Open-source electronic data capture system offered increased accuracy and cost-effectiveness compared with paper methods in Africa

• Published in: Journal of clinical epidemiology Vol. 67; no. 12; pp. 1358 - 1363
• Main Authors: Dillon, David G, Pirie, Fraser, Rice, Stephen, Pomilla, Cristina, Sandhu, Manjinder S, Motala, Ayesha A, Young, Elizabeth H
• Format: Journal Article
• Language: English
• Published: New York, NY Elsevier Inc 01.12.2014; Elsevier; Elsevier Limited
• Subjects: Accuracy; Africa; Analysis. Health state; Biological and medical sciences; Biomedical research; Chronic illnesses; Cost-Benefit Analysis; Data capture; Data collection; Data Collection - economics; Data Collection - methods; Design; Diabetes; Documentation; Electronic questionnaire; Electronics - economics; Epidemiology; Family medical history; General aspects; Humans; Internal Medicine; Internet; Interviews; Licensing; Medical sciences; Methods; Open-source; Original; Programming languages; Public health. Hygiene; Public health. Hygiene-occupational medicine; Questionnaires; Reproducibility of Results; Software; Sub-Saharan Africa; Survey; Surveys and Questionnaires - economics; Validation studies; Sub-Saharan Africa; Africa; Open-source; Data capture; Electronic questionnaire; Survey; Epidemiology; Sub-Saharan Africa; Human; Questionnaire; Methodology; Method; Capture; Source; Medicine; Accuracy; Surveillance; Offer; Health economy; Computer; Paper; Comparative study; Public health; Cost efficiency analysis
• DOI: 10.1016/j.jclinepi.2014.06.012
• ISSN: 0895-4356

Abstract Objectives Existing electronic data capture options are often financially unfeasible in resource-poor settings or difficult to support technically in the field. To help facilitate large-scale multicenter studies in sub-Saharan Africa, the African Partnership for Chronic Disease Research (APCDR) has developed an open-source electronic questionnaire (EQ). Study Design and Setting To assess its relative validity, we compared the EQ against traditional pen-and-paper methods using 200 randomized interviews conducted in an ongoing type 2 diabetes case–control study in South Africa. Results During its 3-month validation, the EQ had a lower frequency of errors (EQ, 0.17 errors per 100 questions; paper, 0.73 errors per 100 questions; P -value ≤0.001), and a lower monetary cost per correctly entered question, compared with the pen-and-paper method. We found no marked difference in the average duration of the interview between methods (EQ, 5.4 minutes; paper, 5.6 minutes). Conclusion This validation study suggests that the EQ may offer increased accuracy, similar interview duration, and increased cost-effectiveness compared with paper-based data collection methods. The APCDR EQ software is freely available ( https://github.com/apcdr/questionnaire ).