Deintensification in older patients with type 2 diabetes: a systematic review of approaches, rates and outcomes.

BACKGROUND: Guideline bodies recommend less strict glycaemic targets in older people with diabetes. It is uncertain whether the benefits of deintensification or de-prescribing, commonly employed by clinicians to achieve the less strict targets, outweighs the harms in these patients. We conducted a systematic review of published evidence, to assess deintensification approaches and rates and evaluate the harms and benefits of deintensification with antidiabetic medication and other therapies amongst older people (≥ 65 years) with type 2 diabetes with or without cardiometabolic conditions. METHODS: We identified relevant studies in a literature search of MEDLINE, Embase, Web of Science, and Cochrane databases to 30 October 2018. Data was extracted on baseline characteristics, details on deintensification, and outcomes and was synthesized using a narrative approach. RESULTS: Ten studies (observational cohorts and interventional studies) with data on 26,558 patients with comorbidities were eligible. Deintensification approaches included complete withdrawal, discontinuation, reducing dosage, conversion, or substitution of at least one medication, but majority of studies were based on complete withdrawal or discontinuation of antihyperglycaemic medication. Rates of deintensification approaches ranged from 13.4% to 75%. Majority of studies reported no deterioration in HbA1c levels, hypoglycaemic episodes falls or hospitalisation on deintensification. On adverse events and mortality, no significant differences were observed between the comparison groups in the majority of studies. CONCLUSION: Available but limited evidence suggests that the benefits of deintensification outweighs the harms in older people with type 2 diabetes with or without comorbidities. Given the heterogeneity of patients with diabetes, further research is warranted on which deintensification approaches are appropriate and beneficial for each specific patient population.