A major purpose of managed care is to control health care costs partly by imposing constraints on physician referral behavior. However, efforts to control costs may have adverse consequences for access and quality. Little is known about the influence of managed care on physician referral behavior and health outcomes. The question is significant, given that over 15 million referrals occur each year in the United States, and that health care reform will likely influence this important part of medical care. The purpose of the study is to examine the effect of managed care on physician referral behavior and health outcomes among pain or depression patients through the following two aims: (l) to determine the effect of managed care on the probability of referral among pain or depression patients in primary care; and (2) to determine the effect of managed care on health outcomes of referred and nonreferred patients in Aim 1. Using a prospective cohort design, actively practicing internists and family/general practitioners in King county (Seattle metropolitan area) will be invited to participate in the study. Financial incentives will be used to increase and sustain physician participation in the two phases of the study [baseline interview ($50), waiting room survey ($100), and referral log ($25); Total incentive+$175]. Physicians who consent to participate will receive a 30 minute telephone interview to collect referral behavior and other physician and practice measures. Within 30 days of each physician interview, a waiting room survey will be conducted in the practice for 2 weeks. Each patient aged 8-64 will be asked to complete a brief screening instrument to identify eligible pain or depression patients, and an ambulatory log will record descriptive data for all patients who consent to participate. One month later a follow-up survey of eligible patients will be conducted by Washington State University to identity referral candidates, or patients with chronic pain or depression who were not referred at the baseline visit. Six months later eligible patients will be asked to complete a follow-up survey; to track the normal course of depression, a 3-month follow-up of depressed patients also will be performed. Primary care physicians will be asked to complete a referral log for all patient referrals to obtain data about the reasons for referral, choice of specialist, and whether the patient was hospitalized. Constraints on physician referral under managed care will be measured through managed care and referral indices.
Aim 1 regression analyses will estimate the direct effects of the managed care and referral indices on the probability of referral and choice of specialist among referral candidates (separate analyses of referral at the waiting room screen also will be conducted).
Aim 2 analyses will estimate the direct and indirect effects of the managed care and referral indices on health outcomes, as well as the direct effect of referral (versus treatment by the primary care physician) on health outcomes.
| Tempel, L W; Perlmutter, J S (1990) Abnormal vibration-induced cerebral blood flow responses in idiopathic dystonia. Brain 113 ( Pt 3):691-707 |
| Perlmutter, J S; Kilbourn, M R; Welch, M J et al. (1989) Non-steady-state measurement of in vivo receptor binding with positron emission tomography: ""dose-response"" analysis. J Neurosci 9:2344-52 |
