Potential Overtreatment and Undertreatment of Type 2 Diabetes Mellitus in Long-Term Care Facilities: A Systematic Review

Objective: To investigate the prevalence, outcomes, and factors associated with potential glycemic overtreatment and undertreatment of type 2 diabetes mellitus (T2DM) in long-term care facilities (LTCFs). Design: Systematic review. Setting and Participants: Residents with T2DM and aged ≥60 years living in LTCFs. Measures: Articles published between January 2000 and September 2020 were retrieved following a systematic search of MEDLINE, EMBASE, Cochrane Library, CINAHL plus, and gray literature. Inclusion criteria were the reporting of (1) potential overtreatment and undertreatment quantitatively defined (implicitly or explicitly) based on hemoglobin A1c (HbA1c) and/or blood glucose; (2) prevalence, outcomes, and associated factors of potential glycemic overtreatment and undertreatment; and (3) the study involved residents of LTCFs. Results: Fifteen studies were included. Prevalence of potential overtreatment (5%–86%, n = 15 studies) and undertreatment (1.4%–35%, n = 8 studies) varied widely among facilities and geographical locations, and according to definitions used. Prevalence of potential overtreatment was 16%–74% when defined as treatment with a glucose-lowering medication in a resident with ≥1 hypoglycemia risk factor or serious comorbidity, together with a HbA1c <7% (n = 10 studies). Potential undertreatment was commonly defined as residents on glucose-lowering medication having HbA1c >8.5% and the prevalence 1.4%–14.8% (n = 6 studies). No studies prospectively measured resident health outcomes from overtreatment and undertreatment. Potential overtreatment was positively associated with use of oral glucose-lowering medications, dementia diagnosis or dementia severity, and/or need for assistance with activities of daily living (n = 2 studies). Negative association was found between potential overtreatment and use of insulin/combined insulin and oral glucose-lowering medication. No studies reported factors associated with potential undertreatment. Conclusions and Implications: The prevalence of potential glycemic overtreatment and undertreatment varied widely among residents with T2DM depending on the definition(s) used in each study. Longitudinal studies examining associations between glycemic management and health outcomes, and the use of consensus definitions of overtreatment and undertreatment are required to establish findings about actual glycemic overtreatment and undertreatment in LTCFs.