Today's News

Thursday, March 1, 2001

Recognize limits of data in TKA outcome studies

Researchers who use patient recall data to derive preoperative status in outcome studies of total knee arthroplasty (TKA) must recognize the limitations of the data, investigators said Wednesday in scientific paper 69.

Their paper compares prospectively collected preoperative data with patient recall of preoperative status three months after TKA. Patients were recruited as part of a prospective observational study of primary TKA for osteoarthritis from the United Kingdom (six centers), the United States (four centers) and Australia (two centers). Preoperative data included: clinical history and physical exam, demographics, socioeconomic status, and two health status measures (WOMAC and SF-36). Three-month data included: WOMAC, SF-36 and recall of preoperative status using items selected from these health status measures.

A total of 884 patients were recruited and recall data was obtained on 778 patients (88 percent). Comparisons of prospective and recalled pain and function items showed poor to fair weighted Kappa (.20-.40). Patients recalled significantly more pain (p< 0.001), but there was random recollection error of the function items.

Correlations between prospective and recalled pain (r = .53) and function (r = .49) summary scores were only moderate and a large proportion of patients (47-62 percent) recalled scores differed by 10 percent or greater from their preoperative scores. Patients over 75 years, depressed patients and patients whose function had deteriorated at three-months compared to preop had significantly poorer recall (p< 0.05).

"Recall of preoperative pain and functional status offers only moderate agreement with what patients report prospectively," the investigators say. "Additionally, subgroups of patients have poorer recall making standardized adjustments to recall data difficult."

Coauthors of the study are Elizabeth A. Lingard, MPH; and Clement B. Sledge, MD, department of orthopaedic surgery, Brigham and Women's Hospital, Boston, Mass. and Elizabeth A. Wright, PhD, RBB Multipurpose Arthritis and Musculoskeletal Diseases Center, Brigham and Women's Hospital, Harvard Medical School, Boston, Mass.

Previous Page
2001 Academy News March 1 Index C

Last modified 14/February/2001 by IS