Decompositional silence: indicator retirement and concealed within-department variation in six years of hospital medical work reports from a Chinese tertiary general hospital
Year: 2026
Type: preprint
Abstract: This project asks how the information a hospital collects about itself changes over time, and what is lost when it does. Using 74 consecutive monthly medical work reports from a Chinese tertiary general hospital (January 2020 to April 2026; 2,219 unit-month observations, 48 reporting units, 42 indicators), we examine whether indicators withdrawn from the report differ systematically from those retained. Each indicator is classified as a "level" — a self-standing quantity — or a "decomposition", whose function is to split another reported figure into interpretable parts. A second analysis exploits the single month in which the hospital reported at ward rather than departmental level, to quantify the within-department variation that aggregation conceals. The classification carrying the primary outcome will be re-made by two blinded, independent raters with third-rater adjudication. The dataset has already been analysed by the lead author and three earlier hypotheses were tested and rejected; this is set out in full in the registration, and only the blinded rating and the specified analysis are prospective. No patient-level data is used.
Authors: Xu Peng, Chengguang Wang, Chaoyang Qu
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7203758155