Table 1 Characteristics of included primary studies

1

Shanghai/Zhang 2010 [13]

CBA

2004–2005/Tertiary A hospital

Inpatient with Shanghai medical insurance/14,000 overall

15 targeted diseases, detail unreported

DRGs payment for insured patients in 1 hospital

FFS payment for uninsured patients in the same hospital

DID analysis, DDD analysis, regression analysis

1) expenditure per admission,2) length of stay, 3) equity of above indicators between insured and uninsured patients

3

Beijing/Jian 2015b (Jian 2015c) [11]

CBA

Jan 2010- Sep 2012/Tertiary A hospital

Inpatient with Beijing basic employee medical insurance/ 318,884 vs. 294,989

108 DRGs with CV < 0.85, detail unreported

DRGs payment, reimbursement ceiling, allowing for 5% annual increase in 6 hospitals

FFS payment in 8 hospitals

DID analysis, regression analysis

1) expenditure per admission, 2) out of pocket payment, 3) length of stay, 4) readmission, 5) equity, cost shifting and patient selection

4

Beijing/Zhang 2015 (Hu 2013, Hu 2014, Wu 2013, Song 2014, Jian 2015a, Tian 2015) [43]

CBA

2012–2013/Tertiary A hospital

Inpatient with urban employee medical insurance/118,091 vs. 120,427

108 DRGs with CV < 0.85, detail unreported

DRGs payment under global budget in 6 hospitals

FFS payment under global budget in 8 hospitals

Comparison of means before and after pilot, no formal statistical test

1) expenditure per admission, 2) out of pocket payment,3) length of stay, 4) 2 weeks readmission, 5) patient selection

9

Changsha, Hunan province/Zhang 2016 [14]

CBA

2013/2 primary,12 secondary and 6 tertiary hospitals

Inpatient with urban employee medical insurance/75 vs. 133

32 groups, detail unreported. Only patients with uncomplicated acute appendicitis were analysed

DRGs payment in 8 hospitals

FFS payment in 12 hospitals

Comparison of means, Student’s t-test, Pearson’s chi-square test

1) expenditure per admission, 2) length of stay,3) use of antimicrobials

5

Beijing/Poon 2017 [12]

CBA

Jan 2010 - Sep 2012/Tertiary hospital

Inpatient with basic employee medical Insurance/1374 vs. unreported

108 DRGs, detail unreported. Patients with acute myocardial infarction were analysed

DRGs payment in 6 hospitals

FFS payment in 8 hospitals

DID analysis, regression analysis

1) expenditure per admission, 2) in hospital mortality, 3) length of stay, 4) prescription of optimal AMI medications at arrival

6

Beijing/Ji 2017 (Zhang LH 2015) [45]

CBA

2011–2015/Secondary hospital

Inpatient with new rural cooperative medical insurance/unreported

All groups (from 560 to 577 at 2011 and 2014)

DRGs payment in 1 hospital

FFS payment in 10 hospitals

Comparison of means, no formal statistical test

1) expenditure per admission, 2) out of pocket payment %, 3) length of stay

7

Harbin, Heilongjiang province/Wang 2015 [39]

ITS

Aug 2010- July 2012/ Secondary A hospital

Inpatient with urban and rural resident medical insurance/213 vs. 251

36 DRG groups, detail unreported. Only patients with cholecystotomy were analysed

1) DRGs payment from Aug 2008, 2) excluding transferred patient or those under deductible or above ceiling of insurance payment in 1 hospital

FFS payment in the same hospital before the reform

t test, interrupted time series analysis

1) expenditure per admission, 2) out of pocket payment, 3) length of stay, 4) floated coding practice

8

Guangxi province/Wu 2015a (Wu 2015b) [41]

ITS

Aug 2010- July 2012/ Secondary A hospital

Inpatient with urban and rural resident medical insurance underwent herniorrhaphy/131 vs. 119

36 DRG groups including 17 diseases and 19 surgical groups, detail unclear.

DRGs payment from Aug 2011 in 1 hospital

FFS payment in the same hospital before the reform

time series analysis (ARIMA)

1) expenditure per admission, 2) out of pocket payment per capita, 3) length of stay

2

Tianjin/Li 2012 [31]

BA

2006–2012/Tertiary A hospital

Inpatient with maternity medical insurance/ 3232 vs. 5712

hospital delivery

DRGs payment from July 2009 in 1 hospital

FFS payment before July 2009 in the same hospital

Comparison of means (Mann-Whitney U test), and rate (chi square test)

1) expenditure per admission,2) length of stay, 3) caesarean rate

10

Lufeng, Yunnan province/Peng 2016 (Li 2013) [44]

BA

2012–2013/County (secondary) hospital

Inpatient with new rural cooperative medical insurance/35272 vs.32369

All patients (432 groups)

DRGs payment from Oct 2012, top 3% patients with highest expense paid by FFS payment, and penalty for readmission within 7 days in 3 hospitals

FFS payment in the same hospitals before the reform

Comparison of means before and after the pilot, no formal test

1) expenditure per admission, 2) length of stay

11

Xiangyun, Yunnan province/Peng 2017 [46]

BA

2014–2015/County (secondary) hospital

Inpatient with new rural cooperative medical insurance/19,479 vs. unreported

All patients (434 groups at 2014, 304 groups at 2015)

DRGs payment in 1 hospital from August, 2014

FFS payment in the same hospital before the reform

Comparison of means, before (2014) and after the pilot (2015), no formal test

1) expenditure per admission, 2) length of stay

12

Yuxi, Yunnan province/Yan 2017 [47]

BA

2015–2016/County (secondary) hospital

Inpatient with new rural cooperative medical insurance/unreported

All patients (493groups)

DRGs payment in 9 hospitals from 2016

FFS payment in the same hospitals before the reform

Comparison of means before and after pilot, no formal test

1) expenditure per admission, 2) length of stay

13

Yuxi, Yunnan province/Zhou 2018 [48]

BA

Jan - Oct 2017/Tertiary hospital

Inpatient with urban employee or resident medical insurance/36,827 vs. unreported

All patients (531 groups)

DRGs payment in 1 hospital

FFS payment under global budget in the same hospital before the reform

Comparison of means, before (2016) and after pilot, no formal test

expenditure per admission