A methodological study of some factors influencing the reporting of symptoms.

The spontaneous occurrence of symptoms that are often regarded as adverse drug reactions was studied by means of open questionnaires or by a 38-item checklist of symptoms. Of the 416 respondents who returned an analyzable questionnaire, only 233 (56 per cent) were healthy (defined as having had no illness and/or having taken no medication in the previous three days). Using the open questionnaires, one or more symptoms were reported by 15 per cent of healthy persons (N equals 155) and by 69 per cent of a group of 118 persons who had been ill and/or who had taken medication in the previous three days. In contrast, using the checklist questionnaires 82 per cent of healthy persons (N equals 78) and 97 percent of persons who had been ill and/or who had taken medication in the previous three days (N equals 65) reported one or more symptoms. The total number of symptoms reported by any one person was three with the open questionnaire and 22 with the checklist. In addition, a greater severity of symptoms was associated with the open questionnaire. Sex, educational level, or presence or absence of previous exposure to clinical trials did not influence the nature and incidence of symptoms reported. However, the severity of symptoms reported was significantly different for the latter two factors. Differences in the frequency with which certain symptoms were reported and the relevance of these and other observations to the design of clinical trial record forms are discussed.