Direct Comparison of Patient-completed and Physician-completed Caprini Scores for Plastic Surgery Patients

We examined the ability of plastic surgery patients to perform their own venous thromboembolism (VTE) risk stratification using a previously validated patient-completed Caprini risk scoring sheet.

Patients' Caprini scores were obtained by an attending physician at an office visit through interview and chart review. Subsequently, patients independently completed a previously validated patient reported scoring sheet. We compared patient and physician reported Caprini scores by each Caprini subquestion and by the overall score. We described discordance of scores using frequencies and proportions, quantified the level of agreement using Cohen's Kappa coefficient, Spearman's correlation coefficient, and the Wilcoxon rank-sum test, and visualize scoring differences using Bland-Altman plots, where appropriate.

We prospectively enrolled 50 patients. Only 24% (n = 12) of patients had exact matches in physician-completed and patient-completed Caprini scores. Among the 76% (n = 38) with discrepancies, 26 received a higher patient-reported score (median = 2 points, range 1-8 points) and 12 received a lower patient-reported score (median = -1.5 points, range -1 to -6 points). Existing venous thromboembolism prophylaxis guidelines support chemical prophylaxis for inpatients with Caprini scores ≥7. Among 38 patients with score discrepancies, 8 (21.1%) would have been incorrectly prescribed chemical prophylaxis and 4 (10.5%) would have been incorrectly denied chemical prophylaxis.

Plastic surgery patients cannot reliably calculate their own 2005 Caprini scores. Reliance on patient completed scores alone would promote ~25% of patients receiving inappropriate prophylaxis strategies.