Data Sheet 1_Development of a nomogram for identifying prevalent medication discrepancy in maintenance hemodialysis patients: a multicenter cross-sectional study.docx

Model performance was evaluated using 1,000-resample bootstrap optimism correction and leave-one-center-out internal-external validation.</p>Results<p>Medication discrepancy was identified in 256 of 328 participants (78.05%). BMQ showed an overall nonlinear association with prevalent medication discrepancy (global P < 0.001; test for nonlinearity P = 0.011). In the primary model, urban residence had an odds ratio (OR) of 1.93 (95% CI 0.99–3.75; P = 0.053) versus rural residence, while monthly income >5,000 CNY had an OR of 0.33 (95% CI 0.15–0.73; P = 0.006) versus <3,000 CNY.

The apparent AUC was 0.723 (95% CI 0.663–0.783) and the bootstrap optimism-corrected AUC was 0.701. Leave-one-center-out validation yielded a pooled AUC of 0.601, Brier score of 0.180, calibration intercept of 0.722, and calibration slope of 0.362, with substantial heterogeneity across centers.</p>Conclusion<p>The three-domain nomogram showed moderate apparent discrimination but substantially attenuated and heterogeneous cross-center performance.

It should be regarded as a development-stage identification model for prevalent medication discrepancy and requires independent external validation, and potentially recalibration, before clinical use.</p>