Chao-Chin Sherina Lee Jui-fen Rachel Lu Chang Gung University, Taiwan. ihea July 11-July 13, 2011

Transcription

1 Chao-Chin Sherina Lee Jui-fen Rachel Lu Chang Gung University, Taiwan ihea July 11-July 13, 2011

2 Motivation Children is the future hope of a country With a declining total fertility rate (TFR) in Taiwan, issue of children health become even more important With NHI coverage, we are interested in studying health equality and equity in health care by children Children are dependents 2

3 Study aims T To further explore the factors influencing the distribution 3

4 Introduction 4

5 Introduction Impacts on children s health and health care use Main care givers Parents Family background and income Health insurance Government medical scheme subsidy policy 5

6 Method: Data Multistage stratified systematic sampling design Probability proportional to size (PPS) on each strata. Western doctor, dentist, prevention care, licensed traditional medicine practitioner (LTMP), emergency room visits, inpatient admission, 6

7 Use concentration index (C) as the measure of relative income-related inequality in use of health care (Wagstaff et al, 1991). Estimates for C and its robust standard error are obtained by running the following convenient regression (Kakwani, Wagstaff et al, 1997): where is the variance of R i and β is equal to C

8 Linear OLS to explain y, total health care utilization rates (plus standard 2-part outcomes as well): Income (lninc) Health need standardizing variables (x) Age, sex and self-assessed health Non-need variables (z) Main care giver s characteristics - including age, sex, education, economic activity status/occupation Geographic region Supply-side - including Herfindahl-Hirschman Index (HHI), health care facilities per 10,000 pop, physicians per 1,000 pop

9 Estimated (partial) demand elasticity of each determinant k was defined as: The total concentration index can be written as: where the first term denotes the partial contribution of income inequality, the second the contribution of health need variables, the third the contribution of all other variables and the last term is the generalized concentration index of ε

10 By indirect standardization, compute an index of horizontal inequity (HI) in health care use by subtracting the need-expected inequality from total inequality: HI=[-2, 2]

11

12 Sample characteristics of sampled children in 2001 in Taiwan (N=3,470) Variables No. of obs. % Child Household income per equivalent adult per month NTD 10,000 and below % NTD 10,001-20,000 1, % NTD 20,001-30, % NTD 30,001-40, % NTD 40,001-50, % NTD 50,001-60, % NTD 60,001-70, % NTD 70,001-80, % NTD 80,001 and above % Self-assess health Good % Common % Bad % Diseases Learning disability % developmental delay % Major care giver Education Informal education % Elementary school % Senior high % High % University and above % Activity status Employed % Non-employed % 12

13 Inequality indices for health Learning disability Pro-poor distribution HI: Developmental delay Pro-poor distribution HI:

14 Inequality indices for volume of utilization 14

15 Decomposition of inequality in volume of utilization 15

16

17 Discussion & Conclusion Compared to the study by Lu et. al (2007) on adults health care utilization that number of visits consumed by children is more than that by adults. Children are not as mature as adults. Subject to unexpected diseases. Socioeconomic characteristics of the care givers matter for children health and use of care Pro-rich tendency in health care Western doctor, dentist, Chinese doctor and emergency room Pro-poor distribution in d learning disability and development delay 17

18 Discussion & Conclusion Future policy directs M E 18

19 Acknowledgements This study was funded by: International Development Research Centre, Canada (IDRC Grant Number ) AusAID (Australian Development Research Award Grant Number 44741) Asian Development Bank (ADB RETA-6515) Analysis for Taiwan was funded by Taiwan National Science Council (NSC H MY3) 19

20 20