A systematic review and meta-analysis examined risk prediction models for peristomal moisture-associated skin damage in patients with enterostomy in China. The researchers looked at 11 models drawn from 10 studies to see how well they predicted skin damage. The incidence rate of this skin damage ranged from 22.8% to 59.1% across these models. The accuracy of the models, measured by AUC values, ranged from 0.812 to 0.914. History of radiotherapy, stoma type, stoma opening height, and surgical wound in the plate area were identified as the strongest predictors.
Despite these findings, all included models exhibited a substantial risk of bias. This uncertainty limits the clinical utility of the models for guiding prophylactic strategies. The methodological quality and applicability of these existing models remain uncertain for real-world use.
Readers should understand that these models offer potential utility but cannot be fully trusted yet. The high risk of bias reduces certainty in the models predictive ability. Do not infer that the identified predictors are causal. Until methodological issues are addressed, these tools should not be used as the sole basis for clinical decisions.