Diabetes beliefs, perceived risk and health behaviours: an embedded mixed-methods analysis from the Richmond Stress and Sugar Study

Abstract

Introduction: Efforts to increase engagement with diabetes prevention programmes largely focus on increasing diabetes awareness, with the logic that risk knowledge will motivate behaviour change. However, the salience of perceived risk as it relates to diabetes prevention is contested. The goal of this cross-sectional, embedded mixed-methods study was to examine the relationships between perceived risk, diabetes beliefs and prevention behaviours among adults at elevated risk of type 2 diabetes.

Methods: Data come from the Richmond Stress and Sugar Study (n=125). Diabetes beliefs (ie, internal, chance, powerful others) were assessed using the Multidimensional Health Locus of Control. Preventive behaviours (eg, changing diet, exercise, tobacco, alcohol) were measured by self-report. Perceived risk of developing diabetes was measured using a probability scale (0%-100%). Logistic and Poisson regression models quantified the relationships between beliefs, behaviours and perceived risk. Qualitative themes regarding challenges and facilitators to preventive behaviours were abstracted from open-ended questions and summarised using content analysis.

Results: Perceived risk of developing diabetes was low (median: 35% likelihood) and only 10% of participants had ever attended a prevention class. None of the diabetes belief domains were significantly associated with either engagement in preventive behaviours or perceived diabetes risk. Perceived diabetes risk was not associated with engagement in preventive behaviours; however, having a family history of diabetes was strongly related to perceived risk (OR: 3.35, 95% CI: 1.42 to 7.86). Qualitative facilitators and barriers of preventive behaviours reflected psychosocial factors (eg, determination, stress, preferences) and resources (eg, social support, time, overall health).

Conclusions: Perceptions of risk and health beliefs are not correlated with engagement in preventive behaviours among adults at clinically elevated risk of diabetes. Awareness campaigns may benefit from incorporating family health history information. Diabetes prevention programmes should address barriers beyond health beliefs to promote engagement in behaviour change.

Keywords: Behavior; Diabetes Mellitus, Type 2; MENTAL HEALTH.

Figure 1. Association between domains of diabetes beliefs predicting perceptions of diabetes risk. Relative OR (unadjusted and adjusted) of the relationship between the three diabetes belief domains (internal, chance and powerful others) predicting perceived risk of developing diabetes (low, medium and high). See online supplemental table 3 for details of the regression models.

Figure 2. Associations between diabetes beliefs (Panel A) and perceived diabetes risk (Panel B) with engagement in diabetes preventive behaviours. Panel A: Incidence rate ratios (IRRs) and 95% CIs of the relationship between diabetes belief domains (internal, chance and powerful others) with number of diabetes preventive behaviours (count: 0–5). Panel B: IRR and 95% CIs of the relationship between perceived risk of developing diabetes (high vs medium; low vs medium) with number of diabetes preventive behaviours (count: 0–5). See online supplemental tables 4 and 5 for the regression models that generated these point estimates.