Page 184 - GHES-2-1

Global Health Econ Sustain Income-related inequality in health

2.4. Robustness check Table 1. Descriptive statistics of the sample

This study performed a set of robustness checks. The Variable Mean (SD)/Percentages
first involved replacing the binary variable in SRH with Health variables
a categorical variable, including five dummy variables: 1, Self-rated health
2, 3, 4, and 5, representing ‘very bad,’ ‘bad,’ ‘so-so,’ ‘good,’
and ‘very good,’ respectively. The second robustness check Bad/fair 56.8
involved replacing the binary variable for ADL ability with Good 43.2
the number of ADL limitations. ADL ability
No ADL limitations 73.6
3. Results Having ADL limitations 26.4

Table 1 presents the summary statistics of the sample. Demographic variables
According to the findings, 43.2% of the participants in Age 85.589 (11.585)
the study sample indicated good SRH, whereas 26.4% Gender
reported limitations in ADL. The average age of the Female 56.2
participants was 85.6 years, and their mean income was Male 43.8
14,540.14 yuan. In addition, 81.7% of the participants
were illiterate. Socioeconomic variables
Income (yuan) 14540.140 (13300.350)
Figure 1 reveals the standardized percentages for SRH
and ADL abilities by income quintile. The rich are more Education
likely to have better SRH. The percentage reporting good Illiteracy 81.7
SRH among the poorest group was 39.99%, whereas it was Elementary school 14.1
43.54% among the richest group. The wealthy are less likely Middle school and above 4.1
to have ADL limitations. The percentage of patients with Having medical insurance
ADL limitations in the poorest group was 7.10%, while it No 11.7
declined to 6.03% in the richest group. Yes 88.3
Table 2 displays the determinants of SRH and ADL Marital status
using the linear probability model, where these coefficients Otherwise 60.8
were used to calculate and decompose the CI in the next Married 39.2
step. Both demographic and socioeconomic factors had a Residence
significant impact on health. In terms of SRH, an increase
in age is associated with a decline in the probability of City 22.2
good SRH. Older men have a higher likelihood of having Town 33.2
good SRH. Income had a significant positive association Rural 44.6
with SRH. The rich are more likely to have good SRH Regions
than the poor. The relationship between educational level North 5.8
and SRH was also significant. Apart from that, residence Northeast 4.2
and regional factors were significantly related to SRH. East 40.5
In particular, living in rural areas significantly increased South Central 36.5
the likelihood of having a good SRH. Compared to those
living in North China, those living in South Central and Northwest 1.1
Southwest China showed a decreased likelihood of having Southwest 11.9
good SRH. N 10,078
The main determinants of SRH were similar to those Notes: The unit of the study sample is the individual. Mean (SD) is
presented for continuous variables, and percentages are presented for
of ADL ability, except for gender and residence. In categorical variables. The income unit is in yuan.
particular, with increasing age, the probability of having Abbreviation: SD: Standard deviation.
ADL limitations increased significantly. Income was
significantly associated with the ability to perform ADLs. with those without education, those with higher
An additional increase in income reduces the likelihood education were more likely to be independent in terms
of ADL limitation. The relationship between education of ADL. Moreover, the region has a close relationship
level and ADL limitation was also significant. Compared with ADL ability.

Volume 2 Issue 1 (2024) 5 https://doi.org/10.36922/ghes.2243