Prasad M, Christie JD, Bellamy S. — University of Pennsylvania
Holmboe ES, Lipner RS, Hess BJ. — American Board of Internal Medicine
Rubenfeld GD. — University of Toronto
Kahn JM. — University of Pittsburgh
Presented: American Thoracic Society Meeting, May 2011
Introduction: Clinical protocols are associated with improved patient outcomes in the ICU yet may affect trainee education. We studied the relationship between fellowship training in ICUs with mechanical ventilation protocols and knowledge about ventilator management, as measured by performance on certification examination questions.
Methods: We surveyed medical ICU directors in U.S. teaching hospitals about protocol availability in three areas of mechanical ventilation management: ventilation liberation, lung-protective ventilation and sedation management. We linked program responses on this survey to each trainee’s results on the 2008 and 2009 ABIM Critical Care Medicine Certification Examination. Programs were classified a priori as either high-intensity protocol programs (≥ 2 of the 3 clinical protocols) or low-intensity protocol programs (<2 protocols). Our primary outcome of interest was performance on the mechanical ventilation questions of the corresponding year’s ABIM examination, estimated as a standardized ability score derived using item-response theory. We used trainee-level linear regression and generalized estimating equations to estimate the relationship between training in a high-intensity protocol program and mechanical ventilation score, adjusting for birth country, medical school country and overall equated first-attempt score on their prior ABIM Internal Medicine Certification Examination.
Results: Of 129 training programs, 90 (70%) responded. Seventy-seven (86%) had protocols for ventilation liberation, 66 (73%) had protocols for sedation, and 54 (60%) had protocols for lung-protective ventilation. Of these programs, 27 (31%) had zero, 19 (22%) had one, 24 (27%) had two, and 18 (20%) had three protocols for at least three years. Therefore, 42 (47%) programs were classified as high-intensity and 46 (53%) as low-intensity, with 304 (55%) and 249 (45%) trainees respectively. A total of 553 examinees from 88 of these programs took the ABIM Critical Care Medicine Certification Exam in 2008 and 2009. In univariate analyses, there was no difference in mechanical ventilation scores in high-protocol compared to low-protocol programs (mean score: 497 ± 6 vs. 497 ± 5; difference in means=0; p=0.995). In multivariable analyses, there was no significant association of training in a high-protocol program with score (adjusted mean difference -5.36; p=0.495). The results were similar in a sensitivity analysis in which we defined protocol availability as a more flexible categorical variable (for 0, 1, 2 or 3 available protocols).
Conclusions: The availability of clinical protocols is not associated with differences in trainee knowledge about mechanical ventilation management. Based on these data, protocols should not be viewed as a barrier to effective mechanical ventilation education.
For more information about this presentation, please contact Research@abim.org.