Page 200 - Effective healthcare cost containment policies Using the Netherlands as a case study - Niek W. Stadhouders
P. 200

Chapter 8
to reduce treatment intensity and health promotion/prevention to reduce demand. Because MCOs may address multiple equilibriums, it may be expected that cost containment is relatively successful. As a consequence, on the payer side market exits and reduced employment, profits, investments and inputs may be expected. For example MCOs may negotiate lower prices for pharmaceuticals, thereby reducing provider input expenses. As provider may negotiate with payers other than MCOs, MCOs change the bargaining condition. On the one hand, due to the reduction in MCO budgets providers may value agreement with traditional plans more, which increases provider bargaining surplus and reduces the budget agreement between traditional plans and providers. On the other hand, demand reductions may increase the unmet need of patients, which implies that patients may value inclusive networks of traditional payers more. This lowers the utility of the payer fallback option, and reduces the bargaining outcome of traditional payers, increasing the provider budget. Increased payer competitiveness due to MCO could therefore have both positive and negative spillover effects on traditional payers. Furthermore, MCO strategies to reduce demand may be focused on health status rather than perceived need, reducing patient utility.
Extending the model would provide insightful modelling possibilities. However, some reservations should be made. The model is designed specifically to assess the effect of cost containment policies, and may not be suited for assessment of other interventions. Secondly, the model remains a simplification of reality. This may suffice for most cost containment policies, but for specific policies, the model may be incomplete. For example, moving from fee-for-service to pay-for-performance payments may be challenging to assess in the current framework. Thirdly, parts of the model are still uncertain, specifically regarding behavioural responses. Although the model provides a set of possible responses, the most likely response may require empirical confirmation. As the model comprises of multiple individual agents in different roles, responses may depend on individual preferences, requiring agent-based model extensions. Lastly, if payers and providers incorporate the outcome of the model into negotiations, the model may suffer from the Lucas critique (Favero and Hendry, 1992).
 Limitations
The model at some points makes simplistic assumptions, necessary to keep the effects tractable, and distil general effects. However, extending the model may improve quantitative predictive capability. Some extensions have been mentioned. An overview of
192
Model extensions




























































































   198   199   200   201   202