Medication Adherence and Cardio-Metabolic Control Indicators among Adult American Indians Receiving Tribal Health Services

点击翻译按钮获取中文摘要

Project Summary/Abstract American Indians (AIs) have the highest prevalence of type 2 diabetes (T2D) of any racial or ethnic group and experience high rates of co-morbidities such as obesity, cardiovascular disease (CVD), and chronic kidney disease (CKD). Uncontrolled cardio-metabolic risk factors--insulin resistance resulting in impaired glucose tolerance, dyslipidemia, and hypertension (HTN)--increase mortality risk. Mortality is significantly reduced by glucose- and lipid-lowering, and antihypertensive medication adherence. Medication adherence is low among AIs living in non-Indian Health Services (IHS) healthcare settings. Virtually nothing is known about the nature and extent of medication adherence among reservation-dwelling AIs who primarily receive their medications without cost from IHS/tribal facilities. Electronic health records (EHR) offer a rich but underutilized data source about medication adherence and its potential to predict Cardio-Metabolic Control Indicators (C-MCI) such as HbA1c, LDL-C (Low Density Lipoprotein), SBP (Systolic Blood Pressure). With the support of Choctaw Nation of Oklahoma (CNO), we will address this oversight by using EHR data generated by this large, state-of- the-art tribal healthcare system to investigate C-MCI. The objective of our R01 application is to characterize the relationships among medication adherence (antihypertensive, glucose- and lipid-lowering drugs) and C- MCI (HbA1c ≤7%, LDL-C <100 mg/dL, and SBP <130 mm Hg), patient demographics (e.g., age, sex, SDOH, residence location) and co-morbidities (e.g., CVD, BMI>30, CKD) as well as the relationship of each C-MCI with patient demographics and co-morbidities from the tribe's EHR (2018-2021) for the 5,970 CNO patients who have T2D. Employing machine learning techniques, we will develop models to predict future (2019-2021) C-MCI based on the previous year medication adherence, patient demographics, co- morbidities, and common labs (e.g., lipid panel). Lastly, key informant interviews will explore facilitators of and barriers to medication adherence within the context of local social determinants of health (SDOH) that are not available in the EHR. Our specific aims are to: (1) Determine the bivariate relationships between (a) medication adherence and C-MCIs, demographics, and co-morbidities; (b) each C-MCI and demographics and co-morbidities; (2) Develop machine-learning models (e.g., random forest, nearest neighbors, others) for predicting future (2019-2021) C-MCI from the previous year medication adherence, demographics, co- morbidities, and common labs; and (3) Identify facilitators of and barriers to medication adherence within the context of SDOH, EHR-derived medication adherence (PDC) and C-MCI (at target, above target, and for HbA1c uncontrolled). We will share our findings with CNO leaders and other stakeholders, who will guide the translation of the results into recommendations for evaluating T2D management and complication prevention programs. Our findings will yield insights to improve medication adherence and C-MCI among AIs, consistent with CNO's State of the Nation's Health Report 2017 goal of reducing T2D and its complications.