify and complete evaluation plan variety of contexts and populations [27–31] and has Lake et al. BMC Health Services Research (2018) 18:396 Page 3 of 18 promotion, behaviour change research methodologies and analysis, are published elsewhere [44], see Additional file 2 intervention development. Monthly meetings (chaired by for interview guide. In brief, we conducted in-depth semi- JS) were held throughout the project, with additional structured interviews to explore factors affecting screening meetings held as-needed, and quarterly progress reports behaviour for young adults with T2D, with an emphasis on provided to the funding body. Throughout the study, the those that were individual-level and modifiable. The study planning team consulted a practicing health psychologist was advertised widely online and in community settings, (CA) with expertise in IM and a track record of developing and recruitment invitations were mailed to eligible mem- and analysing evidence-based health promotion leaflets bers of a leading state diabetes consumer advocacy organ- [33–36]. Additional expert input was provided by represen- isation. All interviews were conducted via telephone by an tatives from key stakeholder organisations, such as the Na- experienced interviewer (AJL), audio-recorded and tional Diabetes Services Scheme (NDSS, an initiative of the transcribed verbatim. All transcripts were checked for Australian Government, which provides free or subsidised accuracy and imported into NVivo10 (QSR Inter- self-management supplies and services to registrants), Op- national Pty Ltd., Doncaster, VIC., Australia, 2012). tometry Australia and key units within Diabetes Victoria. Transcripts were subjected to content analysis (by Patient and public involvement (PPI) is essential for the AJL), with each participant utterance coded for development of high-quality health behaviour change in- behavioural determinants (using an a priori coding terventions [37] and is recommended specifically as a framework informed by the literature [48]), and again strategy for engaging groups at high risk of underutilsation as either ‘facilitator’ or ‘barrier’ dependent upon the of eye healthcare services [38]. In this study, five young context. Twenty percent of transcripts were double- adults with T2D were involved, providing feedback on all coded (by JLB), with high inter-rater reliability of study documentation, piloting the quantitative survey and 99%. Screening determinants were rank-ordered by providing detailed review of the eye health leaflet. frequency of coding (higher frequency of utterances interpreted to indicate higher salience). Conduct mixed methods needs assessment Qualitative interview findings Our study of the literature (summarised in Additional file 1) In brief, ten young adults with T2D (50% women, revealed that, while there was a paucity of research in this aged 29–37 years) were interviewed (average length: specific area, sub-optimal diabetes self-management (in 55 min, range: 31–106 min). Fifty percent had not general) among young people is likely driven by low socio- attended retinal screening previously. Although young economic status [39], low general and health literacy [39], adults with T2D knew of a link between diabetes and low engagement with diabetes self-management education vision loss, they did not have a comprehensive under- [20, 39–41], cultural diversity of the priority population standing of DR or screening (e.g. symptoms, risk [42], optimistic bias and low risk perception [43], life-stage factors, screening guidelines, distinction between demands [44], high rates of diabetes-related distress [40] screening and standard vision checks). Participants re- and complex healthcare needs [45]. ported distress related to having a condition stereo- In our empirical needs assessment studies, we sought to typically associated with older people, and many did determine the relevance of these factors to DR screening spe- not know others of similar age with T2D. Participants cifically, and to identify any additional factors that may facili- indicated that absence of social influence (e.g. prompt- tate or impede this target behaviour. As other researchers ing from significant others, social comparison with others), have found it challenging to recruit young adults with T2D to and low DR risk perception, combined with life-stage research studies [46, 47], several steps were taken to boost re- barriers (e.g. lack of time and finances), negatively im- cruitment in the mixed-methods needs assessment. These pacted screening uptake. Concerned about negative included: giving priority to ease of participant access; judgment by others, and fearing a DR diagnosis, partici- distribution of engaging, cohort-appropriate recruit- pants reported that they did not always disclose their dia- ment invitations with an NDSS and Diabetes Australia betes diagnosis or proactively seek healthcare or social branded cover letter introducing the study; reminder support, thus losing crucial pathways to timely screening invitation after four weeks; age-appropriate incentives uptake. Irrespective of their screening history, young (e.g. entry to a technology-based prize draw), and ex- adults with T2D identified a range of screening bar- tension of recruitment periods until participant regis- riers, suggesting that a cumulation of factors may tration visibly flagged. impact uptake, thus highlighting the need to acknow- ledge and address a broad range of barriers in a tai- In-depth qualitative interviews lored intervention. Qualitative interview procedure Screening facilitators were often conceptualised by par- Detailed description of the study methods and findings, in- ticipants as the opposite of the barriers (e.g. improved, as cluding the participants, procedure, interview guide and opposed to inadequate, knowledge or access to social Lake et al. BMC Health Services Research (2018) 18:396 Page 4 of 18 support). However, the study also highlighted other collectively assessed two behavioural skills constructs (per- screening facilitators: participants compared them- ceived control over screening and overcoming barriers). selves with others experiencing diabetes-related vision Unless otherwise noted, each item was rated on a 7- loss, and were thus influenced to engage in screening point Likert scale (“strongly disagree” to “strongly agree”). due to concerns about the impact that vision loss Individual items were aggregated to provide a composite would have on their lives, including anticipated regret score for each construct, with good internal consistency at the potential impact on their spouses and/or (see Additional file 3). For each, higher scores indicated children. For those who previously attended screen- greater endorsement of the construct measured (e.g. ing, feelings of relief and reassurance facilitated repeat stronger intentions, more positive attitudes). In addition, screening behaviour, with participants expressing in- we collected socio-demographic data to describe the tent to sustain the behaviour and expectation of a sample at baseline. The survey was piloted with young positive outcome (i.e. no DR diagnosis). adult PPI members and representatives from selected stakeholder organisations, who also commented on read- ability, format, accessibility and content; no substantive National online survey changes were required. Online survey procedure Data collection and participants Survey development The survey was conducted nationwide and hosted Using the Information-Motivation-Behavioural skills via a secure online survey platform, Qualtrics™ (Provo, (IMB) model [49] as a foundation, the planning team de- UT, 2014–2015). In Australia, the majority of people veloped a survey designed to identify modifiable behav- with a confirmed diagnosis of diabetes are registered ioural determinants for screening. The IMB model by their health professional with the NDSS [56]. All posits that although information is a key element in young adults with T2D who had been registered on changing behaviour, increasing knowledge and aware- the NDSS in the previous three years (registration ness of a behaviour is not sufficient in itself, and re- date wasusedasa proxyfor diagnosisdate),and quires the integration of motivational and skills who had consented to be contacted for research elements to ensure behaviour change. Use of the IMB (N = 5354) were invited to participate. Exclusion model in behaviour change research requires identifi- criteria included non-English speaking; those aged 40+ cation of deficits in each of the three key areas, to be years, and diagnosis of another type of diabetes. Study addressed in a subsequent intervention. The IMB invitations were managed by the NDSS in order model has been effective both as a framework for to preserve registrant confidentiality, but purposive intervention design [50] and as a predictive model for sampling of those who had not previously screened health-related screening behaviours, such as breast for DR was not possible, due to lack of available data self-examination [51]. on retinal screening status of NDSS registrants. Re- Increasingly used with chronic conditions, the IMB cruitment to the online survey continued for seven model has been validated recently in a model of weeks. diabetes self-care behaviours [52] and medication ad- Statistical analyses herence [53]. Survey items were based on IMB-based Statistical analyses were conducted using SPSS ver- questionnaires previously validated for diabetes self- sion 22 (SPSS Inc., Chicago IL, USA). Univariate management [52, 53], the widely-used Theory of analyses (chi-square and independent measures t- Planned Behaviour [54], and cognitive constructs tests, two-sided) were conducted to explore between- showntoberelevanttoyoung adults with T2D (e.g. group (previous retinal screen: yes/no) differences on optimism/fatalism, social support, risk perception, an- demographic variables and modifiable behavioural de- ticipated regret, self-efficacy) [43, 55]. terminants at the item level (to inform specific inter- In brief, the survey comprised 54 items assessing infor- vention message content). Given the large number of mation/knowledge, motivation and behavioural skills analyses, a conservative p < 0.01 was considered sta- (see Additional file 3 for individual items). Information: tistically significant. 16 items assessed knowledge of the link between dia- Online survey findings betes and vision loss, diabetic retinopathy and retinal Overall, 129 participants (2% of eligible population) screening. Responses scored dichotomously (incorrect / completed the full survey, and their sociodemographic correct). Motivation: 21 items collectively assessed three characteristics are presented in Table 2. Sixty percent attitudinal constructs (attitudes toward screening for were women, average age 34 ± 5 years (range: 19– DR, perception of personal risk, and anticipated regret); 39 years), and 74% had previously screened for DR. No three items assessed normative beliefs and three items significant differences in sociodemographic characteris- assessed intention. Behavioural skills: 11 items tics were found between screening groups. Lake et al. BMC Health Services Research (2018) 18:396 Page 5 of 18 Table 2 Sociodemographic characteristics by screening behaviour was high for both groups. Compared to their non- (N =129) screening counterparts, those who had previously Sociodemographic Retinal screen p-value screened reported greater concern and worry at the pro- characteristics spect of not screening. Participants who had previously No (n = 33) Yes (n = 96) screened were significantly more likely to agree that sig- Age, years 34.39 (33, 37) 34.04 (32, 37) .697 nificant others (i.e. family/friends, healthcare team) Duration, years 1.00 (1.84) 1.69 (1.97) .081 would approve of screening. Intention to screen was Gender: women 15 (45) 62 (65) .084 high among all participants but significantly higher for Primary diabetes management those who had previously screened compared to those Lifestyle only 5 (15) 21 (22) who had not. Medication (not insulin) 23 (70) 64 (67) .652 Those who had screened previously reported signifi- cantly greater confidence on all aspects of behavioural Insulin 5 (15) 11 (11) control over screening (e.g. how to make an appoint- Country of birth: Australian born 18 (55) 66 (69) .206 ment for screening, ability to screen regularly, remember Main language spoken at home: 27 (82) 81 (84) .944 and attend an appointment). No differences were ob- English served between groups on confidence in knowing the Employment status: employed 20 (61) 57 (59) 1.000 steps that can be taken to reduce the risk of DR, al- Socioeconomic status 984.55 (83.52) 991.48 (57.11) .660 though scores were lower for all participants compared Family history of T2D 22 (67) 72 (75) .483 to other behavioural control items. Those who had ≥1 comorbid health condition 25 (76) 75 (79) .891 screened also reported significantly higher confidence in overcoming common screening barriers (including time Depression (PHQ-2) 2.94 (2.48) 2.12 (2.07) .102 and cost, and discussing diabetes and DR with health- Data are number (%), mean (SD), or median (IQR); p-value is Pearson’schi-square or independent t-tests (two-sided); statistical significance p < 0.05 care professionals). Index of Relative Socio-economic Advantage and Disadvantage: lower score indicates relatively greater disadvantage, range 300–1250 Missing data (average 6%, range 2–11%) Summary of key learnings from needs assessment PHQ-2 range 0–6: ≥3 indicating likely depression Key learnings from the literature review, qualitative interviews and quantitative survey are summarised in Behavioural determinants of screening Table 4. The survey identified that compared to their Selected findings for information (knowledge), motiv- non-screening counterparts, those who had previously ation and behavioural skills items are detailed in Table 3 attended screening reported: significantly higher (full detail and construct-level findings provided in knowledge of both DR and retinal screening; more Additional file 3). positive attitudes towards screening; stronger agree- Almost all participants (irrespective of previous ment that significant others would approve of the be- screening behaviour) knew of a link between diabetes haviour; higher intention to screen; greater perceived and vision loss. However, compared to their non- behavioural control (i.e. confidence that they could ar- screening counterparts, those who had previously range and attend screening when due), and greater screened knew that all people with diabetes were at risk confidence in addressing common screening barriers. of DR, the clinically-recommended HbA1c (average The findings suggest that messages highlighting the blood glucose) target for DR prevention, when to initiate prevalence of DR and link between DR and diabetes dur- screening and recommended screening intervals. ation are warranted to prompt reassessment of personal Overall, participants who had screened indicated more risk. Information on modifiable DR risk factors (blood positive attitudes towards the behaviour (e.g. empower- glucose, blood pressure and cholesterol), asymptomatic ing, reassuring and important) than those who had not nature of the condition and screening guidelines are screened. No differences were observed between groups needed to encourage individuals to both reduce DR risk on how pleasant or comfortable the eye check was per- and initiate screening. ceived to be, although scores were lower for all partici- Messages designed to highlight the health and material pants compared to other attitude items. Perception of consequences of screening, including likely positive personal risk of vision problems and DR were moderate emotional consequences, are warranted in order to pro- for all participants with low expectations of a DR diag- mote positive screening attitudes. Findings suggesting nosis in the short term. Although all participants be- that all participants perceived screening as potentially lieved they could not reduce their risk of vision ‘unpleasant’, ‘uncomfortable’ and disruptive to normal problems, those who had screened held this belief more activities are realistic considering that many people ex- strongly. All participants reported negative emotions perienced discomfort and delay from pupil dilation (my- when thinking about not screening, including fear, which driasis) drops. Consequently, positive messages should Lake et al. BMC Health Services Research (2018) 18:396 Page 6 of 18 Table 3 Selected behavioural determinant items by retinal screen (N = 129)* Modifiable behavioural determinants Retinal screen p-value No (n = 33) Yes (n = 96) INFORMATION (KNOWLEDGE) ITEMS Diabetes can lead to vision loss 30 (91) 93 (97) .174 All people with diabetes are at risk of DR 26 (79) 89 (93) .004 Recommended target HbA1c 17 (53) 81 (87) <.001 Initiate eye examinations ‘at diabetes diagnosis’ 5 (15) 42 (45) .004 Screen ‘at least every 2 years’ if no DR present 0 (0) 18 (19) .003 b c MOTIVATION ITEMS An eye health check for DR would be... ...(not) ‘unpleasant’ 3.71 (0.94) 3.86 (1.07) .500 ...reassuring 3.94 (0.96) 4.63 (0.61) <.001 ...important 4.06 (1.06) 4.89 (0.35) <.001 ...empowering 3.10 (1.19) 3.73 (0.97) .004 ...comfortable 3.26 (1.15) 3.68 (1.10) .073 I believe I will develop DR due to my diabetes 4.03 (1.45) 4.14 (1.62) .734 Expect to be diagnosed with DR at next eye check 2.97 (1.47) 2.43 (1.66) .114 Can reduce risk of vision problems... 2.32 (1.44) 1.43 (0.79) .002 If I did NOT have an eye health check for DR, I would feel... ...concerned 5.03 (1.70) 5.88 (1.40) .007 ...fearful 4.48 (1.79) 5.13 (1.70) .073 ...worried 4.65 (1.80) 5.53 (1.47) .007 My family/close friends would approve... 5.94 (1.69) 6.82 (0.80) .008 I plan to attend an eye health check... 4.26 (2.32) 6.76 (0.77) <.001 I intend to have an eye health check... 4.42 (2.32) 6.74 (0.77) <.001 b,e BEHAVIOURAL SKILLS ITEMS How confident are you that you... ...know steps to reduce the risk of developing DR 2.39 (1.17) 3.06 (1.29) .012 ...will remember to have an eye health check… 2.68 (1.35) 4.36 (0.90) <.001 ...can talk to your doctor about your eye health 3.39 (1.28) 4.17 (1.03) .001 ...can find the time to attend an eye health check… 2.74 (1.37) 4.55 (0.75) <.001 ...can afford to pay for the eye health check… 2.68 (1.49) 3.52 (1.48) .008 DR diabetic retinopathy. Data are number (%) of participants who answered each item correctly (Knowledge items); mean (SD) Motivation and Behavioural skills items. p-value is Pearson’s Chi-Square (or Fisher’s exact test if expected cell count< 5), or Independent-samples t-test (two-sided); statistical significance p < 0.01 *Full detail and construct-level findings provided in Additional file 3 Glycated haemoglobin (measure of average blood glucose over the past 8–12 weeks, and indicator of DR risk) Valid n: 121 (motivation items), 120 (behavioural skills items) Item response range: 1 (Strongly disagree) to 5 (Strongly agree) 1 (Strongly disagree) to 7 (Strongly agree) 1 (Not at all confident) to 5 (Extremely confident) be balanced by acknowledgement of the potential for possible unrealistic level of optimism, highlighting the negative consequences related to mydriasis in order to need to emphasise personal susceptibility while provid- maintain credibility. ing information-based content on steps that can be Although moderately high levels of distress in the pri- taken to reduce DR risk. As with many other preventive ority population mean that it is important to avoid direct behaviours, awareness of the potential effectiveness of ‘fear appeal’ messages, low anticipated regret scores for screening followed by subsequent protective action did those who had not screened reinforce the need for mes- not necessarily result in intention formation or priori- sages which emphasise personal susceptibility and de- tisation of preventive intentions. Cognitive dissonance scribe the likely consequences of not screening. induction techniques have been found to have generally Similarly, responses to risk perception items point to a positive effects on changing attitudes, motivations and Lake et al. BMC Health Services Research (2018) 18:396 Page 7 of 18 Table 4 Key lessons learned from needs assessment findings into a logic model for DR screening. The aim of the logic model was to identify the pathways Compared to their older adult counterparts, young adults with T2D have different psychosocial and information needs. There is a lack of behavioural of problem causation moving from determinants, to interventions focused on encouraging screening uptake among young low screening rates and consequent impact on health adults with T2D, indicating that development of a tailored intervention is and quality of life (Fig. 1). warranted. Perceived barriers to and facilitators of screening (which are modifiable and within the scope of the current intervention) include: Context of the intervention and program goals � Knowledge: diabetic retinopathy (awareness of asymptomatic nature of The intervention was to be evaluated and implemented DR, high personal DR risk, modifiable risk factors), screening (consequences of not screening, role of screening in early detection of DR and subsequent in a real-world setting where intervention format and benefit of timely treatment, distinction between standard eye check and delivery medium were dictated by broader policy-level retinal screen) initiatives and a fixed delivery timeline. The intervention � Attitudes: low perception of personal risk, recognition of the benefit of screening was funded by Vision 2020 Australia and grounded � Normative beliefs: awareness of screening approval by significant others, within a suite of Vision Initiative projects collectively de- and screening approval and behaviour of similar others signed to achieve the aims of the Commonwealth gov- � Intention: low prioritisation of target behaviour � Behavioural skills: self-efficacy in overcoming common screening barriers ernment ‘National Framework for Action to Promote to ensure screening attendance (e.g. lack of time or resources), Eye Health and Prevent Avoidable Blindness and Vision engagement with healthcare (sharing diabetes diagnosis, participation in Loss’ [58]. Vision Initiative policy required that the re- diabetes self-management behaviours) source be targeted at the individual-level and delivered directly to eligible young adults with T2D (NDSS regis- health-related behaviour patterns [57]. Consequently, we trants). As such, it was determined by the planning team selected dissonance reduction as a technique that could members who were involved in conception of the study, promote screening motivation. that the most efficient, cost-effective way to meet these Responses to normative behaviour items suggest that criteria was for the intervention to take the form of a messages that provide information about significant leaflet, to be posted to eligible NDSS registrants. Fur- others’ approval are warranted. The findings suggest that thermore, this enabled the leaflet to be included in fu- inclusion of procedural information and messages to ture ‘NDSS starter packs’ for new registrants, and to be promote confidence in knowing steps that can be taken made available online. This decision was supported by to reduce DR risk including how to book and remember previous research, which showed that printed materials a retinal screen, as well as overcoming common barriers are acceptable to young adults with T2D, who give pref- are warranted. Emphasis is required to minimise mis- erence to consistent, centralised information over for- conceptions about some barriers (e.g. inclusion of mes- mat, and who specifically state that the NDSS ‘starter sages which accurately describe the cost and time taken pack’ is a useful resource. for the procedure). With 86% of Australians with T2D registered, the NDSS is considered the “best available source to monitor Logic model of the problem type 2 diabetes in children and young people in Giving consideration to both the qualitative and Australia” [p.36, 56]. However, the NDSS database pri- quantitative needs assessments, we synthesised our marily records registrant postal addresses, which Fig. 1 Logic model of the problem DR: diabetic retinopathy; *Identified in the needs assessment but cannot be modified by the current intervention Lake et al. BMC Health Services Research (2018) 18:396 Page 8 of 18 necessitated the use of a print-based intervention tool Create logic model of change that could be posted to registrants. We developed a logic model of change (Fig. 2) to depict Overall, the purpose of the intervention was to pro- the hypothetical causal pathway from the intervention to mote uptake of screening among young adults with T2D. program outcomes, and anticipated health and quality of Accounting for real-world logistical considerations, the life improvements. Commencing with the intervention, program goal was to develop a leaflet intervention that: we outlined the five modifiable behavioural determinants could be delivered by post, was tailored to young adults (from Fig. 1) and four Performance Objectives (Table 5), with T2D, and included persuasive messaging targeting which were expected to change the measurable behav- behavioural determinants of the target behaviour. ioural outcome. The planning team also acknowledged ex- ternal factors that may affect screening behaviour (i.e. Program outcomes and objectives; logic model of factors that cannot be changed through an individual-level change intervention) in the logic model, even though these were Expected behavioural outcomes and performance objectives beyond the scope of our intervention. The multidisciplinary planning team defined a single, measurable primary outcome when planning for the Create matrix of change objectives subsequent evaluation (uptake of screening for those Once the health behaviours, Performance Objectives and young adults with T2D who had not previously screened determinants were defined, Change Objectives were de- for DR), and multiple secondary outcomes (i.e. change in veloped. Change Objectives are integral to intervention nominated modifiable behavioural screening determinants). content because they represent the behaviour or cogni- Working from the designated program outcomes, and tion being targeted. A sub-group of the planning team informed by the findings of the needs assessment, the generated Change Objectives by creating a matrix, with planning team established the foundation for the inter- modifiable behavioural determinants (in columns) and vention by defining four Performance Objectives (PO, sub-objectives (in rows). Table 5). We increased the specificity of each Perform- Table 6 presents a matrix of Change Objectives for ance Objective by defining sub-objectives, each identify- Performance Objective 3 (Young adults with type 2 dia- ing a behaviour or cognitive process that would promote betes will be motivated to engage in retinal screening). screening uptake. To illustrate, two Change Objectives were generated for sub-objective 3.2 (Understand personal risk of DR), at the intersection with two determinants (knowledge and Table 5 Performance Objectives (e.g. PO.1) and sub-objectives attitudes). The first (K.3.2) sought to improve knowledge (e.g. PO.1.1, PO.1.2, etc.) that risk of DR increases over time, and the second (A.3. Young adults with type 2 diabetes will… 2) sought to change attitudes regarding personal risk PO.1.… demonstrate a clear understanding of diabetic retinopathy (DR) and susceptibility to DR. See Additional file 4 for a PO.1.1 Modifiable and non-modifiable DR risk factors complete matrix of Change Objectives. PO.1.2 Clinical targets for reducing risk of DR Intervention design PO.1.3 Symptoms of DR Intervention themes, components, scope and sequence PO.1.4 Role of DR in vision loss Ensuring that all components of the intervention PO.1.5 Prevalence of DR reflected the needs and preferences of young adults with PO.2…demonstrate clear understanding of retinal screening T2D was a crucial consideration for the planning team. PO.2.1 Role in detecting DR and reducing vision loss The health behaviour change and health communication PO.2.2 Screening procedure and experience literature provided ample foundation on best practice presentation of message content [20, 25, 40, 59–64]. In- PO.2.3 Booking and examination procedure formed by this evidence and the findings of the needs PO.3… be motivated to engage in retinal screening assessment, we developed seven guiding principles for PO.3.1 Prioritise retinal screening leaflet intervention design (Table 7). Consultation with PO.3.2 Understand personal risk of DR young adult PPI members and experts from key stake- PO.3.3 Identify personal barriers to retinal screening holder groups confirmed that these were appropriate PO.3.4 Perceive personal responsibility to engage in screening guiding principles from their perspective. PO.4…proactively engage with the healthcare system and their healthcare team Choose theory and evidence-based behaviour change PO.4.1 Discuss diabetes and eye health with healthcare professionals methods PO.4.2 Understand treatment benefits and options Having established guiding principles, the planning team PO.4.3 Seek more information about diabetes and eye health selected types of theory-based psychological change Lake et al. BMC Health Services Research (2018) 18:396 Page 9 of 18 Fig. 2 Logic model of change T2D: type 2 diabetes, DR: diabetic retinopathy techniques (or change strategies) [65, 66] grouped into significant others’ approval of screening behaviour, provid- six broad change mechanisms designed to ‘boost motiv- ing information about others’ screening behaviour); v) fos- ation and prompt action’ ([65], p.104). The constituent tering a positive screening identity (e.g. providing a techniques (or practical methods) included in the leaflet positive group identity for those engaging in screening); focused on: i) changing beliefs about the benefits of and vi) enhancing self-efficacy (e.g. using persuasive argu- screening (e.g. providing general information on ment to bolster self-efficacy, providing instruction, behaviour-health links, describing likely consequences of prompting barrier identification and planning in relation behaviour); ii) changing risk perception (e.g. emphasising to anticipated barriers, prompting goal setting). personal susceptibility to negative consequences, prompt- ing recipients to assess own risk); iii) changing attitudes associated with screening uptake (e.g. describing likely Intervention development emotional or affective consequences, potentially inducing Draft persuasive message content and leaflet cognitive dissonance among those not intending to act in Develop message content the face of negative consequences); iv) changing (norma- Working from the matrix of Change Objectives, the tive) beliefs about others’ behaviour (e.g. emphasising guiding principles, and theory and evidence-based Table 6 Illustrative matrix of Change Objectives for Performance Objective 3 (PO.3) - Young adults with type 2 diabetes will be motivated to engage in retinal screening Sub-objectives Modifiable behavioural determinants Knowledge Attitudes Normative beliefs Intention Behavioural skills PO.3.1 Prioritise retinal screening NB.3.1 Recognise that I.3.1 Form an similar others have intention to overcome screening prioritise retinal barriers screening PO.3.2 Understand K.3.2 Know that A.3.2 Perceive high personal risk of DR DR risk increases personal risk and over time susceptibility to DR PO.3.3 Identify personal A.3.3 Believe that NB.3.3 See that similar BS.3.3 Be confident in barriers to retinal screening attending screening others face screening one’s ability to identify will relieve fear and barriers (e.g. cost, fear and overcome common guilt and be a of adverse effects) screening barriers positive experience PO.3.4 Perceive personal K.3.4 Know that A.3.4 Adopt personal NB.3.4 Believe that similar BS.3.4 Be confident they responsibility to engage they can take responsibility for others take responsibility have the tools to act on in screening steps to protect retinal screening for their own eye health personal responsibility eye health Determinants: K=Knowledge, A = Attitudes, NB=Normative Beliefs, I=Intention, BS=Behavioural Skills PO performance objective, DR diabetic retinopathy Full matrix of Change Objectives for every Performance Objective is provided in Additional file 4 See Table 5 for full list of Performance Objectives and sub-objectives Lake et al. BMC Health Services Research (2018) 18:396 Page 10 of 18 Table 7 Guiding principles for retinal screening leaflet Assessing readability and suitability intervention design The leaflet was assessed using a combination of an Readability and comprehension: content to be written to acceptable (health) online readability consensus calculator and the Suit- literacy standards, with minimal technical or medical terminology [59, 60]. ability Assessment of Materials (SAM) test, consistent Scope: the scope of intervention messages to be restricted to targeting with best practice [67]. The consensus calculator re- individual-level, modifiable behavioural determinants. ports synthesised results from seven assessment tools Framing: despite long term benefit, retinal screening can be considered (e.g. Flesch Reading Ease formula, Flesch-Kincaid a high-risk behaviour due to the potential for immediate DR diagnosis Grade), to provide two composite scores by grade [61]. Loss-framed messages are effective in promoting engagement with (range: 4–9) and reading level (range: 0–29 ‘very confus- high-risk behaviours and will be used in this leaflet [25]. The majority of headings to be framed as questions to engage the reader while minimising ing’ to 90–100 ‘very easy’)[68]. The SAM test uses six any potentially defensive reaction [62]. evaluation criteria (content, literacy demand, graphics, lay- Sequence: content to follow the logical order of reading. In order to out and type, learning stimulation and motivation, cultural balance loss-framed messages against the high levels of diabetes-related appropriateness) to determine overall suitability [69], with distress and anxiety experienced by young adults with T2D [20, 40], potentially threatening content to be immediately followed by an scores summed and converted to a percentage score and empowering or reassuring statement. classified as ‘not suitable’ (0–39%), ‘adequate’ (40–69%), or Use of quotes: in recognition of the subtle aspects of social influence, ‘superior’ (70–100%). where an individual’s’ beliefs are influenced by those accepted and We excluded the front and back panels of the ‘Who is encouraged by the majority [63], quotes from similar others to be used looking after your eyes?’ leaflet from assessment, as they to reinforce key persuasive messages. All quotes to be sourced verbatim from interview study with descriptors (age and diabetes duration) included included minimal text. For the remaining panels, the to reinforce group membership. median readability consensus grade was 6; median read- Credibility: quote descriptors within the leaflet to reflect demographic ing ease level was 75.6 (fairly easy), and SAM test out- characteristics of the priority population to prompt identification with a come was 75% (superior). credible source. Similarly, logos of leading diabetes and eye health organisations that had contributed to the content to be included to enhance credibility of information. Important yet necessary negative Draft intervention materials information (e.g. discomfort associated with mydriasis, time required to recover clear vision) to be included to provide balance. The planning team selected an 8-panel leaflet design, Graphics and imagery: to reflect the demographic characteristics of the with panels opening outward from the centre, which priority population (e.g. young adults from a range of ethnicities, with and could fit into a standard DL-size envelope. A range of without children). National interpreter symbol to indicate availability of leaflet design options were created in close consult- language assistance services to those with limited English proficiency [64]. ation with a graphic designer who had expertise in producing health promotion materials for people with diabetes. The designs varied in structure, imagery and intervention strategies noted above, a pool of more than organisation, but all adhered to the guiding principles 60 persuasive messages was developed. From this and included consistent messaging. pool, specific change techniques or practical methods were selected to encourage screening. For example, to achieve Change Objective A.3.4 (View retinal screen- Pre-test, refine and produce leaflet ing as a personal responsibility), four potential leaflet Validation and pilot testing heading messages were developed: ‘Eyes: they’re im- The draft leaflet was reviewed by the planning team portant any way you look at it’, ‘Theonlywayto and representatives from key stakeholder organisa- know is to go…’ (verbatim quote), ‘Who is looking tions to confirm that all content was factually accur- after your eyes?’,and ‘Looking at the facts’.All mes- ate and clinically appropriate, and that the resource sages were reviewed by the planning team, and a sub- was likely to meet the project objective. Young adult set selected based on the perceived capacity of the PPI members participated in a thorough piloting message to achieve program goals, target individual process to determine whether: the images and quotes Change Objectives, and satisfy the leaflet guiding were culturally relevant and resonated with the principles. Thus, in the above example, the third op- reader; participants perceived the leaflet would have tion (‘Who is looking after your eyes?’) was selected met their information needs at the time of their T2D because it was phrased as an engaging question, pro- diagnosis; and there were any unintended adverse moting personal responsibility with potential to re- effects in the messaging, imagery or format. Each duce defensive reaction while motivating screening. young adult PPI member received the draft leaflet by A selected example, linking leaflet content to Perform- post and, after reviewing it, participated in a tele- ance Objective 3, is presented in Table 8. Full intervention phone interview during which they commented on map detail for all Performance and Change Objectives is the leaflet’s suitability, responding to questions based provided in Additional file 5. on the SAM criteria [69]. Lake et al. BMC Health Services Research (2018) 18:396 Page 11 of 18 Table 8 Illustrative intervention map linking leaflet content directly back to Performance Objective 3 (PO.3: Young adults with T2D will be motivated to engage in retinal screening) a b Sub-objective (i.e. 3.1) and related Change Objectives Leaflet content (antecedent leaflet text in brackets illustrates context) Panel number PO.3.1 Prioritise retinal screening NB.3.1 Recognise that similar others have overcome (Jenny’s story: before and after the eye health check). “I was scared. 7 screening barriers I was scared of what damage was done…of confronting the fact that my eyesight could be damaged, and of going through the exam and being confronted with what’s there.” I.3.1 Form an intention to prioritise retinal screening (What can I do to protect myself from DR and prevent vision loss?) 5 1. Have a diabetes eye health check. (Note: eye health check listed as Step 1, highest priority) PO.3.2 Understand personal risk of DR K.3.2 Know that DR risk increases over time � The longer you have diabetes the more at risk you are of DR 1 A.3.2 Perceive high personal risk and susceptibility to DR Image: mother and daughter smiling. Child holding hands over 1 mother’s eyes But I’m still young. Am I at risk of DR? Yes you are. Anyone with diabetes can develop DR, which is the 1 leading cause of vision loss for people under 60 years. There are over 34,000 Australians with type 2 diabetes who are 1 under 40 years of age. More than 8500 will already have DR. � The longer you have diabetes the more at risk you are of DR. 1 (Lucas, aged 34, diagnosed with type 2 diabetes 2 years ago) “I didn’t know that I was at risk.” 1 (Jane 25 years, diagnosed with type 2 diabetes 3 years ago) “You might have good vision, you might think that your eyes are 4 absolutely brilliant and there’s no issue. But in the back of your eye, there could be a problem with those little tiny veins that you don’trealise.” PO.3.3 Identify personal barriers to retinal screening A.3.3 Believe that attending screening will relieve fear (Jenny’s story: before and after the eye health check). “It was 7 and guilt and be a positive experience actually quite fun; I don’t know why I put it off. I was really scared going in there, but definitely not now – I’m not fazed by it at all.” NB.3.3 See that similar others face screening barriers (Jenny’s story: before and after the eye health check). “The eye 7 (e.g. cost, fear of adverse effects) drops were a bit uncomfortable and there was a small cost – but I think it’s a wise spend considering what you’re preventing.” BS.3.3 Be confident in one’s ability to identify and (What else do I need to know?) A diabetes eye health check 6 overcome common screening barriers takes about 30 minutes. (What else do I need to know?) It may be free (bulk-billed) or 6 there may be a small fee. (What else do I need to know?) Your optometrist may use eye 6 drops which helps them to see the back of your eye. If you do have eye drops, they may be a little uncomfortable. The drops will also leave you sensitive to light, so bring your sunglasses and be prepared to wait a while for your vision to return to normal PO.3.4 Perceive personal responsibility to engage in screening K.3.4 Know that they can take steps to protect eye health What can I do to protect myself from DR and prevent vision loss? 5 A.3.4 Adopt personal responsibility for retinal screening “I’m a busy person and my family depend on me.” 1 Leaflet heading: Who is looking after your eyes? 3 NB.3.4 Believe that similar others take responsibility for Images: mother and daughter, smiling couple selfie, young man 1,3,5,8 their own eye health of indeterminate cultural origin, Asian female (a.k.a. ‘Jenny’) BS.3.4 Be confident they have the tools to act on personal Leaflet sub-heading: 3 responsibility Your guide to preventing vision loss from diabetes eye disease Protect your sight for life 2 Complete intervention map for all Performance and Change Objectives is provided in Additional file 5 Determinants: K = knowledge, A = attitudes, NB = normative behaviour, I = intention, BS = behavioural skills PO Performance Objective (in bold), DR diabetic retinopathy See Table 6 for illustrative matrix of Change Objectives and Additional file 4 for complete matrix See Fig. 3 for leaflet panels Lake et al. BMC Health Services Research (2018) 18:396 Page 12 of 18 Table 9 Suitability Assessment Materials (SAM) evaluation criteria, young adults’ feedback and changes made to leaflet Sample pilot questions Young adults’ feedback Changes to leaflet Content: Do you think that this leaflet “Key information came through really clearly. I Make ‘Protect your sight for life’ a stand-alone achieves the purpose of the project? didn’t know that early DR doesn’t have any statement and place at top of panel 2, which � Did you learn anything new? symptoms…the doctors tend to focus on signposts location of more information blood glucose, so I knew the 7% (HbA1c) but I didn’t know what the cholesterol target and normal blood pressures were.” ID32 “This leaflet improved my intentions. (DR) is not something you would think could happen to young people.” ID32 “‘Protect your sight for life’ is a powerful statement.” ID36 Literacy demand: Was the length of the leaflet “It only took about 5 minutes to read.” ID33 Discuss whether to include ‘DR’ in leaflet. By acceptable to you? “Language is pretty relaxed which is good for consensus, a decision was made to include it, � How about the number of words? young people.” ID40 but to bold initial definition of diabetic retinopathy � How easy was it to read and understand the “The only thing that caught me was ‘DR’. Did and DR acronym at top of panel 4. information in the leaflet? you mean ‘doctor’ or ‘diabetic retinopathy’?I � Are the words used simple, clear and think you should bold it when it is first informal? defined.” ID32 � Were medical terms defined adequately? Graphics: What do you think of the front “Very professional. Looks like it’s targeted at Bold text ‘When diabetes is first diagnosed’ in panel image? my demographic.” ID36 panel 5. � Are the other images and graphics ‘friendly’ “Maybe bold ‘When diabetes is first Remove graph, which depicted rate of DR and relevant? diagnosed’ so that you hammer home that it’s progression over time. never too early to have an eye check.” ID32 “I’ve never really looked at a graph in a pamphlet. It might appeal to some people, but I don’t know…” ID40 Layout and typography: What do you think of “Main headings need to be in a larger sized the sequence of information? font and bold, and sub-headings in smaller � Is the text type size and font easy to read, or font. Keep the blue colouring.” ID39 could it be easier? “Is there a way that you can make more white � Is the information in the leaflet well-spaced, space? The different colours are more or does it appear cluttered or confusing? attractive.” ID33 Learning stimulation, motivation: Thinking “Jenny’s story is a good thing to have in there. Revisit Performance Objectives to include back to when you first were told that you had Including name, age and diabetes duration understanding the treatment options (PO.4.2), diabetes or when you learnt that diabetes makes the quotes more meaningful.” ID39 populating this through the matrix of Change could affect the eyes – would the leaflet have “Wow, that looks awesome…I didn’t expect to Objectives and into the leaflet content. Add met your information needs at that time? see two smiling faces on the front because more treatment detail to panel 8. � Do you feel that the leaflet is friendly or most diabetes things are all doom and gloom, formal? they’re so terrifying and then you don’t want � Do you feel like you want to read it now or to read them. Whereas, I read this and later? Why? thought, this was a reminder for me to book in for my eye check.”ID40 “I loved the ‘What happens if I had DR’ section. I kept putting off an eye check because I was scared of what would happen. Can you add more about what the treatment is?” ID32 Cultural appropriateness: Do the quotes “I love the pictures; they speak to different Retain multicultural imagery. represent key emotions or experiences that cultural backgrounds.” ID32 you have felt about eye examinations? Was “English is not my first language, but I didn’t the language used throughout the leaflet have any problem reading the leaflet.” ID33 familiar and culturally appropriate to you? � Were there any sections that you found confusing or were unsure about? Feedback from stakeholder reviewers was positive, finalised, leaflet printing was managed by Diabetes with minimal critique offered. Young adult PPI members Victoria (the state agent of the NDSS). gave more considered commentary on what they found useful and what could be improved (Table 9). Where Intervention implementation appropriate, the leaflet was revised to improve content, Planning for program adoption and implementation imagery, readability and cultural acceptability. Once started at study commencement and was heavily Lake et al. BMC Health Services Research (2018) 18:396 Page 13 of 18 influenced by contractual obligations with the funder. ensuring long-term sustainability of the intervention. Fur- These included a one-off statewide distribution of the ther, to enhance reach, an electronic copy of the leaflet was leaflet to all eligible NDSS registrants, timed to coincide made freely available via Diabetes Victoria and Vision 2020 with Vision 2020 Australia public awareness campaign. Australia [70, 71] and promoted to healthcare professionals To protect registrants’ privacy, the NDSS distributed the and members of the priority population. final leaflet (presented in Fig. 3) directly to members of the priority population, on behalf of the planning team. Plans Planning for intervention evaluation are underway for a revision of the NDSS ‘starter pack’ to in- Similarly, evaluation planning started at study commence- clude the eye health leaflet for young adults with T2D, ment. The planning group determined that the best Fig. 3 Who is looking after your eyes? tailored leaflet. ©Vision 2020 Australia, 2018. All rights reserved Lake et al. BMC Health Services Research (2018) 18:396 Page 14 of 18 method of evaluation of the leaflet intervention was a two- Intervention Description and Replication (TIDieR) arm, wait-list randomised controlled trial with screening checklist and guidance [80]. uptake as the primary outcome and change in modifiable behavioural determinant constructs as secondary outcomes. Strengths and limitations The trial, registered with the Australian and New Zealand The key strengths of this work relate to the use of IM, Clinical Trials Registry (ACTRN12614001110673), is now which combines both innovative and traditional interven- complete, and a manuscript is in preparation. tion development activities into an organised, systematic process, and which is consistent with the UK MRC frame- Discussion work for the design and evaluation of complex interven- Uptake of retinal screening from diabetes diagnosis is tions [24]. In the face of limited existing evidence, the crucial for the early identification of DR. In this study, empirical needs assessment, complemented by contribution we undertook the systematic development of an evidence- from the multidisciplinary planning team, key stakeholders based health behaviour change intervention tailored to the and the young adults with T2D PPI group, enabled compre- needs of a priority population, young adults with T2D, hensive exploration of the problem, providing a robust who are at risk of low retinal screening uptake and vision foundation to the intervention. Further, the use of sound loss from DR. theoretical underpinnings, causal modelling, and detailed To date, lack of information on the determinants of pilot and review, provided assurance as to the validity of retinal screening behaviour among young adults with the outcome. As such, the ‘Who is looking after your eyes?’ T2D, and on the elements of individual-level DR screen- leaflet was both evidence-based and sensitive to the needs ing interventions [72], has hampered development of ef- and characteristics of young adults with T2D. fective, targeted intervention strategies for this priority Nonetheless, this study was subject to several limita- population. Further, previous print-based retinal screen- tions. First, the vast majority of studies targeting youth ing interventions have been limited in focus, aiming pri- and young adults with T2D face recruitment challenges marily to increase knowledge and awareness of DR, and [20, 47, 81], and our empirical studies were no different of retinal screening, and neglecting to target other in this respect. Despite numerous steps taken to improve behavioural determinants, such as social norms and in- recruitment, only 10 young adults with T2D participated tentions [72, 73], despite the acknowledged role of psy- in the qualitative study and only 2% of the eligible popu- chosocial factors in health behaviour [74]. lation completed the quantitative online study. The needs assessment described here is the first large- It is likely that recruitment was impacted by a range of scale, mixed-method exploration of modifiable behav- challenges typically specific to young adults with T2D, ioural factors impacting retinal screening behaviour such as social disadvantage, disengagement with existing among young adults with T2D. The findings highlighted services, and complex psychosocial and health needs that many of the clinical and psychosocial barriers to [44, 46, 47, 82]. Furthermore, the needs assessment stud- diabetes self-management faced by young adults with ies were conducted concurrent with a number of other T2D more broadly [18, 20, 40, 41, 75–78], also apply to research projects managed by the NDSS, which may retinal screening. Importantly, when compared to older have contributed to study ‘fatigue’ for this already small adults with T2D, young adults with T2D face both an ac- population (personal communication, D. Rae, National cumulation of barriers to retinal screening, and a num- Inventory Manager, NDSS). Although low sample size ber of uniquely salient barriers and facilitators [44], potentially impacted the generalisability of the needs as- warranting tailored intervention. sessment findings, the response rate for the national sur- Combined with consensus-driven selection of Per- vey was larger than any other conducted to date with formance Objectives, theoretically-grounded change this priority population. methods and comprehensive pilot and review, IM pro- Second, this study was limited to one priority popula- vided the means by which to develop a retinal screening tion where in fact, several populations have been identi- promotion intervention that was both evidence-based fied as at-risk for low retinal screening and vision loss and sensitive to the needs and characteristics of young from DR. These include young adults with T1D, those adults with T2D. However, despite this being a relatively living in socio-economically deprived areas or from mi- simple, single-focus intervention, we shared the experi- nority ethnic and Indigenous populations [83–86], each ence of other programmes, which reported the IM process of which warrant targeted evidence-based intervention, to be both resource and time-intensive [28, 29, 79]. In informed by population-specific needs assessments. particular, we found the high degree of process documen- Finally, many key contextual elements (e.g. interven- tation time-consuming, although we acknowledge that tion level, delivery medium and format) were this activity was crucial for transparency of reporting, and externally prescribed within a broader sphere of real- conforms to key items in the Template for world logistic and contractual limitations. Although Lake et al. BMC Health Services Research (2018) 18:396 Page 15 of 18 unavoidable, this limitation meant that our interven- preliminary evaluation of the ‘Who is looking after your tion was unable to address external factors known to eyes?’ leaflet shows it meets the needs of young adults with impact screening behaviour, such as the cultural di- T2D and its effectiveness in promoting uptake of retinal versity of young adults with T2D, low socioeconomic screening can now be evaluated in a fully-powered RCT. status and lack of English language proficiency, poten- tially limiting effectiveness. Given that NDSS database Additional files strictures limited the intervention to a format suited Additional file 1: Literature study. Literature study procedure and findings. to postal delivery, the leaflet design was suited to the This file describes the procedure and findings of the literature study stated purpose for state-wide implementation. Dia- component of the needs assessment. (DOCX 78 kb) betes Victoria has ensured sustainability and reach of Additional file 2: Interview guide. Interview guide used in qualitative the intervention by regularly updating their resources component of needs assessment. This file presents all interview guide items which comprise the in-depth qualitative interview component of and making an electronic version of the leaflet freely the needs assessment. (DOCX 42 kb) available on its website [70]. Additional file 3: Modifiable behavioural determinants by baseline retinal screen (N = 129). This file presents individual items and findings Future directions from the quantitative online survey component of the needs assessment. (DOCX 22 kb) Recent research suggests that an individual’s beliefs Additional file 4: Matrix of Change Objectives. This file presents the about diabetes and self-management, are most likely to complete matrix of Change Objectives (an illustrative example is provided be influenced early in their diabetes trajectory [87]. in-text in Table 6). The Change Objectives are created at the intersection point Certainly, this appears to be the case for retinal screen- of the five targeted modifiable behavioural determinants (Knowledge, Attitudes, Normative Beliefs, Intention, Behavioural Skills) in columns, ing where, once initiated, the behaviour is generally sus- and sub-objectives (from in-text Table 5), in rows. (DOCX 19 kb) tained [73]. Thus, we recommend targeting individuals Additional file 5: Intervention map linking leaflet content directly back recently diagnosed with T2D via the NDSS, with regis- to Performance Objectives and Change Objectives. This file presents a tration date considered a proxy for date of diabetes diag- complete intervention map (an illustrative example is provided in-text in Table 8). The intervention map links all leaflet content directly back nosis. The leaflet could be used to promote national to the Performance Objectives (specified in-text in Table 5)and the retinal screening programmes in this age group and Change Objectives (Illustrated in-text in Table 6 and presented in full, would be of greatest benefit if translated into additional in Additional file 4.(DOCX 37 kb) languages. Further, this process could be utilised to produce tailored resources designed to increase awareness Abbreviations DR: Diabetic retinopathy; GP: General practitioner; IM: Intervention mapping; and screening for other populations at high-risk of DR MRC: Medical Research Council; NDSS: National Diabetes Services Scheme; (such as young adults with T1D), or for other SAM: Suitability assessment of materials; T2D: Type 2 diabetes mellitus diabetes-related complications which impact young adults with T2D (such as nephropathy and cardiovas- Acknowledgements We thank the people with diabetes who participated in the needs assessments cular disease [88]). and piloting of the leaflet. We thank Virginia Hagger (previously of Diabetes Our experience of the time and resource-intensive na- Victoria) for her involvement in study scoping, prior to CH joining the planning ture of IM reinforces that expressed by others and we team. We thank Nino Soerendata (Diabetes Victoria) for his skilled graphic design work on the leaflet. suggest that undertaking the full IM methodology may not be suitable for all situations. As such, we recom- Funding mend that future programme planners explore alterna- The study was a designated Vision Initiative activity. The Vision Initiative is an integrated health promotion program funded by the Victorian Government and tive options where possible, such as adapting an existing, managed by Vision 2020 Australia. The funding body had no role in design of the effective intervention to their target population. This can study, data collection, analysis or interpretation, or preparation of the manuscript. be enabled by use of a simplified process (IM Adapt), Availability of data and materials which guides decisions regarding selection of appropri- The datasets used in the current study are available from the corresponding author. ate intervention, and components, to adapt [89]. Author’s contributions Conclusions JS, JLB and DT conceived the study. All authors made substantial contributions to study design and intervention development. AJL managed all aspects of the In conclusion, our mixed method needs assessment has study, including conducting the needs assessment (qualitative and quantitative highlighted salient challenges faced by young adults with studies informed by literature review), developing the persuasive messages, T2D and we have demonstrated that IM is a feasible and liaising with stakeholder groups, and piloting the leaflet. AJL led the process of data analysis and interpretation, with substantial input from JLB and JS. CA worthwhile approach to use for the development of an provided expert advice on IM, theoretical basis, the quantitative study and was evidence-based, engaging resource to promote retinal closely involved with content validation of the leaflet. The planning team (AJL, screening to young adults with T2D. This detailed JB, JS, DT, GR, and CH) provided substantial input throughout the project and reviewed and approved materials at all stages. AJL wrote the first draft of this illustration will enable researchers and health promotion manuscript; JLB, JS, GR and CA provided substantial intellectual input through specialists to adopt IM methods when developing reviewing the first and subsequent drafts. All authors approved the final interventions tailored to high-risk groups. Meanwhile, manuscript. Lake et al. BMC Health Services Research (2018) 18:396 Page 16 of 18 Authors’ information 13. American Diabetes Association. Standards of care: microvascular complications AL is a Research Fellow with The Australian Centre for Behavioural Research and foot care. Diabetes Care. 2017;40(Suppl 1):88–98. in Diabetes; a partnership for better health between Diabetes Victoria and 14. National Institute for health and Care Excellence. Type 2 diabetes in adults: Deakin University. This manuscript forms part of the research associated with management NICE guideline [NG28]. 2015. https://www.nice.org.uk/ her PhD. guidance/ng28/chapter/1-Recommendations#managing-complications. Accessed 13 May 2018. Ethics approval and consent to participate 15. Scanlon PH, Stratton IM, Leese GP, Bachmann MO, Land M, Jones C, et al. The studies received ethics approval from the Deakin University Human Research Screening attendance, age group and diabetic retinopathy level at first Ethics Committee (in-depth interview component: 2013–157, quantitative survey screen. Diabet Med. 2016;33(7):904–11. and planned randomised controlled trial evaluation: 2014–156). Participants 16. Wang SY, Andrews CA, Gardner TW, Wood M, Singer K, Stein JD. Ophthalmic provided written informed consent and permission for publication of screening patterns among youths with diabetes enrolled in a large US de-identified quotations at study registration. managed care network. JAMA Ophthalmology. 2017;35(5):432–8. 17. Villarroel MA, Vahratian A, Ward BW. Health care utilization among U.S. adults Competing interests with diagnosed diabetes, 2013. NCHS Data Brief; 2015. www.cdc.gov/nchs/ The authors declare that they have no competing interests. data/databriefs/db183.pdf. Accessed 13 May 2018. 18. Savage S, Dabkowski S, Dunning T. The education and information needs of young adults with type 2 diabetes: a qualitative study. J Nurs Healthcare Publisher’sNote Chronic Illn. 2009;1(4):321–30. Springer Nature remains neutral with regard to jurisdictional claims in 19. Forward H, Hewitt AW, Mackey DA. Missing X and Y: a review of participant published maps and institutional affiliations. ages in population-based eye studies. Clin Exp Ophthalmol. 2012;40(3):305–19. 20. Browne JL, Scibilia R, Speight J. The needs, concerns, and characteristics of Author details younger Australian adults with type 2 diabetes. Diabet Med. 2013;30(5):620–6. 1 2 School of Psychology, Deakin University, Geelong, Australia. The Australian 21. Song SH. Young-onset type 2 diabetes - time to realign clinical priorities. Int Centre for Behavioural Research in Diabetes, Diabetes Victoria, Melbourne, J Diabetes Clin Res. 2015;2:39. Australia. Institute of Health Research, University of Exeter Medical School, 22. MacLennan PA, McGwin G, Heckemeyer C, et al. Eye care use among a 4 5 Exeter, UK. Vision 2020 Australia, Melbourne, Australia. Diabetes Victoria, high-risk diabetic population seen in a public hospital’sclinics. JAMA Melbourne, Australia. Centre for Eye Research Australia, Royal Victorian Eye Ophthalmol. 2014;132(2):162–7. and Ear Hospital, Melbourne 3002, Australia. Ophthalmology, Department of 23. Wilmot E, Idris I. Early onset type 2 diabetes: risk factors, clinical impact and Surgery, University of Melbourne, Melbourne 3010, Australia. AHP Research, management. Ther Adv in Chronic Dis. 2014;5(6):11. Hornchurch, UK. 24. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council Received: 5 September 2017 Accepted: 4 May 2018 guidance. BMJ. 2008;337:a1655. 25. Bartholomew Eldredge LK, Markham CM, Ruiter RAC, Fernandez ME, Kok G, Parcel GS. Planning health promotion programs : an intervention mapping References approach. 4th ed. San Francisco: Jossey-Bass; 2016. 1. Chan JC, Lau ES, Luk AO, Cheung KK, Kong AP, Yu LW, et al. Premature 26. Intervention mapping. https://interventionmapping.com/. Accessed 13 May 2018. mortality and comorbidities in young-onset diabetes: a 7-year prospective 27. Ball GD, Mushquash AR, Keaschuk RA, Ambler KA, Newton AS. Using analysis. Am J Med. 2014;127(7):616–24. intervention mapping to develop the parents as agents of change (PAC) 2. Constantino MI, Molyneaux L, Limacher-Gisler F, Al-Saeed A, Luo C, Wu T, intervention for managing pediatric obesity. BMC Res Notes. 2017;10(1):43. et al. Long-term complications and mortality in young-onset diabetes: type 28. Hurley DA, Murphy LC, Hayes D, Hall AM, Toomey E, McDonough SM, et al. 2 diabetes is more hazardous and lethal than type 1 diabetes. Diabetes Using intervention mapping to develop a theory-driven, group-based complex Care. 2013;36(12):3863–9. intervention to support self-management of osteoarthritis and low back pain 3. Zimmet P, Magliano DJ, Herman W, Shaw J. Diabetes: a 21st century challenge. (SOLAS). Implement Sci. 2016;11:56. Lancet Diabetes & Endocrinology. 2014;2(1):56–64. 29. Gray-Burrows KA, Day PF, Marshman Z, Aliakbari E, Prady SL, McEachan RRC. 4. Harding J, Shaw JE, Peeters A, Davidson S, Magliano DJ. Age-specific trends Using intervention mapping to develop a home-based parental-supervised from 2000–2011 in all-cause and cause-specific mortality in type 1 and type toothbrushing intervention for young children. Implement Sci. 2016;11:61. 2 diabetes: a cohort study of more than one million people. Diabetes Care. 30. Song M, Choi S, Kim S-a, Seo K, Lee SJ. Intervention mapping protocol for 2016;39(6):1018–26. developing a theory-based diabetes self-management education program. 5. Tancredi M, Rosengren A, Svensson AM, Kosiborod M, Pivodic A, Gudbjornsdottir Res Theory Nurs Pract. 2015;29(2):94–112. S, et al. Excess mortality among persons with type 2 diabetes. N Engl J Med. 31. Newby KV, Brown KE, Bayley J, Kehal I, Caley M, Danahay A, et al. Development 2015;373(18):1720–32. of an intervention to increase sexual health service uptake by young people. 6. Song SH, Hardisty CA. Early onset type 2 diabetes mellitus: a harbinger for Health Promot Pract. 2017;18(3):391–9. complications in later years—clinical observation from a secondary care 32. Garba RM, Gadanya MA. The role of intervention mapping in designing cohort. QJM. 2009;102(11):799–806. disease prevention interventions: a systematic review of the literature. PLoS 7. Dart AB, Martens PJ, Rigatto C, Brownell MD, Dean HJ, Sellers EA. Earlier onset of One. 2017;12(3):e0174438. complications in youth with type 2 diabetes. Diabetes Care. 2014;37(2):436–43. 33. Elliott LR, White MP, Taylor AH, Abraham C. How do brochures encourage 8. Al-Saeed AH, Constantino MI, Molyneaux L, D’Souza M, Limacher-Gisler F, walking in natural environments in the UK? A content analysis. Health Promot Luo C, et al. An inverse relationship between age of type 2 diabetes onset Int. 2016;33(2):299–310. and complication risk and mortality: the impact of youth-onset type 2 34. Hill CA, Abraham C. School-based, randomised controlled trial of an evidence- diabetes. Diabetes Care. 2016;39:823–9. based condom promotion leaflet. Psychol Health. 2008;23(1):41–56. 9. Leasher JL, Bourne RAA, Flaxman SR, Jonas JB, Keeffe J, Naidoo K, et al. 35. Abraham C, Southby L, Quandte S, Krahé B, ven der Sluijs W. What's in a Global estimates on the number of people blind or visually impaired leaflet? Identifying research-based persuasive messages in European alcohol- by diabetic retinopathy: a meta-analysis from 1990 to 2010. Diabetes education leaflets. Psychol Health. 2007;22(1):31–60. Care. 2016;39:1643–9. 36. Krahé B, Abraham C, Scheinberger-Olwig R. Can safer-sex promotion leaflets 10. Wong J, Molyneaux L, Constantino M, Twigg SM, Yue DK. Timing is everything: change cognitive antecedents of condom use? An experimental evaluation. age of onset influences long-term retinopathy risk in type 2 diabetes, Br J Health Psychol. 2005;10:203–20. independent of traditional risk factors. Diabetes Care. 2008;31(10):1985–91. 11. Ferris FL. How effective are treatments for diabetic retinopathy? JAMA : the 37. Stewart D, Wilson R, Selby P, Darbyshire J. Patient and public involvement. journal of the American Medical Association. 1993;269(10):1290–1. Ann Oncol. 2011;22(Suppl 7):54–6. 12. Mitchell P and Foran S. Guidelines for the Management of Diabetic Retinopathy. 38. Elam AR, Lee PP. High-risk populations for vision loss and eye care 2008 https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/di15.pdf. underutilization: a review of the literature and ideas on moving forward. Surv Accessed 13 May 2018. Ophthalmol. 2013;58(4):348–58. Lake et al. BMC Health Services Research (2018) 18:396 Page 17 of 18 39. Koelmeyer RL, Dharmage SC, English DR. Diabetes in young adult men: 62. Muller BC, Ritter SM, Glock S, Dijksterhuis A, Engels RC, van Baaren RB. social and health-related correlates. BMC Public Health. 2016;16(S3):63–9. Smoking-related warning messages formulated as questions positively 40. Browne JL, Nefs G, Pouwer F, Speight J. Depression, anxiety and self-care influence short-term smoking behaviour. J Health Psychol. 2014;21(1):60–8. behaviours of young adults with type 2 diabetes: results from the international 63. Cialdini RB, Goldstein NJ. Social influence: compliance and conformity. Annu diabetes management and impact for long-term empowerment and success Rev Psychol. 2004;55:591–621. (MILES) study. Diabet Med. 2014;32(1):133–40. 64. Abraham C, Kools M. Writing health communication: an evidence-based 41. Hessler DM, Fisher L, Mullan JT, Glasgow RE, Masharani U. Patient age: a guide. London: Sage; 2012. neglected factor when considering disease management in adults with 65. Abraham C. In: Abraham C, Kools M, editors. Mapping change mechanisms type 2 diabetes. Patient Educ Couns. 2011;85(2):154–9. onto behaviour change techniques: a systematic approach to promoting 42. Tuomi T, Santoro N, Caprio S, Cai M, Weng J, Groop L. The many faces of behaviour change through text, in: writing health communication: an diabetes: a disease with increasing heterogeneity. Lancet. 2014;383(9922): evidence based guide. London: SAGE; 2012. 1084–94. 66. Abraham C, Michie S. A taxonomy of behavior change techniques used in 43. Reyes-Velazquez W, Sealey-Potts C. Unrealistic optimism, sex, and risk perception interventions. Health Psychol. 2008;27(3):379–87. of type 2 diabetes onset: implications for education programs. Diabetes Spectr. 67. Ryan L, Logsdon MC, McGill S, Stikes R, Senior B, Helinger B, et al. Evaluation 2015;28(1):5–9. of printed health education materials for use by low-education families. 44. Lake AJ, Browne JL, Rees G, Speight J. What factors influence uptake of J Nurs Sch. 2014;46(4):218–28. retinal screening among young adults with type 2 diabetes? A qualitative 68. Scott B. Readability formulas: free readability tools to check for reading levels, study informed by the theoretical domains framework. J Diabetes Complicat. reading assessment and reading grade levels. 2016 http://www.readabilityfor 2017;31(6):997–1006. mulas.com/free-readability-formula-tests.php. Accessed 13 May 2018. 45. Owen KR. Treating young adults with type 2 diabetes or monogenic diabetes. 69. Doak CC, Doak LG, Root JH. Teaching patients with low literacy skills. 2nd Best Pract Res Clin Endocrinol Metab. 2016;30(3):455–67. ed. Philadelphia: J. B. Lippincott; 1996. 46. Zeitler P, Chou HS, Copeland KC, Geffner M. Clinical trials in youth-onset 70. Diabetes Victoria. Who is looking after your eyes? Your guide to preventing type 2 diabetes: needs, barriers, and options. Current Diabetes Reports. vision loss from diabetic retinopathy. https://s3-ap-southeast-2.amazonaws.com/ 2015;15(5):1–8. dv-resources/OrchestraCMS/a1f90000004k5XkAAI.pdf. Accessed 13 May 2018. 47. Nguyen TT, Jayadeva V, Cizza G, Brown RJ, Nandagopal R, Rodriguez LM, et 71. Vision 2020 Australia. Who is looking after your eyes? Your guide to preventing al. Challenging recruitment of youth with type 2 diabetes into clinical trials. vision loss from diabetic retinopathy. 2017 http://www.visioninitiative.org.au/ J Adolesc Health. 2014;54(3):247–54. uploads/assets/files/infosheets/ACBRD_DR%20Age%20under%2040.pdf. 48. Cane J, O'Connor D, Michie S. Validation of the theoretical domains Accessed 13 May 2018. framework for use in behaviour change and implementation research. 72. Lawrenson JG, Graham-Rowe E, Lorencatto F, Presseau J, Burr J, Ivers N, Implement Sci. 2012;7(1):37. Quartilho A, Bunce C, Francis JJ, Grimshaw JM, Peto T, Rice S, Vale L. 49. Fisher WA, Fisher JD, Harman J. The information–motivation–behavioral Interventions to increase attendance for diabetic retinopathy screening skills model: a general social psychological approach to understanding and (Protocol). Cochrane Database of Systematic Reviews 2016, Issue 1. Art. No.: promoting health behavior. In: Sulls J, Walston KA, editors. Social psychological CD012054. https://doi.org/10.1002/14651858.CD012054. foundations of health and illness. Oxford: Blackwell Publishing Ltd; 2003. p. 82–106. 73. Zhang X, Norris SL, Saadine J, Chowdhury FM, Horsley T, Kanjilal S, et al. 50. Fisher J, Fisher WA, Bryan AD, Misovich SJ. Information-motivation-behavioral Effectiveness of interventions to promote screening for diabetic retinopathy. skills model-based HIV risk behavior change intervention for inner-city high Am J Prev Med. 2007;33(4):318–35. school youth. Health Psychol. 2002;21(2):177–86. 74. Michie S, Abraham C, editors. Health psychology in practice, vol. xiv. Malden: 51. Misovich SJ, Martinez T, Fisher JD, Bryan A, Catapano N. Predicting breast Blackwell Pub., British Psychological Society; 2004. p. 416. self-examination: a test of the information-motivation-behavioral skills 75. Waitzfelder B, Pihoker C, Klingensmith G, Case D, Anderson A, Bell RA, et al. model. J Appl Soc Psychol. 2003;33(4):775–90. Adherence to guidelines for youths with diabetes mellitus. Pediatrics. 2011; 52. Osborn CY, Egede LE. Validation of an information-motivation-behavioral 128(3):531–8. skills model of diabetes self-care (IMB-DSC). Patient Educ Couns. 2010; 76. Auslander WF, Sterzing PR, Zayas LE, White NH. Psychosocial resources and 79(1):49–54. barriers to self-management in African American adolescents with type 2 53. Mayberry LS, Osborn CY. Empirical validation of the information-motivation- diabetes: a qualitative analysis. Diabetes Educator. 2010;36(4):613–22. behavioral skills model of diabetes medication adherence: a framework for 77. Brouwer AM, Salamon KS, Olson KA, Fox MM, Yelich-Koth SL, Fleischman intervention. Diabetes Care. 2014;37(5):1246–53. KM, et al. Adolescents and type 2 diabetes mellitus: a qualitative analysis of 54. Fishbein M, Ajzen I. Predicting and changing behaviour: the reasoned action the experience of social support. Clin Pediatr (Phila). 2012;51(12):1130–9. approach. New York: Psychology Press; 2010. 78. Sillars BA, Davis WA, Kamber N, Davis TME. The epidemiology and characteristics 55. Turner KM, Percival J, Dunger DB, Olbers T, Barrett T, Shield JPH. Adolescents’ of type 2 diabetes in urban, community-based young people. Intern Med J. 2010; views and experiences of treatments for type 2 diabetes: a qualitative study. 40(12):850–4. Diabet Med. 2015;32(2):250–6. 79. Greaves CJ, Wingham J, Deighan C, Doherty P, Elliott J, Armitage W, et al. 56. Australian Institute of Health and Welfare. Type 2 diabetes in Australia’s Optimising self-care support for people with heart failure and their caregivers: children and young people: a working paper. In: Diabetes series No.21 2014 development of the rehabilitation enablement in chronic heart failure (REACH- http://www.aihw.gov.au/publication-detail/?id=60129546361. Accessed 13 HF) intervention using intervention mapping. Pilot Feasibility Stud. 2016;2:37. May 2018. 80. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better 57. Freijy T, Kothe EJ. Dissonance-based interventions for health behaviour reporting of interventions: template for intervention description and replication change: a systematic review. Br J Health Psychol. 2013;18:310–37. (TIDieR) checklist and guide. BMJ (Clinical Research Ed). 2014;348:g1687. 58. Commonwealth of Australia. National framework for action to promote eye 81. Speight J, Browne JL, Holmes-Truscott E, Hendrieckx C, Pouwer F. Diabetes health and prevent avoidable blindness and vision loss. 2005 http://www. MILES–Australia (management and impact for long-term empowerment health.gov.au/internet/main/publishing.nsf/content/D3175B31C04E3D72 and success): methods and sample characteristics of a national survey of CA257C750078F76B/$File/frame.pdf. Accessed 13 May 2018. the psychological aspects of living with type 1 or type 2 diabetes in Australian 59. Bailey SC, Brega AG, Crutchfield TM, Elasy T, Herr H, Kaphingst K, et al. adults. BMC Public Health. 2012;12:120. Update on health literacy and diabetes. The Diabetes educator. 2014;40: 82. Nadeau KJ, Anderson BJ, Berg EG, Chiang JL, Chou H, Copeland KC, et al.. 581–604. Youth-onset type 2 diabetes consensus report: current status, challenges, and priorities. Diabetes Care. 2016;39(9):1635–42. 60. Beard E, Clark M, Hurel S, Cooke D. Do people with diabetes understand their clinical marker of long-term glycemic control (HbA1c levels) and does 83. Moreton RBR, Stratton IM, Chave SJ, Lipinski H, Scanlon PH. Factors this predict diabetes self-care behaviours and HbA1c? Patient Educ Couns. determining uptake of diabetic retinopathy screening in Oxfordshire. Diabet 2010;80(2):227–32. Med. 2017;34(7):993–9. 61. Werrij MQ, Ruiter RAC, van't Riet J, de Vries H. Message framing. In: Abraham C, 84. Foreman J, Keel S, Xie J, Van Wijngaarden P, Taylor HR, Dirani M. Adherence Kools M, editors. Writing health communication: an evidence based guide. to diabetic eye examination guidelines in Australia: the National eye Health London: SAGE; 2012. p. 134–43. Survey. Med J Aust. 2017;206(9):402–6. Lake et al. BMC Health Services Research (2018) 18:396 Page 18 of 18 85. Paksin-Hall A, Dent ML, Dong F, and Ablah E. Factors contributing to diabetes patients not receiving annual dilated eye examinations. Ophthalmic Epidemiol. 2013;20(5):281–7. 86. Shi Q, Zhao Y, Fonseca V, Krousel-Wood M, Shi L. Racial disparity of eye examinations among the U.S. working-age population with diabetes: 2002-2009. Diabetes Care. 2014;37(5):1321–8. 87. Skinner TC, Khunti K, Carey ME, Dallosso H, Heller S, Davies MJ. Stability and predictive utility, over 3 years, of the illness beliefs of individuals recently diagnosed with type 2 diabetes mellitus. Diabet Med. 2014;31(10):1260–3. 88. Tryggestad JB, Willi SM. Complications and comorbidities of T2DM in adolescents: findings from the TODAY clinical trial. J Diabetes Complicat. 2015; 29(2):307–12. 89. Highfield L, Hartman MA, Mullen PD, Rodriguez SA, Fernandez ME, Bartholomew LK. Intervention Mapping to Adapt Evidence-Based Interventions for Use in Practice: Increasing Mammography among African American Women. Biomed Res Int. 2015;2015:160103. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Health Services Research Springer Journals

A tailored intervention to promote uptake of retinal screening among young adults with type 2 diabetes - an intervention mapping approach

Free
18 pages

Loading next page...
 
/lp/springer_journal/a-tailored-intervention-to-promote-uptake-of-retinal-screening-among-8k0lIDdUS0
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Public Health; Health Administration; Health Informatics; Nursing Research
eISSN
1472-6963
D.O.I.
10.1186/s12913-018-3188-5
Publisher site
See Article on Publisher Site

Abstract

Background: Young adults (18–39 years) with type 2 diabetes are at risk of early development and rapid progression of diabetic retinopathy, a leading cause of vision loss and blindness in working-age adults. Retinal screening is key to the early detection of diabetic retinopathy, with risk of vision loss significantly reduced by timely treatment thereafter. Despite this, retinal screening rates are low among this at-risk group. The objective of this study was to develop a theoretically-grounded, evidence-based retinal screening promotion leaflet, tailored to young adults with type 2diabetes. Methods: Utilising the six steps of Intervention Mapping, our multidisciplinary planning team conducted a mixed-methods needs assessment (Step 1); identified modifiable behavioural determinants of screening behaviour and constructed a matrix of change objectives (Step 2); designed, reviewed and debriefed leaflet content with stakeholders (Steps 3 and 4); and developed program implementation and evaluation plans (Steps 5 and 6). Results: Step 1 included in-depth qualitative interviews (N = 10) and an online survey that recruited a nationally- representative sample (N = 227), both informed by literature review. The needs assessment highlighted the crucial roles of knowledge (about diabetic retinopathy and screening), perception of personal risk, awareness of the approval of significant others and engagement with healthcare team, on retinal screening intentions and uptake. In Step 2, we selected five modifiable behavioural determinants to be targeted: knowledge, attitudes, normative beliefs, intention, and behavioural skills. In Steps 3 and 4, the “Who is looking after your eyes?” leaflet was developed, containing persuasive messages targeting each determinant and utilising engaging, cohort-appropriate imagery. In Steps 5 and 6, we planned Statewide implementation and designed a randomised controlled trial to evaluate the leaflet. Conclusions: This research provides an example of a systematic, evidence-based approach to the development of a simple health intervention designed to promote uptake of screening in accordance with national guidelines. The methods and findings illustrate how Intervention Mapping can be employed to develop tailored retinal screening promotion materials for specific priority populations. This paper has implications for future program planners and is intended to assist those wishing to use Intervention Mapping to create similar theoretically-driven, tailored resources. Keywords: Type 2 diabetes, Young adults, Diabetic retinopathy, Intervention mapping, Needs assessment, Retinal screening, Young-onset, Health behaviour change * Correspondence: alake@acbrd.org.au School of Psychology, Deakin University, Geelong, Australia The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria, Melbourne, Australia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lake et al. BMC Health Services Research (2018) 18:396 Page 2 of 18 Background been shown to be effective both in identifying determi- Worldwide increase in the prevalence of type 2 diabetes nants and increasing uptake for a range of disease pre- (T2D) in young adults (< 40 years), with its associated vention interventions [32]. Utilising IM, the aim of the considerable morbidity and mortality, is a burgeoning current study was to identify determinants of screening public health concern [1–5]. Adverse phenotype [6], behaviour for young adults with T2D, and develop an sub-optimal glycemic (blood glucose) control and long engaging psycho-educational resource to target these diabetes duration expose young adults with T2D to a factors, designed to promote screening uptake. high lifetime risk of diabetes-related complications [7, 8]. One of the most common is diabetic retinopathy (DR), Method and results which is a leading cause of vision loss and blindness in In this section, IM steps 1–4 are presented in detail, working age adults [9, 10]. followed by summaries of Steps 5 and 6. Method and re- Early detection of DR via retinal screening (‘screening’), sults are reported separately for each step, including il- followed by timely treatment, are crucial factors in pre- lustrative examples of key IM activities (with full detail venting vision loss [11]. Australian national Guidelines for provided in Additional files). Table 1 provides an over- the Management of Diabetic Retinopathy recommend view of each IM step as it was applied to this project. screening uptake at diabetes diagnosis, repeated at least every two years thereafter [12], an interval less frequent Logic model of the problem than that prescribed for adults with T2D in the United Establish and work with a planning group States (US) and United Kingdom (UK) [13, 14]. Unfortu- A six-person multidisciplinary planning team was con- nately however, young adults (aged 18–39 years) are the vened comprising representatives from The Australian least likely to initiate retinal screening in accordance with Centre for Behavioural Research in Diabetes (AJL, JLB, national guidelines and have lower overall screening rates JS); Centre for Eye Research Australia (GR); Diabetes than older adults (aged ≥40 years) or young adults with Victoria (CH); and Vision 2020 Australia (DT). Combined, type 1 diabetes [15–17]. In addition to their low engage- the planning team provided expertise in psychosocial and ment with existing diabetes services [18], additional com- clinical aspects of diabetes and vision loss, health munication challenges exist due to the lack of dedicated programs, hubs or services for young adults with T2D. Table 1 Overview of IM steps and activities applied to the current leaflet development program Thus, there is need for the development of tailored, evidence-based health promotion resources, using an IM steps IM activities application appropriate to the culture and context, in Step 1: Logic model of � Establish and work with a planning group the problem order to encourage screening uptake among this � Conduct mixed-methods needs assessment priority population [19–23]. � Create logic model of the problem Best-practice development of health promotion re- � Describe context of the intervention and sources targets modifiable behavioural determinants for state program goals a clearly specified health behaviour. The UK Medical Step 2: Program outcomes � State expected behavioural outcomes Research Council (MRC) framework for the design and and objectives; logic model and Performance Objectives (PO) evaluation of complex interventions recommends use of of change � Create logic model of change good quality evidence from a range of sources, strong � Create matrix of Change Objectives theoretical underpinnings, causal modelling and a well- Step 3: Program design � Generate program themes, components, designed evaluation [24]. Intervention mapping (IM) is a scope and sequence six-step protocol encompassing MRC elements, which � Choose theory and evidence-based change provides an effective and useful framework for this pur- methods pose [25]. Key activities are: 1) detailed needs assess- Step 4: Program production � Draft persuasive message content and ment, developing causal logic model of the problem, 2) leaflet stating program outcomes and performance objectives, � Pre-test, refine and produce leaflet developing logic model of change, 3) utilising theory and Step 5: Program � Identify program implementers, adopters evidence-based change methods, designing program to implementation and maintainers target identified behavioural determinants, 4) producing, � Design implementation and liaise with pre-testing and refining program with broad stakeholder program implementers input, 5) planning for program implementation, and 6) Step 6: Program evaluation � Write effect and process evaluation planning for evaluation [26]. Intervention mapping has questions been widely used by intervention planners to guide the � Develop measures for assessment development of effective health promotion materials in a � Specify and complete evaluation plan variety of contexts and populations [27–31] and has Lake et al. BMC Health Services Research (2018) 18:396 Page 3 of 18 promotion, behaviour change research methodologies and analysis, are published elsewhere [44], see Additional file 2 intervention development. Monthly meetings (chaired by for interview guide. In brief, we conducted in-depth semi- JS) were held throughout the project, with additional structured interviews to explore factors affecting screening meetings held as-needed, and quarterly progress reports behaviour for young adults with T2D, with an emphasis on provided to the funding body. Throughout the study, the those that were individual-level and modifiable. The study planning team consulted a practicing health psychologist was advertised widely online and in community settings, (CA) with expertise in IM and a track record of developing and recruitment invitations were mailed to eligible mem- and analysing evidence-based health promotion leaflets bers of a leading state diabetes consumer advocacy organ- [33–36]. Additional expert input was provided by represen- isation. All interviews were conducted via telephone by an tatives from key stakeholder organisations, such as the Na- experienced interviewer (AJL), audio-recorded and tional Diabetes Services Scheme (NDSS, an initiative of the transcribed verbatim. All transcripts were checked for Australian Government, which provides free or subsidised accuracy and imported into NVivo10 (QSR Inter- self-management supplies and services to registrants), Op- national Pty Ltd., Doncaster, VIC., Australia, 2012). tometry Australia and key units within Diabetes Victoria. Transcripts were subjected to content analysis (by Patient and public involvement (PPI) is essential for the AJL), with each participant utterance coded for development of high-quality health behaviour change in- behavioural determinants (using an a priori coding terventions [37] and is recommended specifically as a framework informed by the literature [48]), and again strategy for engaging groups at high risk of underutilsation as either ‘facilitator’ or ‘barrier’ dependent upon the of eye healthcare services [38]. In this study, five young context. Twenty percent of transcripts were double- adults with T2D were involved, providing feedback on all coded (by JLB), with high inter-rater reliability of study documentation, piloting the quantitative survey and 99%. Screening determinants were rank-ordered by providing detailed review of the eye health leaflet. frequency of coding (higher frequency of utterances interpreted to indicate higher salience). Conduct mixed methods needs assessment Qualitative interview findings Our study of the literature (summarised in Additional file 1) In brief, ten young adults with T2D (50% women, revealed that, while there was a paucity of research in this aged 29–37 years) were interviewed (average length: specific area, sub-optimal diabetes self-management (in 55 min, range: 31–106 min). Fifty percent had not general) among young people is likely driven by low socio- attended retinal screening previously. Although young economic status [39], low general and health literacy [39], adults with T2D knew of a link between diabetes and low engagement with diabetes self-management education vision loss, they did not have a comprehensive under- [20, 39–41], cultural diversity of the priority population standing of DR or screening (e.g. symptoms, risk [42], optimistic bias and low risk perception [43], life-stage factors, screening guidelines, distinction between demands [44], high rates of diabetes-related distress [40] screening and standard vision checks). Participants re- and complex healthcare needs [45]. ported distress related to having a condition stereo- In our empirical needs assessment studies, we sought to typically associated with older people, and many did determine the relevance of these factors to DR screening spe- not know others of similar age with T2D. Participants cifically, and to identify any additional factors that may facili- indicated that absence of social influence (e.g. prompt- tate or impede this target behaviour. As other researchers ing from significant others, social comparison with others), have found it challenging to recruit young adults with T2D to and low DR risk perception, combined with life-stage research studies [46, 47], several steps were taken to boost re- barriers (e.g. lack of time and finances), negatively im- cruitment in the mixed-methods needs assessment. These pacted screening uptake. Concerned about negative included: giving priority to ease of participant access; judgment by others, and fearing a DR diagnosis, partici- distribution of engaging, cohort-appropriate recruit- pants reported that they did not always disclose their dia- ment invitations with an NDSS and Diabetes Australia betes diagnosis or proactively seek healthcare or social branded cover letter introducing the study; reminder support, thus losing crucial pathways to timely screening invitation after four weeks; age-appropriate incentives uptake. Irrespective of their screening history, young (e.g. entry to a technology-based prize draw), and ex- adults with T2D identified a range of screening bar- tension of recruitment periods until participant regis- riers, suggesting that a cumulation of factors may tration visibly flagged. impact uptake, thus highlighting the need to acknow- ledge and address a broad range of barriers in a tai- In-depth qualitative interviews lored intervention. Qualitative interview procedure Screening facilitators were often conceptualised by par- Detailed description of the study methods and findings, in- ticipants as the opposite of the barriers (e.g. improved, as cluding the participants, procedure, interview guide and opposed to inadequate, knowledge or access to social Lake et al. BMC Health Services Research (2018) 18:396 Page 4 of 18 support). However, the study also highlighted other collectively assessed two behavioural skills constructs (per- screening facilitators: participants compared them- ceived control over screening and overcoming barriers). selves with others experiencing diabetes-related vision Unless otherwise noted, each item was rated on a 7- loss, and were thus influenced to engage in screening point Likert scale (“strongly disagree” to “strongly agree”). due to concerns about the impact that vision loss Individual items were aggregated to provide a composite would have on their lives, including anticipated regret score for each construct, with good internal consistency at the potential impact on their spouses and/or (see Additional file 3). For each, higher scores indicated children. For those who previously attended screen- greater endorsement of the construct measured (e.g. ing, feelings of relief and reassurance facilitated repeat stronger intentions, more positive attitudes). In addition, screening behaviour, with participants expressing in- we collected socio-demographic data to describe the tent to sustain the behaviour and expectation of a sample at baseline. The survey was piloted with young positive outcome (i.e. no DR diagnosis). adult PPI members and representatives from selected stakeholder organisations, who also commented on read- ability, format, accessibility and content; no substantive National online survey changes were required. Online survey procedure Data collection and participants Survey development The survey was conducted nationwide and hosted Using the Information-Motivation-Behavioural skills via a secure online survey platform, Qualtrics™ (Provo, (IMB) model [49] as a foundation, the planning team de- UT, 2014–2015). In Australia, the majority of people veloped a survey designed to identify modifiable behav- with a confirmed diagnosis of diabetes are registered ioural determinants for screening. The IMB model by their health professional with the NDSS [56]. All posits that although information is a key element in young adults with T2D who had been registered on changing behaviour, increasing knowledge and aware- the NDSS in the previous three years (registration ness of a behaviour is not sufficient in itself, and re- date wasusedasa proxyfor diagnosisdate),and quires the integration of motivational and skills who had consented to be contacted for research elements to ensure behaviour change. Use of the IMB (N = 5354) were invited to participate. Exclusion model in behaviour change research requires identifi- criteria included non-English speaking; those aged 40+ cation of deficits in each of the three key areas, to be years, and diagnosis of another type of diabetes. Study addressed in a subsequent intervention. The IMB invitations were managed by the NDSS in order model has been effective both as a framework for to preserve registrant confidentiality, but purposive intervention design [50] and as a predictive model for sampling of those who had not previously screened health-related screening behaviours, such as breast for DR was not possible, due to lack of available data self-examination [51]. on retinal screening status of NDSS registrants. Re- Increasingly used with chronic conditions, the IMB cruitment to the online survey continued for seven model has been validated recently in a model of weeks. diabetes self-care behaviours [52] and medication ad- Statistical analyses herence [53]. Survey items were based on IMB-based Statistical analyses were conducted using SPSS ver- questionnaires previously validated for diabetes self- sion 22 (SPSS Inc., Chicago IL, USA). Univariate management [52, 53], the widely-used Theory of analyses (chi-square and independent measures t- Planned Behaviour [54], and cognitive constructs tests, two-sided) were conducted to explore between- showntoberelevanttoyoung adults with T2D (e.g. group (previous retinal screen: yes/no) differences on optimism/fatalism, social support, risk perception, an- demographic variables and modifiable behavioural de- ticipated regret, self-efficacy) [43, 55]. terminants at the item level (to inform specific inter- In brief, the survey comprised 54 items assessing infor- vention message content). Given the large number of mation/knowledge, motivation and behavioural skills analyses, a conservative p < 0.01 was considered sta- (see Additional file 3 for individual items). Information: tistically significant. 16 items assessed knowledge of the link between dia- Online survey findings betes and vision loss, diabetic retinopathy and retinal Overall, 129 participants (2% of eligible population) screening. Responses scored dichotomously (incorrect / completed the full survey, and their sociodemographic correct). Motivation: 21 items collectively assessed three characteristics are presented in Table 2. Sixty percent attitudinal constructs (attitudes toward screening for were women, average age 34 ± 5 years (range: 19– DR, perception of personal risk, and anticipated regret); 39 years), and 74% had previously screened for DR. No three items assessed normative beliefs and three items significant differences in sociodemographic characteris- assessed intention. Behavioural skills: 11 items tics were found between screening groups. Lake et al. BMC Health Services Research (2018) 18:396 Page 5 of 18 Table 2 Sociodemographic characteristics by screening behaviour was high for both groups. Compared to their non- (N =129) screening counterparts, those who had previously Sociodemographic Retinal screen p-value screened reported greater concern and worry at the pro- characteristics spect of not screening. Participants who had previously No (n = 33) Yes (n = 96) screened were significantly more likely to agree that sig- Age, years 34.39 (33, 37) 34.04 (32, 37) .697 nificant others (i.e. family/friends, healthcare team) Duration, years 1.00 (1.84) 1.69 (1.97) .081 would approve of screening. Intention to screen was Gender: women 15 (45) 62 (65) .084 high among all participants but significantly higher for Primary diabetes management those who had previously screened compared to those Lifestyle only 5 (15) 21 (22) who had not. Medication (not insulin) 23 (70) 64 (67) .652 Those who had screened previously reported signifi- cantly greater confidence on all aspects of behavioural Insulin 5 (15) 11 (11) control over screening (e.g. how to make an appoint- Country of birth: Australian born 18 (55) 66 (69) .206 ment for screening, ability to screen regularly, remember Main language spoken at home: 27 (82) 81 (84) .944 and attend an appointment). No differences were ob- English served between groups on confidence in knowing the Employment status: employed 20 (61) 57 (59) 1.000 steps that can be taken to reduce the risk of DR, al- Socioeconomic status 984.55 (83.52) 991.48 (57.11) .660 though scores were lower for all participants compared Family history of T2D 22 (67) 72 (75) .483 to other behavioural control items. Those who had ≥1 comorbid health condition 25 (76) 75 (79) .891 screened also reported significantly higher confidence in overcoming common screening barriers (including time Depression (PHQ-2) 2.94 (2.48) 2.12 (2.07) .102 and cost, and discussing diabetes and DR with health- Data are number (%), mean (SD), or median (IQR); p-value is Pearson’schi-square or independent t-tests (two-sided); statistical significance p < 0.05 care professionals). Index of Relative Socio-economic Advantage and Disadvantage: lower score indicates relatively greater disadvantage, range 300–1250 Missing data (average 6%, range 2–11%) Summary of key learnings from needs assessment PHQ-2 range 0–6: ≥3 indicating likely depression Key learnings from the literature review, qualitative interviews and quantitative survey are summarised in Behavioural determinants of screening Table 4. The survey identified that compared to their Selected findings for information (knowledge), motiv- non-screening counterparts, those who had previously ation and behavioural skills items are detailed in Table 3 attended screening reported: significantly higher (full detail and construct-level findings provided in knowledge of both DR and retinal screening; more Additional file 3). positive attitudes towards screening; stronger agree- Almost all participants (irrespective of previous ment that significant others would approve of the be- screening behaviour) knew of a link between diabetes haviour; higher intention to screen; greater perceived and vision loss. However, compared to their non- behavioural control (i.e. confidence that they could ar- screening counterparts, those who had previously range and attend screening when due), and greater screened knew that all people with diabetes were at risk confidence in addressing common screening barriers. of DR, the clinically-recommended HbA1c (average The findings suggest that messages highlighting the blood glucose) target for DR prevention, when to initiate prevalence of DR and link between DR and diabetes dur- screening and recommended screening intervals. ation are warranted to prompt reassessment of personal Overall, participants who had screened indicated more risk. Information on modifiable DR risk factors (blood positive attitudes towards the behaviour (e.g. empower- glucose, blood pressure and cholesterol), asymptomatic ing, reassuring and important) than those who had not nature of the condition and screening guidelines are screened. No differences were observed between groups needed to encourage individuals to both reduce DR risk on how pleasant or comfortable the eye check was per- and initiate screening. ceived to be, although scores were lower for all partici- Messages designed to highlight the health and material pants compared to other attitude items. Perception of consequences of screening, including likely positive personal risk of vision problems and DR were moderate emotional consequences, are warranted in order to pro- for all participants with low expectations of a DR diag- mote positive screening attitudes. Findings suggesting nosis in the short term. Although all participants be- that all participants perceived screening as potentially lieved they could not reduce their risk of vision ‘unpleasant’, ‘uncomfortable’ and disruptive to normal problems, those who had screened held this belief more activities are realistic considering that many people ex- strongly. All participants reported negative emotions perienced discomfort and delay from pupil dilation (my- when thinking about not screening, including fear, which driasis) drops. Consequently, positive messages should Lake et al. BMC Health Services Research (2018) 18:396 Page 6 of 18 Table 3 Selected behavioural determinant items by retinal screen (N = 129)* Modifiable behavioural determinants Retinal screen p-value No (n = 33) Yes (n = 96) INFORMATION (KNOWLEDGE) ITEMS Diabetes can lead to vision loss 30 (91) 93 (97) .174 All people with diabetes are at risk of DR 26 (79) 89 (93) .004 Recommended target HbA1c 17 (53) 81 (87) <.001 Initiate eye examinations ‘at diabetes diagnosis’ 5 (15) 42 (45) .004 Screen ‘at least every 2 years’ if no DR present 0 (0) 18 (19) .003 b c MOTIVATION ITEMS An eye health check for DR would be... ...(not) ‘unpleasant’ 3.71 (0.94) 3.86 (1.07) .500 ...reassuring 3.94 (0.96) 4.63 (0.61) <.001 ...important 4.06 (1.06) 4.89 (0.35) <.001 ...empowering 3.10 (1.19) 3.73 (0.97) .004 ...comfortable 3.26 (1.15) 3.68 (1.10) .073 I believe I will develop DR due to my diabetes 4.03 (1.45) 4.14 (1.62) .734 Expect to be diagnosed with DR at next eye check 2.97 (1.47) 2.43 (1.66) .114 Can reduce risk of vision problems... 2.32 (1.44) 1.43 (0.79) .002 If I did NOT have an eye health check for DR, I would feel... ...concerned 5.03 (1.70) 5.88 (1.40) .007 ...fearful 4.48 (1.79) 5.13 (1.70) .073 ...worried 4.65 (1.80) 5.53 (1.47) .007 My family/close friends would approve... 5.94 (1.69) 6.82 (0.80) .008 I plan to attend an eye health check... 4.26 (2.32) 6.76 (0.77) <.001 I intend to have an eye health check... 4.42 (2.32) 6.74 (0.77) <.001 b,e BEHAVIOURAL SKILLS ITEMS How confident are you that you... ...know steps to reduce the risk of developing DR 2.39 (1.17) 3.06 (1.29) .012 ...will remember to have an eye health check… 2.68 (1.35) 4.36 (0.90) <.001 ...can talk to your doctor about your eye health 3.39 (1.28) 4.17 (1.03) .001 ...can find the time to attend an eye health check… 2.74 (1.37) 4.55 (0.75) <.001 ...can afford to pay for the eye health check… 2.68 (1.49) 3.52 (1.48) .008 DR diabetic retinopathy. Data are number (%) of participants who answered each item correctly (Knowledge items); mean (SD) Motivation and Behavioural skills items. p-value is Pearson’s Chi-Square (or Fisher’s exact test if expected cell count< 5), or Independent-samples t-test (two-sided); statistical significance p < 0.01 *Full detail and construct-level findings provided in Additional file 3 Glycated haemoglobin (measure of average blood glucose over the past 8–12 weeks, and indicator of DR risk) Valid n: 121 (motivation items), 120 (behavioural skills items) Item response range: 1 (Strongly disagree) to 5 (Strongly agree) 1 (Strongly disagree) to 7 (Strongly agree) 1 (Not at all confident) to 5 (Extremely confident) be balanced by acknowledgement of the potential for possible unrealistic level of optimism, highlighting the negative consequences related to mydriasis in order to need to emphasise personal susceptibility while provid- maintain credibility. ing information-based content on steps that can be Although moderately high levels of distress in the pri- taken to reduce DR risk. As with many other preventive ority population mean that it is important to avoid direct behaviours, awareness of the potential effectiveness of ‘fear appeal’ messages, low anticipated regret scores for screening followed by subsequent protective action did those who had not screened reinforce the need for mes- not necessarily result in intention formation or priori- sages which emphasise personal susceptibility and de- tisation of preventive intentions. Cognitive dissonance scribe the likely consequences of not screening. induction techniques have been found to have generally Similarly, responses to risk perception items point to a positive effects on changing attitudes, motivations and Lake et al. BMC Health Services Research (2018) 18:396 Page 7 of 18 Table 4 Key lessons learned from needs assessment findings into a logic model for DR screening. The aim of the logic model was to identify the pathways Compared to their older adult counterparts, young adults with T2D have different psychosocial and information needs. There is a lack of behavioural of problem causation moving from determinants, to interventions focused on encouraging screening uptake among young low screening rates and consequent impact on health adults with T2D, indicating that development of a tailored intervention is and quality of life (Fig. 1). warranted. Perceived barriers to and facilitators of screening (which are modifiable and within the scope of the current intervention) include: Context of the intervention and program goals � Knowledge: diabetic retinopathy (awareness of asymptomatic nature of The intervention was to be evaluated and implemented DR, high personal DR risk, modifiable risk factors), screening (consequences of not screening, role of screening in early detection of DR and subsequent in a real-world setting where intervention format and benefit of timely treatment, distinction between standard eye check and delivery medium were dictated by broader policy-level retinal screen) initiatives and a fixed delivery timeline. The intervention � Attitudes: low perception of personal risk, recognition of the benefit of screening was funded by Vision 2020 Australia and grounded � Normative beliefs: awareness of screening approval by significant others, within a suite of Vision Initiative projects collectively de- and screening approval and behaviour of similar others signed to achieve the aims of the Commonwealth gov- � Intention: low prioritisation of target behaviour � Behavioural skills: self-efficacy in overcoming common screening barriers ernment ‘National Framework for Action to Promote to ensure screening attendance (e.g. lack of time or resources), Eye Health and Prevent Avoidable Blindness and Vision engagement with healthcare (sharing diabetes diagnosis, participation in Loss’ [58]. Vision Initiative policy required that the re- diabetes self-management behaviours) source be targeted at the individual-level and delivered directly to eligible young adults with T2D (NDSS regis- health-related behaviour patterns [57]. Consequently, we trants). As such, it was determined by the planning team selected dissonance reduction as a technique that could members who were involved in conception of the study, promote screening motivation. that the most efficient, cost-effective way to meet these Responses to normative behaviour items suggest that criteria was for the intervention to take the form of a messages that provide information about significant leaflet, to be posted to eligible NDSS registrants. Fur- others’ approval are warranted. The findings suggest that thermore, this enabled the leaflet to be included in fu- inclusion of procedural information and messages to ture ‘NDSS starter packs’ for new registrants, and to be promote confidence in knowing steps that can be taken made available online. This decision was supported by to reduce DR risk including how to book and remember previous research, which showed that printed materials a retinal screen, as well as overcoming common barriers are acceptable to young adults with T2D, who give pref- are warranted. Emphasis is required to minimise mis- erence to consistent, centralised information over for- conceptions about some barriers (e.g. inclusion of mes- mat, and who specifically state that the NDSS ‘starter sages which accurately describe the cost and time taken pack’ is a useful resource. for the procedure). With 86% of Australians with T2D registered, the NDSS is considered the “best available source to monitor Logic model of the problem type 2 diabetes in children and young people in Giving consideration to both the qualitative and Australia” [p.36, 56]. However, the NDSS database pri- quantitative needs assessments, we synthesised our marily records registrant postal addresses, which Fig. 1 Logic model of the problem DR: diabetic retinopathy; *Identified in the needs assessment but cannot be modified by the current intervention Lake et al. BMC Health Services Research (2018) 18:396 Page 8 of 18 necessitated the use of a print-based intervention tool Create logic model of change that could be posted to registrants. We developed a logic model of change (Fig. 2) to depict Overall, the purpose of the intervention was to pro- the hypothetical causal pathway from the intervention to mote uptake of screening among young adults with T2D. program outcomes, and anticipated health and quality of Accounting for real-world logistical considerations, the life improvements. Commencing with the intervention, program goal was to develop a leaflet intervention that: we outlined the five modifiable behavioural determinants could be delivered by post, was tailored to young adults (from Fig. 1) and four Performance Objectives (Table 5), with T2D, and included persuasive messaging targeting which were expected to change the measurable behav- behavioural determinants of the target behaviour. ioural outcome. The planning team also acknowledged ex- ternal factors that may affect screening behaviour (i.e. Program outcomes and objectives; logic model of factors that cannot be changed through an individual-level change intervention) in the logic model, even though these were Expected behavioural outcomes and performance objectives beyond the scope of our intervention. The multidisciplinary planning team defined a single, measurable primary outcome when planning for the Create matrix of change objectives subsequent evaluation (uptake of screening for those Once the health behaviours, Performance Objectives and young adults with T2D who had not previously screened determinants were defined, Change Objectives were de- for DR), and multiple secondary outcomes (i.e. change in veloped. Change Objectives are integral to intervention nominated modifiable behavioural screening determinants). content because they represent the behaviour or cogni- Working from the designated program outcomes, and tion being targeted. A sub-group of the planning team informed by the findings of the needs assessment, the generated Change Objectives by creating a matrix, with planning team established the foundation for the inter- modifiable behavioural determinants (in columns) and vention by defining four Performance Objectives (PO, sub-objectives (in rows). Table 5). We increased the specificity of each Perform- Table 6 presents a matrix of Change Objectives for ance Objective by defining sub-objectives, each identify- Performance Objective 3 (Young adults with type 2 dia- ing a behaviour or cognitive process that would promote betes will be motivated to engage in retinal screening). screening uptake. To illustrate, two Change Objectives were generated for sub-objective 3.2 (Understand personal risk of DR), at the intersection with two determinants (knowledge and Table 5 Performance Objectives (e.g. PO.1) and sub-objectives attitudes). The first (K.3.2) sought to improve knowledge (e.g. PO.1.1, PO.1.2, etc.) that risk of DR increases over time, and the second (A.3. Young adults with type 2 diabetes will… 2) sought to change attitudes regarding personal risk PO.1.… demonstrate a clear understanding of diabetic retinopathy (DR) and susceptibility to DR. See Additional file 4 for a PO.1.1 Modifiable and non-modifiable DR risk factors complete matrix of Change Objectives. PO.1.2 Clinical targets for reducing risk of DR Intervention design PO.1.3 Symptoms of DR Intervention themes, components, scope and sequence PO.1.4 Role of DR in vision loss Ensuring that all components of the intervention PO.1.5 Prevalence of DR reflected the needs and preferences of young adults with PO.2…demonstrate clear understanding of retinal screening T2D was a crucial consideration for the planning team. PO.2.1 Role in detecting DR and reducing vision loss The health behaviour change and health communication PO.2.2 Screening procedure and experience literature provided ample foundation on best practice presentation of message content [20, 25, 40, 59–64]. In- PO.2.3 Booking and examination procedure formed by this evidence and the findings of the needs PO.3… be motivated to engage in retinal screening assessment, we developed seven guiding principles for PO.3.1 Prioritise retinal screening leaflet intervention design (Table 7). Consultation with PO.3.2 Understand personal risk of DR young adult PPI members and experts from key stake- PO.3.3 Identify personal barriers to retinal screening holder groups confirmed that these were appropriate PO.3.4 Perceive personal responsibility to engage in screening guiding principles from their perspective. PO.4…proactively engage with the healthcare system and their healthcare team Choose theory and evidence-based behaviour change PO.4.1 Discuss diabetes and eye health with healthcare professionals methods PO.4.2 Understand treatment benefits and options Having established guiding principles, the planning team PO.4.3 Seek more information about diabetes and eye health selected types of theory-based psychological change Lake et al. BMC Health Services Research (2018) 18:396 Page 9 of 18 Fig. 2 Logic model of change T2D: type 2 diabetes, DR: diabetic retinopathy techniques (or change strategies) [65, 66] grouped into significant others’ approval of screening behaviour, provid- six broad change mechanisms designed to ‘boost motiv- ing information about others’ screening behaviour); v) fos- ation and prompt action’ ([65], p.104). The constituent tering a positive screening identity (e.g. providing a techniques (or practical methods) included in the leaflet positive group identity for those engaging in screening); focused on: i) changing beliefs about the benefits of and vi) enhancing self-efficacy (e.g. using persuasive argu- screening (e.g. providing general information on ment to bolster self-efficacy, providing instruction, behaviour-health links, describing likely consequences of prompting barrier identification and planning in relation behaviour); ii) changing risk perception (e.g. emphasising to anticipated barriers, prompting goal setting). personal susceptibility to negative consequences, prompt- ing recipients to assess own risk); iii) changing attitudes associated with screening uptake (e.g. describing likely Intervention development emotional or affective consequences, potentially inducing Draft persuasive message content and leaflet cognitive dissonance among those not intending to act in Develop message content the face of negative consequences); iv) changing (norma- Working from the matrix of Change Objectives, the tive) beliefs about others’ behaviour (e.g. emphasising guiding principles, and theory and evidence-based Table 6 Illustrative matrix of Change Objectives for Performance Objective 3 (PO.3) - Young adults with type 2 diabetes will be motivated to engage in retinal screening Sub-objectives Modifiable behavioural determinants Knowledge Attitudes Normative beliefs Intention Behavioural skills PO.3.1 Prioritise retinal screening NB.3.1 Recognise that I.3.1 Form an similar others have intention to overcome screening prioritise retinal barriers screening PO.3.2 Understand K.3.2 Know that A.3.2 Perceive high personal risk of DR DR risk increases personal risk and over time susceptibility to DR PO.3.3 Identify personal A.3.3 Believe that NB.3.3 See that similar BS.3.3 Be confident in barriers to retinal screening attending screening others face screening one’s ability to identify will relieve fear and barriers (e.g. cost, fear and overcome common guilt and be a of adverse effects) screening barriers positive experience PO.3.4 Perceive personal K.3.4 Know that A.3.4 Adopt personal NB.3.4 Believe that similar BS.3.4 Be confident they responsibility to engage they can take responsibility for others take responsibility have the tools to act on in screening steps to protect retinal screening for their own eye health personal responsibility eye health Determinants: K=Knowledge, A = Attitudes, NB=Normative Beliefs, I=Intention, BS=Behavioural Skills PO performance objective, DR diabetic retinopathy Full matrix of Change Objectives for every Performance Objective is provided in Additional file 4 See Table 5 for full list of Performance Objectives and sub-objectives Lake et al. BMC Health Services Research (2018) 18:396 Page 10 of 18 Table 7 Guiding principles for retinal screening leaflet Assessing readability and suitability intervention design The leaflet was assessed using a combination of an Readability and comprehension: content to be written to acceptable (health) online readability consensus calculator and the Suit- literacy standards, with minimal technical or medical terminology [59, 60]. ability Assessment of Materials (SAM) test, consistent Scope: the scope of intervention messages to be restricted to targeting with best practice [67]. The consensus calculator re- individual-level, modifiable behavioural determinants. ports synthesised results from seven assessment tools Framing: despite long term benefit, retinal screening can be considered (e.g. Flesch Reading Ease formula, Flesch-Kincaid a high-risk behaviour due to the potential for immediate DR diagnosis Grade), to provide two composite scores by grade [61]. Loss-framed messages are effective in promoting engagement with (range: 4–9) and reading level (range: 0–29 ‘very confus- high-risk behaviours and will be used in this leaflet [25]. The majority of headings to be framed as questions to engage the reader while minimising ing’ to 90–100 ‘very easy’)[68]. The SAM test uses six any potentially defensive reaction [62]. evaluation criteria (content, literacy demand, graphics, lay- Sequence: content to follow the logical order of reading. In order to out and type, learning stimulation and motivation, cultural balance loss-framed messages against the high levels of diabetes-related appropriateness) to determine overall suitability [69], with distress and anxiety experienced by young adults with T2D [20, 40], potentially threatening content to be immediately followed by an scores summed and converted to a percentage score and empowering or reassuring statement. classified as ‘not suitable’ (0–39%), ‘adequate’ (40–69%), or Use of quotes: in recognition of the subtle aspects of social influence, ‘superior’ (70–100%). where an individual’s’ beliefs are influenced by those accepted and We excluded the front and back panels of the ‘Who is encouraged by the majority [63], quotes from similar others to be used looking after your eyes?’ leaflet from assessment, as they to reinforce key persuasive messages. All quotes to be sourced verbatim from interview study with descriptors (age and diabetes duration) included included minimal text. For the remaining panels, the to reinforce group membership. median readability consensus grade was 6; median read- Credibility: quote descriptors within the leaflet to reflect demographic ing ease level was 75.6 (fairly easy), and SAM test out- characteristics of the priority population to prompt identification with a come was 75% (superior). credible source. Similarly, logos of leading diabetes and eye health organisations that had contributed to the content to be included to enhance credibility of information. Important yet necessary negative Draft intervention materials information (e.g. discomfort associated with mydriasis, time required to recover clear vision) to be included to provide balance. The planning team selected an 8-panel leaflet design, Graphics and imagery: to reflect the demographic characteristics of the with panels opening outward from the centre, which priority population (e.g. young adults from a range of ethnicities, with and could fit into a standard DL-size envelope. A range of without children). National interpreter symbol to indicate availability of leaflet design options were created in close consult- language assistance services to those with limited English proficiency [64]. ation with a graphic designer who had expertise in producing health promotion materials for people with diabetes. The designs varied in structure, imagery and intervention strategies noted above, a pool of more than organisation, but all adhered to the guiding principles 60 persuasive messages was developed. From this and included consistent messaging. pool, specific change techniques or practical methods were selected to encourage screening. For example, to achieve Change Objective A.3.4 (View retinal screen- Pre-test, refine and produce leaflet ing as a personal responsibility), four potential leaflet Validation and pilot testing heading messages were developed: ‘Eyes: they’re im- The draft leaflet was reviewed by the planning team portant any way you look at it’, ‘Theonlywayto and representatives from key stakeholder organisa- know is to go…’ (verbatim quote), ‘Who is looking tions to confirm that all content was factually accur- after your eyes?’,and ‘Looking at the facts’.All mes- ate and clinically appropriate, and that the resource sages were reviewed by the planning team, and a sub- was likely to meet the project objective. Young adult set selected based on the perceived capacity of the PPI members participated in a thorough piloting message to achieve program goals, target individual process to determine whether: the images and quotes Change Objectives, and satisfy the leaflet guiding were culturally relevant and resonated with the principles. Thus, in the above example, the third op- reader; participants perceived the leaflet would have tion (‘Who is looking after your eyes?’) was selected met their information needs at the time of their T2D because it was phrased as an engaging question, pro- diagnosis; and there were any unintended adverse moting personal responsibility with potential to re- effects in the messaging, imagery or format. Each duce defensive reaction while motivating screening. young adult PPI member received the draft leaflet by A selected example, linking leaflet content to Perform- post and, after reviewing it, participated in a tele- ance Objective 3, is presented in Table 8. Full intervention phone interview during which they commented on map detail for all Performance and Change Objectives is the leaflet’s suitability, responding to questions based provided in Additional file 5. on the SAM criteria [69]. Lake et al. BMC Health Services Research (2018) 18:396 Page 11 of 18 Table 8 Illustrative intervention map linking leaflet content directly back to Performance Objective 3 (PO.3: Young adults with T2D will be motivated to engage in retinal screening) a b Sub-objective (i.e. 3.1) and related Change Objectives Leaflet content (antecedent leaflet text in brackets illustrates context) Panel number PO.3.1 Prioritise retinal screening NB.3.1 Recognise that similar others have overcome (Jenny’s story: before and after the eye health check). “I was scared. 7 screening barriers I was scared of what damage was done…of confronting the fact that my eyesight could be damaged, and of going through the exam and being confronted with what’s there.” I.3.1 Form an intention to prioritise retinal screening (What can I do to protect myself from DR and prevent vision loss?) 5 1. Have a diabetes eye health check. (Note: eye health check listed as Step 1, highest priority) PO.3.2 Understand personal risk of DR K.3.2 Know that DR risk increases over time � The longer you have diabetes the more at risk you are of DR 1 A.3.2 Perceive high personal risk and susceptibility to DR Image: mother and daughter smiling. Child holding hands over 1 mother’s eyes But I’m still young. Am I at risk of DR? Yes you are. Anyone with diabetes can develop DR, which is the 1 leading cause of vision loss for people under 60 years. There are over 34,000 Australians with type 2 diabetes who are 1 under 40 years of age. More than 8500 will already have DR. � The longer you have diabetes the more at risk you are of DR. 1 (Lucas, aged 34, diagnosed with type 2 diabetes 2 years ago) “I didn’t know that I was at risk.” 1 (Jane 25 years, diagnosed with type 2 diabetes 3 years ago) “You might have good vision, you might think that your eyes are 4 absolutely brilliant and there’s no issue. But in the back of your eye, there could be a problem with those little tiny veins that you don’trealise.” PO.3.3 Identify personal barriers to retinal screening A.3.3 Believe that attending screening will relieve fear (Jenny’s story: before and after the eye health check). “It was 7 and guilt and be a positive experience actually quite fun; I don’t know why I put it off. I was really scared going in there, but definitely not now – I’m not fazed by it at all.” NB.3.3 See that similar others face screening barriers (Jenny’s story: before and after the eye health check). “The eye 7 (e.g. cost, fear of adverse effects) drops were a bit uncomfortable and there was a small cost – but I think it’s a wise spend considering what you’re preventing.” BS.3.3 Be confident in one’s ability to identify and (What else do I need to know?) A diabetes eye health check 6 overcome common screening barriers takes about 30 minutes. (What else do I need to know?) It may be free (bulk-billed) or 6 there may be a small fee. (What else do I need to know?) Your optometrist may use eye 6 drops which helps them to see the back of your eye. If you do have eye drops, they may be a little uncomfortable. The drops will also leave you sensitive to light, so bring your sunglasses and be prepared to wait a while for your vision to return to normal PO.3.4 Perceive personal responsibility to engage in screening K.3.4 Know that they can take steps to protect eye health What can I do to protect myself from DR and prevent vision loss? 5 A.3.4 Adopt personal responsibility for retinal screening “I’m a busy person and my family depend on me.” 1 Leaflet heading: Who is looking after your eyes? 3 NB.3.4 Believe that similar others take responsibility for Images: mother and daughter, smiling couple selfie, young man 1,3,5,8 their own eye health of indeterminate cultural origin, Asian female (a.k.a. ‘Jenny’) BS.3.4 Be confident they have the tools to act on personal Leaflet sub-heading: 3 responsibility Your guide to preventing vision loss from diabetes eye disease Protect your sight for life 2 Complete intervention map for all Performance and Change Objectives is provided in Additional file 5 Determinants: K = knowledge, A = attitudes, NB = normative behaviour, I = intention, BS = behavioural skills PO Performance Objective (in bold), DR diabetic retinopathy See Table 6 for illustrative matrix of Change Objectives and Additional file 4 for complete matrix See Fig. 3 for leaflet panels Lake et al. BMC Health Services Research (2018) 18:396 Page 12 of 18 Table 9 Suitability Assessment Materials (SAM) evaluation criteria, young adults’ feedback and changes made to leaflet Sample pilot questions Young adults’ feedback Changes to leaflet Content: Do you think that this leaflet “Key information came through really clearly. I Make ‘Protect your sight for life’ a stand-alone achieves the purpose of the project? didn’t know that early DR doesn’t have any statement and place at top of panel 2, which � Did you learn anything new? symptoms…the doctors tend to focus on signposts location of more information blood glucose, so I knew the 7% (HbA1c) but I didn’t know what the cholesterol target and normal blood pressures were.” ID32 “This leaflet improved my intentions. (DR) is not something you would think could happen to young people.” ID32 “‘Protect your sight for life’ is a powerful statement.” ID36 Literacy demand: Was the length of the leaflet “It only took about 5 minutes to read.” ID33 Discuss whether to include ‘DR’ in leaflet. By acceptable to you? “Language is pretty relaxed which is good for consensus, a decision was made to include it, � How about the number of words? young people.” ID40 but to bold initial definition of diabetic retinopathy � How easy was it to read and understand the “The only thing that caught me was ‘DR’. Did and DR acronym at top of panel 4. information in the leaflet? you mean ‘doctor’ or ‘diabetic retinopathy’?I � Are the words used simple, clear and think you should bold it when it is first informal? defined.” ID32 � Were medical terms defined adequately? Graphics: What do you think of the front “Very professional. Looks like it’s targeted at Bold text ‘When diabetes is first diagnosed’ in panel image? my demographic.” ID36 panel 5. � Are the other images and graphics ‘friendly’ “Maybe bold ‘When diabetes is first Remove graph, which depicted rate of DR and relevant? diagnosed’ so that you hammer home that it’s progression over time. never too early to have an eye check.” ID32 “I’ve never really looked at a graph in a pamphlet. It might appeal to some people, but I don’t know…” ID40 Layout and typography: What do you think of “Main headings need to be in a larger sized the sequence of information? font and bold, and sub-headings in smaller � Is the text type size and font easy to read, or font. Keep the blue colouring.” ID39 could it be easier? “Is there a way that you can make more white � Is the information in the leaflet well-spaced, space? The different colours are more or does it appear cluttered or confusing? attractive.” ID33 Learning stimulation, motivation: Thinking “Jenny’s story is a good thing to have in there. Revisit Performance Objectives to include back to when you first were told that you had Including name, age and diabetes duration understanding the treatment options (PO.4.2), diabetes or when you learnt that diabetes makes the quotes more meaningful.” ID39 populating this through the matrix of Change could affect the eyes – would the leaflet have “Wow, that looks awesome…I didn’t expect to Objectives and into the leaflet content. Add met your information needs at that time? see two smiling faces on the front because more treatment detail to panel 8. � Do you feel that the leaflet is friendly or most diabetes things are all doom and gloom, formal? they’re so terrifying and then you don’t want � Do you feel like you want to read it now or to read them. Whereas, I read this and later? Why? thought, this was a reminder for me to book in for my eye check.”ID40 “I loved the ‘What happens if I had DR’ section. I kept putting off an eye check because I was scared of what would happen. Can you add more about what the treatment is?” ID32 Cultural appropriateness: Do the quotes “I love the pictures; they speak to different Retain multicultural imagery. represent key emotions or experiences that cultural backgrounds.” ID32 you have felt about eye examinations? Was “English is not my first language, but I didn’t the language used throughout the leaflet have any problem reading the leaflet.” ID33 familiar and culturally appropriate to you? � Were there any sections that you found confusing or were unsure about? Feedback from stakeholder reviewers was positive, finalised, leaflet printing was managed by Diabetes with minimal critique offered. Young adult PPI members Victoria (the state agent of the NDSS). gave more considered commentary on what they found useful and what could be improved (Table 9). Where Intervention implementation appropriate, the leaflet was revised to improve content, Planning for program adoption and implementation imagery, readability and cultural acceptability. Once started at study commencement and was heavily Lake et al. BMC Health Services Research (2018) 18:396 Page 13 of 18 influenced by contractual obligations with the funder. ensuring long-term sustainability of the intervention. Fur- These included a one-off statewide distribution of the ther, to enhance reach, an electronic copy of the leaflet was leaflet to all eligible NDSS registrants, timed to coincide made freely available via Diabetes Victoria and Vision 2020 with Vision 2020 Australia public awareness campaign. Australia [70, 71] and promoted to healthcare professionals To protect registrants’ privacy, the NDSS distributed the and members of the priority population. final leaflet (presented in Fig. 3) directly to members of the priority population, on behalf of the planning team. Plans Planning for intervention evaluation are underway for a revision of the NDSS ‘starter pack’ to in- Similarly, evaluation planning started at study commence- clude the eye health leaflet for young adults with T2D, ment. The planning group determined that the best Fig. 3 Who is looking after your eyes? tailored leaflet. ©Vision 2020 Australia, 2018. All rights reserved Lake et al. BMC Health Services Research (2018) 18:396 Page 14 of 18 method of evaluation of the leaflet intervention was a two- Intervention Description and Replication (TIDieR) arm, wait-list randomised controlled trial with screening checklist and guidance [80]. uptake as the primary outcome and change in modifiable behavioural determinant constructs as secondary outcomes. Strengths and limitations The trial, registered with the Australian and New Zealand The key strengths of this work relate to the use of IM, Clinical Trials Registry (ACTRN12614001110673), is now which combines both innovative and traditional interven- complete, and a manuscript is in preparation. tion development activities into an organised, systematic process, and which is consistent with the UK MRC frame- Discussion work for the design and evaluation of complex interven- Uptake of retinal screening from diabetes diagnosis is tions [24]. In the face of limited existing evidence, the crucial for the early identification of DR. In this study, empirical needs assessment, complemented by contribution we undertook the systematic development of an evidence- from the multidisciplinary planning team, key stakeholders based health behaviour change intervention tailored to the and the young adults with T2D PPI group, enabled compre- needs of a priority population, young adults with T2D, hensive exploration of the problem, providing a robust who are at risk of low retinal screening uptake and vision foundation to the intervention. Further, the use of sound loss from DR. theoretical underpinnings, causal modelling, and detailed To date, lack of information on the determinants of pilot and review, provided assurance as to the validity of retinal screening behaviour among young adults with the outcome. As such, the ‘Who is looking after your eyes?’ T2D, and on the elements of individual-level DR screen- leaflet was both evidence-based and sensitive to the needs ing interventions [72], has hampered development of ef- and characteristics of young adults with T2D. fective, targeted intervention strategies for this priority Nonetheless, this study was subject to several limita- population. Further, previous print-based retinal screen- tions. First, the vast majority of studies targeting youth ing interventions have been limited in focus, aiming pri- and young adults with T2D face recruitment challenges marily to increase knowledge and awareness of DR, and [20, 47, 81], and our empirical studies were no different of retinal screening, and neglecting to target other in this respect. Despite numerous steps taken to improve behavioural determinants, such as social norms and in- recruitment, only 10 young adults with T2D participated tentions [72, 73], despite the acknowledged role of psy- in the qualitative study and only 2% of the eligible popu- chosocial factors in health behaviour [74]. lation completed the quantitative online study. The needs assessment described here is the first large- It is likely that recruitment was impacted by a range of scale, mixed-method exploration of modifiable behav- challenges typically specific to young adults with T2D, ioural factors impacting retinal screening behaviour such as social disadvantage, disengagement with existing among young adults with T2D. The findings highlighted services, and complex psychosocial and health needs that many of the clinical and psychosocial barriers to [44, 46, 47, 82]. Furthermore, the needs assessment stud- diabetes self-management faced by young adults with ies were conducted concurrent with a number of other T2D more broadly [18, 20, 40, 41, 75–78], also apply to research projects managed by the NDSS, which may retinal screening. Importantly, when compared to older have contributed to study ‘fatigue’ for this already small adults with T2D, young adults with T2D face both an ac- population (personal communication, D. Rae, National cumulation of barriers to retinal screening, and a num- Inventory Manager, NDSS). Although low sample size ber of uniquely salient barriers and facilitators [44], potentially impacted the generalisability of the needs as- warranting tailored intervention. sessment findings, the response rate for the national sur- Combined with consensus-driven selection of Per- vey was larger than any other conducted to date with formance Objectives, theoretically-grounded change this priority population. methods and comprehensive pilot and review, IM pro- Second, this study was limited to one priority popula- vided the means by which to develop a retinal screening tion where in fact, several populations have been identi- promotion intervention that was both evidence-based fied as at-risk for low retinal screening and vision loss and sensitive to the needs and characteristics of young from DR. These include young adults with T1D, those adults with T2D. However, despite this being a relatively living in socio-economically deprived areas or from mi- simple, single-focus intervention, we shared the experi- nority ethnic and Indigenous populations [83–86], each ence of other programmes, which reported the IM process of which warrant targeted evidence-based intervention, to be both resource and time-intensive [28, 29, 79]. In informed by population-specific needs assessments. particular, we found the high degree of process documen- Finally, many key contextual elements (e.g. interven- tation time-consuming, although we acknowledge that tion level, delivery medium and format) were this activity was crucial for transparency of reporting, and externally prescribed within a broader sphere of real- conforms to key items in the Template for world logistic and contractual limitations. Although Lake et al. BMC Health Services Research (2018) 18:396 Page 15 of 18 unavoidable, this limitation meant that our interven- preliminary evaluation of the ‘Who is looking after your tion was unable to address external factors known to eyes?’ leaflet shows it meets the needs of young adults with impact screening behaviour, such as the cultural di- T2D and its effectiveness in promoting uptake of retinal versity of young adults with T2D, low socioeconomic screening can now be evaluated in a fully-powered RCT. status and lack of English language proficiency, poten- tially limiting effectiveness. Given that NDSS database Additional files strictures limited the intervention to a format suited Additional file 1: Literature study. Literature study procedure and findings. to postal delivery, the leaflet design was suited to the This file describes the procedure and findings of the literature study stated purpose for state-wide implementation. Dia- component of the needs assessment. (DOCX 78 kb) betes Victoria has ensured sustainability and reach of Additional file 2: Interview guide. Interview guide used in qualitative the intervention by regularly updating their resources component of needs assessment. This file presents all interview guide items which comprise the in-depth qualitative interview component of and making an electronic version of the leaflet freely the needs assessment. (DOCX 42 kb) available on its website [70]. Additional file 3: Modifiable behavioural determinants by baseline retinal screen (N = 129). This file presents individual items and findings Future directions from the quantitative online survey component of the needs assessment. (DOCX 22 kb) Recent research suggests that an individual’s beliefs Additional file 4: Matrix of Change Objectives. This file presents the about diabetes and self-management, are most likely to complete matrix of Change Objectives (an illustrative example is provided be influenced early in their diabetes trajectory [87]. in-text in Table 6). The Change Objectives are created at the intersection point Certainly, this appears to be the case for retinal screen- of the five targeted modifiable behavioural determinants (Knowledge, Attitudes, Normative Beliefs, Intention, Behavioural Skills) in columns, ing where, once initiated, the behaviour is generally sus- and sub-objectives (from in-text Table 5), in rows. (DOCX 19 kb) tained [73]. Thus, we recommend targeting individuals Additional file 5: Intervention map linking leaflet content directly back recently diagnosed with T2D via the NDSS, with regis- to Performance Objectives and Change Objectives. This file presents a tration date considered a proxy for date of diabetes diag- complete intervention map (an illustrative example is provided in-text in Table 8). The intervention map links all leaflet content directly back nosis. The leaflet could be used to promote national to the Performance Objectives (specified in-text in Table 5)and the retinal screening programmes in this age group and Change Objectives (Illustrated in-text in Table 6 and presented in full, would be of greatest benefit if translated into additional in Additional file 4.(DOCX 37 kb) languages. Further, this process could be utilised to produce tailored resources designed to increase awareness Abbreviations DR: Diabetic retinopathy; GP: General practitioner; IM: Intervention mapping; and screening for other populations at high-risk of DR MRC: Medical Research Council; NDSS: National Diabetes Services Scheme; (such as young adults with T1D), or for other SAM: Suitability assessment of materials; T2D: Type 2 diabetes mellitus diabetes-related complications which impact young adults with T2D (such as nephropathy and cardiovas- Acknowledgements We thank the people with diabetes who participated in the needs assessments cular disease [88]). and piloting of the leaflet. We thank Virginia Hagger (previously of Diabetes Our experience of the time and resource-intensive na- Victoria) for her involvement in study scoping, prior to CH joining the planning ture of IM reinforces that expressed by others and we team. We thank Nino Soerendata (Diabetes Victoria) for his skilled graphic design work on the leaflet. suggest that undertaking the full IM methodology may not be suitable for all situations. As such, we recom- Funding mend that future programme planners explore alterna- The study was a designated Vision Initiative activity. The Vision Initiative is an integrated health promotion program funded by the Victorian Government and tive options where possible, such as adapting an existing, managed by Vision 2020 Australia. The funding body had no role in design of the effective intervention to their target population. This can study, data collection, analysis or interpretation, or preparation of the manuscript. be enabled by use of a simplified process (IM Adapt), Availability of data and materials which guides decisions regarding selection of appropri- The datasets used in the current study are available from the corresponding author. ate intervention, and components, to adapt [89]. Author’s contributions Conclusions JS, JLB and DT conceived the study. All authors made substantial contributions to study design and intervention development. AJL managed all aspects of the In conclusion, our mixed method needs assessment has study, including conducting the needs assessment (qualitative and quantitative highlighted salient challenges faced by young adults with studies informed by literature review), developing the persuasive messages, T2D and we have demonstrated that IM is a feasible and liaising with stakeholder groups, and piloting the leaflet. AJL led the process of data analysis and interpretation, with substantial input from JLB and JS. CA worthwhile approach to use for the development of an provided expert advice on IM, theoretical basis, the quantitative study and was evidence-based, engaging resource to promote retinal closely involved with content validation of the leaflet. The planning team (AJL, screening to young adults with T2D. This detailed JB, JS, DT, GR, and CH) provided substantial input throughout the project and reviewed and approved materials at all stages. AJL wrote the first draft of this illustration will enable researchers and health promotion manuscript; JLB, JS, GR and CA provided substantial intellectual input through specialists to adopt IM methods when developing reviewing the first and subsequent drafts. All authors approved the final interventions tailored to high-risk groups. Meanwhile, manuscript. Lake et al. BMC Health Services Research (2018) 18:396 Page 16 of 18 Authors’ information 13. American Diabetes Association. Standards of care: microvascular complications AL is a Research Fellow with The Australian Centre for Behavioural Research and foot care. Diabetes Care. 2017;40(Suppl 1):88–98. in Diabetes; a partnership for better health between Diabetes Victoria and 14. National Institute for health and Care Excellence. Type 2 diabetes in adults: Deakin University. This manuscript forms part of the research associated with management NICE guideline [NG28]. 2015. https://www.nice.org.uk/ her PhD. guidance/ng28/chapter/1-Recommendations#managing-complications. Accessed 13 May 2018. Ethics approval and consent to participate 15. Scanlon PH, Stratton IM, Leese GP, Bachmann MO, Land M, Jones C, et al. The studies received ethics approval from the Deakin University Human Research Screening attendance, age group and diabetic retinopathy level at first Ethics Committee (in-depth interview component: 2013–157, quantitative survey screen. Diabet Med. 2016;33(7):904–11. and planned randomised controlled trial evaluation: 2014–156). Participants 16. Wang SY, Andrews CA, Gardner TW, Wood M, Singer K, Stein JD. Ophthalmic provided written informed consent and permission for publication of screening patterns among youths with diabetes enrolled in a large US de-identified quotations at study registration. managed care network. JAMA Ophthalmology. 2017;35(5):432–8. 17. Villarroel MA, Vahratian A, Ward BW. Health care utilization among U.S. adults Competing interests with diagnosed diabetes, 2013. NCHS Data Brief; 2015. www.cdc.gov/nchs/ The authors declare that they have no competing interests. data/databriefs/db183.pdf. Accessed 13 May 2018. 18. Savage S, Dabkowski S, Dunning T. The education and information needs of young adults with type 2 diabetes: a qualitative study. J Nurs Healthcare Publisher’sNote Chronic Illn. 2009;1(4):321–30. Springer Nature remains neutral with regard to jurisdictional claims in 19. Forward H, Hewitt AW, Mackey DA. Missing X and Y: a review of participant published maps and institutional affiliations. ages in population-based eye studies. Clin Exp Ophthalmol. 2012;40(3):305–19. 20. Browne JL, Scibilia R, Speight J. The needs, concerns, and characteristics of Author details younger Australian adults with type 2 diabetes. Diabet Med. 2013;30(5):620–6. 1 2 School of Psychology, Deakin University, Geelong, Australia. The Australian 21. Song SH. Young-onset type 2 diabetes - time to realign clinical priorities. Int Centre for Behavioural Research in Diabetes, Diabetes Victoria, Melbourne, J Diabetes Clin Res. 2015;2:39. Australia. Institute of Health Research, University of Exeter Medical School, 22. MacLennan PA, McGwin G, Heckemeyer C, et al. Eye care use among a 4 5 Exeter, UK. Vision 2020 Australia, Melbourne, Australia. Diabetes Victoria, high-risk diabetic population seen in a public hospital’sclinics. JAMA Melbourne, Australia. Centre for Eye Research Australia, Royal Victorian Eye Ophthalmol. 2014;132(2):162–7. and Ear Hospital, Melbourne 3002, Australia. Ophthalmology, Department of 23. Wilmot E, Idris I. Early onset type 2 diabetes: risk factors, clinical impact and Surgery, University of Melbourne, Melbourne 3010, Australia. AHP Research, management. Ther Adv in Chronic Dis. 2014;5(6):11. Hornchurch, UK. 24. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council Received: 5 September 2017 Accepted: 4 May 2018 guidance. BMJ. 2008;337:a1655. 25. Bartholomew Eldredge LK, Markham CM, Ruiter RAC, Fernandez ME, Kok G, Parcel GS. Planning health promotion programs : an intervention mapping References approach. 4th ed. San Francisco: Jossey-Bass; 2016. 1. Chan JC, Lau ES, Luk AO, Cheung KK, Kong AP, Yu LW, et al. Premature 26. Intervention mapping. https://interventionmapping.com/. Accessed 13 May 2018. mortality and comorbidities in young-onset diabetes: a 7-year prospective 27. Ball GD, Mushquash AR, Keaschuk RA, Ambler KA, Newton AS. Using analysis. Am J Med. 2014;127(7):616–24. intervention mapping to develop the parents as agents of change (PAC) 2. Constantino MI, Molyneaux L, Limacher-Gisler F, Al-Saeed A, Luo C, Wu T, intervention for managing pediatric obesity. BMC Res Notes. 2017;10(1):43. et al. Long-term complications and mortality in young-onset diabetes: type 28. Hurley DA, Murphy LC, Hayes D, Hall AM, Toomey E, McDonough SM, et al. 2 diabetes is more hazardous and lethal than type 1 diabetes. Diabetes Using intervention mapping to develop a theory-driven, group-based complex Care. 2013;36(12):3863–9. intervention to support self-management of osteoarthritis and low back pain 3. Zimmet P, Magliano DJ, Herman W, Shaw J. Diabetes: a 21st century challenge. (SOLAS). Implement Sci. 2016;11:56. Lancet Diabetes & Endocrinology. 2014;2(1):56–64. 29. Gray-Burrows KA, Day PF, Marshman Z, Aliakbari E, Prady SL, McEachan RRC. 4. Harding J, Shaw JE, Peeters A, Davidson S, Magliano DJ. Age-specific trends Using intervention mapping to develop a home-based parental-supervised from 2000–2011 in all-cause and cause-specific mortality in type 1 and type toothbrushing intervention for young children. Implement Sci. 2016;11:61. 2 diabetes: a cohort study of more than one million people. Diabetes Care. 30. Song M, Choi S, Kim S-a, Seo K, Lee SJ. Intervention mapping protocol for 2016;39(6):1018–26. developing a theory-based diabetes self-management education program. 5. Tancredi M, Rosengren A, Svensson AM, Kosiborod M, Pivodic A, Gudbjornsdottir Res Theory Nurs Pract. 2015;29(2):94–112. S, et al. Excess mortality among persons with type 2 diabetes. N Engl J Med. 31. Newby KV, Brown KE, Bayley J, Kehal I, Caley M, Danahay A, et al. Development 2015;373(18):1720–32. of an intervention to increase sexual health service uptake by young people. 6. Song SH, Hardisty CA. Early onset type 2 diabetes mellitus: a harbinger for Health Promot Pract. 2017;18(3):391–9. complications in later years—clinical observation from a secondary care 32. Garba RM, Gadanya MA. The role of intervention mapping in designing cohort. QJM. 2009;102(11):799–806. disease prevention interventions: a systematic review of the literature. PLoS 7. Dart AB, Martens PJ, Rigatto C, Brownell MD, Dean HJ, Sellers EA. Earlier onset of One. 2017;12(3):e0174438. complications in youth with type 2 diabetes. Diabetes Care. 2014;37(2):436–43. 33. Elliott LR, White MP, Taylor AH, Abraham C. How do brochures encourage 8. Al-Saeed AH, Constantino MI, Molyneaux L, D’Souza M, Limacher-Gisler F, walking in natural environments in the UK? A content analysis. Health Promot Luo C, et al. An inverse relationship between age of type 2 diabetes onset Int. 2016;33(2):299–310. and complication risk and mortality: the impact of youth-onset type 2 34. Hill CA, Abraham C. School-based, randomised controlled trial of an evidence- diabetes. Diabetes Care. 2016;39:823–9. based condom promotion leaflet. Psychol Health. 2008;23(1):41–56. 9. Leasher JL, Bourne RAA, Flaxman SR, Jonas JB, Keeffe J, Naidoo K, et al. 35. Abraham C, Southby L, Quandte S, Krahé B, ven der Sluijs W. What's in a Global estimates on the number of people blind or visually impaired leaflet? Identifying research-based persuasive messages in European alcohol- by diabetic retinopathy: a meta-analysis from 1990 to 2010. Diabetes education leaflets. Psychol Health. 2007;22(1):31–60. Care. 2016;39:1643–9. 36. Krahé B, Abraham C, Scheinberger-Olwig R. Can safer-sex promotion leaflets 10. Wong J, Molyneaux L, Constantino M, Twigg SM, Yue DK. Timing is everything: change cognitive antecedents of condom use? An experimental evaluation. age of onset influences long-term retinopathy risk in type 2 diabetes, Br J Health Psychol. 2005;10:203–20. independent of traditional risk factors. Diabetes Care. 2008;31(10):1985–91. 11. Ferris FL. How effective are treatments for diabetic retinopathy? JAMA : the 37. Stewart D, Wilson R, Selby P, Darbyshire J. Patient and public involvement. journal of the American Medical Association. 1993;269(10):1290–1. Ann Oncol. 2011;22(Suppl 7):54–6. 12. Mitchell P and Foran S. Guidelines for the Management of Diabetic Retinopathy. 38. Elam AR, Lee PP. High-risk populations for vision loss and eye care 2008 https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/di15.pdf. underutilization: a review of the literature and ideas on moving forward. Surv Accessed 13 May 2018. Ophthalmol. 2013;58(4):348–58. Lake et al. BMC Health Services Research (2018) 18:396 Page 17 of 18 39. Koelmeyer RL, Dharmage SC, English DR. Diabetes in young adult men: 62. Muller BC, Ritter SM, Glock S, Dijksterhuis A, Engels RC, van Baaren RB. social and health-related correlates. BMC Public Health. 2016;16(S3):63–9. Smoking-related warning messages formulated as questions positively 40. Browne JL, Nefs G, Pouwer F, Speight J. Depression, anxiety and self-care influence short-term smoking behaviour. J Health Psychol. 2014;21(1):60–8. behaviours of young adults with type 2 diabetes: results from the international 63. Cialdini RB, Goldstein NJ. Social influence: compliance and conformity. Annu diabetes management and impact for long-term empowerment and success Rev Psychol. 2004;55:591–621. (MILES) study. Diabet Med. 2014;32(1):133–40. 64. Abraham C, Kools M. Writing health communication: an evidence-based 41. Hessler DM, Fisher L, Mullan JT, Glasgow RE, Masharani U. Patient age: a guide. London: Sage; 2012. neglected factor when considering disease management in adults with 65. Abraham C. In: Abraham C, Kools M, editors. Mapping change mechanisms type 2 diabetes. Patient Educ Couns. 2011;85(2):154–9. onto behaviour change techniques: a systematic approach to promoting 42. Tuomi T, Santoro N, Caprio S, Cai M, Weng J, Groop L. The many faces of behaviour change through text, in: writing health communication: an diabetes: a disease with increasing heterogeneity. Lancet. 2014;383(9922): evidence based guide. London: SAGE; 2012. 1084–94. 66. Abraham C, Michie S. A taxonomy of behavior change techniques used in 43. Reyes-Velazquez W, Sealey-Potts C. Unrealistic optimism, sex, and risk perception interventions. Health Psychol. 2008;27(3):379–87. of type 2 diabetes onset: implications for education programs. Diabetes Spectr. 67. Ryan L, Logsdon MC, McGill S, Stikes R, Senior B, Helinger B, et al. Evaluation 2015;28(1):5–9. of printed health education materials for use by low-education families. 44. Lake AJ, Browne JL, Rees G, Speight J. What factors influence uptake of J Nurs Sch. 2014;46(4):218–28. retinal screening among young adults with type 2 diabetes? A qualitative 68. Scott B. Readability formulas: free readability tools to check for reading levels, study informed by the theoretical domains framework. J Diabetes Complicat. reading assessment and reading grade levels. 2016 http://www.readabilityfor 2017;31(6):997–1006. mulas.com/free-readability-formula-tests.php. Accessed 13 May 2018. 45. Owen KR. Treating young adults with type 2 diabetes or monogenic diabetes. 69. Doak CC, Doak LG, Root JH. Teaching patients with low literacy skills. 2nd Best Pract Res Clin Endocrinol Metab. 2016;30(3):455–67. ed. Philadelphia: J. B. Lippincott; 1996. 46. Zeitler P, Chou HS, Copeland KC, Geffner M. Clinical trials in youth-onset 70. Diabetes Victoria. Who is looking after your eyes? Your guide to preventing type 2 diabetes: needs, barriers, and options. Current Diabetes Reports. vision loss from diabetic retinopathy. https://s3-ap-southeast-2.amazonaws.com/ 2015;15(5):1–8. dv-resources/OrchestraCMS/a1f90000004k5XkAAI.pdf. Accessed 13 May 2018. 47. Nguyen TT, Jayadeva V, Cizza G, Brown RJ, Nandagopal R, Rodriguez LM, et 71. Vision 2020 Australia. Who is looking after your eyes? Your guide to preventing al. Challenging recruitment of youth with type 2 diabetes into clinical trials. vision loss from diabetic retinopathy. 2017 http://www.visioninitiative.org.au/ J Adolesc Health. 2014;54(3):247–54. uploads/assets/files/infosheets/ACBRD_DR%20Age%20under%2040.pdf. 48. Cane J, O'Connor D, Michie S. Validation of the theoretical domains Accessed 13 May 2018. framework for use in behaviour change and implementation research. 72. Lawrenson JG, Graham-Rowe E, Lorencatto F, Presseau J, Burr J, Ivers N, Implement Sci. 2012;7(1):37. Quartilho A, Bunce C, Francis JJ, Grimshaw JM, Peto T, Rice S, Vale L. 49. Fisher WA, Fisher JD, Harman J. The information–motivation–behavioral Interventions to increase attendance for diabetic retinopathy screening skills model: a general social psychological approach to understanding and (Protocol). Cochrane Database of Systematic Reviews 2016, Issue 1. Art. No.: promoting health behavior. In: Sulls J, Walston KA, editors. Social psychological CD012054. https://doi.org/10.1002/14651858.CD012054. foundations of health and illness. Oxford: Blackwell Publishing Ltd; 2003. p. 82–106. 73. Zhang X, Norris SL, Saadine J, Chowdhury FM, Horsley T, Kanjilal S, et al. 50. Fisher J, Fisher WA, Bryan AD, Misovich SJ. Information-motivation-behavioral Effectiveness of interventions to promote screening for diabetic retinopathy. skills model-based HIV risk behavior change intervention for inner-city high Am J Prev Med. 2007;33(4):318–35. school youth. Health Psychol. 2002;21(2):177–86. 74. Michie S, Abraham C, editors. Health psychology in practice, vol. xiv. Malden: 51. Misovich SJ, Martinez T, Fisher JD, Bryan A, Catapano N. Predicting breast Blackwell Pub., British Psychological Society; 2004. p. 416. self-examination: a test of the information-motivation-behavioral skills 75. Waitzfelder B, Pihoker C, Klingensmith G, Case D, Anderson A, Bell RA, et al. model. J Appl Soc Psychol. 2003;33(4):775–90. Adherence to guidelines for youths with diabetes mellitus. Pediatrics. 2011; 52. Osborn CY, Egede LE. Validation of an information-motivation-behavioral 128(3):531–8. skills model of diabetes self-care (IMB-DSC). Patient Educ Couns. 2010; 76. Auslander WF, Sterzing PR, Zayas LE, White NH. Psychosocial resources and 79(1):49–54. barriers to self-management in African American adolescents with type 2 53. Mayberry LS, Osborn CY. Empirical validation of the information-motivation- diabetes: a qualitative analysis. Diabetes Educator. 2010;36(4):613–22. behavioral skills model of diabetes medication adherence: a framework for 77. Brouwer AM, Salamon KS, Olson KA, Fox MM, Yelich-Koth SL, Fleischman intervention. Diabetes Care. 2014;37(5):1246–53. KM, et al. Adolescents and type 2 diabetes mellitus: a qualitative analysis of 54. Fishbein M, Ajzen I. Predicting and changing behaviour: the reasoned action the experience of social support. Clin Pediatr (Phila). 2012;51(12):1130–9. approach. New York: Psychology Press; 2010. 78. Sillars BA, Davis WA, Kamber N, Davis TME. The epidemiology and characteristics 55. Turner KM, Percival J, Dunger DB, Olbers T, Barrett T, Shield JPH. Adolescents’ of type 2 diabetes in urban, community-based young people. Intern Med J. 2010; views and experiences of treatments for type 2 diabetes: a qualitative study. 40(12):850–4. Diabet Med. 2015;32(2):250–6. 79. Greaves CJ, Wingham J, Deighan C, Doherty P, Elliott J, Armitage W, et al. 56. Australian Institute of Health and Welfare. Type 2 diabetes in Australia’s Optimising self-care support for people with heart failure and their caregivers: children and young people: a working paper. In: Diabetes series No.21 2014 development of the rehabilitation enablement in chronic heart failure (REACH- http://www.aihw.gov.au/publication-detail/?id=60129546361. Accessed 13 HF) intervention using intervention mapping. Pilot Feasibility Stud. 2016;2:37. May 2018. 80. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better 57. Freijy T, Kothe EJ. Dissonance-based interventions for health behaviour reporting of interventions: template for intervention description and replication change: a systematic review. Br J Health Psychol. 2013;18:310–37. (TIDieR) checklist and guide. BMJ (Clinical Research Ed). 2014;348:g1687. 58. Commonwealth of Australia. National framework for action to promote eye 81. Speight J, Browne JL, Holmes-Truscott E, Hendrieckx C, Pouwer F. Diabetes health and prevent avoidable blindness and vision loss. 2005 http://www. MILES–Australia (management and impact for long-term empowerment health.gov.au/internet/main/publishing.nsf/content/D3175B31C04E3D72 and success): methods and sample characteristics of a national survey of CA257C750078F76B/$File/frame.pdf. Accessed 13 May 2018. the psychological aspects of living with type 1 or type 2 diabetes in Australian 59. Bailey SC, Brega AG, Crutchfield TM, Elasy T, Herr H, Kaphingst K, et al. adults. BMC Public Health. 2012;12:120. Update on health literacy and diabetes. The Diabetes educator. 2014;40: 82. Nadeau KJ, Anderson BJ, Berg EG, Chiang JL, Chou H, Copeland KC, et al.. 581–604. Youth-onset type 2 diabetes consensus report: current status, challenges, and priorities. Diabetes Care. 2016;39(9):1635–42. 60. Beard E, Clark M, Hurel S, Cooke D. Do people with diabetes understand their clinical marker of long-term glycemic control (HbA1c levels) and does 83. Moreton RBR, Stratton IM, Chave SJ, Lipinski H, Scanlon PH. Factors this predict diabetes self-care behaviours and HbA1c? Patient Educ Couns. determining uptake of diabetic retinopathy screening in Oxfordshire. Diabet 2010;80(2):227–32. Med. 2017;34(7):993–9. 61. Werrij MQ, Ruiter RAC, van't Riet J, de Vries H. Message framing. In: Abraham C, 84. Foreman J, Keel S, Xie J, Van Wijngaarden P, Taylor HR, Dirani M. Adherence Kools M, editors. Writing health communication: an evidence based guide. to diabetic eye examination guidelines in Australia: the National eye Health London: SAGE; 2012. p. 134–43. Survey. Med J Aust. 2017;206(9):402–6. Lake et al. BMC Health Services Research (2018) 18:396 Page 18 of 18 85. Paksin-Hall A, Dent ML, Dong F, and Ablah E. Factors contributing to diabetes patients not receiving annual dilated eye examinations. Ophthalmic Epidemiol. 2013;20(5):281–7. 86. Shi Q, Zhao Y, Fonseca V, Krousel-Wood M, Shi L. Racial disparity of eye examinations among the U.S. working-age population with diabetes: 2002-2009. Diabetes Care. 2014;37(5):1321–8. 87. Skinner TC, Khunti K, Carey ME, Dallosso H, Heller S, Davies MJ. Stability and predictive utility, over 3 years, of the illness beliefs of individuals recently diagnosed with type 2 diabetes mellitus. Diabet Med. 2014;31(10):1260–3. 88. Tryggestad JB, Willi SM. Complications and comorbidities of T2DM in adolescents: findings from the TODAY clinical trial. J Diabetes Complicat. 2015; 29(2):307–12. 89. Highfield L, Hartman MA, Mullen PD, Rodriguez SA, Fernandez ME, Bartholomew LK. Intervention Mapping to Adapt Evidence-Based Interventions for Use in Practice: Increasing Mammography among African American Women. Biomed Res Int. 2015;2015:160103.

Journal

BMC Health Services ResearchSpringer Journals

Published: May 31, 2018

References

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 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

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