Presentation (CTI)

OBJECTIVES: Individualized glycemic targets are recommended for older adults with diabetes, but implementation remains challenging. The 2017 Japanese guidelines promoted individualized HbA1c targets based on cognitive and physical function, comorbidities, and use of drugs with a risk of severe hypoglycemia. This study evaluated whether guideline publication was associated with changes in outcomes, clinical practice, and healthcare expenditures.
METHODS: An interrupted time-series analysis was conducted using Japan’s National Database of Health Insurance Claims. The study included patients aged ≥65 years with diabetes. The outcomes were the monthly incidence of severe hypoglycemia, diabetic ketoacidosis (DKA), and hyperosmolar hyperglycemic state (HHS), as well as antidiabetic prescribing and healthcare expenditures. The intervention point was June 2017, with 14 pre-intervention months and 25 post-intervention months. Level and trend changes were estimated using log-link generalized linear models with quasi-Poisson or negative binomial models, adjusting for seasonality with Fourier terms and including the monthly number of patients as an offset.
RESULTS: The number of patients ranged from 413,322 in April 2016 to 552,776 in June 2019. Guideline publication was not associated with statistically significant immediate level or trend changes in the incidence of severe hypoglycemia. Monthly trends in the incidence of DKA and HHS differed from pre-guideline trends, with incidence rate ratios (IRRs) of 1.01. Healthcare expenditures showed no statistically significant changes. Antidiabetic prescribing showed both level and trend changes after guideline publication. In trend-change estimates, prescribing trends decreased for alpha-glucosidase inhibitors and SGLT2 [IRRs: 0.99 (95% CI: 0.99-0.992) and 0.98 (95% CI: 0.976-0.98)] and increased for insulin preparations [IRRs: 1.01 (95% CI: 1.00-1.01)].
CONCLUSIONS: Guideline publication was associated with changes in antidiabetic prescribing patterns among older adults, although no significant changes were observed in severe hypoglycemia or health expenditures. Ongoing monitoring is warranted to determine whether these practice changes lead to long-term improvements in patient outcomes.