Impact of diagnosis-intervention packet payment reform on pediatric hospital services: an interrupted time series analysis.

Impact of diagnosis-intervention packet payment reform on pediatric hospital services: an interrupted time series analysis - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Health Serv Res . 2026 May 11;26:769. doi: 10.1186/s12913-026-14661-5 Impact of diagnosis-intervention packet payment reform on pediatric hospital services: an interrupted time series analysis Ying Liu 1, 3, # , Jianghui Zhang 1, 2, # , Xiangliang Zhang 4 , Jieling Wu 3 , Wen Chen 1, 2, ✉ Author information Article notes Copyright and License information 1 Department of Medical Statistics, School of Public Health, Sun Yat-Sen University, #74 Zhongshan Road 2, Guangzhou, 510080 China 2 Center for Migrant Health Policy, Sun Yat-Sen University, #74 Zhongshan Road 2, Guangzhou, 510080 China 3 Department of Children’s Health Care, Guangdong Women and Children Hospital, Guangzhou, 511400 China 4 Department of Earth System Science, Tsinghua University, Beijing, 100084 China ✉ Corresponding author. # Contributed equally. Received 2026 Jan 28; Accepted 2026 Apr 27; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13220595  PMID: 42116151 Abstract Background Diagnosis-Intervention Packet (DIP) reform has been widely implemented in China, but its impact in pediatric settings remains unclear. This study evaluated its effects on pediatric admissions, case severity, efficiency, and expenditures in a tertiary maternal and child hospital in Guangzhou. Methods Pediatric inpatient discharge data (0–18 years) from January 2016 to December 2019 were analyzed using an interrupted time series design. Outcomes included total admissions, proportion and number of complex/critical and non-local complex/critical cases, mean length of stay (LOS), and mean total expenditure per case. Subgroup analyses by major disease categories and sensitivity analyses were performed to assess the robustness and heterogeneity of reform effects. Results A total of 93,396 pediatric discharges were included, with 45,493 discharges in the pre-reform period and 47,903 in the post-reform period. The DIP reform was associated with significant long-term changes in pediatric service utilization and expenditure. Specifically, the monthly slope of total pediatric admissions decreased slightly after the reform ( β = −0.006, P < 0.001). Similar decreasing trends were observed for the proportion of complex/critical cases per month ( β = −0.018, P < 0.001), the absolute number of complex/critical cases ( β = −0.017, P < 0.001), the proportion of non-local complex/critical cases ( β = −0.012, P < 0.001), and the absolute number of non-local complex/critical cases ( β = −0.016, P < 0.001). In terms of service efficiency, the post-reform monthly slope of mean LOS showed a reduction after accounting for lag periods, whereas mean total expenditures per case showed a statistically significant decrease ( β = −0.010, P = 0.001). Conclusions The DIP payment reform was associated with decreases in total pediatric admissions, as well as in the proportion and number of complex/critical and non-local complex/critical cases. Mean LOS showed a delayed decrease following the reform, while per-case total expenditures showed a decreasing trend over time. Supplementary Information The online version contains supplementary material available at 10.1186/s12913-026-14661-5. Keywords: Diagnosis-intervention packet, Pediatric hospital, Healthcare reform, Interrupted time series analysis Introduction Pediatric healthcare delivery in China faces mounting pressures amid demographic transition and rising public expectations. Although declining birth rates have reduced the number of children, they may have increased the perceived value placed on child health, while rapid economic development and urbanization have further intensified demand for high-quality pediatric services [ 1 ]. Notably, while China’s basic medical insurance system now covers over 95% of the population in 2024, creating the world’s largest safety net, and the in-reimbursement rate for hospitalization expenses reached 84.8% for urban employees and 68.6% for urban and rural residents that same year [ 2 ], escalating overall healthcare costs continue to strain the sustainability of the social health insurance fund. In response to these fiscal pressures, policymakers have implemented a series of provider payment reforms to improve efficiency and contain expenditures [ 3 ]. Within this context, the Diagnosis–Intervention Packet (DIP) system has emerged as a case-based reimbursement model. First piloted in Guangzhou in 2018, DIP has since been expanded more broadly across China. It represents a shift from retrospective fee-for-service payment toward prospective, standardized reimbursement arrangements [ 4 ]. DIP system was developed to address specific limitations inherent in both traditional fee-for-service models and the direct adoption of international Diagnosis-Related Groups (DRGs) within the Chinese context. While DRGs serve as a robust prospective payment mechanism globally, their initial implementation in China faced challenges related to incomplete clinical coding, heterogeneity in hospital information systems, and difficulties in constructing clinically homogeneous case groups [ 5 ]. To further optimize the systems, the National Healthcare Security Administration introduced the DRG 2.0 and DIP 2.0 schemes in July 2024, refining grouping logic to better align with local clinical practices. A fundamental distinction lies in the granularity and derivation of classification. The DRG 2.0 scheme categorizes cases into 634 refined groups based on pre-defined clinical hierarchies, prioritizing medical logic and resource similarity. In contrast, the DIP 2.0 library utilizes big data analytics on real-world claims to define over 9,520 core disease-intervention combinations, derived directly from the unique pairing of ICD-10 diagnosis codes and ICD-9-CM-3 procedure codes. This data-driven approach allows DIP to capture granular variations in clinical pathways that broader DRG categories might overlook [ 6 ]. To illustrate the operational differences between DRGs and DIP, consider a pediatric inpatient admitted with appendicitis undergoing laparoscopic appendectomy. Under the DRG system, the case is grouped by principal diagnosis, procedure, and complications, receiving a prospective fixed payment regardless of minor procedural variations. In contrast, under the DIP system, the same case would be classified into a more specific diagnosis–intervention combination based directly on the ICD-10 diagnosis code (e.g., K35.x for appendicitis) and the ICD-9-CM-3 procedure code for laparoscopic appendectomy. If additional procedures (e.g., drainage, imaging-guided interventions) or variations in treatment occur, these would generate distinct combinations within the DIP catalogue, each assigned a different relative weight reflecting observed resource consumption in real-world data. Both systems employ mechanisms to ensure fiscal sustainability, though their approaches differ. In DRG, reimbursement is calculated by multiplying the group weight by a fixed base rate, which may be adjusted periodically to align with budget targets. In DIP, reimbursement is the product of the combination’s relative weight and a floating-point value. This point value is determined retrospectively: it is calculated by dividing the region’s total annual budget by the aggregate sum of all points claimed by hospitals in that year. This mechanism inherently enforces budget neutrality, automatically adjusting the monetary value of each point to prevent overspending [ 7 ]. This approach reduces reliance on complex grouping algorithms while enabling more granular classification than many DRG systems. In principle, DIP may incentivize hospitals to improve coding quality, rationalize service provision, and enhance operational efficiency, while supporting expenditure control at the system level [ 8 ]. A growing body of evidence indicates that case-based payment reforms, including DRGs and DIP, can change provider behavior and improve hospital efficiency. In adult and general-hospital settings, DRG-based payment has been associated with shorter length of stay (LOS), lower inpatient expenditures, and reductions in potentially unnecessary admissions [ 9 , 10 ]. Early evaluations of DIP in Chinese pilot cities have similarly reported cost-containment effects, with slower growth in total expenditures after implementation [ 11 ]. Nevertheless, case-based reimbursement may also generate unintended incentives, such as premature discharge, upcoding, or patient selection, which could compromise quality of care or shift financial burdens to patients [ 12 ]. Despite the expanding literature on payment reform, evidence from pediatric hospitals remains limited. Pediatric inpatient care differs from adult care in ways that may modify how case-based payment affects service delivery. First, pediatric admissions are often dominated by acute, self-limited conditions with short baseline LOS, which may leave less room for further reductions under prospective payment incentives [ 13 ]. Second, in the Chinese healthcare system, hospitals are organized into three tiers: primary hospitals provide community-based services including basic medical care, prevention, and rehabilitation; secondary hospitals serve as regional facilities delivering comprehensive care to multiple communities; and tertiary hospitals function as regional or national centers offering highly specialized services and advanced clinical care. Within this tiered system, tertiary pediatric hospitals frequently serve as regional referral centers, particularly for critically ill neonates and children with congenital or rare conditions, generating substantial cross-regional patient flows; reforms may therefore influence not only efficiency and spending but also referral patterns and access to specialized services [ 14 , 15 ]. Consequently, findings from adult populations—where LOS and costs often decline after DRG or DIP implementation—may not be directly generalizable to pediatric settings given differences in disease spectrum, clinical pathways, and referral structures [ 9 , 10 , 12 ]. Equity considerations are also central in pediatrics because children rely on caregivers’ financial capacity, and any unintended increase in medical spending could disproportionately burden families and deter timely care-seeking [ 16 ]. Moreover, pediatric-specific outcomes, remain under-examined in payment reform evaluations. Under case-based reimbursement systems such as DIP, hospitals may have stronger incentives to attract or retain complex pediatric cases with higher case weights, potentially reshaping established referral pathways and the distribution of specialized pediatric services [ 15 ]. Understanding how DIP reform interacts with pediatric referral dynamics is therefore important because shifts in patient flow may affect regional equity, access to specialized services, and families’ financial burden. In response to these gaps, this study evaluates the impact of DIP reform on service, efficiency, and expenditures in a tertiary pediatric hospital, with particular attention to non-local complex/critical cases, to inform pediatric-specific payment policy and contribute evidence relevant to case-based reimbursement reforms more broadly. Methods Data source and study population Data were obtained from a large tertiary maternal and child health hospital in Guangzhou, China, which was among the earliest pilot sites implementing the DIP payment reform in January 2018. The hospital serves as a provincial-level pediatric medical center and provides comprehensive pediatric and perinatal services, including neonatal intensive care, general pediatrics, and multiple pediatric subspecialties. We extracted hospital discharge summary data for pediatric inpatients aged 0–18 years from the hospital electronic medical record system from January 2016 to December 2019. These records are routinely submitted to the local health insurance bureau and undergo standardized data quality checks. We excluded (1) hospitalizations with a length of stay greater than 60 days, which may represent outliers or long-term/chronic rehabilitation admissions, and (2) admissions with a total inpatient expenditure of less than 100 Chinese Yuan (CNY), which may indicate data errors. In total, 786 cases (0.83%) were excluded based on these criteria. Patient characteristics Patient characteristics included age, biological sex (male/female), residence status (urban/rural), and case type (general, acute, complex, critical). Residence status was determined using household registration information. Case severity was assessed using the hospital’s four-tier grading system, which classifies pediatric inpatients as Type A (general/simple cases), Type B (acute/simple emergency cases), Type C (complex cases), or Type D (critical complex cases). Type A patients have straightforward diagnoses, relatively stable conditions, and require no urgent intervention. Type B patients present with acute but uncomplicated conditions that require urgent management, yet vital signs remain stable, and cases are not considered complex or critical. Type C cases involve complicated or unclear diagnoses, treatment difficulties, or serious comorbidities, and are regarded as difficult with relatively poor prognosis. Type D cases represent the most critical and complex patients, with severe, unstable conditions or organ failure requiring emergency interventions. Outcomes The study evaluated outcomes in three domains, all measured at the monthly level. Total service volume was assessed as the total number of pediatric admissions per month. Case mix severity was evaluated using both the proportion and absolute number of complex and critical cases (Type C and D cases) among all discharges. Inter-regional referral of complex cases was assessed using both the proportion and absolute number of non-local complex and critical cases (patients residing outside Guangzhou City). Service efficiency was measured by the monthly mean length of stay (LOS), calculated each month as total inpatient days divided by the number of discharges. Medical expenditures were captured by the monthly mean total expenditure per case, converted to USD. Statistical analysis Continuous variables are summarized as mean ± standard deviation (SD) or median (interquartile range, IQR), as appropriate, and categorical variables as counts and percentages. Between-period comparisons were conducted using Student’s t-test or the Wilcoxon rank-sum test for continuous variables, depending on distributional assumptions, and the chi-square test for categorical variables. We used an interrupted time series (ITS) design to assess changes associated with DIP implementation in January 2018. ITS is a widely used quasi-experimental approach for policy evaluation when randomized controlled trials are infeasible, enabling estimation of both immediate level changes and post-intervention trend changes [ 17 , 18 ]. For each monthly outcome, we fitted generalized linear models with outcome-appropriate distributions, including a Poisson model for total admissions and absolute case counts, a binomial model for monthly proportions, a Gaussian model for mean LOS, and a Gamma model with a log link for expenditure outcomes. The general model specification was: 1 where is the outcome at month ; is the time trend; is an indicator for the post-intervention period (January 2018 onwards). The intercept represents the baseline level of the outcome variable, and is the monthly slope of the outcome variable before the DIP reform. captures the immediate level change after DIP reform, and estimates the monthly slope change compared to the pre-intervention trend. We further estimated the absolute effects of the intervention using a counterfactual framework. For the following outcomes—total monthly pediatric admissions, absolute number of complex/critical cases per month, absolute number of non-local complex/critical cases per month, mean length of stay (LOS, days), and mean total expenditures per case (USD)—two scenarios were constructed at each post-intervention time point: with DIP (DIP = 1) and without DIP (DIP = 0), while holding other covariates constant. Model-based predictions were obtained under both scenarios, and their differences were calculated as the average marginal effect at each time point. Furthermore, the monthly average marginal effects after the intervention were arranged in chronological order to form a series, and a linear regression model was fitted to estimate its temporal trend. The regression coefficient was interpreted as the monthly change after the intervention. Effects were consistently estimated on the response scale across different models. Nonparametric bootstrap (1,000 resamples) was used to estimate 95% confidence intervals. Additionally, for mean total expenditures per case, we aggregated the monthly absolute effects over fixed intervals to provide intuitive estimates of cumulative savings or additional costs (Method S1). In addition to the full-sample analysis of all pediatric discharges, we performed subgroup analyses for the five most frequent primary diagnoses during the study period to examine potential heterogeneity in reform-associated changes. If a hospitalization record contained more than one ICD-10 code corresponding to these diagnoses, the record was included in each relevant diagnosis-specific analysis; therefore, a single record could contribute to multiple subgroups, and subgroup totals do not necessarily sum to the full-sample total. All analyses were conducted in R (version 4.3.0; R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was defined as a two-sided P < 0.05. To assess whether the effect of the DIP reform is sensitive to the assumption of an immediate intervention effect, we applied a kernel-lag approach developed previously by our research team [ 19 ]. This method replaces the abrupt onset of the intervention with a gradual ramp-up, using lag windows of 3 and 6 months. We considered two weighting kernels scaled between 0 and 1 over each window: a uniform kernel (representing a linear increase) and a truncated-normal kernel (representing a smooth S-shaped increase). These kernel weights were applied to both the level change and post-intervention slope terms, replacing the original binary DIP indicator and the time × DIP interaction, while covariates and month fixed effects remained unchanged. More specifically, let denote the start of the intervention (January 2018). For a chosen lag window months and kernel , we defined a ramped post-intervention weight , which was used in place of in both the level and slope terms. For the uniform kernel, the weight is defined as: 2 For the truncated-normal kernel, the ramp is centered at the midpoint of the window, with the spread chosen so that the curve increases from nearly 0 to 1 within the L-month window. Using the standard normal cumulative distribution function , the weight is scaled to [0,1] as follows: 3 4 Results Participant characteristics The study included a total of 93,396 pediatric inpatients from January 2016 to December 2019, with 45,493 cases before the DIP reform (2016–2017) and 47,903 cases after the reform (2018–2019). As shown in Table 1 , the median age of hospitalized children increased slightly after the reform [0.42 (1.94) vs. 0.48 (2.32) years, P < 0.001], whereas the biological sex distribution remained stable, with boys accounting for approximately 63% and girls 37% in both periods. The observed male predominance is largely due to several male-specific or male-predominant diagnoses (e.g., phimosis, hydrocele, cryptorchidism). Additionally, the number of rural patients increased with a corresponding decline in urban patients ( χ 2 = 152.42, P < 0.001). Although the total number of complex and critical cases decreased significantly, the number of non-local complex and critical cases increased significantly. In addition, mean LOS decreased after the reform ( t = 6.86; P < 0.001), while mean total expenditure per case increased ( Z = -28.36; P < 0.001). Table 1. Characteristics of hospitalized patients, 2016–2019 Variables Before DIP reform, 2016–2017 After DIP reform, 2018–2019 t/Z/χ 2 Discharge cases, N 45,493 47,903 Age (years), median (IQR) 0.42 (1.94) 0.48 (2.32) -6.80 ***, a Biological sex, N (%) 0.09 b Male 28,820 (63.35) 30,394 (63.45) Female 16,673 (36.65) 17,509 (36.55) Residence, N (%) 152.42 ***, b Rural 8279 (18.20) 10,263 (21.42) Urban 37,214 (81.80) 37,640 (78.58) Case type, N (%) 7.31 ***, a General 12,701 (27.92) 13,798 (28.92) Acute 12,527 (27.54) 14,372 (30.13) Complex 10,218 (22.46) 9341 (19.58) Critical 10,047 (22.08) 10,196 (21.37) Outcomes Complex and critical cases, N (%) 20,265 (44.54) 19,537 (40.78) 122.74 ***, b Non-local complex and critical cases, N (%) 4147 (9.12) 4697 (9.81) 13.19 ***, b LOS (day), mean ± SD 8.22 ± 7.70 7.87 ± 7.67 6.86 ***, c Mean total expenditure per case (CNY), median (IQR) 6698.95 (7180.45) 7867.13 (9089.21) -28.36 ***, a Mean total expenditure per case (USD), median (IQR) 995.39 (1066.93) 1168.96 (1350.55) -28.36 ***, a Open in a new tab Note: ***, P < 0.001. a, Z test; b, χ² test; c, t test. CNY, Chinese Yuan; USD, United States Dollar; IQR, interquartile range; SD, standard deviation; LOS, length of stay Total service volume As shown in Table 2 ; Fig. 1 A, total monthly pediatric admissions exhibited a significant increasing baseline trend before the DIP reform [ β = 0.008 (0.006 to 0.009); RR = 1.008 (1.006 to 1.009); P < 0.001]. At the time of DIP implementation, there was a modest but statistically significant immediate increase in admissions [ β = 0.067 (0.010 to 0.125); RR = 1.007 (1.010 to 1.133); P = 0.022]. However, the post-intervention monthly trend showed a significant decline relative to the pre-intervention trajectory [ β = −0.006 (− 0.008 to − 0.004); RR = 0.994 (0.992 to 0.996); P < 0.001], corresponding to a counterfactual slope of − 14.13 patients/month (95% CI: −56.8 to 5.51). Table 2. Interrupted time series analyses for outcomes of hospitalized patients before and after the DIP reform Indicators Estimate 95% CI P Total service volume Total monthly pediatric admissions Baseline monthly slope ( β 1) 0.008 0.006, 0.009 < 0.001 Step change ( β 2) 0.067 0.010, 0.125 0.022 Monthly slope change ( β 3) -0.006 -0.008, -0.004 < 0.001 Case mix severity Proportion of complex/critical cases per month Baseline monthly slope ( β 1) -0.010 -0.013, -0.007 < 0.001 Step change ( β 2) 0.652 0.535, 0.768 < 0.001 Monthly slope change ( β 3) -0.018 -0.023, -0.014 < 0.001 Absolute number of complex/critical cases per month Baseline monthly slope ( β 1) 0.002 -0.001, 0.004 0.151 Step change ( β 2) 0.462 0.374, 0.549 < 0.001 Monthly slope change ( β 3) -0.017 -0.020, -0.014 < 0.001 Inter-regional Referral of Complex Cases Proportion of non-local complex/critical cases per month Baseline monthly slope ( β 1) 0.001 -0.004, 0.005 0.819 Step change ( β 2) 0.390 0.235, 0.544 < 0.001 Monthly slope change ( β 3) -0.012 -0.018, -0.007 < 0.001 Absolute number of non-local complex/critical cases per month Baseline monthly slope ( β 1) 0.008 0.004, 0.012 < 0.001 Step change ( β 2) 0.401 0.260, 0.542 < 0.001 Monthly slope change ( β 3) -0.016 -0.022, -0.011 < 0.001 Service Efficiency Mean LOS (days) Baseline monthly slope ( β 1) -0.025 -0.047, -0.002 0.043 Step change ( β 2) 0.661 -0.145, 1.468 0.119 Monthly slope change ( β 3) -0.019 -0.048, 0.011 0.223 Medical expenditures Mean total expenditures per case in USD Baseline monthly slope ( β 1) 0.011 0.007, 0.016 < 0.001 Step change ( β 2) 0.247 0.092, 0.403 0.004 Monthly slope change ( β 3) -0.010 -0.016, -0.004 0.001 Open in a new tab Note: Interrupted time series analyses controlled for patient’s age, sex, residence and seasonality. LOS, length of stay; USD, United States Dollar Fig. 1. Open in a new tab Interrupted time series analysis of DIP payment reform among all hospitalized pediatric inpatients. ( A ) Total monthly admissions; ( B ) Proportion of complex and critical cases; ( C ) Absolute number of complex and critical cases; ( D ) Proportion of non-local complex and critical cases; ( E ) Absolute number of non-local complex and critical cases; ( F ) Mean LOS; ( G ) Mean total expenditures per case. For panel G, predicted cost values are displayed on the original scale after back-transformation from the log scale used in the Gamma regression models Case mix severity The proportion of complex and critical cases showed a significant declining baseline trend prior to the reform [ β = −0.010 (− 0.013 to − 0.007); OR = 0.990 (0.987 to 0.993); P < 0.001; Table 2 ; Fig. 1 B]. At the time of DIP implementation, a substantial immediate increase was observed [ β = 0.652 (0.535 to 0.768); OR = 1.919 (1.708 to 2.156); P < 0.001], followed by a further significant decline in the post-intervention monthly trend compared with baseline [ β = −0.018 (− 0.023 to − 0.014); OR = 0.982 (0.978 to 0.986); P < 0.001]. The absolute number of complex and critical cases did not show a statistically significant baseline trend before the reform [ β = 0.002 (− 0.001 to 0.004); RR = 1.002 (0.999 to 1.004); P = 0.151; Fig. 1 C]. However, a significant immediate increase was observed at the time of DIP implementation [ β = 0.462 (0.374 to 0.549); RR = 1.587 (1.454 to 1.732); P < 0.001]. Subsequently, the monthly trend declined significantly relative to the pre-intervention period [ β = −0.017 (− 0.020 to − 0.014); RR = 0.983 (0.980 to 0.986); P < 0.001], corresponding to − 14.42 patients/month (95% CI: −48.4 to − 3.71). Inter-regional referral of complex cases The proportion of non-local complex and critical cases showed no significant baseline trend prior to the reform [ β = 0.001 (− 0.004 to 0.005); OR = 1.001 (0.996 to 1.005); P = 0.819; Table 2 ; Fig. 1 D]. At the time of DIP implementation, a significant immediate increase was observed [ β = 0.390 (0.235 to 0.544); OR = 1.476 (1.265 to 1.722); P < 0.001], followed by a significant decline in the post-intervention monthly trend [ β = −0.012 (− 0.018 to − 0.007); OR = 0.988 (0.982 to 0.993); P < 0.001]. The absolute number of non-local complex and critical cases showed a significant increasing baseline trend before the reform [ β = 0.008 (0.004 to 0.012); RR = 1.008 (1.004 to 1.013); P < 0.001; Fig. 1 E]. A significant immediate increase was also observed at the time of DIP implementation [ β = 0.401 (0.260 to 0.542); RR = 1.494 (1.297 to 1.720); P < 0.001]. However, the post-intervention monthly trend declined significantly relative to the baseline trend [ β = −0.016 (− 0.022 to − 0.011); RR = 0.984 (0.979 to 0.989); P < 0.001], corresponding to − 6.48 patients/month (95% CI: −21.84 to − 1.75). Service efficiency for all hospitalized patients Mean LOS showed a modest but statistically significant declining baseline trend prior to the DIP reform [ β = −0.025 (− 0.047 to − 0.002); P = 0.043; Table 2 ; Fig. 1 F]. The implementation of the DIP reform was not associated with a statistically significant immediate level change [ β = 0.661 (− 0.145 to 1.468); P = 0.119], nor with a significant change in the post-intervention monthly trend [ β = −0.019 (− 0.048 to 0.011); P = 0.223], with a counterfactual slope of − 0.02 days/month (95% CI: −0.11 to 0.04). Medical expenditures for all hospitalized patients Mean total expenditures per case showed a significant increasing baseline trend before the DIP reform [ β = 0.011 (0.007–0.016); P < 0.001; Table 2 ; Fig. 1 G]. At the time of DIP implementation, there was a significant immediate increase in expenditure [ β = 0.247 (0.092 to 0.403); P = 0.004]. However, the post-intervention monthly trend showed a significant decline relative to the baseline trend [ β = −0.010 (− 0.016 to − 0.004); P = 0.001], corresponding to a counterfactual slope of − 24.17 USD/month (95% CI: −85.95 to − 4.36). As shown in Table S1 , the study hospital experienced additional expenditures in the first two years following DIP implementation compared with the counterfactual scenario, whereas the latter two years were associated with net cost savings. Over the full 4-year study period, the net difference amounted to approximately USD 3.30 million. Subgroup analysis of diseases The top five ICD-10 diagnoses were Q21.1 (atrial septal defect), P59.9 (neonatal jaundice, unspecified), J18.0 (bronchopneumonia, unspecified organism), P23.9 (congenital pneumonia, unspecified), and Q25.0 (patent ductus arteriosus). Overall patterns were broadly consistent with the full sample (Supplementary Table S2 – S6 ), while diagnosis-specific variations were observed. For P59.9, total monthly pediatric admissions showed a significant immediate decrease following DIP implementation [ β = −0.588 (− 0.748 to − 0.428); P < 0.001], accompanied by a significant increase in the post-intervention monthly trend [ β = 0.017 (0.012 to 0.022); P < 0.001]. In contrast, no significant changes were observed in the absolute number of complex and critical cases. Mean LOS increased significantly immediately after the reform [ β = 2.633 (0.182 to 5.085); P = 0.044] and subsequently showed a modest but significant declining trend over time [ β = −0.092 (− 0.172 to − 0.013); P = 0.029]. For J18.0, no significant change in the post-intervention trend of mean total expenditures was observed. For P23.9, neither the immediate level change nor the post-intervention trend in the proportion of non-local complex and critical cases was statistically significant. For Q25.0, no significant immediate change in total monthly pediatric admissions was observed following the reform. Sensitivity analysis Tables S7 – S8 and S9 – S10 present results for the uniform and truncated-normal kernels under 3- and 6-month lag specifications, respectively. For the absolute number of non-local complex/critical cases per month, the intervention was not associated with a statistically significant immediate level change, suggesting that the immediate effect was sensitive to alternative lag structures. For mean LOS, all sensitivity analyses indicated a significant declining baseline trend prior to the intervention, a statistically significant immediate increase following the intervention, and a significant post-intervention monthly decreasing trend. All other findings were consistent with those of the main analysis. Discussion This study applied a quasi-experimental interrupted time series design to evaluate the association between DIP payment reform and pediatric hospital services. We observed an immediate increase in total service volume, case mix severity, and inter-regional referral of complex cases, which attenuated over time. Service efficiency, measured by mean LOS, did not change materially. Mean total expenditure per case increased immediately after implementation and then showed a downward trend thereafter. In sensitivity analyses, mean LOS showed a significant month-by-month decline following the intervention, while other findings were consistent with the primary analysis. Together, these results add to the evidence on case-based payment reform in China and underscore considerations specific to pediatric care. Regarding total service volume, case-mix severity, and inter-regional referral of complex cases, findings from both the primary analysis and sensitivity analyses accounting for policy lag effects indicated that total pediatric admissions declined immediately after the reform, whereas the proportion and number of complex/critical and non-local complex/critical cases increased. This pattern is consistent with prior evidence suggesting that prospective payment reforms can enhance tertiary hospitals’ capacity to admit critically ill patients [ 7 , 20 , 21 ]. However, the subsequent month-to-month decline in these three outcomes suggests that the early increase may have reflected transitional dynamics, such as changes in coding practices, short-term shifts in referral behavior, and the hospital’s adaptation to the new reimbursement incentives. Similar responses have been documented in other health system reforms, where hospitals react strongly at the outset and then stabilize as new routines become embedded [ 22 ]. Compared with studies in adult populations reporting more sustained improvements in case mix after DRG- or DIP-based payment [ 23 , 24 ], pediatric hospitals may be more prone to short-term fluctuations. More importantly, following the reform, both the proportion and absolute number of non-local complex/critical cases demonstrated a gradual month-by-month decline. This pattern may suggest that the DIP payment reform mitigated the “siphoning effect” of provincial tertiary hospitals on critically ill patients, potentially encouraging more patients to seek care within local medical institutions and thereby objectively reinforcing the implementation of the hierarchical medical system. However, comparable empirical evidence remains limited, and further studies are warranted to verify this potential mechanism. In terms of efficiency, we observed no significant immediate or gradual change in LOS in the primary analysis. This contrasts with findings from adult-focused studies, in which DRG- or DIP-based payment has been consistently associated with shorter LOS, likely reflecting stronger incentives for earlier discharge [ 5 , 25 ]. However, in sensitivity analyses incorporating 3- and 6-month lag structures, both an immediate reduction and a significant month-by-month decline in LOS were observed following the reform, suggesting that the effect on efficiency may be delayed rather than absent. This pattern is plausible, as changes in clinical pathways and discharge practices may require a longer adaptation period compared with administrative or coding adjustments. Similarly, a study by Aragón et al. [ 26 ] using 15 years of administrative data in England found that reductions in length of stay under DRG-based payment did not occur immediately but instead increased progressively over time. It is worth noting that several features of pediatric inpatient care may further attenuate or delay this mechanism. First, baseline LOS in children’s hospitals is often relatively short because admissions are dominated by acute infections, injuries, and perinatal conditions that typically require brief inpatient management, leaving limited room for further reduction [ 13 ]. Second, the effect of case-based payment on LOS varies by diagnosis; prior studies have reported larger decreases for elective surgical cases than for acute medical admissions, and the case mix in pediatric hospitals is weighted toward the latter [ 27 ]. Third, pediatric patients—particularly neonates and infants—may require additional observation to ensure clinical stability and safety, making early discharge less feasible than in adult populations [ 28 ]. With respect to expenditures, DIP reform was associated with a “spike-and-decline” pattern, with total spending increasing in the implementation month and then decreasing thereafter. This pattern is consistent with evidence from adult hospitals showing short-term increases during the early adaptation period followed by longer-term cost containment [ 4 , 29 , 30 ]. The initial rise may reflect transitional processes as hospitals adjust coding practices, update documentation, and modify clinical pathways to align with DIP requirements. During this phase, providers may increase diagnostic testing or interventions to ensure appropriate classification, which could temporarily elevate costs [ 31 ]. Our subgroup analyses provide empirical support that the initial rise in expenditures was partly due to enhanced documentation and coding of complex/critical cases. For example, among hospitalized patients with P59.9 (Table S3 ), total admissions decreased immediately after implementation, while both the proportion and absolute number of complex/critical cases, as well as the proportion and absolute number of non-local complex/critical cases, showed significant increases. Therefore, the early rise in costs likely reflects intensified documentation and coding of complex/critical cases, particularly among neonates and other high-severity patient groups. Hospitals may have “caught up” on previously under-recorded high-complexity cases to ensure adequate reimbursement under DIP, temporarily elevating total medical expenditures. In addition, clinicians and administrators may adopt more “defensive” resource use to reduce the risk of under-coding or reimbursement loss under a new payment system [ 32 ]. Cost increases may also partly reflect the release of previously unmet or underprovided services that became reimbursable under standardized rates. The subsequent declines in total spending suggest that, once implementation stabilized, hospitals adapted by rationalizing service delivery and improving operational efficiency. Although DIP reform may strengthen insurance fund sustainability, it can also entail short-term cost shifting to patients and families, which is a substantive concern in settings where affordability influences care-seeking [ 33 ]. Prior studies have shown that higher costs may delay care for children, especially among rural and low-income households [ 34 , 35 ]. Accordingly, policy design should balance cost containment objectives with equitable access to pediatric services. Subgroup analyses of the top five pediatric diagnoses (Q21.1 – atrial septal defect, P59.9 – neonatal jaundice, J18.9 – pneumonia, P23.9 – congenital pneumonia, and Q25.0 – patent ductus arteriosus) generally supported the overall findings but revealed disease-specific variations. For instance, Neonatal jaundice (P59.9) demonstrated a significant post-reform reduction in LOS, consistent with its well-standardized clinical pathways and low resource variability. In contrast, pneumonia (J18.0) showed a decrease in the total monthly pediatric admissions post-intervention. However, the proportion of complex/critical cases per month and the proportion of non-local complex/critical cases per month increased, which may reflect greater case severity and inter-regional referral patterns. Meanwhile, congenital pneumonia (P23.9) and patent ductus arteriosus (Q25.0)—conditions characterized by inherently high resource needs and frequent ICU involvement—showed more stable patterns, suggesting that DIP’s financial incentives may exert less influence on highly protocolized neonatal care. These variations reinforce that pediatric diagnostic groups differ widely in severity distribution, treatment intensity, and coding complexity, and that payment models may require condition-specific adjustments to avoid misalignment between reimbursement and actual resource use [ 11 ]. Our findings have several important policy implications. First, DIP reform may strengthen local treatment capacity. The overall monthly trend in complex and critical cases declined before and after the reform. This downward trajectory did not reverse after implementation, suggesting that the reform did not lead to a sustained increase in provincial referral hospitals’ capacity to manage severely ill pediatric patients. In contrast, no significant baseline trend was observed for the proportion of non-local complex and critical cases prior to DIP implementation. However, it showed a short-term increase at the time of implementation and then declined persistently on a monthly basis. This pattern could indicate reduced cross-regional inflow of severe cases, suggesting more children may have been treated locally. Therefore, future policy efforts should prioritize strengthening pediatric critical care and referral continuity at the prefecture and county levels, as well as improving the tiered delivery system and regional coordination. This will help maintain care quality while reducing unnecessary cross-regional care-seeking burdens. Second, the spike-and-decline pattern of expenditures suggests that transitional support is needed during the early phase of reform. This may include enhanced training for physicians on coding accuracy, and closer monitoring of hospital behavior to detect unintended consequences, such as sudden increases in high-weight diagnostic codes, abrupt shifts in case-mix toward more profitable groups, reductions in admissions of low-weight patients, or unexpected changes in referral patterns that do not align with clinical needs [ 7 , 36 ]. Third, the significant post-reform monthly decline in mean total expenditure is particularly noteworthy in the pediatric context, where healthcare utilization is highly sensitive to household financial constraints, underscoring that equity considerations should remain central to future reform. A large body of research has documented that case-based payment reforms may trigger adaptive provider behaviors during their early implementation phase, including intensified coding, adjustments in case-mix, increased diagnostic testing, or selective admission of profitable patients. For example, Zhang et al. reported clear evidence of patient-selection effects following the introduction of DIP in Chinese hospitals [ 31 ]. Similarly, Cattel and Eijkenaar found that prospective payment systems without adequate oversight may lead to upcoding or unnecessary service use [ 12 ]. Given that DIP standardizes reimbursement and reduces discretionary charging space, it has the potential to improve financial protection, but targeted monitoring is needed to ensure equitable impact across socioeconomic subpopulations. Several limitations should be noted. First, this was a single-center study conducted in a tertiary maternal and child health hospital in Guangzhou. Although the study hospital is broadly representative of high-level pediatric institutions in China, the findings may not be generalizable to other regions, lower-level facilities, or different hospital types. Second, while the interrupted time series design accounts for pre-intervention trends, residual confounding cannot be excluded, including concurrent policy changes, secular system shifts, or seasonal epidemics that may have influenced the outcomes [ 37 ]. In addition, we were unable to explicitly identify or quantify the impact of these factors due to data limitations, and thus their potential influence on the observed trends cannot be fully ruled out. Future studies using multi-center data or incorporating external policy may help to better disentangle these effects. Third, our study relied on hospital discharge summary data, which do not directly contain DIP-specific information. As a result, we were unable to compute DRG-like weighted monthly averages of DIP weights to accurately reflect changes in case mix. Fourth, this study did not evaluate out-of-pocket payments. The main reason is that our dataset includes a substantial number of newborns. At birth, newborns typically have not yet completed insurance registration, so all hospital charges must be initially paid in full by their parents. At this time, the recorded out-of-pocket payments in the case record database effectively equal the total medical expenditures. Although parents can apply for retrospective reimbursement within 1–6 months after birth, the case record database does not update these payments. Therefore, the out-of-pocket proportion in this mixed cohort does not accurately reflect the actual economic burden. Fifth, this study is based solely on discharge data from a single hospital, and the observed outcomes and economic effects reflect only this institution. We did not extrapolate the results to the national level for healthcare expenditures, as doing so could introduce substantial uncertainty and risk misinterpretation or inappropriate citation. Therefore, the findings are primarily applicable to the study hospital, and future research using more comprehensive, multi-center, or nationwide data is needed to evaluate the overall impact of the DIP payment reform. Sixth, we did not evaluate a comprehensive set of safety or quality outcomes. In particular, neonatal mortality events were rare during the study period, precluding reliable statistical inference. Finally, the study period ended in 2019, before the COVID-19 pandemic. Because COVID-19 likely altered pediatric care-seeking, hospital operations, and the policy environment, patterns observed after 2020 may differ from those reported here. Conclusions The DIP payment reform was associated with a transient increase in total service volume, as well as in the proportion and number of complex/critical and non-local complex/critical cases, with these effects attenuating over time. Per-case total spending increased immediately after implementation but subsequently declined, whereas mean length of stay showed a reduction after accounting for different lag periods. These findings suggest that pediatric-tailored refinements to DIP are warranted to balance insurance fund sustainability with family affordability, alongside strengthened financial protection and quality safeguards to maximize the benefits of case-based payment reform in pediatric care. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 (61.1KB, docx) Acknowledgements We thank all members of our study team for their whole‑hearted cooperation and the original authors of the included studies for their wonderful work. Author contributions YL and JZ conceptualized and designed the study, collected data, carried out the initial analyses, and drafted the initial manuscript. XZ designed the study methods, and developed the software used in the analyses. JW critically reviewed and revised the manuscript. WC conceptualized and designed the study, and critically reviewed and revised the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. Funding This study was supported by the Natural Science Foundation of China (Grant number 1: 82473731; Grant number 2: 72074229). Data availability The extracted data will be shared on reasonable request from the corresponding author via email, (mailto: chenw43@mail.sysu.edu.cn). Declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of the Guangdong Provincial Maternity and Child Health Hospital with the approval number 20251019. As anonymized secondary data were used, individual informed consent was waived. The study followed the principles of the Declaration of Helsinki. Consent for publication Not applicable. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Ying Liu and Jianghui Zhang contributed equally to this work. References 1. Meng Q, Mills A, Wang L, Han Q. What can we learn from China’s health system reform? BMJ (Clinical Research Ed). 2019;365:l2349. [ DOI ] [ PMC free article ] [ PubMed ] 2. Statistical bulletin on the development of national healthcare security in 2024. [ https://www.nhsa.gov.cn/art/2025/7/14/art_7_17248.html?spm=5176.28103460.0.0.96a07551XuKHH9 ]. 3. Wang X, Tao Y, Gao S, Feng J, Huang A, Lin L, Cheng H. Effects of DRG/DIP payment reform on hospital pharmacy administration and pharmaceutical services in China: a multicenter cross-sectional study. Front public health. 2025;13:1585279. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Yip W, Fu H, Chen AT, Zhai T, Jian W, Xu R, Pan J, Hu M, Zhou Z, Chen Q, et al. 10 years of health-care reform in China: progress and gaps in Universal Health Coverage. Lancet (London England). 2019;394(10204):1192–204. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Busse R, Geissler A, Aaviksoo A, Cots F, Häkkinen U, Kobel C, Mateus C, Or Z, O’Reilly J, Serdén L, et al. Diagnosis related groups in Europe: moving towards transparency, efficiency, and quality in hospitals? BMJ (Clinical Research Ed). 2013;346:f3197. [ DOI ] [ PubMed ] 6. Policy interpretation of the notice on issuing the 2.0 version grouping schemes for Diagnosis-Related Groups (DRG) and Diagnosis-Intervention Packet (DIP) payment and deeply promoting the relevant work. [ https://www.nhsa.gov.cn/art/2024/7/23/art_105_13316.html ]. 7. Chang J, Chen S, Li A, Yang X, Luo H, Yilamu M, Fu B, Xu N, Liu J, Tian H. Facilitators and barriers to the implementation of DIP payment methodology reform in a public hospital in Guangzhou: a qualitative study based on the implementation of the meta-framework for research (CFIR) framework. Front public health. 2025;13:1569855. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Wang R, Yan J, Zhang X, Qian M, Ying X. Impact of the diagnosis-intervention packet payment reform on provider behavior in China: a controlled interrupted time series study. Int J Health Policy Manage. 2024;13. [ DOI ] [ PMC free article ] [ PubMed ] 9. Luo M, Li H, Li R, Wu Y, Lan Y, Xie S. The impact of DRG payment reform on inpatient costs for different surgery types: an empirical analysis based on Chinese tertiary hospitals. Front public health. 2025;13:1563204. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Meng Z, Hui W, Cai Y, Liu J, Wu H. The effects of DRGs-based payment compared with cost-based payment on inpatient healthcare utilization: A systematic review and meta-analysis. Health Policy. 2020;124(4):359–67. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Zhang Y, Xu SY, Tan GM. Unraveling the effects of DIP payment reform on inpatient healthcare: insights into impacts and challenges. BMC Health Serv Res. 2024;24(1):887. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Cattel D, Eijkenaar F. Value-Based Provider Payment Initiatives Combining Global Payments With Explicit Quality Incentives: A Systematic Review. Med care Res review: MCRR. 2020;77(6):511–37. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Hu X, Lin X, Dai Z, Fang K. Global burden of pediatric fracture (1992–2021) and projections of future disease burden trends. BMC Pediatr. 2025;25(1):416. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Zhu Y, Ding Y, Qin S. Investigating the relationship between social deprivation and health outcomes in China: using spatial regression analysis approach. BMC Public Health. 2025;25(1):951. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Yan X, Dong D, He S, Webster C. Examining Trans-Provincial Diagnosis of Rare Diseases in China: The Importance of Healthcare Resource Distribution and Patient Mobility. Sustainability. 2020;12(13):5444. [ Google Scholar ] 16. Guo Y, Sun J, Hu S, Nicholas S, Wang J. Hospitalization costs and financial burden on families with children with depression: a cross-section study in Shandong Province, China. Int J Environ Res Public Health. 2019;16(19):3526. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Anaby D, Lal S, Huszczynski J, Maich J, Rogers J, Law M. Interrupted time series design: a useful approach for studying interventions targeting participation. Phys Occup Ther Pediatr. 2014;34(4):457–70. [ DOI ] [ PubMed ] [ Google Scholar ] 18. Liu M, Jia M, Lin Q, Zhu J, Wang D. Effects of Chinese medical pricing reform on the structure of hospital revenue and healthcare expenditure in county hospital: an interrupted time series analysis. BMC Health Serv Res. 2021;21(1):385. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Zhang X, Wu K, Pan Y, Zhong W, Zhou Y, Guo T, Yin R, Chen W. Introducing activation functions into segmented regression model to address lag effects of interventions. BMC Med Res Methodol. 2023;23(1):277. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Zhang M, Wang G, Liu H, Wen Y, Chen L. Impact of China’s Diagnosis-Intervention Packet Policy on Hospitalization Costs for Patients With Malignant Tumors: A 2019–2022 Interrupted Time Series Analysis. Risk Manage Healthc policy. 2025;18:655–65. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Mo G, Zhu E, Fang X, Ma J, Kong S, Guo X, Lu Z. Evaluation of the effect of DRG payment policy based on interrupted time series modeling: evidence from a tertiary hospital in Anhui Province. Health Res Policy Syst. 2024;22(1):167. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Barnum H, Kutzin J, Saxenian H. Incentives and provider payment methods. Int J Health Plann Manag. 1995;10(1):23–45. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Xiao P, Liu Z. The impact of DRG/DIP payment reform on medical behavior. In: Proceedings of the 2024 3rd International Conference on Public Health and Data Science: 2024;2024:256–266. 24. Zou K, Li HY, Zhou D, Liao ZJ. The effects of diagnosis-related groups payment on hospital healthcare in China: a systematic review. BMC Health Serv Res. 2020;20(1):112. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Zeng JQ. The pilot results of 47 148 cases of BJ-DRGs-based payment in China. Int J Health Plann Manag. 2019;34(4):1386–98. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Aragón MJ, Chalkley M, Kreif N. The long-run effects of diagnosis related group payment on hospital lengths of stay in a publicly funded health care system: Evidence from 15 years of micro data. Health Econ. 2022;31(6):956–72. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Berkman ND, Chang E, Seibert J, Ali R, Porterfield D, Jiang L, Wines R, Rains C, Viswanathan M. Management of high-need, high-cost patients: a best fit framework synthesis, realist review, and systematic review. 2021. [ PubMed ] 28. Frankel LR, Hsu BS, Yeh TS, Simone S, Agus MSD, Arca MJ, Coss-Bu JA, Fallat ME, Foland J, Gadepalli S, et al. Criteria for Critical Care Infants and Children: PICU Admission, Discharge, and Triage Practice Statement and Levels of Care Guidance. Pediatric critical care medicine: a. Crit Care Soc. 2019;20(9):847–87. journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Gao C, Xu F, Liu GG. Payment reform and changes in health care in China. Soc Sci Med (1982). 2014;111:10–16. [ DOI ] [ PubMed ] 30. Tang S, Brixi H, Bekedam H. Advancing universal coverage of healthcare in China: translating political will into policy and practice. Int J Health Plann Manag. 2014;29(2):160–74. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Zhang X, Tang S, Wang R, Qian M, Ying X, Maciejewski ML. Hospital response to a new case-based payment system in China: the patient selection effect. Health Policy Plann. 2024;39(5):519–27. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Eggleston K, Yip W. Hospital competition under regulated prices: application to urban health sector reforms in China. Int J Health Care Finance Econ. 2004;4(4):343–68. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Xu K, Evans DB, Carrin G, Aguilar-Rivera AM, Musgrove P, Evans T. Protecting households from catastrophic health spending. Health Aff. 2007;26(4):972–83. [ DOI ] [ PubMed ] [ Google Scholar ] 34. Wisk LE, Witt WP. Predictors of delayed or forgone needed health care for families with children. Pediatrics. 2012;130(6):1027–37. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Kraft AD, Quimbo SA, Solon O, Shimkhada R, Florentino J, Peabody JW. The health and cost impact of care delay and the experimental impact of insurance on reducing delays. J Pediatr. 2009;155(2):281–5. e281. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Su M, Zhang Q, Bai X, Wu C, Li Y, Mossialos E, Mensah GA, Masoudi FA, Lu J, Li X, et al. Availability, cost, and prescription patterns of antihypertensive medications in primary health care in China: a nationwide cross-sectional survey. Lancet (London England). 2017;390(10112):2559–68. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002;27(4):299–309. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1 (61.1KB, docx) Data Availability Statement The extracted data will be shared on reasonable request from the corresponding author via email, (mailto: chenw43@mail.sysu.edu.cn). Articles from BMC Health Services Research are provided here courtesy of BMC ACTIONS View on publisher site PDF (1.7 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• SKIP for PA Study: Team and Leadership Level Implementation Support for Collaborative Care#825734
• Data Management and Sharing Plan for: STOP PEDS RCT (Streamlined Treatment of Pulmonary Exacerbations in Pediatrics Randomized Clinical Trial)#207836
• Proceedings from the 2025 Midwest Pediatric Device Consortium showcase featuring software as a medical device.#174523
• Proceedings from the 2025 Midwest Pediatric Device Consortium showcase featuring software as a medical device#170860
• "Reading level is just one component": best practices for cognitive debriefing patient-reported outcome instruments with pediatric populations.#420929