The Application of a Paired Comparisons Technique to Identify Sequence of Recovery After Severe Head Injury

Abstract

The “paired preference technique” PPT as a method for establishing the most likely sequences of recovery in a group of people surviving severe head injury is described. In the PPT every item of observed behaviour is compared with every other item of behaviour in a series of paired comparisons. This is similar to the technique described by Allen and Yen (1979) which they called the “paired comparison method”. We believe this method is suitable for data which are not amenableto moreconventional methods of analysis because of differentnumbers of observations across subjects or because of missing data. We describe the PPT, provide examples from an observational study of 88 head-injured patients, and discuss the strengths and weaknesses of this method of analysis.