Endocrinology · Medical Literature
Latent profiles of self-management behaviours in patients with type 2 diabetes and their associations with metabolic outcomes in China: a multicentre cross-sectional study
Abstract
Background Self-management behaviours are crucial for glycaemic control in type 2 diabetes mellitus (T2DM) but adherence is often poor. Most studies treat diabetes self-management (DSM) behaviours as a single construct, ignoring variations in behavioural patterns and their metabolic impacts. Objectives This study aims to identify distinct DSM behaviour profiles in T2DM patients, explore associated factors, and compare their metabolic outcomes. Design A multicentre cross-sectional study. Setting The study was conducted from March to September 2024 at four tertiary care hospitals. Participants A total of 304 patients diagnosed with T2DM were included. Primary outcomes and measures The primary outcomes were DSM behaviours, self-efficacy, habit strength and metabolic indicators (glycated haemoglobin (HbA1c), fasting blood glucose, lipid profile, blood pressure and body mass index). Latent profile analysis was used to identify distinct DSM behaviour profiles. Univariate analyses (² test or Fisher’s exact test, as appropriate) were performed to identify factors associated with profile membership, followed by multinomial logistic regression analysis. Analysis of variance with post hoc tests was used to compare metabolic indicators across profiles. Results Three distinct profiles were identified: low-engagement (14.0%), moderate-engagement (12.0%) and high-engagement behaviours (74.0%). These profiles were significantly associated with self-efficacy, personal monthly income, medication duration and diabetes-related complications (all p Conclusions Distinct DSM behaviour profiles exist among patients with T2DM and are influenced by psychosocial and clinical factors. The high-engagement behaviour profile was associated with more favourable glycaemic and lipid control, whereas the low- and moderate-engagement behaviour profiles were linked to poorer metabolic outcomes. These findings highlight the need for tailored interventions targeting behaviour-specific p
Clinical relevance
Type 2 diabetes mellitus (T2DM) is a high-burden chronic disease in which self-management behaviour (DSM) is central to glycaemic and metabolic control, yet adherence remains poor and heterogeneous across patients. Prior research has treated DSM as a unidimensional construct, obscuring clinically meaningful subgroups with distinct behavioural patterns and differential metabolic risk. This multicentre Chinese study addresses that gap by applying latent profile analysis to characterise discrete DSM behaviour phenotypes, offering a more precise framework for risk stratification and targeted intervention in a population where T2DM prevalence is rapidly rising.
Key findings
1. Three distinct DSM behaviour profiles were identified among 304 patients with T2DM: low-engagement (14.0% of the cohort), moderate-engagement (12.0%), and high-engagement (74.0%), demonstrating that a substantial minority of patients cluster into suboptimal self-management phenotypes. 2. Profile membership was significantly associated with self-efficacy, personal monthly income, medication duration, and presence of diabetes-related complications (all p-values statistically significant as reported; exact p-values were truncated in the available abstract). 3. The high-engagement profile was associated with more favourable glycaemic outcomes, including lower HbA1c and fasting blood glucose, compared to both the low- and moderate-engagement profiles (specific between-group effect sizes and CIs were not reported in the available abstract). 4. Lipid profile outcomes were also more favourable in the high-engagement profile relative to the low- and moderate-engagement groups, highlighting a multi-domain metabolic benefit of comprehensive DSM behaviour adoption. 5. Psychosocial determinants — specifically self-efficacy and habit strength — were measured alongside metabolic indicators, implicating both behavioural psychology constructs and socioeconomic factors as modifiable drivers of profile membership. 6. Multinomial logistic regression confirmed that profile membership was independently associated with the identified factors after adjustment, though precise odds ratios and confidence intervals were not available in the provided abstract. NOTE: Specific quantitative effect sizes, confidence intervals, p-values, and NNT/NNH could not be included as these were truncated in the available abstract; findings above reflect the reported directional results only.
Practice implications
Clinicians should move beyond a one-size-fits-all approach to diabetes self-management education and instead screen for engagement profile characteristics — particularly low self-efficacy, lower socioeconomic status, longer medication duration, and presence of complications — to identify the approximately 26% of T2DM patients falling into low- or moderate-engagement profiles who are at highest risk for suboptimal glycaemic and lipid outcomes. Targeted, behaviour-specific interventions that address self-efficacy enhancement (e.g., structured problem-solving, motivational interviewing) and habit formation should be prioritised for these higher-risk subgroups rather than applied uniformly. Given that income and complication burden also predicted profile membership, multidisciplinary care models incorporating social work support and complication-focused counselling may be necessary to meaningfully shift patients from low- to high-engagement behavioural profiles.