ORIGINAL ARTICLE Voluntary Physician Switching by Human Immunodeficiency Virus-Infected Individuals A National Study of Patient, Physician, and Organizational Factors Hector P. Rodriguez, PhD, MPH,*† Ira B. Wilson, MD, MSc,†‡ Bruce E. Landon, MD, MBA,§¶ Peter V. Marsden, PhD, and Paul D. Cleary, PhD** Objective: We sought to assess which patient, physician, and organizational factors are related to voluntary physician switching among human immunodeficiency virus (HIV)-infected patients. Design: We analyzed the results from a 3-wave survey of patients conducted by the HIV Cost and Services Utilization Study (HCSUS), a longitudinal study of a nationally representative sample of noninstitutionalized HIV-infected individuals receiving care in the contiguous United States. Physicians providing care and care site directors were surveyed once. Relationships of interpersonal aspects of care, access and continuity, technical quality of care, and physi- cian and site characteristics to voluntary switching were analyzed using multilevel logistic regression models that nested repeated observations within patients, patients within clinicians, and clini- cians within region. Results: Approximately 15% of patients voluntarily changed their usual clinicians during the 2-year study period. In a multivariate model, lower voluntary switching was predicted by patient trust (odds ratio OR= 0.74; 95% confidence interval 95% CI= 0.61– 0.90), physician antiretroviral knowledge (OR = 0.26; 95% CI 0.13– 0.53), moderate (rather than low or high) HIV patient volume at a care site (OR = 0.09; 95% CI = 0.03– 0.31), and Ryan White Care Act funding (OR = 0.27, 95% CI = 0.14 – 0.52). Conclusions: Patients with chronic illnesses may use several mark- ers of specialization and technical quality to make decisions about their care. These results challenge the notion that patients cannot assess the quality of care they receive. Key Words: voluntary physician switching, patient trust, HIV, physician characteristics, organizational characteristics, specialization (Med Care 2007;45: 189 –198) P hysician switching, whether voluntary or involuntary, is disruptive for clinicians, costly for health care systems, 1,2 and can be difficult for patients. It may be impossible to prevent involuntary switching as the result of factors like physician or patient relocation or changes in a patient’s insurance coverage. However, understanding the causes of voluntary physician switching might help to reduce costs related to discontinuity. Few studies have examined predictors of voluntary physician switching. 3–8 All of those studies examined switch- ing in primary care settings and included only patient predic- tors. Those studies found that voluntary switching is linked to interpersonal aspects of the physician–patient relationship, including patient trust, physician communication, and physi- cian knowledge of patients. 6,8 –11 We extend the existing research in 2 important ways. First, we examined voluntary physician switching by patients with a chronic illness, individuals for whom continuity of care might have a greater impact on health outcomes, and where the costs of discontinuities could be greatest. 12–14 Switching may, however, also holds benefits for such patients if they move toward clinicians and care sites that can provide higher-quality care for their condition. Second, because care site and physician specialization are important determinants of care quality in certain conditions, such as infection with human immunodeficiency virus (HIV), 15–19 we asked whether physician or organizational characteristics might be important predictors of voluntary physician switching after accounting for patient factors. To explore these issues, we analyzed data from a national cohort study of patients with HIV infection that documented changes in patients’ primary HIV care providers. We addressed 4 main study questions: 1) Are physician and care site specialization related to voluntary physician switch- ing? 2) Are practice demands, in addition to physician satis- faction and stress, related to voluntary physician switching? From the *PhD Program in Health Policy, Harvard University, Cambridge, Massachusetts; †The Health Institute, Institute for Clinical Research and Health Policy Studies, Tufts-New England Medical Center, Boston, Massachusetts; ‡Department of Medicine, Tufts University, Boston, Massachusetts; §Division of General Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts; ¶Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts; Department of Sociology, Harvard University, Cambridge, Massachusetts; and **Yale School of Public Health, New Haven, Connecticut. Supported by grants from the Robert Wood Johnson Foundation and the Agency for Healthcare Research and Quality (H.S. 10227). Financial support for Dr. Rodriguez was provided by a traineeship from the Agency for Healthcare Research and Quality (5T32HS000055). Reprints: Paul D. Cleary, PhD, Yale School of Public Health, Yale Univer- sity, 60 College Street, P.O. Box 208034, New Haven, CT 06520-8034. E-mail: Paul.Cleary@yale.edu. Copyright © 2007 by Lippincott Williams & Wilkins ISSN: 0025-7079/07/4503-0189 Medical Care • Volume 45, Number 3, March 2007 189