Background Hypoglycaemia is a common complication of insulin therapy. The relationship between hypoglycaemia and mood is not well understood, with conflicting evidence reported in the literature (1,2). Potential explanations for these conflicting data include the variation in hypoglycaemia definitions and thresholds. The strongest associations between hypoglycaemia and mental health appear to relate to severe hypoglycaemia (SH) (3), with impaired awareness of hypoglycaemia (IAH) being a major risk factor (4). Using the dataset with which we previously demonstrated an association between IAH and a mental health burden (5), this study provides an opportunity to explore whether this link operates through increased exposure to hypoglycaemia, using different common definitions of hypoglycaemia. Methods Hypo-METRICS was a prospective, multi-country observational study of the hypoglycaemia experience. Adults with type 1 diabetes (T1D) or insulin-treated type 2 diabetes (T2D) were eligible for inclusion. Participants wore blinded study continuous glucose monitors (CGMs), continued with their usual method of glucose monitoring (CGM or capillary blood glucose CBG), and completed ecological momentary assessment (EMA) through a bespoke app for 10 weeks in real-time (6). Participants completed patient-reported outcome measures (PROMs) at the beginning and end of the study to assess for depression (PHQ-9), anxiety (GAD-7), diabetes distress (PAID) and fear of hypoglycaemia (HFS-II worry) (Electronic Supplementary Material ESM Table 1). Hypoglycaemia was defined as time below range (TBR) 15 minutes, resolving after >15 minutes above the threshold (7). PRH was subdivided as symptomatic (resolved by carbohydrate ingestion), technology-detected (glucose ≤3.9mmol/l on routine monitoring), and total PRH (combined symptomatic and technology-detected episodes) (6). In this post-hoc exploratory analysis, Chi-square, Fisher’s exact, and Mann-Whitney U tests were used to assess for differences in baseline characteristics and PROMs. Unadjusted and adjusted linear regression analyses with the Benjamini-Hochberg procedure were used to examine whether hypoglycaemia was associated with mental health scores after 10 weeks. Age, sex, ethnicity, diabetes duration, education, employment, CGM use and hypoglycaemia awareness status were variables in the adjusted analysis. Results In total, 232 participants with T1D and 310 with insulin-treated T2D consented for inclusion in post-hoc analyses. The baseline characteristics and prospectively-measured hypoglycaemia metrics are presented in Table 1. There were subtle but statistically significant differences in PROMs from week 0 to week 10 (ESM Table 2). There were no significant associations between any hypoglycaemia metric (TBR 0.05) (Figure 1; ESM Tables 3-4). Conclusions In summary, we report no evidence of a significant association between the experience of SDH and PRH documented prospectively over 10 weeks and depression, anxiety, diabetes distress and fear of hypoglycaemia scores at study end in Hypo-METRICS participants with T1D or insulin-treated T2D.
Evans et al. (Tue,) studied this question.