This application addresses broad Challenge Area (01) Behavior, Behavioral Change, and Prevention and specific Challenge Topic, 01-DK-103 Improved understanding of behavioral and social factors related to non- Adherence in people with diabetes. We propose to study patterns and predictors of diabetes patients'failure to utilize referred health services (e.g., standard lab tests, specialty visits, health education). These preventive health services are particularly important in the care of diabetes given the disease complexity, need for continual monitoring, and frequent intensification. This project will inform policy in two areas of importance to the NIH: (1) how to address poor adherence among patients with diabetes and (2) how to reduce health disparities. Findings will help us better understand the potentially high-cost patients who do not adhere to their diabetes treatment plan despite full access to integrated, pre-paid health care and allow identification of barriers to care. The project uses data from an NIH funded study, The Diabetes Study of Northern California (DISTANCE) plus new data captured from the Kaiser Permanente electronic health record (EHR). The study has immediate and longer-term public health implications, given that repeated non-utilization (""""""""persistent non-utilization"""""""") may adversely affect continuity of care, and increase the risk for serious and costly events. This study is feasible within the two-year time frame of the challenge grant award. The investigator team has demonstrated productivity, combines expertise in adherence and diabetes health services research, and is highly experienced in acquiring and analyzing the data involved. ABSTRACT Poor adherence to a medical treatment plan is a serious public health problem in diabetes. While some aspects of adherence, particularly adherence to medications, have been studied extensively, much less is known about adherence to (utilization of) referrals for health services (e.g., standard lab tests, specialty visits, health education). These health services are vital in the care of diabetes given the disease complexity, need for continual monitoring, and frequent intensification. Up until now, quality of and access to care have been traditionally assessed from utilization records. An important limitation of such data is that it cannot differentiate between two causes of non-utilization: 1) the healthcare provider did not offer the care (by prescription or referral) versus 2) non-utilization of offered care. While healthcare providers may assume that their patients will use a health service following a referral, it is virtually unknown to what extent patients fail to do so. Under- utilization in certain subgroups, particularly among minority and socioeconomically disadvantaged patients, has been largely attributed to social disparities in access rather than under-utilization of offered care (i.e., inadequate adherence). While resources are needed to increase access for vulnerable populations, we must consider that there is also sub-optimal uptake of offered services even where access is not at issue. In this study, we take advantage of the electronic health record (EHR) system which captures electronic referral and prescribing within a large, integrated health care delivery system (Kaiser Permanente). The EHR enables us to investigate non-utilization of referred health services. This potentially has great public health importance given that repeated non-utilization (""""""""persistent non-utilization"""""""") may adversely affect continuity of care, and increase the risk for serious and costly events. Understanding the prevalence, social patterns, patient-, provider- and system-level predictors of non-utilization and persistent non-utilization will allow the design of interventions aimed at reducing this form of non-adherence. Findings will also inform a more accurate and valid definition of quality of care and access, and have important applications for our understanding of social disparities in the quality of care. Study subjects will include members of the Kaiser Permanente Northern California Diabetes Registry, a large, well-characterized, ethnically diverse, insured population of managed care patients with diabetes mellitus. Socially disadvantaged patients are well represented in this study population, which has relatively uniform access to and quality of care, unlike most population-based samples. Given that 92 per cent of Americans with diabetes have health insurance, findings from this insured population should have broad public health relevance. Moreover, we will benefit from the rich patient-level data provided by the 20,188 diabetic patients in the DISTANCE cohort. Given this will be the first, large epidemiological study to assess non- utilization of referred care, it will provide a more comprehensive understanding of how nonadherence impacts health and may inform the design of future interventions aimed at reducing health disparities. We have evidence that a small proportion of patients with diabetes within an integrated health plan fail to utilize care following a referral for covered health services (e.g., standard lab tests, specialty visits, health education). These non-utilizing patients lack continuity of care, are at high risk of adverse events, and may end up costing the healthcare systems much more than patients who utilize offered care.
In this study, we will evaluate the prevalence of non-utilization and persistent (repeated) non-utilization for referred health services for diabetes. We will then evaluate social disparities and other patient-, provider- and system-level predictors of non-utilization. Understanding non-utilization will facilitate the design interventions aimed at reducing this form of non-adherence. Findings will also inform a more accurate and valid definition of quality of care and access, and have important application for our understanding of social disparities in the quality of care. This project will help us better understand patients who are """"""""falling through the cracks"""""""" despite full access to integrated health care, and allow healthcare delivery systems to modify their programs to better serve their membership.
Showing the most recent 10 out of 20 publications