The research team set out to address a practical gap in diabetes care. Existing questionnaires tend to measure self-care as a broad construct, aggregating many activities into a single score, and subjective judgements about personal glucose levels have been shown to predict distress more strongly than objective continuous glucose monitoring data. Clinicians, meanwhile, face time-pressured consultations packed with glycaemic metrics but often lack a rapid way to pinpoint which concrete behaviours might be driving a patient’s results. To fill that gap, the investigators used the Behavioural Diabetes Care Record, or BDCR, a questionnaire of 18 items that can be completed in under five minutes, with separate versions tailored to multiple daily injections and continuous subcutaneous insulin infusion pump therapy.

Most BDCR items ask respondents to give a number describing how often a specific behaviour occurred in the preceding 14 days. Questions include how many times per day blood glucose was tested, on how many days a main meal was eaten without administering insulin, and on how many days hypoglycaemia treatment food was carried for most of the day. Pump users are asked about cannula and infusion line changes, while injection users are asked about missed doses. The record also asks individuals to rate, on a scale of one to ten, how well they are achieving their personal diabetes self-care goals and how willing they are to make changes. Previous work has shown the measure responds to behavioural intervention, with medium to large effect sizes, and that improvements in individual behaviours correspond with improvements in HbA1c and distress.

The cross-sectional study recruited 105 adults aged 18 to 40 who had lived with type 1 diabetes for at least six months and were independently managing their condition, drawing participants from consecutive outpatient appointments over eight months. The cohort, 53 percent male with an average age of 27 and average diabetes duration of 12.6 years, had a mean HbA1c of 8.7 percent. Two-thirds used multiple daily injections and about a third used insulin pumps, while 29 percent used some form of continuous glucose monitoring. Alongside the BDCR, participants completed the 17-item Diabetes Distress Scale, the Acceptance and Action Questionnaire-II, reverse scored to index psychological flexibility, and a short form of the Behavioural Activation for Depression Scale. Point-of-care HbA1c values were taken from the medical record.

The behavioural frequencies told a clear story. Participants with HbA1c below 8 percent checked their glucose a median of four times per day, compared with three times among those at or above 8 percent, a statistically significant difference. Days on which prandial insulin was missed were also associated with glycaemic outcome, with a median of zero missed main-meal doses in the better-controlled group versus two in the higher HbA1c group. Among injection users, more frequent insulin administration, around four times per day, was associated with lower HbA1c. Those with better control also detected or tested for hypoglycaemia more often, a median of four times versus two, and were more likely to carry fast-acting treatment food throughout the day.

Distress painted its own distinct picture. Clinically significant diabetes distress, defined as a mean Diabetes Distress Scale score of three or above, was associated with more days on which insulin was skipped for main meals and for snacks, and with less frequent glucose checking before meals. Notably, distress and HbA1c did not move entirely in lockstep: some behaviours separated the groups on one measure but not the other, reinforcing the idea that emotional burden and glycaemic control are related but separable entities. Perceived achievement of personal self-care goals and psychological flexibility were both significantly associated with lower HbA1c and lower distress, while behavioural activation alone was not.

To see whether these signals clustered into recognisable profiles, the researchers applied latent class cluster analysis, guided by statistical fit criteria and input from consumers with lived experience of type 1 diabetes and clinicians from medical, nursing and nutrition disciplines. Three subgroups emerged. Half of participants fell into the first cluster, characterised by near-daily glucose checking, checking before insulin doses, few missed mealtime boluses, ready availability of hypoglycaemia treatment, higher psychological flexibility and strong perceived goal achievement. Only 30 percent of this group had HbA1c at or above 8 percent, and just 4 percent reported clinically significant distress.

The second cluster, about a third of the sample, showed moderate engagement: three-quarters checked glucose daily, but 72 percent had HbA1c at or above 8 percent and roughly 29 percent reported significant distress. The third cluster, nearly 18 percent of participants, stood out as clearly at risk. Ninety-nine percent had HbA1c at or above 8 percent and 72 percent reported clinically significant distress. They checked glucose on only about a third of days, missed main-meal insulin on a mean of nearly six days out of 14, checked before meals only about a quarter of the time, carried hypoglycaemia treatment least often, and recorded the lowest psychological flexibility and goal achievement scores. The authors note that this kind of stratification could allow clinicians to direct attention and resources to those most in need, particularly when glucose levels are suboptimal or distress is evident.

The findings carry implications for how routine consultations are structured. Because the BDCR items are observable and measurable, they can be repeated over time to track behavioural change and response to intervention, and triangulating them with HbA1c and continuous glucose monitoring data may offset some limitations of self-report, such as recall inaccuracy and social desirability bias. The authors suggest the instrument could guide personalised conversations, tailored goals, decisions about diabetes technology, or timely referral to dietitians and clinical psychologists. They also caution that the cross-sectional design limits causal inference and that findings from a tertiary-centre cohort may not generalise to wider populations, including people seen by telehealth or those not attending scheduled appointments.

Even in an era of continuous glucose monitoring and hybrid closed-loop systems, the behaviours highlighted here remain pillars of standard care: checking glucose or a trend arrow before dosing, announcing and administering mealtime insulin, and keeping hypoglycaemia treatment within reach. By showing that a five-minute set of plain questions can flag both metabolic risk and emotional strain, and by linking those patterns to psychological flexibility, the study offers a low-cost foundation for more person-centred, goal-oriented diabetes consultations. The team calls for prospective research to test whether routinely collecting these behavioural data genuinely improves outcomes for adults living with type 1 diabetes.

Subject of Research: Associations between type 1 diabetes self-care behaviours, glycaemic control, diabetes distress and psychological flexibility in adults

Article Title: Frequency of Key Type 1 Diabetes Self‐Care Behaviours Are Associated With Psychological Flexibility, Distress and Glycaemic Outcome

Article References: Nicholas, J. A., Yeap, B. B., Cross, D., Bear, N., & Burkhardt, M. S. (2026). Frequency of Key Type 1 Diabetes Self‐Care Behaviours Are Associated With Psychological Flexibility, Distress and Glycaemic Outcome. Endocrinology, Diabetes & Metabolism, 9(5), Article e70336. https://doi.org/10.1002/edm2.70336

Image Credits: AI Generated

DOI: 10.1002/edm2.70336

Keywords: type 1 diabetes, diabetes self-care, HbA1c, diabetes distress, psychological flexibility, Behavioural Diabetes Care Record, insulin regimen, glucose monitoring, cluster analysis, hypoglycaemia, behavioural intervention, Frequency