Recommendations for Assessing Methodological Quality in Systematic Reviews of Population Pharmacokinetic Models: A Systematic Review

BACKGROUND: Population pharmacokinetic (popPK) models are increasingly used to support model-informed precision dosing owing to their abilities to account for variability in drug exposure. However, there is no accepted/validated risk of bias (RoB) framework tailored to systematic reviews that externally evaluate popPK models. Therefore, this study was conducted to explore how existing systematic reviews on popPK assess model quality and bias and appraise RoB. METHODS: A systematic review was conducted in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines. Systematic reviews that externally evaluated popPK models quantitatively and reported bias/accuracy metrics were searched for on Embase, MEDLINE, PubMed, Web of Science, Cochrane Library, Google Scholar, and CINAHL from inception to November 2025. Data on study selection, external datasets, and model evaluation metrics were extracted. RoB was assessed with RoB in systematic reviews (ROBIS). RESULTS: Twenty-two systematic reviews were included. Considerable variation existed in study selection approaches, external validation datasets, and bias assessment metrics. Prediction error-based metrics were most frequently reported (n = 21), followed by Bayesian forecasting and simulation-based diagnostics (both n = 14). ROBIS assessment indicated recurrent concerns regarding the identification/selection of studies and collection/appraisal of data. Common sources of bias were limitations of external datasets (n = 19; retrospective design, sparse sampling, small sample size), heterogeneity in bioanalytical methods (n = 8), and unaccounted treatment-related factors such as concomitant medication (n = 4). CONCLUSIONS: ROBIS is only partially applicable to systematic reviews on popPK and insufficiently captures external validation-specific issues. An "external validation" domain with signaling questions focused on dataset provenance, adequacy, and assay consistency is proposed for improving transparency, reproducibility, and comparability across future systematic reviews on popPK.