The quality of quality of life studies in general surgical journals

Abstract

Background: There has been a tremendous increase in interest on quality of life in surgical research. An increase in interest does not necessarily translate into better research. This study evaluates surgical articles that claim to measure or make some conclusion on quality of life. Study Design: All articles published in the calendar years 1996 and 1999 that purported to assess quality of life as end points or make some conclusion about quality of life were chosen for review from eight general surgical journals. Articles were assessed for use of a quality of life instrument, type of instrument, validation of the instrument, appropriateness of the instrument for the hypothesis, quality of statistical analysis, and adherence to the Gill and Feinstein criteria. Results: Of the 18 articles published in 1996, 72% used a quality of life instrument. Eighteen instruments were used in 13 studies: 7 generic, 10 disease-specific, and 1 ad hoc. Forty-three percent were validated, 39% were appropriate for the study hypothesis, 39% had correct statistical analysis. The majority did not meet the Gill and Feinstein criteria. Of the 24 studies published in 1999, 63% used a quality of life instrument. Twenty-two instruments were used in 15 studies: 11 generic, 5 disease-specific, and 6 ad hoc. Fifty-five percent were validated, 45% were appropriate, 45% had correct statistical analysis. Once again, the majority did not meet the Gill and Feinstein criteria. Conclusions: Despite the emphasis on quality of life outcomes, a substantial number of studies made errors in conceptually defining quality of life and in use of quality of life instruments. Researchers and journal reviewers need to be better versed on the techniques of quality of life research.