Predictive models for patients treated with or evaluated for peritoneal dialysis: a scoping review protocol

This scoping review will map available research regarding predictive models for clinical and patient-reported outcomes in patients receiving or considered for peritoneal dialysis.

Chronic kidney disease affects over 840 million people worldwide, with peritoneal dialysis offering a home-based kidney replacement therapy. Although evidence demonstrates equivalent long-term survival between dialysis modalities, the individual patient prognosis may favor a particular modality, making personalized outcome prediction essential. Available research evidence is not reflected in guideline recommendations, with clinicians relying on clinical experience rather than evidence-based prediction models. An evaluation of prediction models is required to identify validated tools and guide their clinical implementation.

This review will include patients receiving peritoneal dialysis or who are being considered for initiation, with no restrictions regarding age, modality, or chronicity of disease. The scope will encompass the development, validation, and evaluation of clinical models utilizing multiple predictors for peritoneal dialysis-relevant outcomes. Any health care setting will be included, with no geographic or language restrictions.

MEDLINE (Ovid), Embase, Web of Science Core Collection, and Scopus will be searched, complemented by citation tracking and gray literature searches. Two independent reviewers will conduct title/abstract screening and full-text assessment using predefined criteria. Data extraction will be performed using standardized forms combining CHARMS, TRIPOD+AI, PROBAST, and PROGRESS-Plus. Data synthesis will follow a descriptive quantitative analysis and narrative synthesis, with outcomes categorized using the Standardised Outcomes in Nephrology-Peritoneal Dialysis framework. The analysis will combine outcome mapping, predictor database development, model validation assessment, clinical accessibility evaluation, and population gap identification.

OSF https://osf.io/7dmaq/.