Skip to main content

Table 1 Effects of the PBF program on outcomes at the health facility level

From: Impacts of performance-based financing on health system performance: evidence from the Democratic Republic of Congo

Variable

Health centers

Hospitals

Mean in control group

Impact and CI

% Changeb

N

Mean in control group

Impact and CI

% Changeb

N

Panel a: structural quality

 Basic equipment index

0.54

0.04 (0–0.07)

7%

290

0.75

0.01 (− 0.03 to 0.05)

1%

56

 Essential medicine and consumables index

0.68

0.01 (− 0.04 to 0.07)

2%

290

0.85

 − 0.04 (− 0.11 to 0.03)

 − 5%

56

 Vaccines index

0.62

0.05 (− 0.08 to 0.17)

8%

275

0.13

 − 0.06 (− 0.2 to 0.09)

 − 47%

53

 Family planning products index

0.35

0.19 (0.11–0.27)

55%

290

0.53

0 (− 0.12 to 0.12)

 − 1%

56

 Infrastructure index

0.36

0.02 (− 0.03 to 0.07)

6%

290

0.65

0.04 (− 0.05 to 0.14)

7%

56

 Infection prevention and control index

0.45

0.04 (− 0.03 to 0.11)

9%

264

0.75

0.01 (− 0.11 to 0.13)

2%

55

Average impacta

0.50

0.05 (0.01–0.09)

10%

 

0.72

0 (− 0.03 to 0.03)

0%

 

Panel b: technical process quality

 Antenatal care score

0.91

0.02 (− 0.03 to 0.08)

3%

1482

0.87

0.06 (0–0.13)

7%

283

 IMCI: assessment score

0.56

0.03 (− 0.01 to 0.08)

6%

1631

0.65

0.05 (− 0.01 to 0.1)

7%

364

 IMCI: diagnosis score

0.68

0.05 (− 0.02 to 0.12)

7%

1240

0.59

0.12 (− 0.04 to 0.28)

21%

272

 IMCI: correct treatment

0.64

0.06 (− 0.04 to 0.16)

9%

1304

0.64

 − 0.06 (− 0.17 to 0.06)

 − 9%

294

 IMCI: no unnecessary treatment

0.60

0.07 (− 0.05 to 0.19)

11%

422

0.52

 − 0.04 (− 0.31 to 0.22)

 − 8%

74

 Family planning consultation score

0.67

0.15 (0.06–0.23)

16%

575

0.89

0.06 (− 0.04 to 0.16)

7%

159

 Delivery score

0.60

0.03 (− 0.04 to 0.1)

5%

259

0.67

0.07 (− 0.01 to 0.15)

10%

135

 Postpartum care score

0.07

0.03 (− 0.05 to 0.12)

45%

278

0.02

0.17 (0.02–0.31)

1063%

141

 Newborn care score

0.74

0.04 (− 0.08 to 0.16)

6%

279

0.76

0.11 (− 0.01 to 0.22)

14%

143

Average impacta

0.55

0.04 (0.020.07)

7%

 

0.62

0.06 (0.030.09)

10%

 

Panel c: non-technical process quality

 Antenatal care respect index

0.80

 − 0.03 (− 0.1 to 0.03)

 − 4%

1489

0.81

0.03 (− 0.1 to 0.15)

3%

283

 Antenatal care user satisfaction

0.91

0.01 (− 0.02 to 0.04)

1%

1475

0.87

 − 0.01 (− 0.09 to 0.07)

 − 1%

281

 Child curative care respect index

0.75

0.01 (− 0.07 to 0.1)

2%

1631

0.76

0.08 (− 0.01 to 0.17)

10%

364

 Child curative care user satisfaction

0.88

0 (− 0.04 to 0.04)

0%

1610

0.83

0.01 (− 0.07 to 0.09)

1%

362

 Family planning consultation respect index

0.82

0.09 (0.03–0.16)

12%

575

0.81

0.05 (− 0.06 to 0.16)

6%

159

 Family planning consultation user satisfaction

0.91

0.01 (− 0.04 to 0.06)

1%

563

0.89

0.01 (− 0.06 to 0.07)

1%

155

 Delivery care respect index

0.96

0.04 (0–0.08)

4%

294

0.95

0.01 (− 0.05 to 0.06)

1%

148

Average impacta

0.88

0.02 (− 0.01 to 0.04)

2%

 

0.88

0.02 (− 0.01 to 0.05)

2%

 

Panel d: service fees (in Congolese Francs)

 ANC fee reported by facility

2493

 − 720 (− 1270 to − 170)

 − 29%

289

2540

425 (− 1402 to 2252)

17%

54

 Delivery fee reported by facility

8734

 − 1733 (− 3606 to 141)

 − 20%

288

11,788

 − 1959 (− 5764 to 1847)

 − 17%

56

 Family planning consultation fee reported by facility

574

 − 198 (− 524 to 128)

 − 34%

228

1000

 − 545 (− 1472 to 382)

 − 54%

52

 Curative care fee reported by facility (not specific for child care)

3120

638 (− 37 to 1313)

20%

244

4438

 − 567 (− 4104 to 2970)

 − 13%

16

 ANC fees reported in exit interviews

1976

149 (− 660 to 958)

8%

1489

3221

1098 (− 944 to 3139)

34%

283

 Child curative care fees reported in exit interviews

4997

 − 1006 (− 1897 to − 116)

 − 20%

1631

13,917

 − 1238 (− 4460 to 1984)

 − 9%

364

 Family planning consultation fee reported in exit interviews

704

 − 30 (− 608 to 548)

 − 4%

577

1980

 − 819 (− 2531 to 892)

 − 41%

159

Average impacta

2,355

 − 248 (− 657 to 162)

 − 11%

 

2,574

 − 330 (− 973 to 314)

 − 13%

 

Panel e: fee policies and user satisfaction with fees

 Flat fees

0.51

0.22 (0.08–0.36)

43%

290

0.58

0.14 (− 0.13 to 0.4)

24%

56

 Fees posted

0.63

0.17 (0.06–0.28)

27%

290

0.69

0.29 (0.09–0.48)

41%

56

 Fee exemptions for poor users

0.67

0.19 (0.1–0.29)

29%

287

0.77

0.05 (− 0.16 to 0.27)

7%

56

 Satisfaction with service affordability in antenatal care exit interviews

0.91

0.11 (0.01–0.2)

12%

1266

0.87

0.17 (0.05–0.3)

20%

259

 Satisfaction with service affordability in child care exit interviews

0.88

0.03 (− 0.03 to 0.08)

3%

1560

0.83

0.03 (− 0.15 to 0.2)

3%

359

 Satisfaction with service affordability in family planning exit interviews

0.91

0.12 (− 0.05 to 0.29)

13%

245

0.89

0.19 (− 0.05 to 0.44)

22%

88

Average impacta

0.70

0.13 (0.060.20)

18%

 

0.67

0.15 (0.080.23)

22%

 

Panel f: facility management

 Protocols index

0.43

0.11 (0.03–0.19)

25%

290

0.66

0.08 (− 0.02 to 0.19)

13%

56

 Reporting index

0.76

0.06 (0–0.12)

8%

197

0.93

0.04 (− 0.03 to 0.11)

4%

39

 Posting of infection and control procedures index

0.25

0.19 (0.06–0.32)

76%

264

0.54

0.26 (0.07–0.45)

48%

55

 Mechanism to Seek patients’ opinions

0.60

0.07 (− 0.07 to 0.21)

12%

290

0.54

0.22 (− 0.02 to 0.46)

41%

56

Average impacta

0.57

0.09 (0.040.15)

16%

 

0.73

0.11 (0.020.21)

15%

 

Panel g: providers’ satisfaction

      

 Satisfied with information on own performance

0.77

0.02 (− 0.06 to 0.1)

3%

1129

0.64

0.09 (− 0.07 to 0.25)

14%

275

 Satisfied with level of autonomy

0.79

 − 0.05 (− 0.12 to 0.02)

 − 6%

1146

0.74

0 (− 0.12 to 0.12)

0%

278

 Satisfied with the relationship with the local facility committee

0.74

0.01 (− 0.05 to 0.07)

1%

1019

0.66

0.08 (− 0.04 to 0.21)

13%

225

 Satisfied with support from supervisor

0.75

0.03 (− 0.03 to 0.1)

4%

1153

0.71

0.09 (− 0.04 to 0.22)

12%

274

 Satisfied with recognition received from supervisor

0.79

0.02 (− 0.04 to 0.07)

2%

1163

0.77

0.05 (− 0.04 to 0.14)

6%

277

 Satisfied with reward received from work

0.51

0.02 (− 0.07 to 0.1)

3%

1068

0.43

0.1 (− 0.07 to 0.28)

24%

250

 Satisfied with ability to use skills

0.75

 − 0.03 (− 0.09 to 0.03)

 − 4%

1147

0.69

0.05 (− 0.08 to 0.18)

8%

275

 Satisfied with training opportunities

0.56

 − 0.01 (− 0.11 to 0.08)

 − 2%

1003

0.64

 − 0.04 (− 0.19 to 0.1)

 − 7%

251

 Satisfied with security at the facility

0.86

0.02 (− 0.04 to 0.08)

3%

1080

0.81

0.02 (− 0.08 to 0.11)

2%

265

 Satisfied with work conditions

0.55

 − 0.03 (− 0.12 to 0.06)

 − 5%

1169

0.45

0.03 (− 0.12 to 0.17)

6%

278

 Satisfied with leave

0.66

 − 0.04 (− 0.12 to 0.05)

 − 5%

1015

0.79

 − 0.07 (− 0.17 to 0.02)

 − 9%

267

 Satisfied with work hours

0.72

0.02 (− 0.08 to 0.11)

2%

1166

0.70

0 (− 0.12 to 0.12)

0%

279

 Satisfied with teamwork

0.92

 − 0.01 (− 0.05 to 0.03)

 − 1%

1172

0.89

0.02 (− 0.06 to 0.09)

2%

276

 Satisfied with relationship with facility management

0.86

 − 0.03 (− 0.09 to 0.02)

 − 4%

1138

0.87

 − 0.01 (− 0.09 to 0.06)

 − 2%

263

 Satisfied with income overall

0.26

0 (− 0.09 to 0.1)

0%

1136

0.19

0.04 (− 0.09 to 0.17)

22%

271

 Satisfied with potential for promotion

0.57

0 (− 0.09 to 0.09)

0%

1066

0.57

0.04 (− 0.09 to 0.18)

7%

256

Average impacta

0.75

0 (− 0.02 to 0.01)

0%

 

0.72

0.02 (− 0.01 to 0.05)

3%

 
  1. Notes: analysis of data from the health facility assessments. The impact coefficient is estimated with multivariate regression models in which the outcome is regressed on the PBF treatment, controlling for randomization block (province). Standard errors are clustered at the health zone level. For the indicators in panel b, on technical process quality, the regression models include controls for facility, patient, and health worker characteristics. More information on the regression models is presented in Additional file 1
  2. aAverage impacts are based on random effects meta-analysis pooling all the indicators included in each panel. The weight assigned to each indicator is presented in Additional file 1
  3. bThe “% change” equals the estimated PBF impact divided by the mean in control group. For the meta-analysis coefficient, the mean in control group is computed by weighing each individual indicator with the weights generated by the meta-analysis regression