OBJECTIVE. To empirically estimate the effects that managed care has had on physician and clinical integration in urban hospitals. DATA SOURCES. The 1993 Hospital-Physician Relationship Survey conducted for the Prospective Payment Assessment Commission, augmented with data from a variety of secondary sources. The entire 1,495 responding hospitals were used to construct measures of integration; 591 responding hospitals in urban areas were used for the managed care analysis. STUDY DESIGN. Factor analysis was used to reduce 23 integration variables into 5 physician and 3 clinical integration factors. Two-stage least-squares regression techniques were used to estimate the effects of endogenous managed care. Models were estimated for all urban hospitals and for hospital subsets based upon ownership, multi-hospital system status, and teaching. PRINCIPAL FINDINGS. Other things equal, physician involvement in hospital management and governance increased with managed care involvement; to a lessor degree, the use of physician organization arrangements and other joint ventures also increased. Practice management and support services were lower in hospitals with high managed care activity. Larger hospitals, investor owned, system, and non-teaching hospitals had larger managed care revenues. Managed care revenues were lower in more concentrated hospital markets. CONCLUSIONS. The relationship between managed care and physician and clinical integration is relatively modest. Much of the realignment under managed care has been limited to certain types of efforts. Those efforts can best be described as foundation-building rather than comprehensive or fundamental.