ISOQOL 2004 Symposium Monday, June 28, 2004 9:00 - 10:30 am Measuring Quality of Life for Policy Analysis: Past, Present
and Future The purpose of health care is to extend life expectancy and to improve quality of life. Until the late 1960s, quality of life was rarely discussed or measured. Over the last few decades the growth in quality of life studies has been nearly exponential. Several approaches to quality of life assessment were developed to supply data for comprehensive health policy models. This presentation traces the development of generic approaches to quality of life measurement and reviews applications of the methods in policy analysis. Problems with the models and related measurement approaches will be reviewed and directions for future research will be offered. Bridging the gaps between
conceptualization, application and practice: Optimizing the analysis
of patient-reported outcomes in clinical and health research The emergence of evidence-based methods for medical and public health research has led to an increasing emphasis on the roles of evaluation and inferential statistics to determine the effectiveness of new medical interventions and public health programs. While there has been an exponential rise in research involving measurement and assessment of quality of life and patient-reported outcomes, including the development of thousands of disease-specific instruments, advances in analytical methods dealing with their application to medical and health care evaluation has not been as widely developed. There is great need for such development since the standard statistical tests used in clinical and health research typically ignore measurement error, invoke time-invariance assumptions, assume normality and seldom deal with the required statistical methods for overall hypothesis testing of multiple endpoints. While newer approaches to assessment, such as item banking and computer adaptive testing, promise to aid in the standardization and efficiency of measurement and assessment, these methods will not solve the problems of evaluation, statistical analysis and inference faced by researchers. Unless some major critical analytical issues are addressed, applications of the advances in measurement and assessment to research and practice will be limited. The purpose of this session is to provide guidance for using the scales and indices of quality of life and other patient-reported outcomes in the evaluation of medical interventions and public health programs. Examples will be given to highlight potential areas of ongoing and future research. The specific focus of this presentation will be on procedures for handling multiple outcomes to answer the question "Is this treatment or program beneficial to the patient or participant?" A review of several methods will be given including, alpha adjustment, latent regression models, restricted hypothesis testing and overall statistical tests to address this issue.
|