Plastic Surgery

Pediatric pressure‑injury risk models show discrimination but poor applicability and bias

Twelve pediatric PI prediction models showed AUCs ranging from 0.612 to 0.978 but most lacked proper calibration and external validation. All models were rated high risk of bias and 10 had major applicability concerns, limiting clinical generalizability. Common methodological flaws included low events‑per‑variable, poor handling of missing data, and inappropriate variable categorization. Until multicenter validation, age stratification, and clinical‑utility assessment are performed, do not rely on these models for routine bedside risk stratification. 1

Pivotal medical device trials rarely embed equity in design or analysis

Scoping review of 74 pivotal device studies found age and sex were usually reported but rarely used in subgroup analyses. Only 18.9% performed age-based and 14.8% sex-based subgroup analyses; race/ethnicity was reported in 35.1%. Explicit EDI framing was present in only 2.7% of trials, and none used population benchmarking or CONSORT‑Equity. Interpret device trial results cautiously for underrepresented groups and request representativeness data during regulatory or procurement decisions. 2

Protocol for a living RCT systematic review on communicating results to participants

This living systematic review protocol will compare randomized trials of communication tools for sharing research results with participants and people with lived experience. Primary outcomes include satisfaction, objective and perceived understanding, and ease of use. Monthly automated searches and prospective living updates are planned to incorporate new RCT evidence over time. Until RCT evidence is synthesized, prefer simple, tested plain‑language summaries and measure participant comprehension locally. 3