Assessing Scoring Systems for Athletic Patients
Abstract & Commentary
Synopsis: All 4 knee outcome scales that were assessed satisfied criteria for reliability, validity, and responsiveness and were all considered acceptable for use in clinical research. The Activities of Daily Living Scale of the knee outcome survey did have some advantages over the other scoring systems.
Source: Marx RG, et al. J Bone Joint Surg Am. 2001;83A(10):1459-1469.
Although many knee scoring systems are available to evaluate athletic patients, little information is available to evaluate effectiveness of the measurement properties of these scales, or to support the use of one questionnaire over another. Marx and associates evaluated 4 commonly used knee rating systems including the Lysholm scale, the Cincinnati knee-rating scale, the American Academy of Orthopaedic Surgeons sports knee-rating scale (originally included in MODEMS), and the Activities of Daily Living scale of the knee outcome survey. Only athletic patients with a Tegner score of ³ 4 were included. An effort was made to get a cross-section of knee problems to include both operative and nonoperative diagnoses. This was in response to the issue that many prior publications had focused on use of a more homogenesis population of just ligament injuries. Only the subjective, outcomes component of each scoring system was used to measure true disabilities (things patients could or could not do) rather than examination components to measure impairments (loss of motion or stability). These were compared to a standardly accepted outcomes measure, the Short Form-36.
Each of the scales was assessed for reliability, validity, and responsiveness. Patients were recruited while in the waiting room before seeing a sports medicine physician. Numbers of patients were determined by power analysis. Test-retest reliability was assessed on a subset of patients that had a stable diagnosis that was not expected to change. These patients were administered the questionnaires about 1 week apart to determine whether the test results would be reliability reproduced for a stable diagnosis. The validity, an assessment of whether the instrument actually measures what it is intended to measure, was determined by the baseline responses of 133 patients. The opinions of 5 sports medicine orthopaedic surgeons as well as the subjective opinions of the patients were used to assess the validity of the scoring systems and compared to the SF-36. Responsiveness was determined with a different set of 42 patients who were administered the questionnaire before and after treatment to determine whether the test could assess a measured improvement.
Their findings confirmed that all 4 scales satisfied their criteria for reliability, validity, and responsiveness, and, therefore, all were acceptable for use in clinical research. All of the scales correlated with each other as well as with the SF-36 and subjective opinions of the patients and clinicians.
Comment by David R. Diduch, MS, MD
For those of us who have attempted to perform clinical research and determine which rating system is optimal, this paper is a welcome addition to the literature. Unfortunately, this does not attempt to compare the different scoring systems to arrive at a definite best rating scale. However, it does validate the use of all 4 of these scales by determining that they satisfy criteria for reliability, validity, and responsiveness. This is important. It is then left to the researcher to choose among these scoring systems. It would be helpful to have included the International Knee Documentation Committee scoring system, as this is widely used as well.
While they did not attempt to demonstrate an advantage of one system over another, they do conclude at the end of the paper that they recommend the Activities of Daily Living scale of the knee outcome survey for several reasons. One is that the scale is well understood by patients and could be completed in a short period of time due to its clear wording. It includes 17 multiple choice questions divided into 2 sections, one for symptoms and one for functional disability. It seemed to effectively evaluate a wide variety of symptoms and could be easily applied to various knee diagnoses at once. Ironically, this is probably the one used least often of the group.
The AAOS Sports Knee Rating scale was originally developed for MODEMS and was the most cumbersome with a total of 23 questions that are divided up into 5 subscales. This was found to be too complicated with some responses that could not be answered and, therefore, had to be dropped and scored as missing. This is the scoring system that has been least validated in the literature previously. This was not endorsed. The Lysholm scoring system consisted of 8 items and was originally designed to focus on knee ligament surgery outcomes. Its limitation appears to be application to a wide variety of diagnoses. The Cincinnati knee rating system had 11 components, which are part of a larger scoring system that also included physical exam elements that were not included in this analysis for reasons above. The problem they noted with the Cincinnati scoring system was the increased measurement variability in the responses that produced a somewhat decreased reliability.
This paper is helpful to clearly outline what each of these scoring systems do well and not do well, and also to establish their effectiveness in clinical research. Although they did not intend to compare one scoring system to another, they do provide some helpful recommendations and endorse the Activities of Daily Living Scale. Further studies that specifically focus on one scoring system to another would be extremely useful additions to the literature.
Dr. Diduch, Associate Professor, Department of Orthopaedic Surgery, University of Virginia School of Medicine, Charlottesville, VA, is Editor of Sports Medicine Reports.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content