Mind the information gap: fertility rate and use of cesarean delivery and tocolytic hospitalizations in Taiwan

Health Economics Review

Table 7 Probit estimates for equation (5): the effects of declining fertility and health information gap on the probability of having tocolytic hospitalizations, 1997-2004 (Base outcome: having no tocolytic hospitalizations)^a

	Specification 1		Specification 2
	(Treatment group: female physicians and female relatives of physicians; Comparison group: other women)		(Treatment group: high socioeconomic status women; Comparison group: low socioeconomic status women;)
Variables	Coef.	Robust Std. Err.	Coef.	Robust Std. Err.
Log of lagged ob/gyn per 100 births	0.174***	0.038	0.339***	0.091
Log of lagged ob/gyn per 100 births × Information^b	-0.008	0.134	-0.293	0.206
Information^b	-0.103*	0.057	-0.304	0.164
Log GFR	0.966	0.681	-1.127	0.81
*Patients' characteristics*
Age	0.027***	0.001	0.025***	0.001
Insurable wage (÷10²)	-0.0002***	0.0002	0.0003	0.0002
Having a major disease card	0.016	0.018	0.012	0.049
Having pregnancy-associated hospitalizations before	0.521***	0.006	0.693***	0.009
Previous year's inpatient expenses	0.0001***	0.0002	0.0001***	0.0001
*Hospitals' characteristics*
Public	-0.155***	0.01	-0.187***	0.025
Private non-profit	-0.214***	0.108	-0.403*	0.196
Medical center	0.127	0.219	0.188	0.254
Regional Hospital	0.113***	0.012	0.050***	0.002
District Hospital	0.045***	0.007	0.100***	0.015
Teaching Hospital	0.068***	0.011	0.048*	0.027
Bed size (÷10²)	-0.007***	0.001	-0.007	0.002
*Ob/gyn characteristics*
Ob/gyn age	-0.002	0.002	0.002	0.004
Ob/gyn gender	0.020*	0.01	0.066**	0.029
Constant	-2.390***	0.079	-8.413***	0.16
Number of observations	1,941,935		1,770,654
Log likelihood	-181,362.22		-199,483.47

a The regression includes a full set of time and regional dummies.
b Information is a dummy variable and information = 1 indicates medically-informed individuals.
* Statistically significant at the 10% level.
** Statistically significant at the 5% level.
*** Statistically significant at the 1% level.
The marginal effect of the interaction term "Log of lagged ob/gyn per 100 births × Information" on the probability of having toocolytic hospitalizations (for specification 1): $\begin{gathered} [\begin{gathered} (Pr (L O B B I R T H = - 0.2910312, I = 0)) \\ - (Pr (L O B B I R T H = - 0.6134288, I = 0)) \end{gathered}] - [\begin{gathered} (Pr (L O B B I R T H = - 0.2910312, I = 1)) \\ - (Pr (L O B B I R T H = - 0.6134288, I = 1)) \end{gathered}] \\ = 0.0001165 \end{gathered}$
Standard error for the marginal effect obtained by bootstrapping: 0.0005519
The marginal effect of the interaction term "Log of lagged ob/gyn per 100 births × Information" on the probability of having toocolytic hospitalizations (for specification 2): $\begin{gathered} [\begin{gathered} (Pr (L O B B I R T H = - 0.2910312, I = 0)) \\ - (Pr (L O B B I R T H = - 0.6134288, I = 0)) \end{gathered}] - [\begin{gathered} (Pr (L O B B I R T H = - 0.2910312, I = 1)) \\ - (Pr (L O B B I R T H = - 0.6134288, I = 1)) \end{gathered}] \\ = 0.0001728 \end{gathered}$
Standard error for the marginal effect obtained by bootstrapping: 0.0004824

ISSN: 2191-1991