Factors associated with tocolytic hospitalizations in Taiwan: evidence from a population-based and longitudinal study from 1997 to 2004

Table 3 Probit model for the use of tocolytic hospitalizations

Variables	Probit model	Marginal effect
Constant term	-3.028 (0.423)**	-2.326 (0.074)**
*Individual factors*
Age	0.026 (0.001)**	-0.010 (0.0002)**
Wage	0.004 (0.0001)**	0.001 (0.0001)**
Having a major disease card	0.015 (0.018)	0.061 (0.065)
Having pregnancy-associated hospitalizations before	0.428 (0.006)**	0.665 (0.018)**
Previous year's inpatient expenses	0.001 (0.0002)**	0.027 (0.001)**
*Institutional factors*
Bed size	0.004 (0.001)**	0.007 (0.001)**
Ownership
Public	0.059 (0.010)**	-0.020 (0.004)**
Private non-profit	-0.102 (0.010)**	-0.042 (0.004)**
Proprietary (Reference)
Accreditation status
Medical center	0.392 (0.021)**	0.013 (0.007)**
Regional hospital	0.102 (0.017)**	0.021 (0.006)**
District hospital	0.113 (0.013)**	0.015 (0.005)**
Clinic (Reference)
Teaching status
Teaching	0.128 (0.011)**	0.050 (0.004)**
Non-teaching (Reference)
*Physician factors*
Male gender	-0.012 (0.010)	-0.004 (0.002)
Age	0.003 (0.0005)**	0.019 (0.001)**
*Area factor*
Logged regional fertility rate	-0.539 (0.112)**	-0.221 (0.042)**
Observations	1,979,311	1,979,311
Log-likelihood	-148,762.7

Robust standard errors are in parentheses. The three models also include a full set of time and regional dummies. *Statistically significant at the 10% level. **Statistically significant at the 5% level.

ISSN: 1471-2393