Inpatient surgical activity and hospital charges during early implementation of diagnosis intervention packet payment in Inner Mongolia a city level difference in differences analysis
Diagnosis-Intervention Packet (DIP) payment in China assigns points to case groups defined by diagnosis and procedure and pays hospitals from a regional budget with a floating point value. We tested three hypotheses for early DIP payment: budget pressure lowers charges and procedures per admission (H1); procedure-defined grouping raises procedures per admission and lowers the average fee per procedure (H2); and competition for budget share raises admissions (H3). We used published statistics for 66 city-years from 11 cities or leagues in Inner Mongolia, 2016–2021: three pilots that began DIP payment in 2021 and eight regions that adopted it in 2022. City and year fixed-effects difference-in-differences models were estimated for six outcomes, with a wild cluster bootstrap enumerating all 2,048 Rademacher assignments as primary inference. Synthetic difference-in-differences, permutation and leave-one-city-out analyses were sensitivity checks; results with p < 0.10 were reported as suggestive signals with small-sample-corrected p values. The estimates favoured H2. Recorded procedures rose by 11,983 per city-year (+13.7%; 95% confidence interval [CI], −4,503 to 34,578; p = 0.272), and synthetic difference-in-differences gave 11,502 (approximate p = 0.062). Procedures per 100 admissions rose by 2.14 (+10.4%; p = 0.328), reaching p = 0.030 when Ordos was omitted (0.141 after small-sample correction). The fee per recorded procedure fell by CNY 280.77 (−11.7%; p = 0.164), reaching p = 0.044 without Ulanqab (corrected p = 0.060). H3 was not supported: admissions fell by 4.5%, the interval excluded increases above 3.4%, and hospital-only admissions gave wild bootstrap p values of 0.098 to 0.104. Charges per discharge changed by −3.1% (95% CI, −16.5% to 11.9%). No primary estimate reached p < 0.05; all five signals were suggestive. Early DIP payment in Inner Mongolia did not expand admissions or cut charges per discharge by more than about a sixth. Suggestive signals consistently pointed to more procedures per admission at a lower fee per procedure, a procedure-intensity response that patient-level data should test.
Authors
- Shuyu Wen (ORCID: https://orcid.org/0000-0002-2859-3659)
- Chao Zhang (ORCID: https://orcid.org/0000-0002-5957-2287)
- Sixian Du (ORCID: https://orcid.org/0009-0009-9804-4645)
- Xinping Zhang
Institutions
- Wuhan Union Hospital (CN)
- Huazhong University of Science and Technology (CN)
Publication Details
- Journal
- BMC Health Services Research
- Published
- 2026-11-19
- DOI
- https://doi.org/10.1186/s12913-026-16199-y
- Primary Topic
- Healthcare Systems and Reforms
- Type
- article
- Field-Weighted Citation Impact
- 0.00