Reliability of bimanual pelvic examinations performed in emergency departments

Abstract

OBJECTIVE:To test the reliability of bimanual pelvic examinations performed in emergency departments by emergency medicine physicians.DESIGN:Prospective observational study; 2 examiners each recorded various pelvic examination findings on 186 patients.SETTING:A private university hospital and a public county hospital staffed by attending emergency medicine physicians who share an emergency medicine residency program.SUBJECTS:Senior resident (3rd or 4th year) and attending emergency physicians.MAIN OUTCOME MEASURES:Percentage of agreement and percentage of positive agreement for cervical motion tenderness, uterine tenderness, adnexal tenderness, adnexal mass, and uterine size (within 2 cm).RESULTS:The agreement ranged between 71% and 84%, but the percentage of positive agreement was much lower, ranging from 17% to 33%. Agreement for uterine size, within 2 cm, was 60%.CONCLUSION:The findings of bimanual pelvic examinations performed by emergency physicians in an emergency department have poor interexaminer reliability.