Service collaboration and hospital cost performance - Direct and moderating effects

Background: Growing reliance on service provision through systems and networks creates the need to better understand the nature of the relationship between service collaboration and hospital performance and the conditions that affect this relationship. Objective: We examine 1) the effects of service provision through health systems and health networks on hospital cost performance and 2) the moderating effects of market conditions and service differentiation on the collaboration-cost relationship. Research Design: We used moderated regression analysis to test the direct and moderating effects. Data on 1368 private hospitals came from the 1998 AHA Annual Survey, Medicare Cost Reports, and Solucient. Measures: Service collaboration was measured as the proportion of hospital services provided at the system level and at the network level. Market conditions were measured by the levels of managed care penetration and competition in the hospital's market. Results: The proportion of hospital services provided at the system level had a negative relationship with hospital cost. The relationship was curvilinear for network use. Degree of managed care penetration moderated the relationship between network-based collaboration and hospital cost. Conclusion: The benefits of service collaboration through systems and networks, as measured by reduced cost, depend on degree of collaboration rather than mere membership. In loosely structured collaborations such as networks, costs reduce initially but increase later as the extent of collaboration increases. The effect of network-based collaboration is also tempered by managed care penetration. These effects are not seen in more tightly integrated forms such as systems.