Table 3 Summary of results according different model specifications and estimation techniques—(Models M1 to M5)
Probability of being hospitalized | Length of hospital stay | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Linear probability model | Zero truncated Poisson† | |||||||||||||
Coeff (DID) | 95%CI | IRR (DID) | 95%CI | |||||||||||
Changes in the matching variables: | M2 | -0,02 | -0,05 | - | 0,02 | 0,72 | *** | 0,66 | - | 0,79 | ||||
Socioeconomics: age, gender, partnership, employment situation, coverage eligibilty | ||||||||||||||
Changes in the matching method: | ||||||||||||||
Exact matching | M3 | 0,01 | 0,42 | - | 1,29 | 0,73 | *** | 0,65 | - | 0,93 | ||||
k-nearest (k = 5) | M4 | -0,01 | -0,04 | - | 0,01 | 0,74 | *** | 0,69 | - | 0,79 | ||||
Changes in the Two Part version model: | M5 | 0,82 | *** | 0,74 | - | 0,91 | ||||||||
Estimation of a ZIP model (results for the second part) | ||||||||||||||