Abstract
BACKGROUND
Sickle cell disease (SCD) affects ~100,000 people in the US and is associated with multi-system acute and chronic complications that shorten lifespan. While most pediatric patients in the US survive beyond age 18, the risk of death increases dramatically during late adolescence and young adulthood. This surge is due partly to insufficient access to comprehensive sickle cell and subspecialty care. As a result, patients with episodic SCD care have high rates of ED use and hospitalization, incurring substantial cost. Emergency physicians, hospitalists, and primary care physicians are frequently responsible for these highly complex patients and studies suggest that they lack necessary SCD knowledge and training.
Question banks have emerged as the most widely used educational resource among medical students and residents studying for licensing exams. Medical trainees favor question banks over all other learning platforms, including lecture, to learn new material and for review. Question banks feature "highest-yield" topics and may be viewed as the most important clinical knowledge to carry into practice.
Our study aims to examine SCD-related content in the most popular commercial question bank to determine how this important teaching platform serves to educate medical trainees about the complex pathophysiology and scope of SCD.
METHODS
We obtained access to three question banks used by trainees to prepare for the current USMLE Step 1, Step 2, and Step 3 exams. The keyword "sickle cell" was entered into a search feature. Inclusion criteria consisted of question vignettes that included a patient with SCD. Exclusion criteria comprised questions featuring patients with sickle cell trait or other hemoglobinopathy and non-SCD questions with SCD-related answer choices. Resulting questions were analyzed for disease-related topic and patient age.
Questions were assigned to one of the following categories, selected due to high incidence of clinically significant complications per NHLBI guidelines: vasooclusive crisis (VOC)/pain, fever/infection, stroke, acute chest syndrome (ACS), spleen, renal, priapism, or hepatobiliary. Questions outside of these categories were designated "other." Patient age was categorized by group: birth to 5 years, 6-10 years, 11-15 years, 16-20 years, and 20 years or older.
For clinical vignettes with multiple questions in series, questions were categorized individually while the patient was counted once (ex: 5-year-old with VOC in first question, develops ACS in second question: one "VOC/pain," one "ACS," and one patient 0-5 years). Test questions that were unrelated to the clinical vignette were categorized by the patient's clinical history (ex: 12-year-old frequently hospitalized for pain and observed sickle-shaped RBCs on peripheral smear most likely has what sequence in the beta-globin gene? Categorized as "VOC/pain" and patient 11-15 years).
RESULTS
There were 9041 questions total in the three banks. "Hematology & Oncology" questions comprised 5.3% (478/9041) of content and less than 1% were tagged "sickle cell" (0.55%, 50/9041). However, SCD rarely formed the question content (0.24%, 22/9041). A total of 20 patients with SCD were presented in questions.
Pediatric patients were featured in almost every question (90.0%, 18/20) and the median patient age was 7.5 years. Two questions involved adult patients (M, F; age 23) and both presented to the ER with pain. Acute VOC pain, including dactylitis, was the most common complaint (36.4%, 8/22), followed by patients with fever or infection (27.3%, 6/22). Acute chest syndrome and splenic dysfunction (from fibrosis) were each tested once and acute stroke was tested twice (Table 1). There were no questions about renal failure, priapism, hepatobiliary complications, or splenic sequestration.
CONCLUSIONS
Our study found a significant proportion of questions focused on acute SCD pain in pediatric patients. Adult SCD patients were largely omitted from this question bank's content. Given that medical trainees favor question banks over all other learning platforms, this finding may suggest a limited scope of understanding of this disease.
No relevant conflicts of interest to declare.
