Campus Alert: Find the latest UMMS campus news and resources at umassmed.edu/coronavirus

Search Close Search
Search Close Search
Page Menu

Anthony Nunes PhD received funding for an NIH R21 entitled “Diabetes Treatments and Hypoglycemic-Related Adverse Events in Nursing Home Residents”

Assistant Professor Anthony Nunes PhD received funding for an NIH R21 entitled “Diabetes Treatments and Hypoglycemic-Related Adverse Events in Nursing Home Residents”

National, contemporary, longitudinal studies of the pharmacologic management of diabetes mellitus in US nursing homes do not exist, with the most recent national data on use of glucose lowering medications in nursing homes being nearly a decade old. As reported in the American Diabetes Association’s consensus report on diabetes in older adults, “There are essentially no directly applicable clinical trial data on glucose control for large segments of the older diabetic patient population.” This dearth of observational or clinical evidence, consistent across disease areas, has been coined the geriatric pharmacoparadox because older adults are in most need of an evidence base for medication decisions, but are rarely the target of evidence generation. The need to foster diabetes research in older adults was highlighted by the Diabetes Mellitus Interagency Coordinating Committee (DMICC) and described in the NIDDK Recent Advances & Emerging Opportunities (2019). Specifically, the DMICC recommendations include studies assessing the impact of cognitive and functional impairments, multimorbidity, polypharmacy, and risk of hypoglycemia-related outcomes in long-term care settings.

These priorities were central to our proposed scope of research. Consistent with the NIH Diabetes in America report, the American Diabetes Association recommends tailoring goals and treatment approaches based on a careful evaluation of comorbidities and overall health. These recommendations support the trend towards personalized medicine. No studies have assessed the feasibility or effect of implementing personalized antidiabetic treatments in the nursing home population.

Using a national, contemporaneous data source from the Centers for Medicare and Medicaid Services allows us to study virtually all US nursing home residents with type 2 diabetes (T2DM). Specifically, this study employs linked Minimum Dataset (MDS) 3.0 assessments and Medicare Part A and D claims (2011-2016) to examine current antidiabetic treatment practices in nursing home resident from 2011 to 2016

The specific aims of this study are to:
1. identify and characterize resident phenotypes and antidiabetic treatment practices among nursing home residents
2. Identify and describe clinically relevant phenotypes of nursing home residents with diabetes
3. Quantify the risk of hospitalized outcomes (overall, hypoglycemia-specific, and hypoglycemia-related)
associated with treatment groupings and evaluate effect measure modification by resident phenotypes

Extensive descriptive analyses will guide the implementation of statistical models (including the appropriate parameterization of covariates), assist in the checking of model assumptions, and
inform the interpretation of model results. The analytic approach includes the use of robust techniques such as machine learning (Aim 2), and propensity scores (Aim 3).

We hypothesize that:
1) antidiabetic treatment practices will be heterogenous and consist of higher risk regimens (e.g., sulfonylureas, glyburide) and lower risk regimens (e.g., DPP-4i, SGLT2i);
2) discrete phenotypes will be identifiable and representative of comorbid complexity, polypharmacy, prognosis, cognitive impairment, and functional limitations; and 3) clinically meaningful effect heterogeneity will be observed between phenotypes when quantifying the association between antidiabetic treatment practices and risks of hospitalization.