Diabetes mellitus (DM) is one of the most prominent non-communicable disease epidemics of the twenty-first century. According to the International Diabetes Federation (IDF), 537 million people aged 20–79 years had diabetes in 2021, or 10.5% of this population. The IDF predicts that this number will increase to 643 million by 2030 and 783 million by 2045 [1]. These numbers conceal significant regional differences: South-East Asia and the Western Pacific bear the largest absolute burden, while the Middle East and North Africa (MENA) region, which includes all six GCC countries, recorded the highest adult diabetes prevalence globally at 18.1% (age-adjusted comparative prevalence, 2021) [1].   The chronic and multisystemic nature of diabetes necessitates sustained and active engagement by patients in their own care—commonly termed diabetes self-management or self-care. Self-care encompasses a spectrum of behaviours including adherence to medication, dietary modification, regular physical activity, structured blood glucose monitoring, foot inspection, tobacco abstinence, and engagement with healthcare services for complication screening [2]. Improved glycaemic and blood-pressure control, to which these behaviours contribute, reduces the risk of microvascular complications (retinopathy, nephropathy, neuropathy), macrovascular events (myocardial infarction, stroke), diabetic foot ulcers and premature mortality [3–5].   Despite decades of research confirming the efficacy of self-care interventions—particularly structured diabetes self-management education (DSME) programmes [2,6–8] and physical activity [9]—uptake and practice remain suboptimal in all regions, though for different reasons. In Western healthcare systems, reported barriers include healthcare fragmentation, medication cost (especially in the USA) and psychosocial burden [10]. In the GCC, dietary transition, sedentary lifestyles, the clinical complexity introduced by Ramadan fasting, and access differences between national citizens and expatriate workers have been described [11]. Asia is remarkably diverse: resource-constrained rural areas in South and Southeast Asia coexist with technologically sophisticated urban centres in China, South Korea and Japan; in lower-income settings, out-of-pocket costs can limit access to guideline-recommended care [12].   The rapid development of digital health solutions—telemedicine, continuous glucose monitoring (CGM), mobile health (mHealth) apps and AI-assisted clinical decision support—is another factor. These technologies are changing self-care opportunities, but their adoption is uneven: it is most prevalent in Western countries and parts of East Asia, expanding in GCC metropolitan centres, and nascent or limited in low-resource Asian settings because of cost and connectivity [13–15].   Several reviews have examined diabetes self-management within a single region, population or intervention type. Thirunavukkarasu and Alsaidan synthesised self-management behaviours across GCC countries in a narrative review [11]; Maina et al. mapped complication-prevention self-management practices in low- and middle-income countries [12]; Hawthorne et al. evaluated culturally appropriate education for ethnic minority groups, predominantly in Western countries [16]; and Piragine et al. pooled adherence to oral antidiabetic drugs globally without region-specific estimates [17]. Other reviews have addressed specific interventions—DSME [2,8], mobile health [13–15] and telemedicine [18]—or conceptual models of self-care [19]. To our knowledge, none of these (i) placed Asian, GCC/MENA and Western evidence side by side within a single prespecified framework, (ii) separated intervention-efficacy evidence from regional implementation evidence, or (iii) examined cultural, religious and economic modifiers across all three spheres. The present review addresses this gap. Because the three spheres are internally heterogeneous, we defined them at country level and report findings by subgroup where possible; the cross-regional comparisons that follow are predominantly indirect and are interpreted accordingly.   Importance of the Study Understanding why and how self-care varies between settings is more than an intellectual endeavour. Complications such as amputation, end-stage renal disease and vision loss consume a disproportionate share of health budgets and have serious effects on affected people and their families. The IDF estimated direct health expenditure attributable to diabetes at USD 966 billion globally in 2021 [1]. Comparative analysis can support the adaptation of proven interventions to new settings, help avoid the repetition of failures, and inform culturally competent clinical guidelines. For GCC health systems undergoing transformation under national visions such as Saudi Arabia's Vision 2030, and for resource-constrained Asian health systems that are expanding capacity, such evidence is needed.

Study Design and Reporting This is a systematic review with narrative synthesis, reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [20] (completed checklist in Supplementary Appendix S8) and the Synthesis Without Meta-analysis (SWiM) reporting guideline [21]. Because study designs, populations, self-care measures and outcomes were heterogeneous, quantitative pooling was not undertaken; findings were synthesised narratively after systematic identification, screening and appraisal of sources.   Protocol and Registration No review protocol was prospectively registered in PROSPERO, the Open Science Framework or any other registry, and no protocol was published. Without a registered protocol, post hoc changes to methods cannot be ruled out; this is acknowledged as a limitation (Section 9).   Information Sources and Search Strategy PubMed/MEDLINE, Scopus and Web of Science Core Collection were searched on 1 September 2024 for records published from 1 January 2000 to the search date. The search was not updated after this date. The strategy combined four concept blocks—diabetes, self-care, geography, and complication/outcome or digital-health terms—using MeSH and free-text terms in PubMed and equivalent title/abstract/keyword fields in Scopus and Web of Science. The core concept structure was: ("diabetes mellitus" OR "type 2 diabetes" OR "T2DM") AND ("self-care" OR "self-management" OR "patient education" OR "adherence") AND ("Asia" OR "South Asia" OR "Southeast Asia" OR "East Asia" OR "Gulf Cooperation Council" OR "GCC" OR "Saudi Arabia" OR "United Arab Emirates" OR "Middle East" OR "Western countries" OR "Europe" OR "North America") AND ("complication prevention" OR "glycemic control" OR "HbA1c" OR "foot care" OR "retinopathy" OR "nephropathy" OR "digital health" OR "mHealth"). The full, database-specific strategies—with field tags, controlled vocabulary, limits, exact search dates and the number of records retrieved from each database—are reported in Supplementary Appendix S1.   Search Sensitivity Requiring the fourth concept block (complication/outcome or digital-health terms) increases precision but may reduce sensitivity for self-care studies that do not mention these terms in indexed fields. To mitigate this, the reference lists of all included sources and of key reviews [2,12,16,17] were hand-searched. Residual risk of missed studies is acknowledged as a limitation (Section 9).   Grey Literature The websites of the World Health Organization (www.who.int/publications and the IRIS repository), the International Diabetes Federation (www.idf.org; www.diabetesatlas.org) and the National Institute for Health and Care Excellence (www.nice.org.uk/guidance) were searched in September 2024, alongside the database searches, using the terms "diabetes", "self-care", "self-management", "education" and "Ramadan". Guidelines, atlases and technical reports were eligible if they were issued by a national or international organisation, described their evidence base or development method, and made recommendations or reported data on self-care, education or complication prevention in adults with diabetes. The grey-literature search log is given in Supplementary Appendix S2.   Publications After the Search Cut-Off Three publications dated 2025 [11,19,22] appeared in the reference list of the original submission. They were not identified through the systematic search, which ended on 1 September 2024. They are retained only as contextual references, are labelled as such in Supplementary Table S4 (Part B), and do not contribute to the evidence synthesis, the tables of findings or the certainty ratings.   Eligibility Criteria and Geographic Definitions Studies were included if they: (i) involved adult participants (≥18 years) with type 1 or type 2 diabetes; (ii) reported on self-care behaviours, patient education, awareness or complication-prevention strategies; (iii) were conducted in, or explicitly reported data from, Asia, GCC/MENA or Western countries (Europe, North America, Australia, New Zealand); (iv) were published in English between January 2000 and September 2024; and (v) were primary research studies (randomised or non-randomised trials, cohort, cross-sectional or instrument-validation studies), systematic reviews, meta-analyses or scoping reviews, or evidence-based guidelines/consensus statements from recognised professional bodies. Studies were excluded if they focused exclusively on pharmacological interventions without a self-care component, lacked geographic specificity, studied populations without diabetes, were narrative or conceptual reviews, editorials or commentaries, or if the full text was unavailable. Conference abstracts without full data were also excluded.   Geographic Definitions Countries were assigned to spheres a priori at country level (full list in Supplementary Appendix S3). Asia: East Asia (China, including Hong Kong SAR; Taiwan; Japan; Republic of Korea), South Asia (e.g., India, Pakistan, Bangladesh, Sri Lanka, Nepal) and Southeast Asia (e.g., Vietnam, Thailand, Malaysia, Indonesia, the Philippines, Singapore). GCC/MENA: the six GCC member states (Saudi Arabia, United Arab Emirates, Kuwait, Qatar, Bahrain, Oman) form the core group; other countries of the IDF MENA region (e.g., Egypt, Jordan, Palestine, Iran) form a separate "wider MENA" subgroup, whose findings are reported separately and are not attributed to the GCC. Western countries: countries of Europe, the United States, Canada, Australia and New Zealand. Countries spanning geographic or cultural classifications were assigned by their IDF region: Iran (IDF MENA region) was assigned to the wider-MENA subgroup, and Turkey (IDF Europe region) was treated as a transcontinental case reported separately and not used to characterise either Western or MENA findings. Multinational studies were classified as "multi-regional", and their findings were attributed to a sphere only when results were reported by country.   Study Selection Records were de-duplicated with reference-management software. Titles and abstracts, and then full texts, were screened independently by two reviewers against the eligibility criteria; disagreements were resolved by discussion and, where necessary, by a third author. Inter-reviewer agreement was not formally quantified (e.g., with Cohen's kappa), which is acknowledged as a limitation (Section 9). At revision, every source retained after full-text screening was re-assessed against the prespecified criteria (eligibility re-audit); the sources excluded at this stage are listed with reasons in Supplementary Appendix S9.   Data Extraction Data were extracted with a standardised form that was piloted on an initial sample of sources and refined before use. At revision, all extracted data were re-checked against the full text or published abstract of each source and corrected where necessary. Extracted variables were: first author and year; country, sphere and subgroup; study design and evidence stream; setting; population (diabetes type, age, sample size); self-care domain(s); measurement instrument; intervention and comparator (where applicable); outcomes and effect estimates; period of data collection; and appraisal results. The extracted characteristics of every source are given in Supplementary Table S4.   Evidence Streams and Avoidance of Double-Counting Included sources were allocated to three evidence streams that were analysed separately: Stream A, primary studies; Stream B, systematic reviews, meta-analyses and scoping reviews; and Stream C, guidelines and consensus statements. Stream B was used to answer intervention-efficacy questions; Stream A was used to describe regional practice, implementation and determinants. When a primary study was also included in an included systematic review, its results were not counted again within the same evidence body (overlap was checked by comparing reference lists). Stream C documents were used to define recommended standards (e.g., screening intervals) and to frame recommendations, not as evidence of effect. References cited only for background are identified as contextual in Supplementary Table S4 (Part B) and were not used to derive findings or certainty ratings.   Quality Appraisal Methodological quality was appraised with design-specific tools: the Cochrane RoB 2 tool for randomised trials [23], JBI critical appraisal checklists for cohort, analytical cross-sectional and instrument-validation studies [24], AMSTAR 2 for systematic reviews [25], and AGREE II domains for guidelines [26]. Appraisal was completed at the revision stage, after study selection, against the published criteria of each tool; the judgement for every source is reported in Supplementary Table S4.   Data Synthesis Meta-analysis was not possible because of variation in study designs, populations and outcome measures. Synthesis followed the SWiM guideline [21] and the ESRC guidance on narrative synthesis [27] in six steps. (1) Grouping: findings were grouped by self-care domain—(a) awareness and education; (b) medication adherence; (c) diet; (d) physical activity; (e) foot care and complication screening; (f) glucose monitoring and digital health; (g) cultural, religious, economic and psychosocial factors; (h) complication prevention—and, within each domain, by sphere and subgroup. (2) Standardised metric: for intervention questions, the mean difference in HbA1c (percentage points) reported by included systematic reviews or trials was used; for practice and adherence questions, proportions were tabulated as reported, with the instrument used, and were not pooled. (3) Vote counting based on statistical significance was not used. (4) Conflicting findings were examined against study design, risk of bias, measurement instrument, setting and period; conflicts that could not be explained are reported as inconsistency. (5) Direct versus indirect comparison: a cross-regional comparison was classified as direct only when a single study measured self-care in more than one sphere using the same methods (e.g., the DAWN study [10]); all other contrasts are labelled indirect and interpreted cautiously. (6) Strength and consistency of evidence were summarised with GRADE.   Certainty of Evidence Certainty was rated with GRADE [28] for eleven prespecified outcome–comparison evidence bodies (Section 3.10), not for regions as a whole. Bodies of randomised evidence (or systematic reviews of trials) started at high certainty and observational evidence at low certainty, and were downgraded for risk of bias, inconsistency, indirectness (including transfer of findings across regions), imprecision or publication bias, or upgraded for large effects or a dose–response gradient. Where no single pooled estimate was available, ratings followed the approach of Murad et al. for narrative syntheses [29]. Downgrading and upgrading decisions are explained in Section 3.10 and Supplementary Appendix S5.

Study Selection The database searches retrieved 5,305 records. After removal of duplicates, screening of titles, abstracts and full texts against the eligibility criteria, and supplementary hand-searching of reference lists and grey literature, 52 sources were retained and cited in the original submission. At revision, these 52 sources were re-assessed against the prespecified criteria (eligibility re-audit), and 14 were excluded with reasons: published after the search cut-off (n = 3) [11,19,22]; ineligible design, such as commentary, conceptual or basic-science review or modelled epidemiological estimates (n = 5) [30–34]; population without diabetes or topic not diabetes-specific (n = 4) [35–38]; outside the three spheres (n = 1) [39]; and pharmacological intervention without a self-care component (n = 1) [40] (Figure 1; Supplementary Appendix S9). Thirty-eight sources were therefore included: 19 primary studies (Stream A), 13 systematic reviews, meta-analyses or scoping reviews (Stream B) and six guidelines or consensus statements (Stream C). The excluded sources are still cited where they provide background and are labelled as contextual references (Supplementary Table S4, Part B). The numbers of duplicates and of records excluded at the title/abstract and initial full-text stages were not retained in a form that allows exact reporting; this is acknowledged as a limitation (Section 9).   Adapted from the PRISMA 2020 flow diagram template [20]. 52 sources retained after full-text screening − 14 excluded at the eligibility re-audit = 38 included sources.   Characteristics of Included Sources Table 1 summarises the evidence base by stream and sphere; Supplementary Table S4 lists every source with its design, country, population, sample size, self-care measure, outcomes, main findings and appraisal. Of the 19 primary studies, five were from Asia (China, Vietnam, Bangladesh, Republic of Korea), two from GCC member states (both Saudi Arabia), four from the wider MENA subgroup (Egypt, Palestine, Iran), one from Turkey, six from Western countries (USA, UK) and one was multi-regional (DAWN, 13 countries). Nine were cross-sectional surveys, four instrument-validation studies, four randomised or controlled trials and two cohort studies. All 13 systematic reviews synthesised trials or studies from Western or multiple regions; none was restricted to Asia or the GCC. GCC-specific conclusions therefore rest on limited direct evidence.   Table 1: Composition of the Included Evidence Base by Evidence Stream and Geographic Sphere (Number of Sources) Evidence stream Asia GCC (core) Wider MENA Turkey Western Multi-regional / global Total A. Primary studies 5 2 4 1 6 1 19 B. Systematic reviews, meta-analyses, scoping reviews 0 0 0 0 4 9 13 C. Guidelines / consensus statements 0 0 0 0 4 2 6 Total 5 2 4 1 14 12 38 Western Reviews: [2,7,16,41]; Multi-Regional Reviews: [12–15,17,18,42–44]; Western Guidelines: [3,9,45,46]; Multi-Regional Guidance: [47,48]. Individual sources are listed in Supplementary Table S4   Epidemiological Context and Diabetes Burden The diabetes burden is unequally distributed worldwide, and this epidemiological heterogeneity shapes self-care priorities. The GCC is part of the MENA region, which had the highest age-adjusted comparative prevalence of all IDF regions in 2021 (18.1%), compared with 11.9% in North America and the Caribbean, 10.0% in South-East Asia, 9.9% in the Western Pacific and 7.0% in Europe [1]. The largest national numbers of adults with diabetes are found in China, India and Pakistan [1]. IDF regions do not map exactly onto the spheres used here (for example, the Western Pacific region includes both China and Australia, and the MENA region includes non-GCC countries), so these figures describe context rather than sphere-specific estimates. Western countries generally report lower prevalence; however, rising rates of type 2 diabetes and prediabetes linked to obesity challenge this pattern [1]. A descriptive cross-regional overview of several self-care domains is given in Table 2; it summarises the available evidence and its sources and is not intended as a ranking of regions.   Table 2: Descriptive Overview of Diabetes Self-Care Indicators in Asia, GCC/MENA and Western Countries, with Sources Self-Care Domain Asia GCC/MENA Western Countries Diabetes prevalence, age-adjusted, 20–79 Years [1] 10.0% (South-East Asia region); 9.9% (Western Pacific region) 18.1% (MENA region; includes non-GCC countries) 7.0% (Europe); 11.9% (North America and Caribbean) Medication adherence (study-level; instruments differ; not comparable across columns) No included Asian study measured adherence directly. Global pooled estimate across 156 studies: 54% adherent (95% CI 51–58%) [17] Wider MENA: 57.3% adherent (MMAS-8; Palestine, n = 405) [49]; 30.5% high and 51.3% medium adherence (Egypt, n = 302) [50]. No GCC-core estimate in included studies No single-study estimate in included sources; see global pooled estimate [17] Structured diabetes education Variable: community programmes in parts of East Asia; limited in many low-resource settings [12,51] No included GCC evaluation of DSME; mostly hospital-based delivery described in contextual sources Embedded in national programmes (e.g., DESMOND) [6,52]; benefits attenuate over time [6,8] Digital health / mHealth adoption Rapid uptake; messaging/app platforms widely used Increasing; government e-health drives Mature; CGM, apps, telehealth integrated[3,53] Dietary self-care Traditional diets; high-carbohydrate staples challenge adherence Energy-dense traditional and processed foods; Ramadan fasting a key factor [48,54] Evidence-based diet counselling available[3,45] Physical activity Low in urban populations; cultural barriers Low; sedentary lifestyle predominant (contextual [11]) Guidelines established [9]; many adults do not meet recommendations [9] Foot care practices Poor awareness; resource constraints [12] Wider MENA: 75.2% unsatisfactory foot self-care (older adults, Egypt, n = 230) [55]; no GCC-core estimate High standard: podiatry integrated [47] Cultural/religious influences Collectivist norms; family support. Islamic practices (fasting, prayer times) integrated into care [48,54] Individualistic; secular care model Healthcare system model Mixed public–private; variable access Predominantly state-funded for nationals Universal health coverage (Europe); insurance (USA) Complication screening regularity Irregular; urban–rural gap wide[12] Improving; Vision 2030 driving change Annual screening standard in guidelines [3,56] Within-sphere heterogeneity High-income East Asia vs lower-income South/Southeast Asia; urban vs rural GCC nationals vs expatriate workers; differences between GCC states; GCC vs wider MENA Universal-coverage systems vs insurance-based USA; socioeconomic and ethnic-minority gaps [16] CGM: Continuous Glucose Monitoring, DSME: Diabetes Self-Management Education, GCC: Gulf Cooperation Council, MENA: Middle East and North Africa, MMAS-8: 8-item Morisky Medication Adherence Scale. Entries without a citation are descriptive health-system context, not derived from included studies and not graded. Values in different columns come from different populations, instruments and years and must not be compared directly. Amputation-rate ranges and region-level GRADE labels shown in the previous version have been removed: the amputation ranges could not be traced to a source with a defined population and period, and certainty is now rated per outcome (Section 3.10). The source of every numerical value is listed in Supplementary Appendix S6   Diabetes Self-Management Education and Awareness Structured DSME is the most consistently evidenced intervention for improving self-care behaviours and glycaemic outcomes [2,7,8]. In Western countries, DSME is embedded within national guidelines: the UK's DESMOND and Diabetes Manual programmes [6,52], the USA's national DSMES standards [3], and equivalent European frameworks represent mature, community-based delivery models. In a meta-analysis of 31 randomised trials, Norris et al. found that DSME reduced HbA1c by 0.76 percentage points (95% CI 0.34–1.18) more than control at immediate follow-up, with smaller effects at longer follow-up [8]. In the DESMOND cluster trial, the HbA1c difference was no longer significant at 3 years (−0.02; 95% CI −0.22 to 0.17; 604 participants with biomedical data), although improvements in illness beliefs persisted [6].   No included primary study evaluated DSME delivery in a GCC member state; the two included Saudi studies assessed diabetes distress [57] and population disease burden [58]. In the wider MENA subgroup, lower disease knowledge and weaker beliefs about the necessity of medicines were associated with non-adherence in Palestine [49], and patient education level and physician participation in diabetes training were associated with better glycaemic control in Egyptian primary care [50]. A recent narrative review, published after our search cut-off and therefore not included, describes predominantly hospital-based education services in GCC countries [11].   Asian evidence was heterogeneous. In six urban districts of Chengdu, western China, 46% of 364 surveyed patients showed "good" self-management behaviour, 45% "fair" and 6% "poor"[51], and a Korean instrument-development study (n = 473) produced a locally validated self-management scale [59], no included Asian study evaluated a DSME programme. In a Cochrane review of 11 trials, culturally appropriate health education improved short-term glycaemic control compared with usual care in ethnic minority groups, mostly in Western countries [16]. These trials compared adapted education with usual care rather than with generic programmes, so they do not show that adapted programmes outperform generic ones.   Medication Adherence In all three spheres, medication non-adherence is associated with poorer glycaemic control [17,49]. Piragine et al. pooled 156 studies published in 2011–2022 and found that 54% (95% CI 51–58%) of patients were adherent to oral antidiabetic drugs, and that good adherence was associated with good glycaemic control (OR 1.33; 95% CI 1.17–1.51); adherence was measured with administrative data, self-report questionnaires or pill counts, and region-specific pooled estimates were not reported [17]. Complexity of regimens, side effects, cost (especially in the USA) and health literacy are linked to non-adherence [10,49]. Long-term UKPDS follow-up showed a legacy effect of early intensive glucose control, including metformin, on microvascular and cardiovascular outcomes [60].   In the wider MENA subgroup, 57.3% of 405 patients in Palestine were adherent on the MMAS-8 [49], and in Egyptian primary care 30.5% showed high and 51.3% medium adherence [50]. No included primary study from a GCC member state reported medication adherence. A specific set of clinical guidelines addresses the difficulties of taking medication during Ramadan, including rescheduling of insulin and the risk of hypoglycaemia with sulfonylureas. The IDF and Diabetes and Ramadan (DAR) International Alliance practical guidelines provide risk stratification and medication-adjustment advice [48,54].   No included primary study from Asia measured medication adherence directly. In Bangladesh and Vietnam, diabetes distress—including regimen-related distress—was common (52.5% of 259 and 29.4% of 517 patients with moderate-to-high distress, respectively) [61,62]. In low- and middle-income settings, cost of and access to medicines are recurrent barriers to self-management [12].   Dietary Self-Care Dietary modification is a foundational pillar of diabetes self-management, yet it is shaped by cultural food norms, making generalisation across regions difficult [16,45]. In Western countries, evidence-based dietary guidance—including the Mediterranean diet, low-glycaemic-index approaches and structured meal planning—is reflected in guidelines [3,45]. In a meta-analysis of 14 trials (356 participants), low-glycaemic-index diets reduced HbA1c by 0.43 percentage points more than high-glycaemic-index diets [42]. National dietary guidelines in the UK (NICE NG28), USA (ADA Standards) and major European nations provide standardised nutrition recommendations integrated into DSME programmes [3,56].   In GCC countries, contextual sources describe how dietary self-care is challenged by the nutritional transition from traditional diets to energy-dense, processed food environments [11]. Ramadan—a 29–30-day period of dawn-to-sunset fasting with altered meal timing and often large caloric intake at Iftar—requires specific dietary and pharmacological management [48,54]. Current guidance for Muslim patients recommends, balanced Suhoor meals, portion control at Iftar, and avoiding calorie-dense desserts [48,54].   High-carbohydrate staple diets are commonly described as a challenge to dietary self-care in Asian settings, although no included study quantified this directly. In Chinese adults with impaired glucose tolerance—that is, before the onset of diabetes—the Da Qing study reported 6-year reductions in diabetes incidence of 31%, 46% and 42% in the diet, exercise and combined arms, respectively [63]; this is prevention evidence and only indirectly relevant to self-care in established diabetes. Culturally tailored education, often including dietary components, improved short-term glycaemic outcomes compared with usual care [16]; direct comparisons with generic dietary advice are lacking.   Physical Activity In people with diabetes, regular exercise lowers insulin resistance, improves glycaemic management and lowers cardiovascular risk [9]. The ADA recommends at least 150 minutes of moderate-to-vigorous aerobic activity per week plus resistance training on two or more days a week [9]. Many adults with type 2 diabetes do not meet these recommendations [9].   In GCC countries, high summer temperatures, car-dependent urban design, sedentary occupations and, for some women, limited access to public exercise spaces are described as barriers to physical activity [11]; no included study from a GCC member state measured physical activity or evaluated an exercise programme.   In Asia, physical activity patterns are heterogeneous. Traditional practices such as Tai Chi, yoga and walking are culturally embedded in many Asian communities and represent accessible, low-cost options. However, rapid urbanisation has reduced incidental physical activity, and access to safe recreational environments differs between high-income East Asian cities and lower-income South/Southeast Asian settings. In a 16-week controlled trial in China (49 adults aged 60–69 years with type 2 diabetes, compared with non-exercising controls), supervised aerobic exercise at the intensity of maximal fat oxidation improved glycaemic control, body composition and physical capacity [64].   Foot Care Practices Diabetic foot disease—neuropathic and ischaemic ulcers, infection and amputation—is among the most serious and costly complications of diabetes [47]. The IWGDF guidance recommends daily foot inspection, appropriate footwear, prompt wound management and regular professional foot examination [47]; it notes, however, that trial evidence for patient education alone in preventing ulcers is limited [47].   In Western countries, integrated podiatry services, annual foot screening in primary care and multidisciplinary foot-care teams are standard [47,56]. In UKPDS, intensive glucose control and tight blood-pressure control reduced microvascular endpoints [4,5]. In an observational meta-analysis, glycaemic control was examined in relation to diabetic foot ulcer outcomes [43].   No included primary study from a GCC member state measured foot-care practice. A recent narrative review published after our search cut-off reports that most GCC patients had inadequate foot-care knowledge [11]. In the wider MENA subgroup, 75.2% of 230 older adults with type 2 diabetes in Egypt had unsatisfactory foot self-care practice [55].   In Asian settings, foot-care awareness has been reported to be low in rural populations and among patients with low health literacy [12]. Walking barefoot in parts of South and Southeast Asia, poorly fitted footwear and agricultural foot injuries are commonly cited as additional contextual risks. Community-based foot-care education and low-cost monofilament neuropathy screening have been implemented in resource-limited settings.   Digital Health and mHealth in Diabetes Self-Care The role of digital technologies in supporting diabetes self-care has grown markedly over the past decade, accelerated by the COVID-19 pandemic [13–15,18]. Table 3 summarises the principal self-care interventions, separating evidence of efficacy (what the intervention achieves in trials) from evidence of regional implementation and adoption (where and how it is used).   In Western countries, CGM is now part of standard care for type 1 diabetes and increasingly for insulin-treated type 2 diabetes [3]; in the DIAMOND trial, CGM reduced HbA1c more than usual care in adults with type 1 diabetes using injections [65]. In a meta-analysis of 42 RCTs, telemedicine produced a modest additional HbA1c reduction compared with usual care [18]. In a US cohort of 27,207 adults with diabetes, practices using electronic health records achieved higher composite standards of care and outcomes than paper-based practices [53].   Government-backed digital health platforms, such as Saudi Arabia's Seha and the United Arab Emirates' NABIDH, have expanded digital health infrastructure in GCC nations; their effect on self-care was not evaluated in any included study. A meta-analysis of mobile-phone interventions reported a pooled HbA1c reduction of about 0.5 percentage points [13], suggesting potential for low-cost, population-level impact; no included trial was conducted in a GCC member state. The COVID-19 pandemic accelerated the use of teleconsultation in settings where preferences for in-person care had previously limited it.   In Asia, China's messaging-based health platforms, South Korea's integrated digital health systems and India's expansion of telemedicine after COVID-19 represent a "leapfrogging" pattern. Hou et al. pooled 14 RCTs (1,360 participants) and found that diabetes apps reduced HbA1c by 0.49 percentage points (95% CI 0.30–0.68; moderate-certainty evidence) [14], and Wang et al. reported generally favourable effects of mHealth interventions on diabetes self-management [15]. In a meta-analysis of 28 RCTs in general adult populations, consumer wearable trackers increased daily steps by approximately 627 [36]; because participants were not restricted to people with diabetes, this evidence is indirect and is cited as context only.   Table 3: Diabetes Self-Care Interventions: Efficacy Evidence Versus Regional Implementation/Adoption Evidence (Restructured) Intervention Efficacy evidence (Design; Effect on HbA1c or Main Outcome) [refs] Certainty (Table 4) Implementation / adoption: Asia Implementation / adoption: GCC / wider MENA Implementation / adoption: Western Structured DSME SR of 72 RCTs [2]; MA of 31 RCTs: −0.76 pp (95% CI 0.34–1.18) at end of intervention, attenuating thereafter [8]; DESMOND: no difference at 3 y [6]; NMA [7] Moderate No included programme evaluation; community programmes in parts of East Asia No included GCC evaluation National programmes (DESMOND, Diabetes Manual) [6,52] Culturally adapted education Cochrane review, 11 RCTs (ethnic minority groups): short-term HbA1c improvement vs usual care [16] Low Not evaluated in included studies Not evaluated in included studies Trials mainly among ethnic minorities in UK/USA [16] Mobile phone / app support MA: −0.5 pp [13]; MA of 14 RCTs (n = 1,360): −0.49 pp (95% CI 0.30–0.68) [14]; SR [15] Moderate Trials contributed to the MAs [13,14]; widespread app/messaging platforms No included GCC trial; government digital platforms Widely used [15] Telemedicine MA of 42 RCTs: modest HbA1c reduction vs usual care [18] Moderate Post-COVID expansion Post-COVID expansion Integrated in many systems [18] SMBG (non-insulin-treated T2DM) Cochrane review: small reduction at ≤6 months, not sustained at 12 months [41] Low Cost barriers in LMICs [12] Not measured in included studies Standard of care where indicated [3] CGM (T1DM on injections) DIAMOND RCT (n = 158): greater HbA1c reduction than usual care (≈0.6 pp at 24 weeks) [65] Moderate Cost barriers Cost barriers Recommended in guidelines [3] Low-GI dietary advice MA of 14 trials (n = 356): −0.43 pp vs high-GI diets [42] Low Staple-based diets Ramadan-specific advice [48,54] Integrated in guidelines [3,45] Structured exercise Position statement [9]; 16-week controlled trial in 49 older Chinese adults: improved glycaemic control [64] Low Trial in China [64] Not evaluated in included studies Guideline-based [9] Foot-care education IWGDF: limited trial evidence for education alone [47]; observational MA on glycaemic control and DFU outcomes [43] Very low Low awareness reported [12] Wider MENA: 75.2% unsatisfactory practice (Egypt) [55] Integrated podiatry [47] Structured 1:1 education / psychosocial support Cluster RCT (n = 245): no significant HbA1c difference; less diabetes distress, greater confidence [52] Low Distress common (Bangladesh, Vietnam) [61,62] Distress measured (Saudi Arabia, Iran) [57,66] Trial in UK primary care [52] Pre-Ramadan education and risk stratification Practice guidance; supporting studies observational [48,54] Very low Muslim-majority Asian countries [48] Core regional relevance [48,54] Muslim minority populations [48] CGM: Continuous Glucose Monitoring, DSME: Diabetes Self-Management Education, GI: Glycaemic Index, LMICs: Low- And Middle-Income Countries, MA: Meta-Analysis, NMA: Network Meta-Analysis, pp: Percentage Points, RCT: Randomised Controlled Trial, SMBG: Self-Monitoring of Blood Glucose, SR: Systematic Review. Entries without a citation are descriptive implementation context, not derived from included studies and not graded. The previous version of this table combined trial, review, guideline and general statements in one framework and included citations that did not match the reference list (e.g., Norris 2002, Clair 2015, Turner 1998); these have been corrected   Certainty of Evidence Table 4 presents the certainty of evidence for eleven prespecified outcome–comparison evidence bodies. Certainty was moderate for the effects of DSME, mobile-health support, telemedicine and CGM on HbA1c; low for culturally appropriate education, SMBG, low-glycaemic-index diets and structured exercise; and very low for foot-care education, pre-Ramadan education and all cross-regional differences in self-care practice.   Table 4: Summary of Findings and Certainty of Evidence (GRADE) for Prespecified Outcome–Comparison Evidence Bodies Outcome / comparison Evidence body [refs] Summary of effect Starting level Downgrading / upgrading Certainty 1. DSME vs usual care → HbA1c SRs/MA of RCTs [2,7,8]; cluster RCT [6] −0.76 pp at end of intervention; smaller at longer follow-up; no difference at 3 y in DESMOND High −1 inconsistency (heterogeneous effects, attenuation over time) Moderate 2. Culturally appropriate education vs usual care → HbA1c (ethnic minority groups) Cochrane review of 11 RCTs [16] Short-term (3–6 month) improvement; long-term data lacking High −1 risk of bias; −1 indirectness (mainly Western ethnic-minority populations; not tested in Asian/GCC settings) Low 3. Mobile phone / app support vs usual care → HbA1c MAs of RCTs [13,14]; SR [15] About −0.5 pp (−0.49, 95% CI 0.30–0.68) High −1 risk of bias (unblinded trials, short follow-up) Moderate 4. Telemedicine vs usual care → HbA1c MA of 42 RCTs [18] Modest reduction High −1 inconsistency (heterogeneous interventions) Moderate 5. SMBG vs no SMBG (non-insulin T2DM) → HbA1c Cochrane review [41] Small reduction at ≤6 months; not sustained at 12 months High −1 risk of bias; −1 imprecision/clinical relevance Low 6. CGM vs usual care (T1DM on injections) → HbA1c One RCT, n = 158 [65] ≈0.6 pp greater reduction at 24 weeks High −1 imprecision (single trial); indirect for T2DM and non-Western settings Moderate 7. Low-GI vs high-GI diet → HbA1c MA of 14 trials, n = 356 [42] −0.43 pp High −1 risk of bias; −1 imprecision (small, short trials) Low 8. Structured exercise → glycaemic control Position statement [9]; one non-randomised controlled trial, n = 49 [64] Improved glycaemic control and body composition at 16 weeks Low No further downgrading; direction consistent with position statement [9] Low 9. Foot-care education → ulcer / amputation Guidance [47]; observational MA [43] Limited evidence of effect of education alone High/Low −1 risk of bias; −1 imprecision; indirectness Very low 10. Pre-Ramadan education/risk stratification → hypoglycaemia and adverse events Practice guidance [48,54]; observational studies Fewer adverse events reported; magnitude uncertain Low −1 risk of bias (observational, unadjusted) Very low 11. Cross-regional differences in self-care practice, adherence and distress Cross-sectional studies [49–51,55,57–59,61,62,66,67]; DAWN [10] Wide variation within and between spheres; no consistent regional ranking Low −1 indirectness (indirect comparisons); −1 inconsistency (different instruments, populations, years) Very low pp: Percentage Points. GRADE: High: Further Research Very Unlikely to Change Confidence in the Estimate, Moderate: Likely to Have an Important Impact, Low: Very Likely to Have an Important Impact, Very Low: Any Estimate Is Very Uncertain. Regional GRADE labels used in the previous version (e.g., "High" for Western countries) have been withdrawn. Domain-by-domain judgements are given in Supplementary Appendix S5   Cultural, Religious, and Economic Factors of Self-Care Cultural, religious and economic factors may influence whether recommendations are understood, accepted and enacted. In the included sources, evidence on these factors is mostly cross-sectional or comes from guidance documents; it describes associations and contexts rather than established causal determinants, and each factor varies substantially within regions. Table 5 lists each factor with the basis of its evidence and its within-region variation; the adaptation strategies are suggestions that, unless a reference is given, have not been tested in the included studies.   In the GCC and Muslim-majority Asian countries, fasting during Ramadan poses a clinically significant cultural-religious consideration. Approximately 50 million people with diabetes observe Ramadan annually, although high-risk patients are advised that they are medically exempt [48]. The IDF-DAR risk-stratification framework supports individualised management [54]. Current guidance favours glucose-lowering agents with a low risk of hypoglycaemia and individualised dose adjustment; sodium-glucose cotransporter-2 (SGLT2) inhibitors require attention to hydration because of the risk of volume depletion (and, rarely, ketoacidosis) during fasting, and glucagon-like peptide-1 (GLP-1) receptor agonists are preferably established well before Ramadan because of gastrointestinal adverse effects [54,68]. These are guideline-based recommendations rather than findings of this review. Pre-Ramadan structured education is recommended, supported mainly by observational studies [48,54].   Family members may support medication adherence, dietary change and healthcare engagement, or may reinforce unhealthy habits [10]. Culturally appropriate education programmes, some of which involved family members, improved short-term glycaemic control compared with usual care, mainly among ethnic minority populations in Western countries [16]; whether family involvement itself drives benefit has not been tested in the included studies. Family structures also differ widely within Asia and the GCC—for example, between urban nuclear and extended households, and between national families and expatriate workers living apart from their families.   Economic factors intersect with cultural ones. In lower-income settings, out-of-pocket costs for medicines and monitoring are a recurrent barrier to self-management [12], although their magnitude varies widely between and within countries. Generic medication policies, subsidy programmes and community health worker-delivered education are commonly proposed responses [12]; their effects on self-care were not evaluated in the included studies.   Table 5: Cultural and Religious Factors that May Influence Diabetes Self-Care: Evidence Basis, Within-Region Variation and Suggested Adaptations Factor Reported or proposed impact Evidence basis (included sources) Within-region variation Suggested adaptation Ramadan fasting (GCC/Asia/Muslim communities worldwide) Altered meal timing, medication dosing, hypoglycaemia risk; a large majority (≈79% in a large multinational survey) of Muslims with T2DM fast [48] Practice guidance; observational studies [48,54] Fasting duration varies by latitude and season; exemptions depend on individual risk Individualised pre-Ramadan medical review; medication adjustment per guideline, with individual risk–benefit assessment for any agent; IDF-DAR risk stratification [54] Family involvement (Asia, GCC) Family may enable adherence or reinforce unhealthy traditional diets Indirect: family components in culturally adapted education trials [16]; not tested in Asian/GCC included studies Urban nuclear vs extended households; nationals vs expatriate workers living without family Family-inclusive education sessions; involve key decision-maker (to be evaluated) Traditional/complementary medicine use (Asia) Patients may delay evidence-based care; herb–drug interactions possible No included study Varies by country, ethnicity and health-system integration Non-judgmental enquiry at every visit; integrate into shared decision-making Gender norms and privacy (GCC, parts of Asia) Women may avoid mixed-gender clinics; may reduce screening uptake No included study; contextual [11] Differs between and within countries and over time Same-gender healthcare providers; female diabetes educator programmes Fatalistic health beliefs 'God's will' framing may reduce self-efficacy No included study Not specific to any region or religion; varies by individual Spiritually sensitive counselling; community religious leader engagement High-carbohydrate staple diets (Asia: rice, noodles) Postprandial hyperglycaemia challenging to control Low-GI trial evidence [42]; regional diet not quantified in included studies Staple foods differ widely across Asia Portion-control strategies; glycaemic index education; culturally adapted plate models Economic disparities (South/Southeast Asia) Out-of-pocket medication costs reduce adherence Scoping review [12]; cross-national barriers [10] Large differences between and within countries Generic medication access; government subsidy advocacy; CHW programmes Ultra-processed food transition (GCC) Rapid nutritional transition associated with obesity and T2DM incidence Contextual [11] Differs by GCC state and population group National food labelling policy; taxation on sugar-sweetened beverages Stigma around diabetes (all regions Concealment of diagnosis; delays in care-seeking Not measured directly in included studies Present in all regions; magnitude unknown Anti-stigma community campaigns; peer support groups High physical inactivity (GCC) Hot climate, car-dependent culture, sedentary work Contextual [11] Differs by sex, nationality and occupation Indoor exercise facilities; employer wellness programmes; step-counting apps DM: Diabetes Mellitus, IDF-DAR: International Diabetes Federation and Diabetes and Ramadan International Alliance, SGLT2i: Sodium-Glucose Cotransporter-2 Inhibitor, GLP-1 RA: Glucagon-Like Peptide-1 Receptor Agonist, CHW: Community Health Worker. Contextual": Background Source not Included in the Synthesis. Suggested adaptations are expert/policy suggestions unless a reference is given   Complication Prevention Strategies: A Cross-Regional Assessment Preventing the major complications of diabetes—nephropathy, retinopathy, peripheral neuropathy, cardiovascular disease and diabetic foot disease—requires sustained self-care integrated with clinical monitoring. Table 6 summarises guideline-recommended screening intervals, key prevention strategies and reported regional gaps for each major complication domain.   UKPDS remains the landmark evidence for intensive glycaemic control, demonstrating a 25% risk reduction in microvascular endpoints with intensive glucose control [4], further benefit from tight blood-pressure control [5] and a legacy effect at 10-year follow-up [60]. Additional benefits from cholesterol management and, more recently, cardioprotective glucose-lowering medications such as SGLT2 inhibitors and GLP-1 receptor agonists were shown by subsequent trials [68]. Smoking cessation is associated with a lower risk of cardiovascular disease in adults with diabetes, despite post-cessation weight gain [69]. Annual complication screening is recommended by guidelines [3,56] and, in some Western countries, integrated into primary care quality frameworks and pay-for-performance schemes.   Late detection of complications in GCC countries has been described in contextual sources [11]. The nationwide SAUDI-DM household survey documented a substantial burden of diabetes and abnormal glucose metabolism in the Saudi population [58]. National initiatives such as Vision 2030's diabetes component and Saudi Arabia's Healthy Saudi Initiative aim to address this through community screening and primary care capacity-building.   In Asian contexts, the shortage of ophthalmologists, nephrologists and podiatrists relative to the high prevalence of diabetes, especially in South and Southeast Asia, limits complication prevention [12]. Task-shifting approaches, in which trained nurses, pharmacists and community health workers deliver standardised complication screening, are proposed as affordable and scalable options. India's National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Diseases and Stroke (NPCDCS) is a large-scale attempt to integrate complication screening into primary healthcare.   Table 6: Evidence-Based Complication Prevention Strategies and Regional Implementation Gaps Complication Screening Interval (guideline-recommended) Key Prevention Strategy Regional Gap Identified Diabetic foot ulcer / amputation Annual foot examination Offloading, foot care education, podiatry referral GCC: Podiatry access inequitable; Asia: Awareness low [12] Diabetic nephropathy Annual eGFR + urine ACR SGLT2i, RAAS blockade, BP control < 130/80 mmHg Asia: Late-stage presentation common; GCC: Improving Diabetic retinopathy Annual dilated fundus exam Optimal HbA1c & BP; anti-VEGF for PDR; laser Asia/GCC: Ophthalmology access gaps Cardiovascular disease Annual lipid profile, ECG Statin therapy, antiplatelet, GLP-1 RA, SGLT2i GCC: Underuse of cardioprotective agents Diabetic peripheral neuropathy Annual 10-g monofilament Glycaemic control, pain management, balance exercises All regions: Underdiagnosed in primary care Hypoglycaemia (iatrogenic) Continuous monitoring preferred Individualised HbA1c targets; patient education on rescue glucose Western: CGM adoption rising; Asia/GCC: Cost barriers Diabetic ketoacidosis (T1DM) Sick-day rules education Sick-day rules; insulin never stopped; ketone testing Asia/GCC: Sick-day rule knowledge low Hypertension-related complications Every visit BP measurement ACEi/ARB first line; dietary sodium restriction GCC: Sodium-rich traditional diet challenge Egfr: Estimated Glomerular Filtration Rate; ACR: Albumin-To-Creatinine Ratio; CGM: Continuous Glucose Monitoring; PDR: Proliferative Diabetic Retinopathy; RAAS: Renin-Angiotensin-Aldosterone System; BP: Blood Pressure. Screening intervals based on ADA [3], and NICE NG28 [56]. Regional gaps without a citation are descriptive context, not derived from included studies and not graded   Swot Analysis: Regional Self-Care Landscapes The SWOT analysis in Table 7 is an interpretive, author-generated summary intended to structure policy discussion. It is not graded evidence, and each entry generalises across countries that differ substantially; it should be read alongside the within-region heterogeneity described in Section 8.2.   Table 7: SWOT Analysis of Diabetes Self-Care Systems: Asia, GCC, and Western Countries (Interpretive: Not Graded Evidence) SWOT Domain Asia GCC Western Countries Strengths Strong family/community networks; rapid digital adoption; large study populations Government-funded healthcare for nationals; Vision 2030 digital health investment; low smoking rates Mature DSME infrastructure; robust RCT evidence base; comprehensive guidelines Weaknesses Urban–rural access gap; limited podiatry/specialist services; cost barriers Ethnic minority (expatriate) health inequity; cultural barriers to mixed-gender care; high obesity rates Healthcare cost barriers (USA); insulin pricing; complacency in high-income populations Opportunities mHealth scale-up; community health worker integration; regional guidelines adaptation National NCD strategies; telemedicine expansion; Ramadan intervention protocols Precision medicine; CGM/artificial pancreas; personalised nutrition algorithms Threats Rapid nutritional transition; pharmaceutical access inequity; growing non-communicable disease burden Sedentary lifestyle; ultra-processed food industry; limited academic research output Health misinformation online; ageing population DSME: Diabetes Self-Management Education, CGM: Continuous Glucose Monitoring, NCD: non-communicable disease. Author-generated interpretive summary, entries are not graded evidence and do not apply uniformly to all countries in a sphere

Principal Findings This systematic review included 38 sources on diabetes self-care in Asia, the GCC/wider MENA and Western countries. Interventions with moderate-certainty evidence of benefit on glycaemic control—structured DSME, mobile-health support and telemedicine—were tested mainly in Western or multi-regional trials. Evidence on regional practice and implementation came largely from cross-sectional studies using different instruments, and only two included primary studies came from GCC member states. Across the evidence base, the dominant pattern is heterogeneity: self-care practices, adherence and diabetes distress varied as much within spheres as between them.   Heterogeneity Within Regions Each sphere is internally diverse. Asia spans high-income East Asian health systems with universal insurance and lower-income South and Southeast Asian settings with high out-of-pocket spending and large urban–rural gaps [12]; the included Asian studies (China, Republic of Korea, Vietnam, Bangladesh) cannot represent the continent as a whole. Within the GCC, national health systems, financing arrangements and the size of expatriate populations—who may have different entitlements, languages and living arrangements from nationals—differ between member states, and the wider MENA countries (Egypt, Palestine, Iran) differ again in income and health-system organisation. Western countries range from universal tax-funded systems to the insurance-based system of the USA, with persistent socioeconomic and ethnic-minority gaps [16]. Region-level labels therefore mask important variation, and the findings of this review should be interpreted at the level of countries and settings rather than continents.   Direct Versus Indirect Comparisons Only one included study, the DAWN study, measured psychosocial barriers to self-care in several spheres (13 countries in Asia, Europe, North America and Australia) with a common method [10]. All other cross-regional contrasts in this review are indirect: they compare studies conducted in different populations (clinic attenders, community samples, older adults), with different instruments (e.g., MMAS-8, the Summary of Diabetes Self-Care Activities measure [70], different language versions of the Diabetes Distress Scale), in different years. Apparent differences may therefore reflect measurement and sampling rather than true differences in self-care. Piragine et al. illustrate the point: adherence estimates depended on whether administrative data or self-report were used [17]. For this reason, the previous version's statements that Western countries have the "highest" adherence and the GCC the "lowest" have been removed; the evidence does not support a ranking.   Comparison with Existing Reviews Our findings are consistent with a recent GCC narrative review, published after our search cut-off, which synthesised 88 articles from 2020–2025 and reported fair but highly variable medication adherence, self-monitoring of blood glucose ranging from 14% to over 90% across studies, and generally inadequate foot-care knowledge [11]. They also agree with Maina et al., who described cost, access and health-literacy barriers to complication-prevention practices in low- and middle-income countries [12]. Unlike these single-region reviews, the present review places the three spheres within one framework and separates efficacy evidence from implementation evidence, which makes explicit how little direct comparative evidence exists.   Recommendations The following recommendations combine the findings of this review with guideline and expert opinion. Each is labelled E when it is supported by included evidence (with its certainty from Table 4) or P when it is an expert or policy recommendation based on guidelines or author judgement.   E: DSME, moderate; cultural adaptation, low, P: family-inclusive and Ramadan components. Implement culturally adapted DSME programmes and evaluate them locally, with consideration of family-inclusive models in Asian and GCC settings where appropriate, and integration of Ramadan-specific counselling for Muslim patients. Programmes developed elsewhere should be adapted and evaluated before scale-up [2,8,16] E: mHealth Efficacy, Moderate, P: Population scale-up and equity frameworks Scale mHealth interventions, prioritising SMS-based reminders and low-bandwidth applications in resource-constrained settings. Governments in Asia and the GCC should establish digital health equity frameworks so that people without smartphone access or digital literacy are not excluded [13,14,18] P: Establish Ramadan diabetes management protocols as a standard component of national diabetes guidelines in the GCC and Muslim-majority Asian nations. Pre-Ramadan diabetes clinics, and structured risk stratification using IDF-DAR criteria should be institutionalised, with medication adjustment according to current guidelines and an individual risk–benefit assessment for any agent used during fasting, including SGLT2 inhibitors and GLP-1 receptor agonists [54,68] E: low-GI diets, low, P: materials and food-environment policy, create culturally appropriate nutritional counselling materials in regional languages and dialects that include local staple foods, traditional food systems and familiar meal patterns. Individual-level counselling may be complemented by food-environment policies such as taxes on sugar-sweetened beverages, front-of-pack labelling and public-institution food standards [42,45] P: Supporting Evidence Very Low, Address foot care as a priority in Asia and the GCC through training of primary care clinicians in annual foot examination, community foot-care education campaigns, and podiatry workforce development in GCC health systems [47,55] P: Implement task-shifting models for complication screening in under-resourced Asian settings, with trained nurses, pharmacists and community health workers delivering standardised retinopathy, nephropathy, neuropathy and foot screening [12] P: Research recommendation Commission comparative effectiveness research on self-care interventions across the three spheres. Head-to-head trials evaluating the same DSME model in Western, GCC and Asian settings, and multinational surveys using common, validated self-care and adherence instruments, would substantially advance the evidence base [7,10] P Invest in anti-stigma campaigns to address diabetes-related social stigma, which may discourage diagnosis disclosure, care-seeking and medication adherence in all three regions   LIMITATIONS This review has several limitations. (1) No registered protocol: the review was not prospectively registered, so post hoc methodological decisions cannot be excluded. (2) Language and databases: only English-language publications and three databases were searched; Embase, CINAHL and regional databases (e.g., the Index Medicus for the Eastern Mediterranean Region, CNKI, the Western Pacific Region Index Medicus) were not, which may under-represent Arabic- and Asian-language evidence. (3) Search structure: requiring complication/outcome or digital-health terms may have reduced sensitivity for relevant self-care studies, only partly mitigated by hand-searching. (4) Currency: the search was not updated after 1 September 2024, so more recent studies, including GCC studies summarised in a 2025 review [11], were not included. (5) Heterogeneous designs and populations: included sources differed in design, setting, population and period. (6) Secondary evidence: inclusion of systematic reviews and guidelines alongside primary studies risks double-counting; we mitigated this with separate evidence streams and overlap checks. (7) Indirect comparisons: almost all cross-regional comparisons were indirect. (8) Publication bias could not be assessed formally. (9) No quantitative pooling was possible, and the narrative synthesis involves judgement. (10) Variable definitions of self-care: instruments and cut-offs for adherence, self-management and distress varied. (11) Reviewer bias and incomplete process records: selection, extraction and appraisal involved judgement; screening was performed in duplicate, but inter-reviewer agreement was not quantified, the numbers of records excluded at the title/abstract and initial full-text stages were not retained, and methodological appraisal and GRADE ratings were completed at the revision stage rather than prospectively. (12) Sparse GCC evidence: only two included primary studies were from GCC member states, and some descriptions of regional health systems rely on contextual sources that were not graded.

This systematic review found that diabetes self-care awareness, practice and complication-prevention activity vary considerably both between and within Asia, the GCC/wider MENA and Western countries. Structured DSME and mobile-health support improve glycaemic control in trials (moderate-certainty evidence), but these trials were conducted mainly in Western or multi-regional settings, and their effects in GCC and many Asian settings remain uncertain. Because almost all cross-regional contrasts were indirect—drawn from studies with different populations, instruments and periods—this review does not establish a ranking of regions or health systems. The evidence instead points to heterogeneity and context-specific considerations, including Ramadan fasting, cost and access barriers, health-system organisation, and family and gender dynamics, each of which varies within regions.   Context-adapted self-care programmes should be evaluated locally with standardised outcome measures, and multinational studies using common validated instruments are needed to allow direct comparison. Strengthening primary research in GCC member states is a particular priority.   Conflicts of Interest The authors declare no conflicts of interest.   Data Availability No original datasets were generated for this review. All data sources are cited in the reference list, and the extracted data are provided in Supplementary Appendices S4–S6.   Supplementary Material Supplementary Appendices S1–S9 (database-specific search strategies; grey-literature search log; country-level geographic classification; characteristics and appraisal of included sources and list of contextual references; GRADE evidence profile; sources of numerical estimates; reference audit log; PRISMA 2020 checklist; list of sources excluded at the eligibility re-audit) are provided in a separate file.

Supplementary Material Diabetes Self-Care in Asia, the Gulf Cooperation Council and Western Countries: A Systematic Review with Narrative Synthesis Manuscript JPMS332246 — Supplementary Appendices submitted in response to reviewer comments   Contents Database-specific search strategies Grey-literature search log Country-level classification of geographic spheres Characteristics and methodological appraisal of included sources (Part A) and contextual references (Part B) GRADE evidence profile: domain-by-domain judgements Sources of numerical estimates in Table 2 and values removed Reference audit log PRISMA 2020 checklist Sources excluded at the eligibility re-audit, with reasons   Appendix S1. Database-specific search strategies The syntax below implements, for each database, the four concept blocks reported in Section 2.3 of the manuscript. All three databases were searched on 1 September 2024 for records published from 1 January 2000 to the search date. Limits: English language; humans (PubMed); article and review document types (Scopus, Web of Science). The search was not updated after 1 September 2024. The four concept blocks are those reported in Section 2.3 of the manuscript.   S1: 1 PubMed/MEDLINE (Interface: pubmed.ncbi.nlm.nih.gov) Number Search string #1 "Diabetes Mellitus"(Mesh) OR "Diabetes Mellitus, Type 2"(Mesh) OR "diabetes mellitus"(tiab) OR "type 2 diabetes"(tiab) OR T2DM(tiab) #2 "Self-Care"(Mesh) OR "Self-Management"(Mesh) OR "Patient Education as Topic"(Mesh) OR "Medication Adherence"(Mesh) OR "self-care"(tiab) OR "self-management"(tiab) OR "patient education"(tiab) OR adherence(tiab) #3 "Asia"(Mesh) OR "Middle East"(Mesh) OR "Saudi Arabia"(Mesh) OR "United Arab Emirates"(Mesh) OR "Europe"(Mesh) OR "North America"(Mesh) OR Asia(tiab) OR "South Asia"(tiab) OR "Southeast Asia"(tiab) OR "East Asia"(tiab) OR "Gulf Cooperation Council"(tiab) OR GCC(tiab) OR "Saudi Arabia"(tiab) OR "United Arab Emirates"(tiab) OR "Middle East"(tiab) OR "Western countries"(tiab) OR Europe(tiab) OR "North America"(tiab) #4 "Diabetes Complications"(Mesh) OR "Glycated Hemoglobin"(Mesh) OR "Diabetic Foot"(Mesh) OR "Diabetic Retinopathy"(Mesh) OR "Diabetic Nephropathies"(Mesh) OR "Telemedicine"(Mesh) OR "complication prevention"(tiab) OR "glycemic control"(tiab) OR HbA1c(tiab) OR "foot care"(tiab) OR retinopathy(tiab) OR nephropathy(tiab) OR "digital health"(tiab) OR mHealth(tiab) #5 #1 AND #2 AND #3 AND #4 #6 #5 AND ("2000/01/01"(dp) : "2024/09/01"(dp)) AND English(lang) AND humans(mh)   S1: 2 Scopus (Interface: Scopus.Com, Advanced Search) Number Search string 1 TITLE-ABS-KEY("diabetes mellitus" OR "type 2 diabetes" OR T2DM) 2 TITLE-ABS-KEY("self-care" OR "self-management" OR "patient education" OR adherence) 3 TITLE-ABS-KEY(Asia OR "South Asia" OR "Southeast Asia" OR "East Asia" OR "Gulf Cooperation Council" OR GCC OR "Saudi Arabia" OR "United Arab Emirates" OR "Middle East" OR "Western countries" OR Europe OR "North America") 4 TITLE-ABS-KEY("complication prevention" OR "glycemic control" OR HbA1c OR "foot care" OR retinopathy OR nephropathy OR "digital health" OR mHealth) 5 1 AND 2 AND 3 AND 4 AND PUBYEAR > 1999 AND PUBYEAR < 2025 AND LANGUAGE(english) AND DOCTYPE(ar OR re)   S1: 3 Web of Science Core Collection (Interface: webofscience.com; All Editions) Number Search String 1 TS = ("diabetes mellitus" OR "type 2 diabetes" OR T2DM) 2 TS = ("self-care" OR "self-management" OR "patient education" OR adherence) 3 TS = (Asia OR "South Asia" OR "Southeast Asia" OR "East Asia" OR "Gulf Cooperation Council" OR GCC OR "Saudi Arabia" OR "United Arab Emirates" OR "Middle East" OR "Western countries" OR Europe OR "North America") 4 TS = ("complication prevention" OR "glycemic control" OR HbA1c OR "foot care" OR retinopathy OR nephropathy OR "digital health" OR mHealth) 5 #1 AND #2 AND #3 AND #4; Publication years 2000–2024 (to 1 Sept 2024); Language: English; Document types: Article, Review Total records retrieved from the three databases: 5,305 (before de-duplication)   Appendix S2: Grey-Literature Search Log Source URL / repository searched Date searched Search terms Outcome World Health Organization www.who.int/publications; IRIS (iris.who.int) September 2024 diabetes AND (self-care OR self-management OR education) No document met the eligibility criteria for inclusion as a source International Diabetes Federation www.idf.org (resources); www.diabetesatlas.org September 2024 diabetes self-care; self-management; education; Ramadan IDF Diabetes Atlas estimates used for epidemiological context [1,30]; IDF-DAR Ramadan guidance used to frame recommendations [54]; not counted as included sources National Institute for Health and Care Excellence www.nice.org.uk/guidance September 2024 type 2 diabetes; type 1 diabetes; diabetic foot NICE NG28 used to define recommended standards [56]; not counted as an included source Reference lists of included sources and key reviews [2,12,16,17] Backward citation screening September 2024 n/a Eligible sources identified in this way are within the 52 sources retained after screening (Figure 1) Eligibility of guidelines and reports: issued by a national or international organisation; describes its evidence base or development method; reports recommendations or data on self-care, education or complication prevention in adults with diabetes; published January 2000–September 2024; English language   Appendix S3: Country-level Classification of Geographic Spheres Sphere / subgroup Countries and territories Rule applied Included sources Asia – East Asia China (incl. Hong Kong SAR, Macao SAR), Taiwan, Japan, Republic of Korea, Mongolia UN M49 Eastern Asia [51] China; [64] China; [59] Republic of Korea Asia – South Asia India, Pakistan, Bangladesh, Sri Lanka, Nepal, Bhutan, Maldives, Afghanistan UN M49 Southern Asia (excluding Iran – see below) [62] Bangladesh Asia – Southeast Asia Vietnam, Thailand, Malaysia, Singapore, Indonesia, Philippines, Myanmar, Cambodia, Lao PDR, Brunei, Timor-Leste UN M49 South-eastern Asia [61] Vietnam GCC (core) Saudi Arabia, United Arab Emirates, Kuwait, Qatar, Bahrain, Oman GCC member states [57] Saudi Arabia; [58] Saudi Arabia Wider MENA (non-GCC) Egypt, Jordan, Palestine, Lebanon, Syria, Iraq, Yemen, Iran, Libya, Tunisia, Algeria, Morocco, Sudan IDF MENA region, excluding GCC; reported separately, not attributed to GCC [49] Palestine; [55] Egypt; [50] Egypt; [66] Iran Transcontinental Turkey IDF Europe region; geographically and culturally transcontinental; reported separately [67] Turkey Western countries European Union and EFTA countries, United Kingdom, United States, Canada, Australia, New Zealand Europe (IDF Europe region excluding Turkey and Central Asian states), North America, Australasia [53,65,69,70] USA; [6,52] UK Multi-regional Studies or reviews spanning more than one sphere Attributed to a sphere only when results reported by country [10] DAWN (13 countries); reviews [12–15,17,18,42–44] Central Asian states (e.g., Kazakhstan, Uzbekistan) and Israel: no eligible sources were identified; had any been found, they would have been reported separately as transcontinental/other   Appendix S4: Characteristics and Methodological Appraisal of Sources Ref Source Stream Design Country / sphere Population; n Self-care domain / measure Outcomes and main findings Appraisal 6 Khunti 2012 (DESMOND) A Cluster RCT, 3-year follow-up UK / Western Newly diagnosed T2DM; 824 randomised, 207 practices; 604 with 3-year biomedical data Structured group education vs usual care; HbA1c, illness beliefs No HbA1c difference at 3 years (−0.02; 95% CI −0.22 to 0.17); illness beliefs improved RoB 2: some concerns (attrition) 10 Peyrot 2005 (DAWN) A Cross-national survey 13 countries in Asia, Europe, North America, Australia / multi-regional 5,104 adults with diabetes; 3,827 providers Psychosocial problems and barriers to self-management Psychosocial problems common across countries; only direct cross-regional comparison in this review JBI: moderate concern (sampling varied by country) 49 Sweileh 2014 A Cross-sectional survey Palestine (Nablus) / wider MENA Adults with T2DM, governmental primary care clinic; n = 405 Medication adherence (MMAS-8); BMQ; MDKT 57.3% adherent; lower knowledge, weaker necessity beliefs and greater concerns associated with non-adherence JBI: moderate concern (single site) 50 Azzam 2021 A Cross-sectional survey Egypt (Mansoura) / wider MENA Adults with diabetes, 24 PHC facilities; n = 302 Glycaemic control; medication adherence; facility and physician factors 30.5% high and 51.3% medium adherence; education, exercise, guideline use and physician training associated with control JBI: moderate concern 51 Huang 2014 A Cross-sectional survey China (Chengdu) / Asia Adults with T2DM, 6 urban districts; n = 364 Self-management knowledge, beliefs, efficacy, support and behaviour (28-item scale) 46% "good", 45% "fair", 6% "poor" self-management behaviour JBI: moderate concern (urban sample; local instrument) 52 Sturt 2008 A Cluster RCT UK / Western Adults with T2DM in primary care; n = 245 Diabetes Manual 1:1 structured education; HbA1c, distress, confidence No significant HbA1c difference; reduced distress and increased confidence RoB 2: some concerns 53 Cebul 2011 A Retrospective cohort USA (Cleveland) / Western Adults with diabetes, 46 practices; n = 27,207 EHR use; composite standards of care and outcomes EHR practices achieved higher standards of care and outcomes JBI: moderate concern (non-random practice selection) 55 Abo-Eata 2022 A Descriptive cross-sectional Egypt (Tanta) / wider MENA Older adults with T2DM, university diabetes clinic; n = 230 Foot self-care practice; health locus of control 75.2% unsatisfactory foot self-care practice JBI: moderate concern (single site; descriptive) 57 Batais 2021 A Instrument validation Saudi Arabia (Riyadh) / GCC Adults with T2DM, university medical city; n = 109 Arabic Diabetes Distress Scale (SADDS-17) α = 0.848; test–retest 0.78; distress scores elevated JBI: moderate concern (small single-site sample) 58 Al-Rubeaan 2014 (SAUDI-DM) A National household survey Saudi Arabia / GCC Population-based sample; n = 18,034 Diabetes and abnormal glucose metabolism; impact Substantial burden of diabetes and abnormal glucose metabolism JBI: low concern 59 Lee 2020 A Instrument development Republic of Korea / Asia Adults with T2DM, 3 university hospitals and 1 health centre; n = 473 17-item Diabetes Self-Management Scale Six-factor structure; good fit and internal consistency JBI: low concern 61 Huynh 2021 A Cross-sectional survey Vietnam (Ho Chi Minh City) / Asia Adults with T2DM, 3 district hospitals; n = 517 Diabetes Distress Scale (DDS-17) 23.6% moderate, 5.8% high distress; associated with age, duration, HbA1c, medication type JBI: moderate concern 62 Kamrul-Hasan 2022 A Cross-sectional survey Bangladesh / Asia Adults with T2DM, endocrinology clinics; n = 259 DDS-17; PHQ-9 52.5% distress (29.7% moderate, 22.8% high) JBI: moderate concern 64 Jiang 2020 A Controlled trial vs non-exercising controls (randomisation not reported) China / Asia Adults aged 60–69 y with T2DM; n = 49 (24 women, 25 men) Supervised aerobic exercise at maximal fat-oxidation intensity, 1 h/day, 3 days/week, 16 weeks; glycaemic control, body composition Improved glycaemic control, body composition, lipid profile and physical capacity vs controls JBI (quasi-experimental): moderate concern (allocation method not reported) 65 Beck 2017 (DIAMOND) A Randomised controlled trial USA / Western Adults with T1DM on multiple daily injections; n = 158 CGM vs usual care; HbA1c at 24 weeks Greater HbA1c reduction with CGM (≈0.6 pp difference) RoB 2: some concerns (open label) 66 Azadbakht 2020 A Cross-sectional survey Iran / wider MENA Older adults with T2DM, Tehran; sample size not reported in the indexed abstract record Diabetes distress Prevalence of distress and predictors (comorbidity, body mass index, disease duration, sex, sedentary lifestyle) examined JBI: moderate concern (cross-sectional; details limited in indexed record) 67 Telci Caklili 2020 A Instrument validation Turkey (Istanbul) / transcontinental Adults with T2DM, university clinic; n = 205 Turkish DDS-17 α = 0.874; 30% with score ≥3; weak correlation with HbA1c JBI: moderate concern 69 Clair 2013 A Prospective cohort (Framingham Offspring) USA / Western Adults with and without diabetes; n = 3,251 Smoking cessation, weight change, CVD Cessation associated with lower CVD risk despite weight gain; estimates in diabetes subgroup imprecise JBI: low concern 70 Toobert 2000 A Instrument validation (7 studies) USA / Western Adults with diabetes; n = 1,988 across 7 studies Summary of Diabetes Self-Care Activities (SDSCA) Revised SDSCA with acceptable reliability and validity JBI: low concern 2 Norris 2001 B SR of RCTs Mainly USA / Western 72 RCTs of self-management training in T2DM Knowledge, SMBG, diet, glycaemic control Positive short-term effects; long-term evidence limited AMSTAR 2: low 7 Pillay 2015 B SR and network MA Mainly Western / Western 132 RCTs of behavioural programmes for T2DM HbA1c and other outcomes Programmes with >10 contact hours generally reduced HbA1c by >0.4 pp; brief DSME without added support gave little benefit AMSTAR 2: high 12 Maina 2023 B Scoping review LMICs (incl. Asia) / multi-regional Studies of complication-prevention self-management in T2DM Self-management practices, barriers Cost, access and health-literacy barriers commonly reported Scoping review: reporting adequate (AMSTAR 2 not applicable) 13 Liang 2011 B MA of RCTs Multi-regional RCTs of mobile-phone interventions HbA1c Pooled reduction ≈0.5 pp; larger in T2DM AMSTAR 2: low 14 Hou 2016 B SR/MA of 14 RCTs with GRADE Multi-regional 1,360 participants HbA1c −0.49 pp (95% CI 0.30–0.68); moderate-certainty evidence AMSTAR 2: moderate 15 Wang 2017 B SR Multi-regional Studies of mHealth for obesity and diabetes Self-management outcomes Generally favourable effects; heterogeneous designs AMSTAR 2: critically low 16 Hawthorne 2008 B Cochrane SR/MA Mainly UK/USA / Western 11 RCTs in ethnic minority groups Culturally appropriate education; HbA1c, knowledge Short-term HbA1c improvement vs usual care AMSTAR 2: high 17 Piragine 2023 B SR/MA Global / multi-regional 156 studies (87 cohort, 69 cross-sectional); >10 million patients Adherence to oral antidiabetic drugs 54% adherent (95% CI 51–58%); adherence associated with glycaemic control (OR 1.33) AMSTAR 2: moderate 18 Tchero 2019 B MA of 42 RCTs Multi-regional RCTs of telemedicine in diabetes HbA1c Modest reduction vs usual care AMSTAR 2: low 41 Malanda 2012 B Cochrane SR/MA Mainly Europe / Western RCTs of SMBG in non-insulin-treated T2DM HbA1c, well-being Small reduction at ≤6 months; not sustained at 12 months AMSTAR 2: high 42 Brand-Miller 2003 B MA of RCTs Multi-regional 14 studies; n = 356 Low- vs high-GI diet; HbA1c −0.43 pp AMSTAR 2: low 43 Lane 2020 B SR/MA of observational studies Multi-regional Observational studies of patients with DFU Glycaemic control and ulcer outcomes Associations examined; observational evidence AMSTAR 2: moderate 44 Sartore 2023 B MA Multi-regional Studies of HbA1c variability in T2DM Macro-/microvascular complications HbA1c variability associated with complications AMSTAR 2: low 3 ADA 2024 C Clinical practice guideline USA / Western — Standards of care incl. screening intervals, CGM Recommended standards AGREE II: high 9 Colberg 2016 C Position statement USA / Western — Physical activity ≥150 min/week aerobic + resistance training AGREE II: moderate 45 ADA 2023 (Section 5) C Clinical practice guideline USA / Western — DSMES, nutrition, physical activity, smoking cessation Standards for health behaviour support AGREE II: high 46 Gómez-Peralta 2024 C Expert consensus proposal Spain / Western — Phenotype- and stage-based management Expert management proposal AGREE II: low 47 Bus 2016 (IWGDF) C International guidance International / multi-regional — Foot-ulcer prevention Foot inspection, footwear, education; limited evidence for education alone AGREE II: high 48 Ahmed 2020 C Practice update / consensus UK-based, international scope / multi-regional — Ramadan and diabetes Risk stratification, education, medication adjustment AGREE II: low–moderate DFU = diabetic foot ulcer; EHR = electronic health record; GI = glycaemic index; LMICs = low- and middle-income countries; MA = meta-analysis; pp = percentage points; SR = systematic review. Part A. Included sources (n = 38). Stream A = primary study; B = systematic review/meta-analysis/scoping review; C = guideline/consensus statement. Appraisal tools: RoB 2 (randomised trials); JBI checklists (cohort, cross-sectional, validation studies); AMSTAR 2 overall confidence (systematic reviews; scoping review appraised for reporting only); AGREE II overall (guidelines). Appraisal was completed at the revision stage   Part B: Contextual references (n = 14) and references added at revision (n = 18). These are cited for background or methods only. They are not included sources and did not contribute to the tables of findings or certainty ratings. Ref Source Type Reason not included as a source 11 Thirunavukkarasu 2025 Narrative review of GCC self-management Published after search cut-off; narrative review 19 Puzhakkal 2025 Conceptual review Published after search cut-off; conceptual (ineligible design) 22 Mogale 2025 Primary-care barriers study Published after search cut-off; South Africa (outside spheres) 30 Saeedi 2019 IDF Atlas 9th-edition estimates Epidemiological background; superseded by 2021 estimates [1] 31 Obici 2003 Basic-science review Not a self-care study 32 Al-Dwaikat 2023 Concept analysis Conceptual (ineligible design) 33 Zimmet 2001 Commentary on epidemic Epidemiological background; narrative 34 Colagiuri 2005 Commentary on epidemic Epidemiological background; narrative 35 Schwingshackl 2015 MA of Mediterranean diet and risk of diabetes Population without diabetes (criterion i) 36 Brickwood 2019 MA of wearable trackers Population not restricted to diabetes 37 Greenhalgh 2004 Organisational innovation review Not diabetes-specific 38 Caron 2023 Health promotion overview Not diabetes-specific 39 Gill 2009 Regional perspective Sub-Saharan Africa (outside spheres); narrative 40 Zoungas 2014 (ADVANCE-ON) RCT follow-up Pharmacological only; no self-care component 1, 4, 5, 8, 54, 56, 60, 63, 68 IDF 2021; IDF-DAR 2021; Norris 2002; UKPDS 33/38; Holman 2008; Pan 1997; ADA/EASD 2022; NICE NG28 Added at revision To support specific statements with correct primary sources (reference audit, Appendix S7) 20, 21, 23–29 PRISMA 2020; GRADE; ESRC narrative synthesis; SWiM; Murad 2017; AMSTAR 2; RoB 2; JBI; AGREE II Methods references Reporting and appraisal methods   Appendix S5: GRADE Evidence Profile: Domain-By-Domain Judgements Outcome / comparison Sources Design (start) Risk of bias Inconsistency Indirectness Imprecision Publication bias Upgrade Certainty Explanation 1. DSME → HbA1c [2,6–8] RCTs (high) 0 −1 0 0 Undetected None Moderate Effects vary between trials and attenuate after the intervention ends; DESMOND showed no 3-year difference. 2. Culturally appropriate education → HbA1c [16] RCTs (high) −1 0 −1 0 Undetected None Low Trials at unclear risk of bias; populations mainly ethnic minorities in Western countries; not tested in Asian or GCC health systems. 3. Mobile phone / app → HbA1c [13–15] RCTs (high) −1 0 0 0 Undetected None Moderate Unblinded trials with short follow-up; consistent direction of effect. 4. Telemedicine → HbA1c [18] RCTs (high) 0 −1 0 0 Undetected None Moderate Heterogeneous interventions and effect sizes. 5. SMBG (non-insulin T2DM) → HbA1c [41] RCTs (high) −1 0 0 −1 Undetected None Low Small effect not sustained beyond 6 months; clinical relevance uncertain. 6. CGM (T1DM, injections) → HbA1c [65] RCT (high) 0 n/a 0* −1 n/a None Moderate Single trial; *indirect for T2DM and non-Western settings. 7. Low-GI diet → HbA1c [42] RCTs (high) −1 0 0 −1 Undetected None Low Small, short trials. 8. Structured exercise → glycaemic control [9,64] Non-randomised controlled trial (low) 0 n/a 0 0 n/a None Low One 16-week controlled trial (n = 49) with non-exercising controls; direction of effect consistent with the ADA position statement. 9. Foot-care education → ulcer/amputation [43,47] Mixed −1 0 −1 −1 Undetected None Very low Limited trial evidence for education alone; observational data on glycaemic control only. 10. Pre-Ramadan education → adverse events [48,54] Observational (low) −1 0 0 0 Undetected None Very low Uncontrolled or unadjusted observational studies. 11. Cross-regional differences in practice/adherence/distress [10,49–51,55,57–59,61,62,66,67] Observational (low) 0 −1 −1 0 Undetected None Very low Indirect comparisons across different instruments, populations and years; only DAWN offers a direct comparison. Ratings follow GRADE [28] and, where no single pooled estimate was available, Murad et al. [29]. "0" = not serious; "−1" = serious; "−2" = very serious   Appendix S6: Sources of Numerical Estimates in Table 2 and Values Removed Table 2 row Value Source Population and year Status Prevalence – MENA 18.1% IDF Diabetes Atlas 10th ed. [1] Adults 20–79 y; age-adjusted comparative prevalence; 2021; MENA region incl. non-GCC countries Replaces "~17.3%" (source not traceable) Prevalence – South-East Asia 10.0% [1] As above; IDF SEA region (incl. India, Bangladesh, Sri Lanka) Replaces "~11.9%" Prevalence – Western Pacific 9.9% [1] As above; IDF WP region (incl. China, Japan, Korea, Vietnam, Australia) New Prevalence – Europe 7.0% [1] As above; IDF Europe region Replaces "~10.8%" Prevalence – North America & Caribbean 11.9% [1] As above; IDF NAC region New Adherence – global 54% (95% CI 51–58%) Piragine 2023 [17] 156 studies, 2011–2022; mixed measures Replaces regional ranges Adherence – Palestine 57.3% adherent Sweileh 2014 [49] n = 405; MMAS-8; 2012 data collection New Adherence – Egypt 30.5% high; 51.3% medium Azzam 2021 [50] n = 302; primary care, Mansoura New Foot care – Egypt 75.2% unsatisfactory Abo-Eata 2022 [55] n = 230 older adults; Tanta New Adherence ranges 52–68% (Asia); 42–61% (GCC); 65–78% (Western) Previously attributed to [6,7,11,17,49,52] — Removed: cited sources do not report these ranges Amputation rates 3.1–6.4; 4.5–8.1; 1.9–3.8 per 1,000 Previously attributed to [30,47,50,57,66] — Removed: not reported in cited sources; no defined population or period Regional GRADE labels Moderate / Low-moderate / High — — Removed: certainty now rated per outcome (Table 4) Other numerical statements corrected in the text: Norris DSME effect (0.74% → 0.76 pp, 95% CI 0.34–1.18, attributed to Norris 2002 [8]); mHealth pooled effect (0.49 pp attributed to Hou 2016 [14], not Wang 2017); wearable trackers (1,200 steps/24 RCTs → ≈627 steps/28 RCTs [36], general population); Da Qing (33% → 31%, 46% and 42% by arm [63]); UKPDS (25% microvascular risk reduction with intensive therapy [4], not "per 1% HbA1c"). Unverifiable figures were removed: Hong Kong adherence 64.3%; WHO STEPwise inactivity >70%; household medication spending 20–30%; SAUDI-DM comorbidity 39.8%.   Appendix S7: Reference Audit Log Ref / Location Issue Identified Action Taken [3] ADA Standards 2024 Incorrect/incomplete DOI "10.2337/dc24-Sint"; outdated title ("Standards of Medical Care") Corrected to "Standards of Care in Diabetes—2024", Diabetes Care 2024;47(Suppl 1):S1–S321, with supplement URL and Introduction DOI 10.2337/dc24-SINT Original refs 8–9 (Puzhakkal 2025; Sweileh 2014) Line break in the DOI of the first entry made it run into the second Reformatted as separate entries [45] "Supple 1"; cited in text for IDF statistics it does not contain Corrected to "Suppl 1"; IDF statistics now cited to [1] [55], [50], [57], [32], [43] Non-Vancouver formatting; garbled author list in [43] (e.g., "Library W.M., Core D.M., Therapy E."); wrong journal ("HHS Public Access") Reformatted; [43] corrected to J Diabetes Complications 2020;34(10):107638 [7] Pillay 2015 Cited as Cochrane review CD010967 Corrected to Ann Intern Med 2015;163(11):848–60 (132 RCTs; verified against the published record) [46] "ofthe"; "1 Clin Med" Corrected to "of the"; "J Clin Med" [19], [22], [11] Published 2025, after search cut-off Reclassified as contextual references (Methods 2.3; Appendix S4 Part B) Old Table 2 Cited works absent from reference list or with wrong details (Norris 2002; Liang Diabetes Care 2011; Schwingshackl Nutrients 2017; Bus 2016 pp.136–61; Sturt 2010; Clair BMJ 2015; Turner 1998; Tchero JMIR 2019) Table restructured; citations now point to numbered reference-list entries; Norris 2002 and UKPDS 38 added [5,8] Text: "Al-Jaghbeer and colleagues" [11] Name does not match [11] Sentence removed; [11] described as a post-cut-off narrative review Text: "Al Mansouri and colleagues" [61]; "Ebrahim and colleagues" [67] [61] and [67] are diabetes-distress studies Attributed to Ahmed 2020 [48] and IDF-DAR 2021 [54] Text: "Al Hayek and colleagues" [66] [66] is an Iranian distress study Sentence removed Text: Da Qing [32]; UKPDS [12,43,59]; "Wang 0.49%" [15,38]; DSME [18,35,41,47,65]; PA [11,14,22,55]; barriers [70]; digital [12,69] Citation does not support the statement Re-cited to [63]; [4,5,60]; [14]; [2,6–8]; [9]; [10]; [13–15] Text: NICE NG28 [59] [59] is a Korean scale study NICE NG28 added as [56] Text: "[,20]" Typographical error Corrected to [12] Text: "Table 6 summarises…" No Table 6 for this content Corrected to Table 3 All in-text citations Checked against reference list for accuracy, relevance, date and support of claim Unsupported claims removed or reworded; descriptive context marked § in tables All references have been renumbered consecutively in order of first citation in the revised manuscript (1–70). Bracketed numbers in this log refer to the renumbered reference list