The effects of diagnosis-intervention packet payment on healthcare in Chinese public hospitals: a systematic review and meta-analysis

doi:10.21203/rs.3.rs-4875814/v1

Download PDF

Article

The effects of diagnosis-intervention packet payment on healthcare in Chinese public hospitals: a systematic review and meta-analysis

https://doi.org/10.21203/rs.3.rs-4875814/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

To control the rapid growth of health expenditures, China developed an innovative case-based payment system for inpatient care called the diagnosis-intervention packet (DIP) payment. However, the effectiveness of DIP payment in public hospitals is mixed. This study aims to conduct a systematic review and meta-analysis to investigate the effects of DIP-based payment on healthcare in terms of quantity, efficiency, quality, and expenditure of public hospitals in China. We searched PubMed, Embase, the Cochrane Library, Web of Science, EBSCO, ProQuest, Opengrey, the China National Knowledge Infrastructure, Wanfang, cqvip, and China Biology Medicine disc from October 2003 to December 2023. Twenty studies were identified and examined, including six non-randomized controlled trials, ten controlled before-after studies, and four interrupted time series studies. The meta-analysis revealed that DIP payment was effective in improving efficiency by shortening the length of stay. There was no significant overall effect on the quantity of healthcare. Based on the results of in-hospital mortality and readmission within 30 days, quality was partially improved. The meta-analysis did not show a significant decrease in total expenditures per case, but there were some positive changes in the structure of hospitalization expenditures. There is insufficient evidence to establish a definitive relationship between DIP-based payment and healthcare in Chinese public hospitals. Future research should explore the long-term effects of the DIP payment reform. Before applying and implementing the DIP-based payment, policymakers need to consider its positive and negative outcomes. This can help policymakers make better decisions on provider payment reform to improve hospital healthcare.

Health sciences/Health care/Health care economics

Health sciences/Health care/Health policy

Health sciences/Health care/Health services

Diagnosis-intervention packet

provider payment method

public hospitals

China

systematic review

meta-analysis

Health spending has increased substantively in many countries. Between 1995 and 2016, health spending grew at an annual rate of 4.0%¹. Global health spending is projected to reach $10.6 trillion in 2030 and further increase to $15.0 trillion by 2050¹. Cost containment has become a major concern for contemporary healthcare systems that are aimed at providing more economical and efficient healthcare². Hospitalization expenses are the largest component of healthcare expenditures in all countries regardless of their income level, so the burden of cost containment is continually shifted onto controlling the expenses of hospitalized members^3,4. According to the published findings, the Asia-Pacific region is increasingly capturing the vast majority share of global medical supply and demand⁵. Moreover, underlying health spending patterns in many low- and middle-income countries (LMICs) are causing a sustainability crisis⁶. To address the issues above, countries have sought to allocate healthcare resources, control expenditure growth, and achieve health system goals by implementing provider payment reforms⁷.

Over the past decades, China’s healthcare spending has risen steadily. Total expenditure reached RMB 8.4 trillion in 2022, more than doubling from 2015⁸. Public hospitals are the main body of China’s healthcare delivery system. Regarding the distribution of medical resources and health services, public hospitals accounted for 80% of the national resources⁹ and provided more than 84% of the hospital discharges¹⁰. To control medical costs and improve healthcare accessibility in public hospitals, the Chinese government launched the national Diagnosis Related Group (DRG) pilot program in 2019. Although the effectiveness of DRG-based payment in reducing healthcare expenditures has been demonstrated¹¹, its rollout and implementation have been slow since patient classification requires high-quality data from medical record front pages and the judgment of physicians. This challenge is also wide and common in LMICs with limited resources¹².

With the aim of regulating provider behaviors and cost management, China developed a new case-based payment system for inpatient care under the regional global budget, replacing retrospective payments with prospective payments for hospital charges. The system, a simplified version of DRG payment, is called the diagnosis-intervention packet (DIP)¹². Unlike the DRG-based payment, DIP payment primarily uses the first four digits of principal diagnosis codes (International Classification of Diseases, Tenth Revision, ICD-10) and procedure codes (International Classification of Diseases, Ninth Revision, Clinical Modification, ICD-9-CM3) for case classification, while patient demographic and administrative characteristics (such as age, gender, discharge status, complications and comorbidities) are not considered to classify patients. According to historical cost data, relative weights are assigned to patient groups to indicate the consumption of market resources close to different groups. Relative weights are then converted to payments based on the market’s global insurance budget¹³. Furthermore, the DIP classification circumvents the major challenges of DRG-based payment¹². Because of its unique characteristics in patient classification¹⁴, DIP payment requires lower data quality requirements and less reliance on physician judgment¹³. Each coordinating region dynamically adjusts and expands the disease categories according to the actual situation. For the uneven distribution of medical resources in China, DIP payment can be widely promoted and implemented in a short time¹⁴.

Several empirical studies have analyzed the effects of the DIP payment reform. However, there is a lack of systematic reviews to comprehensively evaluate the impacts of DIP-based payment due to the inconsistency of regions, participants, times, results, and conclusions in different studies. Therefore, the objective of this study was to systematically review the effects of DIP-based payment on healthcare with regard to quantity, efficiency, quality, and expenditure in Chinese public hospitals, which may be more relevant and helpful to inform health policy making for other LMICs in the same scenario.

The systematic review was reported according to the latest Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines¹⁵ and registered on the International Prospective Register of Systematic Reviews (PROSPERO: CRD42024503159)¹⁶.

Search strategy

Eleven health and social science databases were searched from October 2003 to December 2023, including PubMed, Embase, the Cochrane Library, Web of Science, EBSCO, ProQuest, Opengrey, the China National Knowledge Infrastructure, Wanfang, cqvip, and China Biology Medicine disc. Reference lists of the included studies and relevant reviews were also screened for additional studies. The search strategy was firstly designed in PubMed using MeSH and free-text terms and adapted for other databases with Boolean operators and truncations. The complete search strategies can be found in Supplementary Appendix 1.

Eligibility criteria

Studies that met the following criteria were eligible for inclusion. (1) Type of participants: public hospitals in China. (2) Type of studies: study designs approved by the Cochrane Effective Practice and Organization of Care Group (EPOC) which included randomized controlled trials (RCTs), non-randomized controlled trials (non-RCTs), controlled before-after (CBA) studies, interrupted time series (ITS) studies and repeated measures studies. ITS studies were included only if there was a clearly defined point in time when the intervention occurred and at least three data points before and three after the intervention. (3) Type of interventions: diagnosis-intervention packet payment that can be applied in tertiary, secondary or primary care settings to reimburse inpatient services. (4) Type of comparisons: any other payment methods, such as fee-for-service (FFS), salary, pay-for-performance, or their combination. (5) Types of outcomes: healthcare outcomes in quantity, efficiency, quality, and expenditure.

Study selection

Two reviewers (T.X. and L.J.) screened the titles and abstracts of the articles based on the inclusion and exclusion criteria. Full texts were obtained and screened when the articles were relevant or unsure in the first step. The entire process was screened independently by two reviewers (T.X. and L.J.). Any disagreements on inclusion were resolved by discussion with a third reviewer (C.L.).

Data extraction

Two reviewers (T.X. and L.J.) separately carried out data extraction using a standardized data collection form¹⁷ and cross-checked for errors. The information we mainly extracted included first author, publication year, study design, study period, implementation condition, participants, and outcomes. A third reviewer (C.L.) investigated discrepancies. The data were managed in Excel and Word.

Assessment of risk of bias

Two reviewers (T.X. and L.J.) independently described what was reported in the study, commented on the description, and judged the risk of bias. For non-randomized controlled trials, we assessed the quality using the Risk of Bias In Non-randomized Studies of Interventions (ROBINS-I)¹⁸. The quality assessment consists of seven domains. Then, an overall risk of bias rating was generated: low, moderate, serious, critical, and no information¹⁹. For randomized controlled trials, controlled before-after studies, and interrupted time series studies, criteria recommended by EPOC were used to assess the risk of bias²⁰. The study included was classified as having a high risk of bias, low risk of bias or unclear. If all domains were assessed as ‘low risk’ for the outcome in a study, the overall assessment was categorized as ‘low risk.’ If one or more key criteria were scored as ‘unclear,’ the summary assessment was considered to have an unclear risk of bias. If one or more key areas were rated as ‘high risk,’ the overall assessment was classified as having a high risk of bias. Variation in opinion between review authors (T.X., L.J. and C.L.) was resolved through discussion.

Data synthesis

We conducted a structured synthesis and categorized the comparisons and outcomes. We calculated the risk ratios (RR) with 95% confidence intervals (CIs) for dichotomous outcomes and standardized mean difference (SMD) with 95% CIs for continuous outcomes. The formula used to convert from absolute to relative difference change is shown in Supplementary Appendix 2. For CBA studies, the effect estimates derived from difference in differences (DID) model were considered step changes (see Supplementary Appendix 3)²¹. For ITS studies, the results were presented as changes along two dimensions: step change (indicating the immediate effect of DIP payment) and change in slope (reflecting the sustained effect of DIP payment), which were summarized in separate analyses (see Supplementary Appendix 4). We conducted a meta-analysis of the included studies if more than one study had similar comparisons and outcomes. A fixed-effect model was used for meta-analysis if there was no statistical heterogeneity between studies (P ≥ 0.05 or I² ≤ 50%), and a random-effect model was used if heterogeneity was statistically significant (P < 0.05 and I² > 50%). Heterogeneity between results was assessed with the Higgins I^{2 22}. The results are reported as pooled effects and forest plots. The funnel plot and Begg’s test were used to show the publication bias if data allowed²³. The cut-off point for statistical significance was P ≤ 0.05. The analyses were conducted by using Review Manager 5.4.

The search yielded 2890 records from the databases and websites. After removing duplicates, 1767 records were title and abstract screened and 36 records were full-text screened. Eventually, twenty studies were included for further investigation, of which five studies were published in English and fifteen were in Chinese (Fig. 1). Sixteen full-text articles were excluded for study participants, study design, data availability, and outcome indicators. The characteristics of the excluded studies are reported in Supplementary Appendix 5.

Characteristics of included studies

The characteristics of the included studies are summarized in Table 1. Among the twenty studies, there were six non-RCTs, ten CBA studies, and four ITS studies. Almost all studies (n = 19) were published between 2019 and 2023. At the hospital level, there were ten studies conducted in tertiary hospitals, two studies in secondary hospitals, two studies in secondary and tertiary hospitals, and four studies in primary, secondary and tertiary hospitals. The data for eleven studies were sourced from hospital information systems and eight studies with data from healthcare security administration. DIP-based systems varied in implementation conditions, including implementation details and implementation time. Nine studies reported the full adoption of DIP payment, while eleven studies reported the implementation of a localized pilot. The interventions ranged from 12 to 24 months, and eleven studies lasted 24 months. All studies compared DIP payment with FFS payment methods under the global budget. In total, one study (5%) reported changes in healthcare quantity, sixteen studies (80%) in efficiency, four studies (20%) in quality and eighteen studies (90%) in expenditure.

Risk of bias in included studies

Of the six non-RCTs, one study was assessed as low risk of bias, while the risk of bias for confounding from baseline differences and selection of participants was unclear in the other five studies. None of the ten CBA studies reported information on random sequence generation and allocation concealment. Four studies lacked sufficient information regarding outcome measurements, and seven studies did not include adequate information on baseline characteristics. Among the ITS studies, one study had a high risk of bias that incomplete outcome data might not have been adequately addressed. The assessments of risk of bias for the included studies can be found in Supplementary Appendix 6.

Effects of interventions

As shown in Fig. 2 and Supplementary Appendix 7, the study reported the following outcomes for the characteristics and estimated effects of DIP payment on healthcare in public hospitals: (a) Quantity, including annual number of discharges; (b) Efficiency, including length of stay (LOS); (c) Quality, including in-hospital mortality, readmission within 30 days, operation associated infection and postoperative complication; (d) Expenditure, including total expenditures per case, out-of-pocket payments per case, proportion of the out-of-pocket payments per case, drug expenditures per case, proportion of the drug expenditures per case, medical consumables expenditures per case, treatment expenditures per case, service expenditures per case, examination expenditures per case and other expenditures per case.

Quantity outcomes

In order to evaluate the quantity, the annual number of discharges was measured. Only one CBA study reported the quantity outcomes³⁵. According to Supplementary Appendix 7, there was no significant correlation between DIP payment and annual number of discharges compared with FFS.

Efficiency outcomes

For the purpose of assessing the efficiency outcomes, length of stay was evaluated in the studies. Sixteen studies evaluated changes in LOS after DIP payment, including four non-RCTs^24–27, nine CBA studies^{13,30–34,36−38} and three ITS studies^12,40,41. Eleven studies reported reduction of LOS (Supplementary Appendix 7)^{12,24–27,30.31,33,34,37,40}. However, one CBA¹³ and one ITS study⁴¹ found that the LOS increased after piloting DIP payment. Another CBA study³⁶ found mixed results in LOS, i.e., DIP payment led to an increase of 0.228 days in the typical diagnosis and a decrease of 0.568 days in the typical surgical procedures.

Quality outcomes

In order to assess the quality, measures such as in-hospital mortality, readmission within 30 days, operation associated infection, and postoperative complication were evaluated in the included studies. Three studies reported in-hospital mortality, of which one CBA study³⁶ showed that the rate decreased, while other two ITS studies^12,40 showed that the rate increased after the implementation of the DIP. Furthermore, each outcome indicator had only one CBA study reporting the association between DIP payment and them which were the readmission within 30 days³⁶, operation associated infection¹³, and postoperative complication¹³. According to Supplementary Appendix 7, it could be observed that these outcome indicators decreased after the DIP payment.

Expenditure outcomes

For the purpose of evaluating the expenditure outcomes, eight indicators were assessed in the studies, such as total expenditures per case, out-of-pocket payments per case, etc. Eighteen of the examined studies, namely six non-RCTs^24–29, eight CBA studies^{13,30–34,37,38} and four ITS studies^12,39–41, evaluated changes in total expenditures per case after the implementation of DIP payment. The reported effects of DIP payment on the total expenditures per case were mixed. Eleven studies found a decrease in total expenditures per case^{24–30,33,34,37,39} while two CBA^13,31 and two ITS studies^12,40 reported an increase. Another one ITS study⁴¹ indicated that total expenditures per case in tertiary hospitals keep decreasing, opposite to that in secondary hospitals. More detailed information on other expenditure indicators can be seen in Fig. 2 and Supplementary Appendix 7.

Meta-analysis

Considering the similarity in the intervention contents of DIP payment, participants and outcomes, the meta-analysis in this review included ten studies with six evaluation indicators. A summary of the meta-analysis results is shown in Table 2. The related forest plots for the six outcome indicators are displayed in Fig. S1-Fig. S6 (see Supplementary Appendix 8).

First, eight studies^{12,13,24–27,36,40} with 14547405 participants reported the essential data on LOS. These studies showed that DIP payment was effective for decreasing LOS (pooled effect: SMD = -0.30, 95%CI = -0.37 to -0.23, Z = 8.14, P<0.001) when compared with FFS payment (Fig. S1).

Second, three studies^12,36,40 with 8713540 participants displayed the effects of DIP-based payment on in-hospital mortality. There was a slight decrease (pooled effect: SMD = -0.02, 95%CI = -0.03 to -0.01, Z = 4.72, P <0.001) (Fig. S2).

Third, nine studies^{12,13,24–29,40} with 14483706 participants assessed the effects of DIP payment on total expenditures per case. As seen in Fig. S3, the meta-analysis did not show a significant decrease (pooled effect: SMD = -0.02, 95%CI = -0.05 to 0.00, Z = 1.61, P = 0.11). For Out-of-pocket payments per case, three studies^26,29,40 with a total of 3829180 participants were included in the meta-analysis (Fig. S4). The results found that DIP payment was associated with a significant decrease (pooled effect: SMD = -4.09, 95% CI = -5.68 to -2.49, Z = 5.02, P < 0.001). In addition, the meta-analysis for drug expenditures per case^24–26,28 and proportion of the drug expenditures per case^24,25 both showed that compared to FFS payment, DIP payment was effective for decreasing drug expenditures per case (pooled effect: SMD = -1.56, 95% CI = -2.22 to -0.90, Z = 4.65, P < 0.001) (Fig. S5) and proportion of the drug expenditures per case (pooled effect: SMD = -0.81, 95% CI = -1.34 to -0.28, Z = 2.99, P = 0.003) (Fig. S6). Due to significant statistical heterogeneity between studies (P < 0.05 and I² > 50%), a random-effect model was used for meta-analysis.

Additionally, for studies that could not be used for meta-analysis, the effects of DIP payment were categorized and summarized according to the study design. Tables 3 and 4 show the descriptive results for ITS and CBA studies, and all outcome indicators were continuous variables.

Table 1

Characteristics of included studies
Study	Study design	Hospital level	Data sources	Implementation condition of DIP	Study period	Participants		Outcomes
Study	Study design	Hospital level	Data sources	Implementation condition of DIP	Study period	Control group	Intervention group	Outcomes
Zheng, 2017²⁴	non–RCT	Tertiary hospital	Hospital information system	Pilot in Qingyuan, Guangdong Province, January 2014	Pre: 2013.1–2013.12 Post: 2016.1–2016.12	19 hospitalized patients with appendicitis, 18 hospitalized patients with senile cataracts, 13 hospitalized patients with inguinal hernia	20 hospitalized patients with appendicitis, 17 hospitalized patients with senile cataracts, 13 hospitalized patients with inguinal hernia	②Efficiency; ④Expenditure
Ding et al, 2019²⁵	non–RCT	Tertiary hospital	Hospital information system	Pilot in Yinchuan, Ningxia Hui Autonomous Region, January 2015	Pre: 2014.1–2014.12 Post: 2016.1–2016.12	200 hospitalized patients with cerebral infarction, 200 hospitalized patients with non-insulin-dependent diabetes mellitus	200 hospitalized patients with cerebral infarction, 200 hospitalized patients with non-insulin-dependent diabetes mellitus	②Efficiency; ④Expenditure
Wang et al, 2022²⁶	non–RCT	Tertiary hospital	Hospital information system	Pilot in Suqian, Jiangsu Province, June 2021	Pre: 2020.6–2021.5 Post: 2021.6–2022.6	219 discharges with pterygium	230 discharges with pterygium	②Efficiency; ④Expenditure
Wu, 2022²⁷	non–RCT	Tertiary hospital	Hospital information system	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: before January 1, 2018 Post: after January 1, 2018	38000 discharges before 2018	34490 discharges after 2018	②Efficiency; ④Expenditure
Liu et al, 2022²⁸	non–RCT	Secondary hospital	Hospital information system	Pilot in Hubei Province, July 2021	Pre: 2020.6–2021.6 Post: 2021.7–2022.6	40 hospitalized patients with traumatic craniocerebral injury	40 hospitalized patients with traumatic craniocerebral injury	④Expenditure
Chen, 2022²⁹	non–RCT	Secondary hospital	Hospital information system	Pilot in Yichang, Hubei Province, November 2020	Pre: 2020.5–2021.5 Post: 2021.6–2022.5	120 hospitalized patients in the endocrinology department	68 hospitalized patients in the endocrinology department	④Expenditure
Zheng et al, 2020³⁰	CBA	Tertiary hospital	Hospital information system	Pilot in Xiamen, Fujian Province, July 2016	Pre: 2014.7.1–2016.6.30 Post: 2016.7.1–2018.6.30	Other medical insurance patients of a tertiary hospital in Xiamen, Fujian Province	Local medical insurance patients of a tertiary hospital in Xiamen, Fujian Province	②Efficiency; ④Expenditure
Zheng et al, 2020³¹	CBA	Tertiary hospital	Hospital information system	Pilot in Xiamen, Fujian Province, July 2016	Pre: 2014.7.1–2016.6.30 Post: 2016.7.1–2018.6.30	All discharges with a diagnosis containing a malignant tumor paid for by other medical insurance of a tertiary hospital	All discharges with a diagnosis containing a malignant tumor paid for by Xiamen medical insurance of a tertiary hospital	②Efficiency; ④Expenditure
Yang et al, 2021³²	CBA	Tertiary hospital	Hospital information system	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2016.1–2017.12 Post: 2018.1–2019.12	169569 discharges paid for by other medical insurance	117264 discharges paid for by Guangzhou medical insurance	②Efficiency; ④Expenditure
Qian et al, 2021¹³	CBA	Primary, secondary and tertiary hospitals	Healthcare Security Administration of Guangzhou	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2016.1–2017.12 Post: 2018.1–2019.12	2398936 uninsured discharges from 309 hospitals in Guangzhou	5760026 insured discharges from 309 hospitals in Guangzhou	②Efficiency; ③Quality; ④Expenditure
Hu et al, 2022³³	CBA	Tertiary hospital	Hospital information system	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2016.1–2017.12 Post: 2018.1–2019.12	7717 discharges with cardiovascular diseases paid for by other medical insurance	57073 discharges with cardiovascular diseases paid for by Guangzhou medical insurance	②Efficiency; ④Expenditure
Lai et al, 2022³⁴	CBA	NR	Health information centers of Guangzhou and Beijing	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2017.7.16–2017.12.31 Post: 2018.1.1–2019.12.31	1% insured local patients in Beijing from July 16, 2017 to December 31, 2018	2% insured local patients in Guangzhou from July 16, 2017 to December 31, 2018	②Efficiency; ④Expenditure
Lin et al, 2023³⁵	CBA	Primary, secondary and tertiary hospitals	Healthcare Security Administration and Health Commission of Hubei Province and Yichang	Pilot in Yichang, Hubei Province, November 2020	Pre: 2018–2020 Post: 2021–2022	192 public hospitals in non-pilot cities, Hubei Province	26 public hospitals in Yichang, Hubei Province	①Quantity
Zhang et al, 2023³⁶	CBA	Secondary and tertiary hospitals	Healthcare Security Administration of Shanghai	Pilot in Shanghai, July 2019	Pre: 2017.7–2019.6 Post: 2019.7–2021.6	Hospitalized patients covered by the urban employee basic medical insurance scheme in 8 tertiary general public hospitals and 19 secondary public hospitals (13 general hospitals, 5 Chinese medicine hospitals, and 1 specialized hospital) in 5 districts that have not launched the pilot DIP payment scheme	Hospitalized patients covered by the urban employee basic medical insurance scheme in 2 tertiary general public hospitals and 11 secondary public hospitals (5 general hospitals, 3 Chinese medicine hospitals, and 3 specialized hospitals) in 2 districts that have launched the pilot DIP payment scheme	②Efficiency; ③Quality
Ye et al, 2023³⁷	CBA	Tertiary hospital	Hospital information system	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2017.7.15–2017.12.31 Post: 2018.1.1–2019.3.31	Hospitalized patients with cerebrovascular diseases in a large tertiary hospital paid for by other medical insurance	Hospitalized patients with cerebrovascular diseases in a large tertiary hospital paid for by Guangzhou medical insurance	②Efficiency; ④Expenditure
Deng et al, 2023³⁸	CBA	Tertiary hospital	Healthcare Security Administration of Heilongjiang	Pilot in Jiamusi, Hegang and Yichun, Heilongjiang Province, December 2021	Pre: 2021.1–2021.12 Post: 2022.1–2022.12	4 tertiary hospitals in the non-national DIP pilot cities of Jixi and Shuangyashan, including 2 general hospitals and 2 specialized hospitals	4 tertiary hospitals in the national DIP pilot cities of Jiamusi, Hegang, and Yichun, including 2 general hospitals and 2 specialized hospitals	②Efficiency; ④Expenditure
Han et al, 2021³⁹	ITS	NR	Healthcare Security Administration of Zhuhai	Fully adopt in Zhuhai, Guangdong Province, January 2018	Pre: 2016.1–2017.12 Post: 2018.1–2019.12	All discharges in Zhuhai, Guangdong Province, from 2016 to 2017	All discharges in Zhuhai, Guangdong Province, from 2018 to 2019	④Expenditure
Chen et al, 2023⁴⁰	ITS	Primary, secondary and tertiary hospitals	Healthcare Security Administration of Guangzhou	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2016.1.1–2017.12.31 Post: 2018.1.1–2019.12.31	1721889 hospitalized patients covered by the urban employee basic medical insurance scheme in all contracted hospitals in Guangzhou, Guangdong Province, from 2016 to 2017	2106654 hospitalized patients covered by the urban employee basic medical insurance scheme in all contracted hospitals in Guangzhou, Guangdong Province, from 2018 to 2019	②Efficiency; ③Quality; ④Expenditure
Ding et al, 2023⁴¹	ITS	Secondary and tertiary hospitals	Healthcare Security Administration of Dongying	Pilot in Shandong Province, January 2021	Pre: 2019.1–2020.12 Post: 2021.1–2022.4	Discharges from secondary and tertiary hospitals in Dongying, Shandong Province, from January 2019 to December 2020	Discharges from secondary and tertiary hospitals in Dongying, Shandong Province, from January 2021 to April 2022	②Efficiency; ④Expenditure
Tang et al, 2023¹²	ITS	Primary, secondary and tertiary hospitals	Healthcare Security Administration of Guangzhou	Fully adopt in Guangzhou, Guangdong Province, January 2018	Pre: 2016.1.1–2017.12.31 Post: 2018.1.1–2019.12.31	1988492 discharges from 235 public hospitals in Guangzhou, Guangdong Province, from 2016 to 2017	2832538 discharges from 235 public hospitals in Guangzhou, Guangdong Province, from 2018 to 2019	②Efficiency; ③Quality; ④Expenditure
Notes: non-RCT: non-randomized trials; CBA: controlled before-after studies; ITS: interrupted time series studies; NR: not reported; Primary hospitals: mainly township hospitals in rural China and community health centers in urban areas.

Table 2

Summary of meta-analysis results for diagnosis-intervention packet
Outcomes				Meta-analysis			Heterogeneity
Outcomes		Included study (samples)	Participants	Pooled effects (95%CI)	P	Effect models	I²	P
Efficiency	Length of stay	Eight studies^{12,13,24–27,36 40}	14547405	-0.30 (-0.37, -0.23)	< 0.001	Random	100%	< 0.001
Quality	In-hospital mortality	Three studies^12,36,40	8713540	-0.02 (-0.03, -0.01)	< 0.001	Random	93%	< 0.001
Expenditure	Total expenditures per case	Nine studies^{12,13,24–29,40}	14483706	-0.02 (-0.05, 0.00)	0.110	Random	99%	< 0.001
	Out-of-pocket payments per case	Three studies^26,29,40	3829180	-4.09 (-5.68, -2.49)	< 0.001	Random	99%	< 0.001
	Drug expenditures per case	Four studies^24–26,28	1429	-1.56 (-2.22, -0.90)	< 0.001	Random	96%	< 0.001
	Proportion of the drug expenditures per case	Two studies^24,25	900	-0.81 (-1.34, -0.28)	0.003	Random	90%	< 0.001

Table 3

Results of controlled before-after studies
Outcomes	Study	Number of patients(n)	Baseline level: mean	Mean change from baseline: mean	DID estimates (SE)	Relative difference change^a, change in level
Quantity	Annual number of discharges
	Lin et al, 2023³⁵	NR	C: 26867.56	C: 40.49	-0.090	-0.0005%
		NR	I: 18585.54	I: -270.62
Efficiency	Length of stay
	Zheng et al, 2020³⁰	NR	C: 9.55	C: -0.38	-0.403 (0.086)	-4.22%
		NR	I: 9.55	I: -0.79
	Zheng et al, 2020³¹	NR	C: 10.79	C: -0.93	-0.263 (0.164)	-2.49%
		NR	I: 10.58	I: -1.31
	Yang et al, 2021³²	C: 169569	C: 7.00	C: 0.00	NR	NA
		I: 117264	I: 8.00	I: -1.00
	Qian et al, 2021¹³	C + I: 8158962	NR	NR	0.080 (0.050)	NA
	Hu et al, 2022³³	C: 7717	C: 6.33	C: -0.02	-0.053	-0.82%
		I: 57073	I: 6.50	I: -0.40
	Lai et al, 2022³⁴	C: 26279	NR	NR	-0.450 (0.196)	NA
		I: 13666
	Zhang et al, 2023³⁶
	Typical Diagnosis	C + I: 175411	NR	NR	0.228 (0.158)	NA
	Typical Surgical Procedures	C + I: 8889	NR	NR	-0.568 (0.297)	NA
	Ye et al, 2023³⁷	C: 3065	NR	NR	-0.110	NA
		I: 806	NR	NR
	Deng et al, 2023³⁸	NR	C: 12.52	C: -1.15	NR	NA
		NR	I: 10.01	I: -0.60
Quality	In-hospital mortality
	Zhang et al, 2023³⁶
	Typical Diagnosis	C + I: 175420	NR	NR	-0.005 (0.003)	NA
	Typical Surgical Procedures	C + I: 8890	NR	NR	-0.011 (0.003)	NA
	Readmission within 30 days
	Zhang et al, 2023³⁶
	Typical Diagnosis	C + I: 175420	NR	NR	-0.020 (0.009)	NA
	Typical Surgical Procedures	C + I: 8890	NR	NR	-0.049 (0.020)	NA
	Operation associated infection
	Qian et al, 2021¹³	C + I: 4002641	NR	NR	-0.002 (0.001)	NA
	Postoperative complication
	Qian et al, 2021¹³	C + I: 4002641	NR	NR	-0.003 (0.007)	NA
Expenditure	Total expenditures per case
	Zheng et al, 2020³⁰	NR	C: 13754.32	C: 1212.37	-761.120 (164.787)	-6.07%
		NR	I: 12548.19	I: 451.25
	Zheng et al, 2020³¹	NR	C: 17340.40	C: -510.53	54.640 (343.275)	+ 0.34%
		NR	I: 16055.52	I: -455.89
	Yang et al, 2021³²	C: 169569	C: 9952.00	C: 22.00	NR	NA
		I: 117264	I: 10177.00	I: -321.00
	Qian et al, 2021¹³	C + I: 4867273	NR	NR	0.043 (0.012)	NA
	Hu et al, 2022³³	C: 7717	C: 50807.66	C: 1681.04	-0.114	-232.37%
		I: 57073	I: 49060.62	I: -2076.38
	Lai et al, 2022³⁴	C: 26279	NR	NR	-0.952 (0.020)	-4.80%
		I: 13666
	Ye et al, 2023³⁷	C: 3065	NR	NR	-1657.870	NA
		I: 806	NR	NR
	Deng et al, 2023³⁸	NR	C: 8762.32	C: 1142.83	NR	NA
		NR	I: 8681.23	I: -295.3
	Out-of-pocket payments per case
	Zheng et al, 2020³⁰	NR	C: 13754.31	C: 1212.39	-1067.220 (139.653)	-20.05%
		NR	I: 5070.33	I: 144.67
	Zheng et al, 2020³¹	NR	C: 17340.40	C: -510.53	451.190 (302.434)	+ 7.27%
		NR	I: 6203.74	I: -59.34
	Hu et al, 2022³³	C: 7717	C: 18031.87	C: 883.52	NR	NA
		I: 57073	I: 26510.11	I: -2104.67
	Proportion of the out-of-pocket payments per case
	Zheng et al, 2020³⁰	NR	C: 1.000	C: 0.000	-0.028 (0.001)	-6.15%
		NR	I: 0.455	I: -0.028
	Zheng et al, 2020³¹	NR	C: 1.00	C: 0.00	0.000 (0.003)	0.00%
		NR	I: 0.38	I: 0.00
	Drug expenditures per case
	Zheng et al, 2020³⁰	NR	C: 4290.04	C: -208.39	-274.130 (60.860)	-6.60%
		NR	I: 4150.71	I: -482.52
	Zheng et al, 2020³¹	NR	C: 6433.12	C: -861.06	-385.650 (137.473)	-5.70%
		NR	I: 6763.09	I: -1246.71
	Lai et al, 2022³⁴	C: 26279	NR	NR	-0.857 (0.035)	-14.27%
		I: 13666
	Ye et al, 2023³⁷	C: 3065	NR	NR	-35.800	NA
		I: 806	NR	NR
	Medical consumables expenditures per case
	Zheng et al, 2020³⁰	NR	C: 2925.45	C: 420.65	-140.220 (77.862)	-6.26%
		NR	I: 2241.27	I: 280.43
	Zheng et al, 2020³¹	NR	C: 3124.89	C: -260.42	102.210 (113.969)	+ 4.45%
		NR	I: 2295.54	I: -158.21
	Service expenditures per case
	Lai et al, 2022³⁴	C: 26279	NR	NR	-0.973 (0.024)	-2.68%
		I: 13666
Notes: ^aRelative difference change accounting for baseline for baseline data (see Supplementary Appendix 2); DID: differences in differences; C: control group; I: intervention group; NR: not reported; NA; not analyzed.

Table 4

Results of interrupted time series studies
Outcomes	Study	Number of patients(n)	Coefficient(95%CI) for intercept: baseline level	Coefficient(95%CI) for baseline trend	Coefficient(95%CI) for change in level: step change	Coefficient(95%CI) for change in trend: slope change	Relative difference change^a: change in level	Relative difference change^a: change in slope
Efficiency	Length of stay
	Chen et al, 2023⁴⁰	3828543	-3.287 (-12.478, 5.903)	-0.051 (-0.079, -0.023)	0.271 (-0.019, 0.561)	-0.008 (-0.051, 0.035)	-8.24%	+ 0.24%
	Ding et al, 2023⁴¹
	Tertiary hospitals	NR	9.89 (9.73, 10.05)	-0.02 (-0.03, -0.01)	0.09 (-0.11, 0.30)	-0.05 (-0.07, -0.02)	+ 0.91%	-0.51%
	Secondary hospitals	NR	8.57 (8.10, 9.04)	0.00 (-0.03, 0.03)	0.44 (-0.26, 1.14)	0.04 (-0.04, 0.12)	+ 5.13%	+ 0.47%
	Tang et al, 2023¹²	4821030	-1.163 (-8.101, 5.775)	-0.060 (-0.079, -0.042)	0.309 (0.060, 0.558)	0.022 (0.001, 0.043)	-26.57%	-1.89%
Quality	In-hospital mortality
	Chen et al, 2023⁴⁰	3828543	-2.359 (-7.588, 2.871)	-0.015 (-0.033, 0.003)	-0.084 (-0.278, 0.110)	0.005 (-0.026, 0.036)	+ 3.56%	-0.21%
	Tang et al, 2023¹²	4821030	-6.787 (-9.621, -3.953)	-0.011 (-0.020, -0.002)	-0.017 (-0.129, 0.095)	0.001 (-0.009, 0.010)	+ 0.25%	-0.01%
Expenditure	Total expenditures per case
	Han et al, 2021³⁹	NR	16273.978 (13996.516, 18551.439)	485.019 (325.630, 644.408)	-1697.568 (-4822.299, 1427.164)	212.024 (-13.387, 437.434)	-10.43%	+ 1.30%
	Chen et al, 2023⁴⁰	3828543	7.820 (6.908, 8.732)	-0.003 (-0.006, 0.000)	0.041 (0.008, 0.074)	0.005 (0.000, 0.010)	+ 0.52%	+ 0.06%
	Ding et al, 2023⁴¹
	Tertiary hospitals	NR	11.21 (10.97, 11.46)	0.08 (0.06, 0.11)	-0.61 (-1.50, 0.27)	-0.15 (-0.22, -0.09)	-5.44%	-1.34%
	Secondary hospitals	NR	4.94 (4.81, 5.08)	0.04 (0.02, 0.05)	0.17 (-0.15, 0.48)	-0.01 (-0.03, 0.01)	+ 3.44%	-0.20%
	Tang et al, 2023¹²	4821030	9.690 (8.757, 10.623)	-0.006 (-0.009, -0.003)	0.056 (0.021, 0.092)	0.011 (0.008, 0.014)	+ 0.58%	+ 0.11%
	Out-of-pocket payments per case
	Chen et al, 2023⁴⁰	3828543	-2.228 (-20.673, 16.217)	0.016 (-0.015, 0.047)	-0.298 (-0.661, 0.065)	-0.011 (-0.074, 0.051)	+ 13.38%	+ 0.49%
	Proportion of the out-of-pocket payments per case
	Ding et al, 2023⁴¹
	Tertiary hospitals	NR	45.02 (44.50, 45.53)	0.02 (-0.02, 0.07)	0.61 (-0.30, 1.53)	-0.17 (-0.23, -0.11)	+ 1.35%	-0.38%
	Secondary hospitals	NR	38.88 (38.57, 39.19)	-0.06 (-0.09, -0.03)	-0.32 (-1.13, 0.48)	-0.09 (-0.15, -0.03)	-0.82%	-0.23%
	Medical consumables expenditures per case
	Han et al, 2021³⁹	NR	138.425 (133.671, 143.179)	0.496 (0.163, 0.828)	-2.195 (-8.717, 4.328)	-0.336 (-0.807, 0.134)	-1.59%	-0.24%
	Treatment expenditures per case
	Han et al, 2021³⁹	NR	148.337 (144.117, 152.556)	0.874 (0.578, 1.169)	6.616 (0.827, 12.405)	-0.980 (-1.397, -0.562)	+ 4.46%	-0.66%
	Service expenditures per case
	Han et al, 2021³⁹	NR	150.626 (147.415, 153.837)	-0.479 (-0.704, -0.254)	-0.503 (-4.909, 3.903)	-0.361 (-0.679, -0.043)	-0.33%	-0.24%
	Other expenditures per case
	Han et al, 2021³⁹	NR	41.833 (38.751, 44.916)	0.786 (0.570, 1.002)	-3.070 (-7.299, 1.159)	-0.464 (-0.769, -0.159)	-7.34%	-1.11%
Notes: ^aRelative difference change accounting for baseline for baseline data (see Supplementary Appendix 2); C: control group; I: intervention group; NR: not reported.

To the best of our knowledge, this is the first systematic review and meta-analysis to examine the effects of DIP-based payment on healthcare in Chinese public hospitals with regard to quantity, efficiency, quality and expenditure. We examined not only the desired effects of DIP payment reform but also the unintended effects. Integrating the currently available data, the findings indicated that partial quality improvements accompanied efficiency increases. There was no overall reduction in the quantity of healthcare. Total expenditures per case did not show a significant change, but some compelling evidence for positive outcomes was identified regarding the structure of hospitalization expenditures.

The results showed a slight decrease in the annual number of discharges after the DIP reform without significant differences, so it did not have an impact on the annual number of discharges³⁵. The overall change in the quantity of healthcare remained stable in public hospitals. This might be attributed to the effective monitoring and evaluation mechanisms of the DIP reform, which weakened the potential behaviors of increasing service volume³⁴. This finding was consistent with the results of Shanghai and Guangzhou, where intelligent supervision controlled irrational medical behaviors in healthcare institutions⁴². Instead of being unprofitable, medical institutions were at risk of losses due to differences in penalties.

LOS is an important indicator for evaluating efficiency²⁶. The pooled estimates of the meta-analysis suggested that DIP-based payment reduced the LOS, which aligned with the policy expectations and was theoretically plausible⁴³. Compared with the FFS payment, DIP-based payment provided public hospitals with stronger financial incentives to improve efficiency by shortening the LOS. Considering that LOS was another reflection of hospital costs, the reduction of the LOS might have a positive impact on the total expenditures per case, leading to higher productivity and profits for the hospitals. It was worth noting that decreasing LOS might yield unintended consequences, such as early discharges, patient selection, and cream skimming². However, from the results of the readmission within 30 days in this study, no problems were identified in this regard. Therefore, efficiency outcomes need to be viewed cautiously and ensure that the DIP payment reform did not lead to efficiency degradation. It encouraged further large-scale, adequately powered, and multicenter studies to assess the effects of DIP payment reform⁴⁴.

In terms of DIP payment reform’s impact on quality outcomes, indicators such as in-hospital mortality, readmission within 30 days, operation associated infection, and postoperative complication were investigated. The present meta-analysis results demonstrated that in-hospital mortality was slightly reduced. Only a limited number of studies investigated the other three quality indicators. The results suggest that the readmission within 30 days was slightly reduced. The changes in the operation associated infection rate in response to the policy reform were insignificant and negligible¹³. Moreover, although the postoperative complication rate decreased by 0.2 percentage points, the estimate was also not statistically significant. It was impossible to definitively answer whether the quality altered after the DIP payment reform². To begin with, none of the studies attributed the change in the quality of healthcare to DIP implementation because determining the intermediate and long-term impacts on the quality outcomes required ongoing monitoring over the years and decades. Besides, due to some variations in medical resources, payment scopes, and complexity factors before and after the DIP payment reform, this might not reflect actual changes in quality indicators³⁶. At last, the quality of healthcare was evaluated through multi-dimensional indicators, and individual indicators could not reflect the whole picture.

The meta-analysis results showed no significant relationship between DIP payment and total expenditures per case. The possible reasons for this included: on the one hand, the policy was in operation for a relatively short period, resulting in insignificant statistical differences in the short term; on the other hand, the actual reimbursement rate was influenced by policy adjustments and hospital grades³⁸. This meant that it was not possible to derive whether there was a substantial change in total expenditures per case. Longstanding and well-known distortions under FFS, such as over-treatment and over-prescription, had contributed to irrationalities in the structure of hospitalization expenditures⁴⁵. Transitioning from FFS payment to DIP payment, although there was no significant turnaround in total expenditures per case, the structure of hospitalization expenditures was partially optimized. Out-of-pocket payments per case, drug expenditures per case, proportion of the drug expenditures per case, service expenditures per case, examination expenditures per case, and other expenditures per case showed a downward trend. To prevent medical institutions from shifting costs to patients, the assessment indicator of proportion of the out-of-pocket payments per case was set. This indicator decreased as a whole, indirectly reflecting the positive effects of DIP payment reform. Despite the improvement in the expenditure structure, it was still necessary to strengthen the control of priority cost categories and reduce medical consumables and treatment expenditures.

The only relevant review examined the impacts of DIP payment reform on the operation of medical insurance funds, medical institutions, patients, and resource flow. It relied on simple descriptive statistics to analyze the effects of DIP-based payment, with a lack of in-depth comparative analysis and limiting findings in informing policy⁴⁶. In addition, most studies used simple before-after designs and lacked statistical methods, thus failing to provide reliable evidence for evaluating interventions^47,48. In order to involve the available high-quality studies, we only included study designs recommended by EPOC. Econometric statistical models, such as the difference-in-differences approach and interrupted time series model, were considered more scientifically rigorous methods for accessing policy effects. Thus, our study might provide more robust evidence for evaluating the impacts of DIP payment on public hospitals.

Although some studies did not meet our inclusion criteria and no meta-analysis was performed, we summarized some common themes from these studies. In short, efficiency was improved, and DIP payment reform had potential cost-saving implications by reducing LOS. The difference in readmission rates was not statistically significant, whether it increased or decreased. Previous studies also indicated that when different disease categories were selected for the study, the effects of DIP reform were different or even opposite, which was also related to the characteristics of disease categories. The conclusions about total expenditures per case were not consistent. Most studies showed that total expenditures per case decreased to varying degrees. However, some studies observed an increase in the total expenditures per case and the growth rate gradually decreased after the reform^49,50. Drug expenditures per case, proportion of the out-of-pocket payments per case, and other expenditures per case were decreased, and the expenditure structure was optimized, reducing the direct economic burden of patients. The evidence generated here might be helpful and adaptive to other LMICs with similar contexts and goals in health reform. Additional high-quality studies were needed to expand the evidence base of DIP payment reform.

However, this study had several limitations. Firstly, some unpublished studies might be missing in this study, even though as many eligible studies as possible were searched and included. Secondly, substantial heterogeneity hindered the meta-analysis, which could be attributed to variations in sample characteristics, intervention settings, and policy details among different studies despite using a random-effect model to account for it. Thirdly, although we followed the EPOC guidelines to include the current best designs, no randomized trials were found and the quality of evidence was generally low. Fourthly, there is a lack of discussion on the impact of different hospital ownership types. The variations in hospital ownership may be ascribed to responding differently to the incentives created by the provider payment reforms, which are likely to impact healthcare structure, process, and outcomes⁵¹. Future research can be explored as more relevant empirical studies are conducted. Finally, this study only identified the short-term effects of the DIP payment reform across public hospitals. Considering the lag in policy implementation and effectiveness feedback, the long-term effects deserve further investigation.

While it is difficult to draw definitive conclusions from our review about the links between DIP-based payment and healthcare in Chinese public hospitals, there was preliminary evidence that DIP payment could modestly improve the efficiency of healthcare by reducing LOS. The overall quantity remained unchanged, but the quality of healthcare was partially improved regarding in-hospital mortality and readmission within 30 days. There were some positive changes in the structure of hospitalization expenditures, though total expenditures per case did not significantly change. Our findings remind policymakers to consider its positive and negative outcomes when intending to implement the new DIP payment. As healthcare gradually moves towards value-based payment, multiple payment methods can leverage their respective strengths and deeply integrate into a better payment system.

Data availability statement

The datasets supporting the conclusions of this study are included within the article and its supplementary information.

Ethical Approval

Ethical approval for this type of study is not required by our institute.

The study was reported according to the latest Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and registered on the International Prospective Register of Systematic Reviews (PROSPERO: CRD42024503159).

Competing interests

The authors declare no conflicts of interest.

Author contributions

Conception or design of the study: T.X., L.J., C.L. and X.W. Study selection: T.X. and L.J. Data collection: T.X. and L.J. Data analysis and interpretation: T.X. and C.L. Drafting of the manuscript: T.X. Critical revision of the article: L.J., C.L. and X.W. Final approval of the version to be submitted: All authors (T.X., L.J., C.L. and X.W.) read and approved the final manuscript.

Funding

This work was supported by the Health Commission of Shandong Province [grant number WZY202390].

Banach, M., Friedman, S. & Srinivasan, R.G. Past, present, and future of global health financing: a review of development assistance, government, out-of-pocket, and other private spending on health for 195 countries, 1995–2050. Lancet 398, 956–956. https://doi.org/10.1016/S0140-6736(19)30841-4 (2021).
Barouni, M., Ahmadian, L., Anari, H.S. & Mohsenbeigi, E. Investigation of the impact of DRG based reimbursement mechanisms on quality of care, capacity utilization, and efficiency: A systematic review. Int. J. Healthc. Manag. 14, 1463–1474. https://doi.org/10.1080/20479700.2020.1782663 (2020).
World Health Organization. The world health report: health systems financing: the path to universal coverage‎. https://iris.who.int/handle/10665/44371 (2010).
Cylus, J. & Irwin, R. The challenges of hospital payment systems. EuroObserver 12, 1–12 (2010).
Jakovljevic, M., Groot, W. & Souliotis, K. Editorial: health care financing and affordability in the emerging global markets. Front. Public. Health. 4, 2. https://doi.org/10.3389/fpubh.2022.1054409 (2016).
Souliotis, K. et al. Informal payments in the Greek health sector amid the financial crisis: old habits die last. Eur. J. Health. Econ. 17, 159–170. https://doi.org/10.1007/s10198-015-0666-0 (2016).
Roberts, M.J. Getting health reform right: a guide to improving performance and equity. https://doi.org/10.1093/acprof:oso/9780195371505.001.0001 (2004).
National Health Commission of the People's Republic of China. China Health Statistics Yearbook in 2022. http://www.nhc.gov.cn/mohwsbwstjxxzx/tjtjnj/202305/6ef68aac6bd14c1eb9375e01a0faa1fb/files/b05b3d958fc546d98261d165cea4adba.pdf (2023).
Wu, A. Study on the relationship between stakeholder orientation and performance in public non-profit hospitals. https://www.xueweilunwen.com/doc/1647091 (2008).
Statistical Information Center of the National Health Commission. National healthcare services from January to August 2023. http://www.nhc.gov.cn/mohwsbwstjxxzx/s7967/202401/d6fd0c655fc04e0585ce143e014adc1a.shtml (2024).
Jian, W. et al. Payment reform pilot in Beijing hospitals reduced expenditures and out-of-pocket payments per admission. Health. Aff. 34, 1745–1752. https://doi.org/10.1377/hlthaff.2015.0074 (2015).
Tang, X. et al. Variations in the impact of the new case-based payment reform on medical costs, length of stay, and quality across different hospitals in China: an interrupted time series analysis. BMC. Health. Serv. Res. 23, 568. https://doi.org/10.1186/s12913-023-09553-x (2023).
Qian, M. et al. The pilot of a new patient classification-based payment system in China: The impact on costs, length of stay and quality. Soc. Sci. Med. 289, 114415. https://doi.org/10.1016/j.socscimed.2021.114415 (2021).
Ying, Y. DlP & DRG: similarities and differences. China. Health. Insur.1, 39–42. https://doi.org/10.19546/j.issn.1674-3830.2021.1.009 (2021).
Page, M.J. et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ 372, n160. https://doi.org/10.1136/bmj. n160 (2021).
Xu, T. et al. The effects of diagnosis-intervention packet on the quantity, efficiency, quality, and expenditure of public hospitals in China: a systematic review and meta-analysis. https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID =503159 (2024).
Cochrane Effective Practice and Organization of Care. EPOC Resources for review authors: Data collection form. https://epoc.cochrane.org/resources/epoc-specifc-resources-review-authors (2017).
Sterne, J.A.C. et al. ROBINS-I: a tool for assessing risk of bias in non-randomized studies of interventions. BMJ 355, 1–7. https://doi.org/10.1136/bmj.i4919 (2016).
Schünemann, H.J. et al. GRADE guidelines: 18. How ROBINS-I and other tools to assess risk of bias in nonrandomized studies should be used to rate the certainty of a body of evidence. J. Clin. Epidemiol. 111, 105–114. https://doi.org/10.1016/j.jclinepi. 2018.01.012 (2019).
Higgins, J.P.T. et al. Chapter 8: Assessing risk of bias in a randomized trial. In: Higgins JPT, Thomas J, Chandler J et al. (eds). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 updated February 2022. Cochrane. (2023).
Lechner, M. The estimation of causal effects by difference-indifference methods. Found. Trends. Econometrics. 4, 165–224. https://doi.org/10.1561/0800000014 (2011).
Higgins, J.P.T. & Thompson, S.G. Quantifying heterogeneity in a meta-analysis. Stat. Med. 21, 1539–1558. https://doi.org/10.1002/sim.1186 (2002).
Begg, C.B. & Mazumdar, M. Operating characteristics of a rank correlation test for publication bias. Biometrics 50, 1088–1101. https://doi.org/10.2307/2533446 (1994).
Zheng, M. The impact of diagnosis-intervention packet on hospital management. J. Imaging. Res. Med. Appl. 1, 162–164. https://doi.org/10.3969/j.issn.2096-3807.2017. 18.103 (2017).
Ding, L. et al. The practice and reflection on the diagnosis-intervention packet of a tertiary hospital in Yinchuan. Ningxia. Med. J. 41, 281–282. https://doi.org/10.13621/j.1001-5949.2019.03.0281 (2019).
Wang, Y., He, R. & Bai, H. Evaluation of the effect of implementing DIP payment in a hospital in Suqian. Jiangsu. Health. Syst. Manag. 33, 1664–1666 (2022).
Wu, T. A study on the implementation effect of diagnosis-intervention packet. Money. China. 20, 80–82. https://doi.org/10.19887/j.cnki.cn11-4098/f.2022.20.009 (2022).
Liu, Z. et al. An analysis of the Impact of Healthcare Payment Reform on Average Costs and Patient Satisfaction of Neurosurgical Inpatients. China. Health. Insur. 12, 95–97. https://doi.org/10.19546/j.issn.1674-3830.2022.12.018 (2022).
Chen, F. A study on the effect of diagnosis-intervention packet on the treatment outcomes of endocrinology patients. China. Health. Insur. 11, 92–94. https://doi.org/10.19546/j.issn.1674-3830.2022.11.018 (2022).
Zheng, X., Kang, Q. & Chen, X. Based on the method of difference-in-difference, the performance of payment by disease score under global budget control evaluated from the perspective of designated hospital. Jiangsu. Health. Syst. Manag. 31, 622–626. https://doi.org/10.3969/j.issn.1005-7803.2020.05.021 (2020).
Zheng, X., Kang, Q. & Chen, X. Difference-in-difference method for evaluating the influence of disease score-based payment on hospitalization expenses of patients with malignant tumors. Chin. J. Med. Manag. Sci. 10, 17–22. https://doi.org/10.3969/j.issn. 2095-7432.2020.02.003 (2020).
Yang, Y. et al. Study on the Cost Control Effect of “Payment by Disease Score”. Health. Econ. Res. 38, 36–39 (2021).
Hu, F. et al. Evaluation of the impact of Diagnosis-Intervention Packet on patients with cardiovascular diseases from the perspective of difference-in-difference method. Mod. Prev. Med. 49, 3181–3186. https://doi.org/10.20043/j.cnki.MPM.202202296 (2022).
Lai, Y. et al. Hospital response to a case-based payment scheme under regional global budget: The case of Guangzhou in China. Soc. Sci. Med. 292, 114601. https://doi.org/10.1016/j.socscimed.2021.114601 (2022).
Lin, K. et al. Analysis on the Prevention and Control Strategies of lncreasing Related Weight Behavior in Public Hospitals in Yichang City Under DIP Payment Method. China. Health. Insur. 6, 60–66. https://doi.org/10.19546/j.issn.1674-3830.2023.6.009 (2023).
Zhang, X. et al. Impact of Diagnosis-Intervention Packet payment in S City on the healthcare quality of diseases concerned by the national public hospital performance assessment. Chin. J. Health. Policy. 16, 44–50. https://doi.org/10.3969/j.issn.1674-2982.2023.04.007 (2023).
Ye, D. et al. Evaluation of the implementation effect of diagnosis-intervention packet (DIP) based on differences-in-differences method. Soft. Sci. Health. 37, 71–74. https://doi.org/10.3969/j.issn.1003-2800.2023.04.017 (2023).
Deng, Q. et al. An Empirical Study on the Implementation Effect of Diagnosis-Intervention Packet in Pilot Cities Based on the Method of Difference-in-Difference. Chin. Hosp. Manag. 43, 6–9 (2023).
Han, F. et al. Phased study on the impact of diagnosis-intervention packet payment on hospitalization expenses. Chin. J. Hosp. Admin. 37, 647–652. https://doi.org/10.3760/cma.j. cn111325-20201120-02125 (2021).
Chen, Y. et al. How do inpatients' costs, length of stay, and quality of care vary across age groups after a new case-based payment reform in China? An interrupted time series analysis. BMC. Health. Serv. Res. 23, 160. https://doi.org/10.1186/s12913-023-09109-z (2023).
Ding, Y. et al. The impacts of diagnosis-intervention packet payment on the providers' behavior of inpatient care-evidence from a national pilot city in China. Front. Public. Health. 11, 1069131. https://doi.org/10.3389/fpubh. 2023.1069131 (2023).
Xu, S. et al. The big data diagnosis-intervention packet payment method: experience from Shanghai and Guangzhou. Chin. J. Hosp. Admin. 37, 186–190. https://doi.org/10.3760/cma.j. cn111325-20201009-02012 (2021).
Meng, Z. et al. The effects of DRGs-based payment compared with cost-based payment on inpatient healthcare utilization: A systematic review and meta-analysis. Health. Policy. 124, 359–367. https://doi.org/10.1016/j.healthpol.2020.01.007 (2020).
Chen, Y. et al. Impact of diagnosis-related groups on inpatient quality of health care: A systematic review and meta-analysis. Inquiry 60, 1–16. https://doi.org/10.1177/0046958 0231167011 (2023).
Hui, E.C. The contemporary healthcare crisis in China and the role of medical professionalism. J. Med. Philos. 35, 477–492. https://doi.org/10.1093/jmp/jhq031 (2010).
Tao, Y., Chen, C. & Han, Y. Study on the effect of Diagnosis-Intervention Packet and its enlightenment to DRG pilot hospitals in China. Chin. Hosp. 25, 16–19. https://doi.org/10.19660/j.issn.1671-0592.2021.6.05 (2021).
Luangasanatip, N. et al. Comparative efficacy of interventions to promote hand hygiene in hospital: systematic review and network meta-analysis. BMJ 351, h3728. https://doi.org/10.1136/bmj.h3728 (2015).
Biglan, A., Ary, D. & Wagenaar, A.C. The value of interrupted time-series experiments for community intervention research. Prev. Sci. 1, 31–49. https://doi.org/10.1023/A: 1010024016308 (2000).
Gan, L. et al. Study on cost analysis and cost control strategy of oncology patient hospitalization based on DIP. Chin. Hosp. Manag. 43, 41–44 (2023).
Cao, J. et al. Evaluating the implementation performance of Diagnosis-Intervention Packet (DIP): an empirical study in Qingyuan City based on synthetic control method. Chin. J. Health. Policy. 15, 31–37. https://doi.org/10.3969/j.issn.1674-2982.2022.12.005 (2022).
He, J. et al. A comparative study on medical quality among hospitals of different ownerships. Chin. Health. Qual. Manag. 24, 18–20. https://doi.org/10.13912/j.cnki.chqm. 2017.24.4.08 (2017).

No competing interests reported.

SupplementalFile.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

The effects of diagnosis-intervention packet payment on healthcare in Chinese public hospitals: a systematic review and meta-analysis

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Quantity outcomes

Efficiency outcomes

Quality outcomes

Expenditure outcomes

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1