Complication-attributable admission episodes after shunt surgery for idiopathic normal pressure hydrocephalus in Japan: a scenario-based micro-costing analysis of hospital margins under DPC/PDPS and fee-for-service

Shunt-related complications after surgery for idiopathic normal pressure hydrocephalus (iNPH) generate repeat admissions, procedures, and bed use, but their provider-side financial consequences depend on the reimbursement rule. Evidence on hospital accounting margins under Japan’s mixed Diagnosis Procedure Combination/Per-Diem Payment System (DPC/PDPS) remains limited. This study compared modeled hospital margins under DPC/PDPS and a fee-for-service (FFS) comparator and, as a secondary analysis, estimated infection-specific prevention thresholds. A deterministic scenario-based micro-costing model was constructed from the hospital perspective. The primary analysis comprised 10 mutually exclusive complete complication pathways: suspected infection without immediate shunt removal (A1), a two-admission deep-infection pathway consisting of shunt removal followed by staged reconstruction (A23), two subdural-lesion pathways, four mechanical-revision pathways, one abdominal-complication pathway, and one seizure pathway. A2 and A3 were analyzed as component admissions of A23 but were not counted as additional independent pathways. Revenue and cost were calculated separately for each admission and then summed for A23. One-way, two-way, joint-scenario, and length-of-stay (LOS) sensitivity analyses were performed. Under DPC/PDPS, all 10 complete pathways had positive base-case margins, ranging from Japanese yen (JPY) 35,230 to JPY 231,428. Under FFS, 9 of 10 pathways had negative margins, ranging from JPY − 284,283 to JPY − 29,951; total shunt revision (C4) was the only positive pathway, at JPY 7,859. Within A23, the A2 component margin was JPY − 8,145 under DPC/PDPS and JPY − 295,282 under FFS, whereas A3 was JPY 239,573 and JPY 10,999, respectively. The A23 FFS margin was JPY − 284,283, calculated by summing the two admission-level ledgers, including admission-specific inpatient and other FFS revenue items. At the base ward cost, the A23 DPC/PDPS margin fell from JPY 420,088 at a relative per-diem coefficient of 1.2 to JPY 42,768 at 0.8; it became JPY − 447,142 when ward cost was JPY 55,000/day. Extending A23 from 35 to 49 bed-days changed its DPC/PDPS margin from JPY 231,428 to JPY − 29,582. For A1 plus A23, the DPC/PDPS accounting margin of JPY 275,359 was offset at an opportunity cost of JPY 6,556 per occupied bed-day. Reimbursement design substantially changed modeled provider margins. Positive accounting margins under DPC/PDPS did not establish that complication admissions were financially desirable or that the payment system was policy-optimal, because local ward cost, LOS, and capacity constraints could reverse the result. The estimates should be recalibrated with institution-specific cost and claims data.