Attrition and certification failure remain persistent challenges in paramedic education, reducing workforce readiness and straining training programs. Despite widespread use of selective admissions criteria, little empirical evidence exists regarding the predictive validity of selective admissions examinations in paramedicine. The aim of this study is to evaluate whether preadmission academic performance and scores on the Fisdap Paramedic Entrance Exam predict successful program completion and first-attempt certification. A repeated cross-sectional study was conducted using records from one paramedic program (N = 109). Program completion was defined as on-time graduation with first-attempt success on the NREMT examination. Three logistic regression models were tested: academic predictors, Fisdap entrance exam subscales, and a combined model. Model performance was evaluated using likelihood-ratio chi-square tests, Nagelkerke R 2 , and AUC. Both the academic model (R 2 = 0.15; AUC = 0.69, 95% CI 0.57–0.81; LR χ 2 (4) = 11.19, p = 0.025) and the Fisdap model (R 2 = 0.28; AUC = 0.77, 95% CI 0.68–0.87; LR χ 2 (3) = 21.59, p < 0.001) were statistically significant. Institutional grade point average (GPA) and anatomy grade positively predicted completion in the academic model, while only the reading subscale was a significant positive predictor in the Fisdap model. In the combined model (R 2 = 0.26; AUC = 0.77, 95% CI 0.68–0.86; LR χ 2 (2) = 19.79, p < 0.001), the Fisdap composite score remained a significant positive predictor, whereas the GPA did not. The combined model demonstrated a sensitivity of 0.81 and a specificity of 0.43. Entrance exam performance, particularly reading comprehension, demonstrated moderate predictive validity beyond prior academic achievement and showed stronger associations with program completion than did institutional GPA. These findings support using cognitive entrance exam data to assess readiness and complement holistic admissions processes in paramedic education. Multi-site studies are needed to confirm generalizability and refine evidence-based selection and support practices.
Joseph M. McTaggart (Mon,) studied this question.