Since most radiologists do not meet with patients, questionnaires often substitute for face-to-face interviews to collect patients' history. We report the effect of direct radiologist-patient interviews on the quality of patient history recorded for musculoskeletal magnetic resonance imaging.
Magnetic resonance imaging questionnaires completed by outpatients were separated into 2 cohorts: (1) imaging center (IC) forms with no radiologist interviews; (2) hospital (H) forms with radiologist-patient interviews. Three blinded radiologists independently scored each questionnaire for quality on a 5-point scale. A separate quantitative analysis was also performed. The unpaired t test, Fischer exact test, and χ2 test were used to compare the cohorts.
The mean score of the H cohort among reviewers was superior to the IC cohort: 3.79 (±0.98) versus 3.04 (±1.00), P < .0001. Each reviewer also independently found the H cohort to be of higher quality for patient history, P < .0001. For the IC cohort, 7.8% of questionnaires did not report a single symptom versus 0.0% in the H cohort, P = .0331. Also, the IC cohort recorded symptoms in 2 or less words more often than the H cohort, P < .0001.
Brief radiologist-patient interviews are superior for obtaining a higher quality of patient history for musculoskeletal MRI than patient questionnaires alone.
Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine, Baltimore (Drs Davis, Mulligan, and Resnik); and Angelina Diagnostic Radiology Associates, Lufkin, Texas (Dr Moszkowicz).
Correspondence: Derik L. Davis, MD, Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine, 22 S. Greene St, Baltimore, MD 21201 (email@example.com).
Abstract presented at the Radiological Society of North America 99th Scientific Assembly and Annual Meeting, Chicago, Illinois, December 4, 2013.
The authors thank Carly Stokum, MPH, for her assistance with statistical analyses in the study.
There is no source of funding to disclose. All authors of this manuscript have no conflicts of interest to declare.