Jan 1, 1992
Published in: Measuring Functioning and Well-Being: The Medical Outcomes Study Approach / edited by Anita L. Stewart and John E. Ware, Jr. (Durham, N.C.: Duke University Press, 1992), Chapter 20, p. 345-371
Posted on RAND.org on January 01, 1992
The goal of MOS was to develop a set of measures appropriate for diverse populations that could be used as a standard for comparisons of populations with different chronic health problems, as well as for healthy populations. The MOS samples represent a wide range of ages, including many elderly. Some MOS patients were relatively well, having only mild hypertension; others were quite ill, having severe and complicated heart disease and other comorbid problems. Many MOS patients had depression, either alone or in combination with a medical problem. This chapter summarizes the underlying conceptual framework for MOS, discusses the methods of constructing MOS measures, describes the four sets of MOS measures, explains the state-of-the-art advances of MOS measures, identifies ways of selecting measures from the MOS sets and of preparing a questionnaire, and discusses future directions for the MOS. Comprehensive tables of definitions and statistics are included. Each MOS measure is as unique as possible. Investigators can be assured that redundancy is less likely than when scales are chosen from different instruments. The comprehensiveness of a health survey is crucial, because little is known about which concepts are most affected by disease and treatment and which are most important. The MOS is more comprehensive than any other well-known health surveys. However, it does have gaps: It needs to directly assess changes in health over time and to address issues of missing data (e.g., people who are more depressed are less likely to return self-administered instruments).