Background
Methods
Study time and settings
Indicators | Dangyang | Zhijiang |
---|---|---|
Location | Contiguous to Zhijiang | Contiguous to Dangyang |
Administrative districts | 10 towns | 9 towns |
Population(thousand) | 486 | 496 |
Aging population rate (%) | 18.7 | 15.2 |
Urbanization rate (%) | 50.0 | 54.0 |
Per capita annual income (urban/rural, yuan) | 27,433/16928 | 23,675/15,285 |
Participant recruitment
Patient sampling
Physician contract
Intervention
SHI increasing outpatient expenditure reimbursement ratio
Health management services provided by contracted township hospital physicians
Project oversight
Data collection
Statistical analysis
Propensity score matching (PSM) model [26‐29]
Difference-in-differences (DID) model [30‐32]
-
\( \hat{\upbeta_0} \): Mean outcome for the control group at baseline.
-
\( \hat{\upbeta_0} \)+\( \hat{\upbeta_1} \): Mean outcome for the control group at follow-up.
-
\( \hat{\upgamma} \): The single difference between treatment and control groups at baseline.
-
\( \hat{\upbeta_0} \)+\( \hat{\upgamma} \): Mean outcome for the treatment group at baseline.
-
\( \hat{\upbeta_0} \)+\( \hat{\upbeta_1} \)+\( \hat{\upgamma} \)+\( \hat{\updelta} \): Mean outcome for the treatment group at follow-up.
-
\( \hat{\updelta} \): The DID or impact of the intervention.
-
ε i :The random error.
Ethical issues
Results
Valid observation subjects through PSM
Variables | Before PSM | Standardized difference (%) |
P
| Variables | After PSM | Standardized difference (%) |
P
| ||
---|---|---|---|---|---|---|---|---|---|
Intervention group (n = 2083) | Control group (n = 3152) | Intervention group (n = 1673) | Control group (n = 1673) | ||||||
Sex (male, %) | 49.1 | 51.9 | −5.6 | 0.048 | Sex (male, %) | 52.4 | 51.7 | 1.4 | 0.678 |
Age (year) | 66.4 ± 7.9 | 62.9 ± 11.4 | 35.7 | 0.021 | Age (year) | 65.0 ± 5.1 | 64.5 ± 7.3 | 7.9 | 0.307 |
Annual income (yuan) | 12,901 ± 20,705 | 13,865 ± 19,244 | −4.8 | 0.143 | Annual income (yuan) | 12,805 ± 14,429 | 12,863 ± 12,050 | −0.4 | 0.992 |
Diastolic pressure (mmHg) | 115.6 ± 33.2 | 111.9 ± 39.0 | 10.2 | 0.039 | Diastolic pressure (mmHg) | 114.9 ± 29.7 | 112.6 ± 34.6 | 7.1 | 0.420 |
Systolic pressure (mmHg) | 198.0 ± 65.7 | 201.1 ± 68.2 | −4.6 | 0.122 | Systolic pressure (mmHg) | 197.5 ± 39.5 | 197.2 ± 30.1 | 0.9 | 0.855 |
Frequency of per capita annual medical service utilization
Outcome variable (y) | Baseline (1673 vs. 1673) | Follow-up (1673 vs. 1673) | |||||
---|---|---|---|---|---|---|---|
Control
\( \left({\widehat{\upbeta}}_0\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+\widehat{\gamma}\right) \)
| Diff (BL)
\( \left(\widehat{\gamma}\right) \)
| Control
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1+\widehat{\gamma}+\widehat{\updelta}\right) \)
| Diff (FU)
\( \left({\widehat{\gamma}}_0+\widehat{\updelta}\right) \)
| DIFF-IN-DIFF
\( \left(\widehat{\updelta}\right) \)
| |
Outpatient | 4.180 | 4.072 | −0.108 | 4.161 | 7.353 | 3.192 | 3.3 |
P
| 0.115 | 0.041 | 0.008 | ||||
Inpatient | 0.063 | 0.125 | 0.062 | 0.087 | 0.074 | −0.013 | −0.075 |
P
| 0.000 | 0.095 | 0.000 | ||||
Total medical services | 4.243 | 4.197 | −0.046 | 4.248 | 7.427 | 3.179 | 3.225 |
P
| 0.862 | 0.003 | 0.001 |
Per capita annual medical expenditure
Outcome variable (y) | Baseline (1673 vs. 1673) | Follow-up (1673 vs. 1673) | |||||
---|---|---|---|---|---|---|---|
Control
\( \left({\widehat{\upbeta}}_0\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+\widehat{\gamma}\right) \)
| Diff (BL) \( \left(\widehat{\gamma}\right) \) | Control
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1+\widehat{\gamma}+\widehat{\updelta}\right) \)
| Diff (FU)
\( \left({\widehat{\gamma}}_0+\widehat{\updelta}\right) \)
| DIFF-IN-DIFF
\( \left(\widehat{\updelta}\right) \)
| |
Outpatient | 269.2 | 255.4 | −13.8 | 263.8 | 331.2 | 67.4 | 81.2 |
P
| 0.452 | 0.095 | 0.048 | ||||
Inpatient | 818.1 | 1167.6 | 349.5 | 910.9 | 785.0 | −125.9 | −475.4 |
P
| 0.009 | 0.003 | 0.027 | ||||
Total | 1087.3 | 1423.0 | 335.7 | 1174.7 | 1116.2 | −58.5 | −394.2 |
P
| 0.015 | 0.043 | 0.013 |
SHI per capita annual reimbursement expenditure
Outcome variable (y) | Baseline (1673 vs. 1673) | Follow-up (1673 vs. 1673) | |||||
---|---|---|---|---|---|---|---|
Control
\( \left({\widehat{\upbeta}}_0\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+\widehat{\gamma}\right) \)
| Diff (B)\( \left(\widehat{\gamma}\right) \) | Control
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1+\widehat{\gamma}+\widehat{\updelta}\right) \)
| Diff (FU)
\( \left(\widehat{\gamma}+\widehat{\updelta}\right) \)
| DIFF-IN-DIFF
\( \left(\widehat{\updelta}\right) \)
| |
Outpatient | 33.3 | 33.6 | 0.3 | 34.7 | 125.3 | 90.6 | 90.3 |
P
| 0.951 | 0.031 | 0.018 | ||||
Inpatient | 384.5 | 642.2 | 257.7 | 475.5 | 450.6 | −24.9 | −282.6 |
P
| 0.029 | 0.830 | 0.016 | ||||
Total | 417.8 | 675.8 | 258.0 | 510.2 | 575.9 | 65.7 | −192.3 |
P
| 0.005 | 0.048 | 0.010 |
Patient per capita annual out-of-pocket expenditure
Outcome variable (y) | Baseline (1673 vs. 1673) | Follow-up (1673 vs. 1673) | |||||
---|---|---|---|---|---|---|---|
Control
\( \left({\widehat{\upbeta}}_0\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+\widehat{\gamma}\right) \)
| Diff (BL)
\( \left(\widehat{\gamma}\right) \)
| Control
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1+\widehat{\gamma}+\widehat{\updelta}\right) \)
| Diff (FU)
\( \left(\widehat{\gamma}+\widehat{\updelta}\right) \)
| DIFF-IN-DIFF
\( \left(\widehat{\updelta}\right) \)
| |
Outpatient | 235.9 | 221.8 | −14.1 | 229.1 | 205.9 | −23.2 | −9.1 |
P
| 0.452 | 0.395 | 0.518 | ||||
Inpatient | 433.6 | 525.4 | 91.8 | 435.4 | 334.4 | − 101 | −192.8 |
P
| 0.049 | 0.043 | 0.027 | ||||
Total | 669.5 | 747.2 | 77.7 | 664.5 | 540.3 | − 124.2 | −201.9 |
P
| 0.085 | 0.041 | 0.022 |
Health outcomes
Outcome variable (y) | Baseline (1673 vs. 1673) | Follow-up (1673 vs. 1673) | |||||
---|---|---|---|---|---|---|---|
Control
\( \left({\widehat{\upbeta}}_0\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+\widehat{\gamma}\right) \)
| Diff (BL)
\( \left(\widehat{\gamma}\right) \)
| Control
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1\right) \)
| Intervention
\( \left({\widehat{\upbeta}}_0+{\widehat{\upbeta}}_1+\widehat{\gamma}+\widehat{\updelta}\right) \)
| Diff (FU)
\( \left(\widehat{\gamma}+\widehat{\updelta}\right) \)
| DIFF-IN-DIFF
\( \left(\widehat{\updelta}\right) \)
| |
SBP | 114.9 | 112.6 | −2.3 | 114.6 | 109.4 | −5.2 | −2.9 |
P
| 0.855 | 0.043 | 0.011 | ||||
DBP | 197.5 | 197.2 | 0.3 | 203.6 | 196.0 | −7.6 | − 7.9 |
P
| 0.420 | 0.095 | 0.508 |