Since data integrity and quality is a basic concern in clinical trials, we examined the validity of an electronic versus paper method of collecting symptom information from subjects. The aim of this study was to determine the correlation between an electronic, interactive voice response system (IVRS) and paper self-assessment measure.

Subjects enrolled in an asthma clinical trial investigating a novel treatment were asked to complete a twice daily electronic and validated paper assessment tool. Symptom scores, frequency of beta-2 agonist use, and peak expiratory flow were recorded. For each subject, paired observations were graphed on a scatter plot with a 45 degree reference line. Correlation coefficients of each variable were calculated. Separately, mean difference scores between corresponding observations were evaluated by paired t-tests.

As a group, the electronic and paper diary scores across subjects were highly correlated (p<0.001) and exhibited minimal overall bias or “shift” in the reported values between the two methods. Subject's individual scatter plots provided straightforward examination of their accuracy of data reporting and allowed for discrepancies to be easily detected.

The hypothesis that the mode of data collection (paper versus electronic) has a biasing effect was not supported. Electronic diary data collection (IVRS) appeared appropriate compared to a validated paper and pencil version. Distinct advantages with electronic diaries such as time and date stamping, increased legibility and completeness, and automated database construction are discussed.