Cost distribution and Monte Carlo simulation of financial risk under DRG-based prospective payment for laparoscopic cholecystectomy in patients with gallbladder polyps: a single-center retrospective study.

BACKGROUND

Under China's Diagnosis-Related Group (DRG) prospective payment system, hospitals bear the financial risk of costs that exceed the fixed reimbursement standard, making case-level cost distribution and the probability of loss critical to departmental sustainability for high-volume, low-complexity procedures such as laparoscopic cholecystectomy.

OBJECTIVE

However, the probability distribution of hospitalization costs and the quantitative financial risk faced by a single institution operating under a fixed DRG payment standard for the group HS10B (laparoscopic cholecystectomy without common bile duct exploration) remain unclear.

METHODS

This retrospective single-center study included 475 consecutive patients who underwent laparoscopic cholecystectomy (surgery code 51.2300) for gallbladder polyps (DRG group HS10B) over a five-year period (2020-2024; annual case volume ranged from 50 to 136). Total hospitalization cost, medication cost, and material cost were extracted; other cost was derived as total cost minus medication cost minus material cost. Distributional properties were characterized using Kolmogorov-Smirnov normality testing and comparative fitting of normal, log-normal, and Gamma distributions (Akaike information criterion [AIC] and goodness-of-fit testing). Monte Carlo simulation (20,000 iterations), combining parametric resampling from the fitted Gamma distribution and non-parametric bootstrap resampling of the observed cohort, was used to estimate the probability distribution of case-level and aggregate financial margin relative to the local DRG reimbursement standard of ¥12,592.

RESULTS

Mean total hospitalization cost was ¥10,293.17 (SD 1,442.22; median ¥10,263.40, IQR ¥9,258.10-11,174.65), which was right-skewed (skewness 0.54) and best represented by a Gamma distribution (Kolmogorov-Smirnov p = 0.819). Twenty-two of 475 cases (4.63%) incurred costs exceeding the ¥12,592 breakeven threshold. Monte Carlo simulation estimated a mean per-case margin of ¥2,302.55 (95% simulation interval, -¥686.76 to ¥4,928.44) and a case-level loss probability of 6.2%. Bootstrap resampling of the observed cohort (n = 475) projected a mean aggregate margin of ¥1,091,888 (95% CI ¥1,029,920-1,153,715), with no net cohort-level loss observed among the 20,000 simulated resamples (upper one-sided 95% bound, 0.015%); an annual projection assuming 95 cases/year yielded a mean annual margin of ¥218,354 (95% CI ¥190,780-245,908), again with no simulated annual loss (upper bound, 0.015%); results were consistent when total cost was modeled as log-normal instead of Gamma. Total cost differed significantly across admission year (Kruskal-Wallis p < 0.001), declining from a mean of ¥10,172 in 2020 to ¥9,329 in 2024, with the empirical case-level loss rate falling correspondingly from 7.4 to 0% over the same interval.

CONCLUSION

Inpatient cost for HS10B laparoscopic cholecystectomy follows a right-skewed distribution adequately described by a Gamma model; in this single-center cohort, case-specific cost volatility at the individual level was accompanied by a positive aggregate margin at the departmental level, consistent with a risk-pooling effect at this case volume. Hospital administrators should transition from conventional mean-centric accounting toward distribution-aware financial surveillance to better anticipate tail risk under prospective payment systems.