Introduction
Type 2 diabetes mellitus is one of Australia’s most pressing chronic conditions. The Australian Institute of Health and Welfare (2023) estimates that approximately 1.3 million Australians live with diagnosed diabetes, the large majority of whom have type 2 diabetes, and prevalence continues to rise with population ageing and increasing rates of obesity. Diabetes Australia (2021) identifies day-to-day self-management, encompassing blood glucose monitoring, dietary regulation, physical activity, medication adherence and foot care, as the foundation of effective control, because the person living with the condition makes the great majority of clinically relevant decisions between infrequent consultations. Effective self-management is associated with improved glycaemic control and a reduced risk of complications, yet it depends heavily on knowledge, confidence and sustained behaviour change.
Australia is among the most culturally and linguistically diverse (CALD) nations in the world. The Australian Bureau of Statistics (2022) reports that more than one quarter of the population was born overseas and that over 5.5 million people speak a language other than English at home. The burden of diabetes is not evenly distributed across this population; several CALD communities, including people from South Asian, Middle Eastern and some East Asian backgrounds, experience elevated rates of type 2 diabetes and poorer glycaemic outcomes than the Australian-born population. Structural and communication barriers, including limited access to language-concordant education, low confidence in navigating the health system, and self-management resources designed for English-speaking, health-literate users, compound this disadvantage.
Digital health has been positioned as a partial remedy. The National Diabetes Services Scheme (NDSS), delivered on behalf of the Australian Government, and the Australian Digital Health Agency both promote digital self-management tools, and smartphone ownership is now near universal across most Australian demographic groups. Mobile applications (apps) can deliver reminders, structured education, self-monitoring and tailored feedback at low marginal cost. However, most evaluated diabetes apps are developed in English and assume a level of health literacy that many CALD users do not have, raising the concern that digital health may widen rather than narrow existing inequities, a phenomenon sometimes described as the digital health divide.
Problem statement
Despite growing evidence that mobile self-management apps can support glycaemic control, this evidence is drawn overwhelmingly from English-speaking, digitally literate samples. Culturally and linguistically diverse Australians, who carry a disproportionate share of the type 2 diabetes burden, are rarely represented in app trials, and few apps are meaningfully adapted beyond surface translation. It is therefore unknown whether a culturally and linguistically adapted app is feasible to deliver, acceptable to CALD users, and capable of supporting the behavioural mechanisms that underpin self-management in this population. Establishing feasibility is a necessary precondition for any larger effectiveness trial.
Aim and research questions
The aim of the proposed study is to evaluate the feasibility and acceptability of a culturally and linguistically adapted smartphone app for type 2 diabetes self-management among adults from selected CALD communities in Australia, and to estimate the parameters needed to design a definitive randomised controlled trial. The study addresses three research questions:
- Is it feasible to recruit, randomise and retain adults with type 2 diabetes from selected CALD communities into a trial of a culturally adapted self-management app?
- What is the acceptability and pattern of engagement of the adapted app among CALD users, and how do these vary by language group and health literacy?
- What are the preliminary effects of the app on diabetes self-efficacy, self-care behaviour and glycaemic control (HbA1c) that could inform the sample size of a future definitive trial?
Literature review
Mobile health for type 2 diabetes
Systematic review evidence indicates that mobile self-management interventions can produce modest but clinically meaningful improvements in glycaemic control. Hou et al. (2016), in a meta-analysis of fourteen randomised trials, reported a mean reduction in HbA1c of approximately 0.5 percentage points favouring app-based self-management over usual care, with larger effects among people with type 2 diabetes than type 1. Whitehead and Seaton (2016), reviewing self-management apps across long-term conditions, found that the most effective interventions combined self-monitoring with tailored feedback and an explicit behavioural framework, rather than simply presenting information. These reviews converge on two points: apps work best when they actively support behaviour, and effect sizes are heterogeneous, partly because engagement tends to decay over time. Sustained engagement, rather than initial download, is the principal implementation challenge.
Critically, the samples underpinning this evidence base are seldom representative of CALD populations. Trials are typically conducted in English, exclude participants who cannot complete English-language instruments, and report little about cultural tailoring. The external validity of pooled effect estimates for Australian CALD communities is therefore uncertain, which motivates population-specific feasibility work before effectiveness can be assumed.
Health literacy and cultural adaptation
Health literacy, defined by Nutbeam (2008) as the personal and cognitive resources that enable individuals to access, understand and use health information, is a central mediator of self-management. Low health literacy is more prevalent among people with limited English proficiency and is independently associated with poorer glycaemic control. An app intended for CALD users must therefore reduce, rather than assume, literacy demands, for example through plain-language content in community languages, audio and pictographic formats, and culturally familiar dietary examples.
Cultural adaptation is more than translation. Resnicow et al. (1999) distinguish surface structure, the matching of materials to observable characteristics such as language and imagery, from deep structure, which reflects the cultural, social and historical values that shape how health and illness are understood. Interventions that address only surface structure risk being comprehensible but not persuasive. Effective adaptation for Australian CALD communities requires genuine engagement with community members, bilingual health workers and, consistent with national multicultural health frameworks, professionally accredited interpreters and translators. The Australian Diabetes Educators Association (ADEA) similarly emphasises culturally responsive education delivered by credentialled diabetes educators.
Theoretical framework
To move beyond describing whether an app works towards understanding why, the study is underpinned by the COM-B model of behaviour (Michie et al., 2011), which proposes that behaviour (B) is the product of capability (C), opportunity (O) and motivation (M). Capability refers to the psychological and physical capacity to perform a behaviour, including knowledge and skills; opportunity refers to external factors, including physical access and the social and linguistic environment; and motivation refers to the reflective and automatic processes that energise behaviour, including beliefs and confidence. The model is well suited to CALD self-management because it locates behaviour not only in the individual but in their opportunity structure, which is precisely where language and cultural barriers operate. Self-efficacy, a core motivational construct in Bandura’s (1997) social cognitive theory and a consistent predictor of self-management (Lorig & Holman, 2003), is nested within the motivation component. Figure 1 maps the adapted app’s components onto the three COM-B sources and their hypothesised pathway to glycaemic control.
Methodology
Design
A pilot randomised feasibility trial with an embedded qualitative acceptability sub-study is proposed, using a mixed-methods, sequential design. The approach follows the Medical Research Council framework for developing and evaluating complex interventions (Craig et al., 2008), in which feasibility and piloting precede a definitive trial, and it adopts the feasibility objectives articulated by Thabane et al. (2010): assessing recruitment, randomisation, retention, data completeness and intervention delivery rather than confirming effectiveness. Consistent with feasibility aims, the trial is not powered for a definitive clinical endpoint, and preliminary effect estimates are used only to inform a future sample size.
Setting and participants
The study will be conducted in partnership with community health services and general practices in culturally diverse areas of Greater Western Sydney and metropolitan Melbourne. A target sample of 160 adults will be recruited, stratified across three language groups (for example, Arabic, Vietnamese and Simplified Chinese speaking communities) to reflect major CALD diabetes populations. Eligible participants will be adults aged 18 years and over with a diagnosis of type 2 diabetes, an HbA1c of 7.5% or above in the preceding three months, ownership of a compatible smartphone, and primary use of one of the study languages. People with severe cognitive impairment or a life-limiting illness will be excluded. Randomisation will be 1:1 to the adapted app plus usual care or to usual care alone, using computer-generated allocation concealed from recruiters.
Intervention
The intervention is a culturally and linguistically adapted smartphone app providing in-language structured education, blood glucose and activity self-monitoring, medication and appointment reminders, and tailored feedback messages. Content will be adapted through a structured process of translation and independent back-translation, review by bilingual diabetes educators, and pre-testing with community members, addressing both surface and deep structure. Participants in both arms will continue to receive usual care through their general practitioner and, where relevant, NDSS-registered services. App access will be provided for six months.
Outcomes and measures
The primary outcomes are feasibility indicators, while secondary outcomes capture the behavioural and clinical signals of interest. Table 1 summarises the outcome domains, instruments and assessment schedule. Self-efficacy will be assessed with a validated diabetes self-efficacy scale, self-care behaviour with the Summary of Diabetes Self-Care Activities measure (Toobert et al., 2000), and health literacy with a validated health literacy questionnaire administered at baseline as a covariate. Engagement will be measured objectively through app analytics.
Table 1: Feasibility, behavioural and clinical outcome measures and assessment schedule.
| Domain | Measure or instrument | Assessment points | Outcome type |
|---|---|---|---|
| Recruitment and retention | Screening and enrolment logs | Continuous; 6 months | Primary (feasibility) |
| Data completeness | Proportion of complete outcome data | 6 months | Primary (feasibility) |
| Engagement | App analytics (active days, logins) | Continuous | Process |
| Diabetes self-efficacy | Validated diabetes self-efficacy scale | Baseline, 3, 6 months | Behavioural |
| Self-care behaviour | Summary of Diabetes Self-Care Activities (Toobert et al., 2000) | Baseline, 6 months | Behavioural |
| Glycaemic control | HbA1c (%), point-of-care or pathology | Baseline, 6 months | Clinical |
| Health literacy | Validated health literacy questionnaire | Baseline | Covariate |
| Acceptability | Semi-structured interviews | 6 months | Qualitative |
Data analysis
Feasibility outcomes will be reported descriptively with 95% confidence intervals around recruitment and retention rates, assessed against pre-specified progression criteria (for example, retention of at least 80%). Preliminary between-group differences in HbA1c, self-efficacy and self-care will be estimated using analysis of covariance adjusting for baseline values, and reported as effect sizes with confidence intervals rather than as hypothesis tests, consistent with feasibility guidance. Qualitative interview data will be analysed thematically, with transcripts translated and coded to explore acceptability, perceived cultural fit and barriers to engagement, and the findings will be integrated with the quantitative results.
Ethical considerations
The study will be conducted in accordance with the National Statement on Ethical Conduct in Human Research (National Health and Medical Research Council, 2018) and will not commence before approval from a registered Human Research Ethics Committee (HREC). Working with CALD participants raises specific ethical obligations. Informed consent materials will be provided in each community language, and consent will be supported by professionally accredited interpreters rather than by family members, to protect confidentiality and voluntariness. Translation of instruments and app content will follow recognised forward and back-translation procedures to preserve meaning and avoid the risk of misinformation. Data will be stored securely and de-identified, and participants will be able to withdraw at any time without any effect on their standard care.
Project timeline
The study is planned over 18 months. Table 2 sets out the phases, key activities and indicative months, with deliberate overlap between recruitment and follow-up to maintain momentum across sites.
Table 2: Indicative 18-month project timeline.
| Phase | Key activities | Months |
|---|---|---|
| 1. Ethics and set-up | HREC application, cultural and linguistic adaptation, translation and back-translation, interpreter and consent protocols, staff training | 1-4 |
| 2. Site and app readiness | Community and service partnerships, app pre-testing with community members, recruitment materials prepared in the study languages | 3-6 |
| 3. Recruitment and randomisation | Screening, informed consent, baseline assessment, 1:1 allocation | 6-11 |
| 4. Intervention and follow-up | Six-month app access per participant, engagement monitoring, 3 and 6 month assessments | 7-16 |
| 5. Qualitative sub-study | Semi-structured interviews, transcription, translation and coding | 13-17 |
| 6. Analysis and dissemination | Statistical and thematic analysis, integration of findings, reporting | 16-18 |
Significance
The study addresses a clear equity gap. Culturally and linguistically diverse Australians bear a disproportionate share of the type 2 diabetes burden yet are underrepresented in the digital health evidence base. By testing whether a genuinely adapted app is feasible to deliver and acceptable to users, and by grounding the intervention in the COM-B framework, the study will generate the recruitment, retention and preliminary effect parameters needed to design a definitive, adequately powered trial. The findings will be directly relevant to the NDSS, primary care providers and policymakers seeking to ensure that investment in digital health narrows rather than widens inequities, and they will contribute Australian evidence to a literature that remains dominated by English-speaking samples.
Limitations
Several limitations are acknowledged. As a feasibility trial, the study is not powered to demonstrate clinical effectiveness, and preliminary effect estimates must be interpreted with caution. Recruitment from three language groups in two cities may limit generalisability to other CALD communities and to rural and regional settings. Reliance on smartphone ownership may exclude the most digitally marginalised individuals, potentially understating the very divide the study seeks to address. Self-reported behavioural measures are subject to social desirability bias, which is partially mitigated by objective app-usage and HbA1c data. Finally, blinding of participants is not possible in a behavioural trial, although outcome assessment and analysis will be conducted by staff independent of the allocation.
Conclusion
Type 2 diabetes places a heavy and unequal burden on Australia’s culturally and linguistically diverse communities, and current digital self-management tools have largely been designed and evaluated without them. This proposal sets out a theoretically grounded and ethically robust pilot randomised feasibility trial of a culturally and linguistically adapted self-management app. By establishing whether such an intervention can be delivered, is acceptable, and shows promise for improving self-efficacy and glycaemic control, the study will provide the evidence and parameters required to justify and design a full-scale trial, and it will help ensure that the benefits of digital health reach the Australians who stand to gain the most.
References
Australian Bureau of Statistics. (2022). Cultural diversity of Australia: Information paper. ABS.
Australian Institute of Health and Welfare. (2023). Diabetes: Australian facts. AIHW.
Bandura, A. (1997). Self-efficacy: The exercise of control. W. H. Freeman.
Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: The new Medical Research Council guidance. BMJ, 337, a1655.
Diabetes Australia. (2021). Type 2 diabetes. Diabetes Australia.
Hou, C., Carter, B., Hewitt, J., Francisa, T., & Mayor, S. (2016). Do mobile phone applications improve glycemic control (HbA1c) in the self-management of diabetes? A systematic review, meta-analysis, and GRADE of 14 randomized trials. Diabetes Care, 39(11), 2089-2095.
Lorig, K. R., & Holman, H. (2003). Self-management education: History, definition, outcomes, and mechanisms. Annals of Behavioral Medicine, 26(1), 1-7.
Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42.
National Health and Medical Research Council. (2018). National statement on ethical conduct in human research. NHMRC.
Nutbeam, D. (2008). The evolving concept of health literacy. Social Science & Medicine, 67(12), 2072-2078.
Resnicow, K., Baranowski, T., Ahluwalia, J. S., & Braithwaite, R. L. (1999). Cultural sensitivity in public health: Defined and demystified. Ethnicity & Disease, 9(1), 10-21.
Thabane, L., Ma, J., Chu, R., Cheng, J., Ismaila, A., Rios, L. P., Robson, R., Thabane, M., Giangregorio, L., & Goldsmith, C. H. (2010). A tutorial on pilot studies: The what, why and how. BMC Medical Research Methodology, 10, 1.
Toobert, D. J., Hampson, S. E., & Glasgow, R. E. (2000). The summary of diabetes self-care activities measure: Results from 7 studies and a revised scale. Diabetes Care, 23(7), 943-950.
Whitehead, L., & Seaton, P. (2016). The effectiveness of self-management mobile phone and tablet apps in long-term condition management: A systematic review. Journal of Medical Internet Research, 18(5), e97.