Continuity of care in older adults with multiple chronic conditions

Thesis (Ph.D.)--University of Washington, 2016-08

Nearly three out of four people over 65 years of age in the United States (U.S.), or greater than 35 million older adults, have two or more chronic conditions. People with multiple chronic conditions (MCCs) pose a significant challenge to the health care system because they are at greater risk for morbidity and mortality, utilize more health care services, and are vulnerable to poor quality care. Despite their considerable needs, this patient population is often excluded from clinical and health services research. Thus, the evidence base for best practices of care for this population is lacking, which contributes to poor outcomes of care. Continuity of care (COC) is an important process of care that prior studies have shown is associated with greater patient satisfaction, fewer emergency department (ED) visits and hospitalizations, a reduced risk of mortality, and lower costs of care in older adults. Studies specifically examining patients with MCCs have also demonstrated associations between higher COC and fewer duplicated medications, fewer ED visits and hospitalizations, and a lower risk of death. Thus, COC is a recommended component of high quality care for patients with MCCs. There are, however, a number of limitations to the current COC literature for older adults with MCCs. Although patients with MCCs would seem to be particularly vulnerable to care fragmentation, the association between morbidity burden and COC has not been explored. Also missing from the literature are studies exploring the relationships between COC and patient-reported measures of health status. Another shortcoming is the limited exploration of provider type in COC studies. Finally, differences in benefits conferred by continuity with an individual provider as compared with a medical practice are unclear. In this dissertation, I aim to fill the gaps in the COC literature by testing: 1) whether there is an association between morbidity burden and COC at the provider and practice-level among older adults who primarily saw a PCP and older adults who primarily saw a specialist for their medical care; 2) whether there is an association between provider and practice-level COC and functional status in a population of older adults with MCCs, and whether any observed associations were moderated by the type of provider the patient primarily saw for their health care visits; and 3) whether there is an association between provider and practice-level COC and health care expenditures in a population of older adults with MCCs, and whether any observed associations were moderated by the type of provider the patient primarily saw for their health care visits. In the first study of this dissertation, I found that multimorbidity is an independent risk factor for lower COC. The magnitude of the association was such that people with high levels of morbidity burden would be expected to experience a decrease in continuity that was clinically meaningful, and could impact their clinical outcomes. In the second study, I found that neither provider nor practice-level continuity was significantly associated with functional status decline. However, in subgroup analyses, I observed that specialty care continuity was significantly associated with a lower odds of functional status decline among patients seeing primarily a specialty care provider. Finally, in the third study, I found a significant association between higher continuity and lower expenditures that was irrespective of provider type and provider or practice levels. Results also suggested the lower costs may have arisen from lower rates of emergency department visits and hospitalizations among those with higher COC. Our findings lend further support for the value of COC, a process of care that should be encouraged, particularly among high morbidity patients who are at risk of greater care fragmentation. They also provide insight into the possible effects of delivery system reform efforts that emphasize COC. Our results suggest that COC may provide benefits in terms of costs and utilization, though not necessarily patient-reported health outcomes. They also suggest that emphases on different levels of continuity might not produce appreciable differences in terms of lowered costs, but differences may arise for patient-reported health outcomes. Finally, they indicate the importance of provider type considerations when thinking about care continuity.