Effects of health poverty alleviation project from the perspective of vulnerability to poverty: evidence from five Chinese prefectures

Figures & data

Table 1. Variables.

Category	Variable	Variable code	Definition	Mean	SD
Dependent variable	Vulnerability to poverty	vp	Measured results	0.37	0.03
Independent variable	Eligibility for HPAP	hpap	Yes = 1, No = 0	0.43	0.50
Characteristics of the household head	Household income per capita	Income_h	The logarithm of household net income (RMB)/the number of family members	8.53	1.08
	Age	age	Age of the head of the household	47.31	11.81
	Has spouse	spouse	Yes = 1, No = 0	0.84	0.37
	Education level	edu	No education = 1, Primary school = 2, Junior high school = 3, Senior high school/Secondary vocational School and above = 4	2.38	0.90
	Health status	health	Good = 1, Average = 2, Poor = 3	1.66	0.75
Characteristics of the household	Household size	size	The number of family members	4.04	1.40
	The prevalence of severe illness or chronic illness	twp	The total number of individuals who have severe illness or chronic illness/The number of family members	0.13	0.23
	The proportion of medical expenditure	med_exp	Medical expenditure/Household expenditure	0.11	0.17
	The presence of out-of-home workers	worker_ooh	Yes = 1, No = 0	0.48	0.50
	The proportion of the household workforce	wf_rate	The proportion of aged 15–59	0.61	0.22
	The arable land area per capita	land	Household arable land area (mu)/The number of family members	1.25	1.51
Characteristics of the village	Per capita income	income_v	The per capita income of the village	8.76	0.91
	Coverage distance of the clinic	clinic	Within 2 km = 1, 2–4 km = 2, Over 4 km = 3	2.16	0.79
	Ecological fragility	eco	Yes = 1, No = 0	0.27	0.44

Figure 1. Locations of study areas.

Table 2. Survey data of five Chinese prefectures compared with that of the sixth National health Service survey (%).

	Survey			Sixth National Health Service Survey
	Average	Middle China Anqing, Enshi	West China Liangshan, Shangluo, Linxia	Average	Middle China	West China
Two-week prevalence	30.4	34.3	27.1	32.2	31.8	32.3
Two-week consultation rate	22.3	25.2	19.9	24.0	21.4	25.2
Health check attendance in the past year	18.1	20.2	16.3	47.2	44.9	47.7
Basic medical insurance coverage proportion	87.2	84.8	89.1	96.8	96.4	97.6
BMI
Underweight (BMI < 18.5)	6.7	6.1	7.2	7.9	7.7	9.3
Normal (BMI = 18.5–23.9)	62.5	61.7	63.2	54.8	54.8	56.1
Overweight (BMI = 24–27.9)	26.2	28.2	24.6	28.9	29.2	26.9
Obese (BMI ≥ 28)	4.6	4.0	5.0	8.4	8.3	7.7

Table 3. Vulnerability to poverty by region.

Area	Vulnerability to poverty	Vulnerability proportion(%)
Linxia (n = 177)	0.435	31.1
Enshi (n = 306)	0.320	11.8
Anqing (n = 272)	0.310	12.5
Shangluo (n = 358)	0.371	19.3
Liangshan (n = 182)	0.460	37.9
Total (n = 1295)	0.367	2.2

Table 4. Hierarchical linear regression estimation results.

Variables	Average	Poor	Non-poor
Fixed effects
Intercept	0.723	0.735	0.716
Household variables
Eligibility for HPAP	−0.018***	−0.016***	−0.020***
Age	0.001***	0.001***	0.001***
Has spouse	−0.113***	−0.115***	−0.111***
Education level	−0.069	−0.069***	−0.069***
Health status	0.001	0.003*	0.000
Population size	0.074***	0.077***	0.072***
The prevalence of severe illness or chronic illness	0.112***	0.115***	0.111***
The proportion of medical expenditure	0.319***	0.327***	0.300***
Presence of out-of-home workers	−0.001	−0.001	−0.002*
The proportion of the household workforce	−0.067***	−0.075***	−0.062***
Per capita arable land area	0.004***	0.005***	0.003***
Village-level variables
Per capita income	−0.062***	−0.064***	−0.060***
Coverage distance of the clinic	0.033***	0.033***	0.032***
Ecological fragility	0.091***	0.094***	0.087***
Random effects
Random term	0.007	0.007	0.006
Intercept reliability estimation	0.017	0.016	0.017

*p < 0.1, **p < 0.05, ***p < 0.01.

Table 5. The effect of HPAP.

	Analysis method	Treated	Control	ATT	T-value
Global	K-nearest neighbour matching	0.359	0.367	−0.018***	−1.76
	Caliper matching	0.350	0.380	−0.030***	−3.43
	Kernel matching	0.349	0.368	−0.019***	−2.03
Poor	K-nearest neighbour matching	0.432	0.441	−0.009	−0.48
	Caliper matching	0.432	0.434	−0.002	−0.10
	Kernel matching	0.432	0.452	−0.020	−1.17
Non-poor	K-nearest neighbour matching	0.308	0.320	−0.012	−1.07
	Caliper matching	0.308	0.350	−0.042***	−3.92
	Kernel matching	0.308	0.328	−0.020***	−1.98

*p < 0.1, **p < 0.05, ***p < 0.01.

Table 6. Mediating effect model estimation results.

Group 1: Human capital growth	Model 1	Model 2	Model 3
HPAP	−0.019*** (0.001)	0.390*** (0.064)	−0.018*** (0.001)
Human capital growth	\	\	−0.001* (0.001)
Control variable	Control	Control	Control
Group 2: Out-of-pocket medical ratio	Model 4	Model 5	Model 6
HPAP	−0.023*** (0.005)	−0.021*** (0.006)	−0.021*** (0.006)
Out-of-pocket medical ratio	\	\	0.101*** (0.101)
Control variable	Control	Control	Control

*p < 0.1, **p < 0.05, ***p < 0.01; standard errors are in parentheses.