Development and validation of a geriatric knowledge test for medical students

OBJECTIVES: To assesses the reliability and validity of a geriatrics knowledge test designed for medical students. DESIGN: Cross-sectional studies. SETTING: An academic medical center. PARTICIPANTS: A total of 343 (86% of those sampled) medical students participated in the initial study, including 137 (76%) first-year, 163 (96%) third-year, and 43 (86% of those sampled) fourth-year students in the 2000-2001 academic year. To cross-validate the instrument, another 165 (92%) third-year and 137 (76%) first-year students participated in the study in the 2001-2002 academic year. MEASUREMENTS: An 18-item geriatrics knowledge test was developed. The items were selected from a pool of 23 items. An established instrument assessing the clinical skills of medical students was included in the validation procedure. RESULTS: The instrument demonstrated good reliability (Cronbach alpha=0.80) and known-groups and concurrent validity. Geriatrics knowledge scores increased progressively with the higher level of medical training (mean percentage correct=31.3, 65.3, and 66.5 for the first-year, third-year, and fourth-year classes, respectively, P<.001). A significant (P<.01) relationship was found between the third-year students' geriatrics knowledge and their clinical skills. Similar results, except the relationship between knowledge and clinical skills, were found in the cross-validation study, supporting the reliability and known-groups validity of the test. CONCLUSION: The 18-item geriatrics knowledge test demonstrated sound reliability and validity. The average scores of the student groups indicated substantial room for growth. The relationship between geriatrics knowledge and overall clinical skills needs further investigation.