Skip to main content

Table 4 Supplementary robustness check: Clarifying potential self-selection bias of transport and health care need (combining Heckman two-stage procedure and multiple logit model)

From: Family support and transport cost: understanding health service among older people from the perspective of social-ecological model

 

Dependent variables: Class of health care facilities visited [multiple logit model, REF = first tier (community healthcare centers, township hospital, health care service station, village clinic/private clinic)]

 

Model 1

Model 2

 

second tier (specialized hospital, Chinese medicine hospital)

third-tier (general hospital)

second tier (specialized hospital, Chinese medicine hospital)

third-tier (general hospital)

 

Coef.

Robust S.E.

95% CI

Coef.

Robust S.E.

95% CI

Coef.

Robust S.E.

95% CI

Coef.

Robust S.E.

95% CI

Independent variables

 Transport cost in health service seeking [ex-post, having paid]

1.031 **

0.090

[0.854, 1.207]

1.019 **

0.074

[0.875, 1.164]

1.036 **

0.090

[0.859, 1.213]

1.030 **

0.075

[0.884, 1.177]

 Family support [face to face]

0.339 **

0.103

[0.138, 0.541]

0.184 **

0.077

[0.034, 0.335]

      

 Family support [Online connection]

      

0.354 *

0.138

[0.083, 0.624]

0.328 **

0.104

[0.125, 0.531]

 Disability

−1.207 **

0.420

[−2.031, −0.384]

−0.976 **

0.308

[−1.579, − 0.373]

− 1.217 **

0.420

[−2.039, − 0.394]

− 0.986 **

0.310

[− 1.594, − 0.378]

 Disability ×Transport cost in health service seeking [ex-post, having paid]

0.408 *

0.180

[0.055, 0.762]

0.409 **

0.154

[0.106, 0.711]

0.418 *

0.183

[0.060, 0.776]

0.420 **

0.157

[0.112, 0.728]

Demography

 Age

−0.007

0.014

[− 0.035, 0.021]

0.028 **

0.010

[0.008, 0.047]

−0.008

0.014

[−0.036, 0.020]

0.029 **

0.010

[0.009, 0.050]

 Gender

  Male

REF.

  

REF.

  

REF.

  

REF.

  

  Female

−0.264

0.228

[−0.711, 0.182]

0.166

0.173

[−0.174, 0.505]

−0.269

0.229

[−0.719, 0.181]

0.158

0.176

[−0.188, 0.503]

 Health insurance

−0.426

0.510

[−1.426, 0.574]

−0.598

0.367

[−1.318, 0.122]

−0.500

0.510

[−1.500, 0.500]

−0.634

0.369

[−1.357, 0.089]

 Financial situations

0.054

0.064

[−0.072, 0.180]

0.050

0.047

[−0.041, 0.142]

0.058

0.065

[−0.069, 0.185]

0.051

0.048

[−0.042, 0.145]

 Chronic conditions

0.144

0.079

[−0.011, 0.299]

0.161 **

0.053

[0.057, 0.265]

0.145

0.080

[−0.010, 0.301]

0.165 **

0.054

[0.060, 0.271]

 Self-rated health

−0.104

0.128

[−0.355, 0.147]

−0.148

0.082

[−0.308, 0.013]

− 0.107

0.128

[− 0.357, 0.143]

−0.148

0.082

[−0.309, 0.012]

 Inverse Mills ratio (adjusted according to health care need)

65.104

175.500

[− 278.871, 409.078]

−100.521

136.470

[−367.996, 166.954]

87.438

178.278

[− 261.981, 436.857]

−83.035

141.805

[− 360.967, 194.896]

Intercept

−23.929

50.083

[−122.091, 74.232]

24.282

38.968

[−52.093, 100.657]

−30.500

50.938

[− 130.338, 69.337]

17.709

40.466

[−61.604, 97.022]

Num. of non-missing observations

1373

     

1373

     

Wald-statistics

282.18

     

272.89

     

[p-value]

[0.000]

     

[0.000]

     
  1. *p < 0.05, **p < 0.01. Robust standard errors are reported