Page 113 - Effective healthcare cost containment policies Using the Netherlands as a case study - Niek W. Stadhouders

Do managed competition and active purchasing go hand in hand?
is 0.75. Next, based on market share of each insurer, a weighted average contracting index ( ) is obtained for each provider, ranging from zero (not contracted for any type of care in any insurer plan) to one (contracted for all types of care in all insurer plans).
The results show that selective contracting is used extensively in the hospital sector, with contracting indices of 0.88 and 0.53 for hospitals and ITCs, respectively. Selective contracting is more likely for smaller hospitals/ITCs (correlation coefficient= 0.40, Appendix figure 5.5a). However, high use of selective contracting may not translate to high market share reallocations. For example, providers may compensate budget loss from one insurer by budget increases from other insurers. Furthermore, under Dutch regulations, non-contracted care need still be reimbursed, although at a lower reimbursement rate. Lastly, selective contracting may be used as a threat to accept stringent contract requirements without effect on the provider’s total budget. Therefore, we also research the effect of selective contracting on hospital budgets. The underlying assumption is that when an insurer excludes a provider from his network, the budget of that provider is reduced. Vice versa, if an insurer adds a provider to the plan network, the budget of that provider is expected to increase. The change in contracting index is related to the change in the hospital budget by using multivariate linear regression.
(4)
Here, the difference in market share of provider between 2015 and 2016 is explained by the size of provider and differences in the contracting index of provider . The extent of selective contracting changed on average by 13% between 2015 and 2016 (Appendix figure 5.5b). However, no significant relation was found between changes in selective contracting and changes in market share (table 5.4), which does not change when we correct for size. The results suggest that despite extensive use of selective contracting, no effects on provider
budget reallocations are found. This is consistent with the results of the main analysis.
123
0.00039 (0.00026) 0.0004 (0.0003) 0.00040 (0.0003) 0.0421*** (0.0113) 0.00704 (0.0197) 1.5090* (0.7502) -0.00005 (0.00007 -0.0002** (0.0001) -0.0002 (0.0001)
Table 5.4: Relation between changes in market share and selective contracting, 2015-2016
Contracting index
Size
Size
2
F( 1,253) = 2.26 F( 2,252) = 7.07 F( 3,251) = 7.59 R2=0.0042 R2=0.0625 R2=0.0701
Constant
N=255
Note: robust standard errors are in parentheses; sign. *<5%;**<1%;***<0.1%
N=255 N=255
105