Frontiers in medicine · 2026
Drug-cost trajectories after national volume-based procurement in Chinese public hospitals: a 31-province interrupted time-series analysis.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Mao Jing, Liu Guangqi
Аннотация
BACKGROUND: China's national volume-based procurement (NVBP) policy reduced prices for selected medicines, but national aggregate drug-cost trends are exposed to concurrent reforms and nonlinear prepolicy trajectories. We assessed public-hospital drug-cost trajectories around 2019 and explored hospital revenue structure.
METHODS: This uncontrolled ecological interrupted time-series study used a balanced panel of 31 province-level units from 2011 through 2024 (434 province-years). Province fixed-effects segmented regressions estimated immediate level and annual trend changes in per-encounter drug costs and drug-cost shares. Sensitivity analyses examined nonlinear and piecewise pretrends, non-drug costs as a control series, calendar-year fixed effects in drug-versus-non-drug comparative models, consumer-price-index deflation, multiple testing, and influential years. Revenue structure was described in 35 observations from 4 data-available provinces, with wild cluster bootstrap inference reported only as exploratory.
RESULTS: No primary outcome showed a clear immediate 2019 level change (all P > .05). Under the prespecified single-linear-pretrend model, annual trends changed by -1.67 Chinese yuan (CNY) per outpatient visit [95% confidence interval (CI), -2.70 to -0.63] and -164.53 CNY per inpatient discharge (95% CI, -199.80 to -129.26). In the linear model, the model-implied 2024 differences were -5.31 CNY per outpatient visit (95% CI, -11.70 to 1.07; P = .100) and -753.18 CNY per inpatient discharge (95% CI, -924.76 to -581.60; P < .001). However, results were highly sensitive to pretrend form: with a 2016-2018 piecewise pretrend, corresponding changes were 1.64 CNY (95% CI, 0.27 to 3.02) and 17.11 CNY (95% CI, -33.46 to 67.68). Inpatient non-drug cost also changed substantially (-409.50 CNY/year; 95% CI, -483.03 to -335.97), and the inpatient drug-versus-non-drug differential trend change was not significant after calendar-year fixed effects (P = .488). Drug-cost shares remained below 2011 levels but declined more slowly after 2019 in all specifications. The 4-province revenue trend-change interval crossed zero with wild cluster bootstrap inference (P = .055). Conclusion: The single-linear ITS suggested post-2019 changes in aggregate drug-cost trajectories, but nonlinear pretrends and synchronous non-drug changes materially weakened their specificity to NVBP. These annual ecological data support temporal description, not attribution of the observed breakpoint to NVBP alone.
METHODS: This uncontrolled ecological interrupted time-series study used a balanced panel of 31 province-level units from 2011 through 2024 (434 province-years). Province fixed-effects segmented regressions estimated immediate level and annual trend changes in per-encounter drug costs and drug-cost shares. Sensitivity analyses examined nonlinear and piecewise pretrends, non-drug costs as a control series, calendar-year fixed effects in drug-versus-non-drug comparative models, consumer-price-index deflation, multiple testing, and influential years. Revenue structure was described in 35 observations from 4 data-available provinces, with wild cluster bootstrap inference reported only as exploratory.
RESULTS: No primary outcome showed a clear immediate 2019 level change (all P > .05). Under the prespecified single-linear-pretrend model, annual trends changed by -1.67 Chinese yuan (CNY) per outpatient visit [95% confidence interval (CI), -2.70 to -0.63] and -164.53 CNY per inpatient discharge (95% CI, -199.80 to -129.26). In the linear model, the model-implied 2024 differences were -5.31 CNY per outpatient visit (95% CI, -11.70 to 1.07; P = .100) and -753.18 CNY per inpatient discharge (95% CI, -924.76 to -581.60; P < .001). However, results were highly sensitive to pretrend form: with a 2016-2018 piecewise pretrend, corresponding changes were 1.64 CNY (95% CI, 0.27 to 3.02) and 17.11 CNY (95% CI, -33.46 to 67.68). Inpatient non-drug cost also changed substantially (-409.50 CNY/year; 95% CI, -483.03 to -335.97), and the inpatient drug-versus-non-drug differential trend change was not significant after calendar-year fixed effects (P = .488). Drug-cost shares remained below 2011 levels but declined more slowly after 2019 in all specifications. The 4-province revenue trend-change interval crossed zero with wild cluster bootstrap inference (P = .055). Conclusion: The single-linear ITS suggested post-2019 changes in aggregate drug-cost trajectories, but nonlinear pretrends and synchronous non-drug changes materially weakened their specificity to NVBP. These annual ecological data support temporal description, not attribution of the observed breakpoint to NVBP alone.