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

项目总结/摘要 美国印第安人（AI）在任何种族或族裔群体中的2型糖尿病（T2 D）患病率最高， 患有肥胖症、心血管疾病（CVD）和慢性肾脏疾病等合并症的比例很高 病（CKD）。不受控制的心脏代谢危险因素-胰岛素抵抗导致血糖受损 耐受性、血脂异常和高血压（HTN）--增加死亡风险。死亡率显著降低， 降糖降脂和抗高血压药物依从性。药物依从性低， 居住在非印度卫生服务（IHS）医疗机构的AI。事实上，我们对自然界的 以及主要接受药物治疗的保留地居住的AI的药物依从性程度 无需IHS/部落设施的费用。电子健康记录（EHR）提供了丰富但未充分利用的数据源 关于药物依从性及其预测代谢控制指标（C-MCI）的潜力， 如HbA 1c、LDL-C（低密度脂蛋白）、SBP（收缩压）。在乔托的支持下 国家的俄克拉荷马州（CNO），我们将解决这一疏忽，通过使用EHR数据产生的这个大，国家的- 最先进的部落医疗系统来调查C-MCI。我们的R 01应用程序的目标是表征 药物依从性（降压药、降糖药和降脂药）与C- MCI（HbA 1c ≤ 7%，LDL-C <100 mg/dL，SBP <130 mm Hg），患者人口统计学（例如，年龄，性别，SDOH， 居住位置）和共病（例如，CVD、BMI>30、CKD）以及各C-MCI之间的关系 根据部落EHR（2018-2021年）中5，970名CNO的患者人口统计资料和合并症 T2 D患者。利用机器学习技术，我们将开发模型来预测未来 （2019-2021年）C-MCI基于前一年的药物依从性，患者人口统计学，共 发病率，和常见的实验室（例如，脂质组）。最后，关键的线人访谈将探讨促进者， 以及在当地社会健康决定因素（SDOH）背景下的药物依从性障碍， 在EHR中不可用。我们的具体目标是：（1）确定（a） 药物依从性和C-MCI、人口统计学和合并症;（B）每种C-MCI和人口统计学，以及 共病;（2）开发机器学习模型（例如，随机森林，最近邻，其他） 预测未来（2019-2021年）C-MCI从前一年的药物依从性，人口统计学，共 发病率和常见的实验室;（3）确定促进者和障碍，以药物依从性内 SDOH、EHR衍生的药物依从性（PDC）和C-MCI（达到目标、高于目标和 HbA 1c不受控制）。我们将与CNO领导人和其他利益相关者分享我们的发现，他们将指导 将结果转化为评价T2 D管理和并发症预防的建议 程序.我们的研究结果将产生见解，以改善AI中的药物依从性和C-MCI， CNO的国家健康报告2017年的目标是减少T2 D及其并发症。

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.