A method for imputing missing questionnaire data
Jenkinson C., Harris R., Fitzpatrick R.
© Cambridge University Press 2011. Introduction When conducting research with questionnaire-based outcome instruments, it is inevitable that some data will be missing. Rigorous analyses of the data quality of patient-completed questionnaires are becoming more common, but an area that can cause concern in surveys using health status measures, and especially in clinical trials, is that of missing data. Increasingly, trials use results from self-reported health status instruments as primary endpoints, and missing data from such measurements can potentially reduce analytic power and be a source of bias. The purpose of this chapter is to evaluate a simple missing data algorithm for the imputation of missing dimension scores on the 39-item Parkinson’s Disease Questionnaire (PDQ-39). The Parkinson’s Disease Questionnaire (PDQ-39) is the most widely used disease-specific measure of health status in Parkinson’s disease and has been recommended as the most comprehensively validated for competing PD-specific outcome measures. The instrument was developed in the United Kingdom but has been translated into more than 30 languages, including Spanish, American English, French, German, Polish, and Japanese, and has been used in both single-country and cross-cultural trials. Typically, the instrument has been shown to have relatively low levels of missing data, but inevitably with questionnaire-based outcome instruments, some data will be missing. Data Imputation One of the most widely used techniques for data imputation currently used within the social sciences is the “expectation maximization” (EM) computational algorithm for multiple imputation.