BMC Medical Research Methodology (May 2020)

Frequently repeated measurements -our experience of collecting data with SMS

  • I. Axén,
  • I. Jensen,
  • E. Butler Forslund,
  • B. Grahn,
  • V. Jørgensen,
  • C. H. Opava,
  • L. Bodin

DOI
https://doi.org/10.1186/s12874-020-01013-y
Journal volume & issue
Vol. 20, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background As technology is advancing, so are the possibilities for new data collection methods in research, potentially improving data quality and validity of the results. In Sweden, a system using frequent repeated data collection using text messages, SMS Track, has been used in clinical research for more than a decade. In this paper, compliance with repeated text message questions was examined across five different studies, i.e. if compliance was 1: associated with study-specific factors (age or gender of the subjects, the condition, its’ severity or course, i.e. improvement, relapse or steady state) and/or. 2: associated with the methodology itself (the question being asked, the frequency and number of questions, duration of data collection, initial compliance or the management of the system). Methods Descriptive comparisons were done across five studies. Three studies were collecting weekly responses over at least 52 weeks (“Weekly studies”) and were used to investigate the effect of age, sex and pain severity on compliance, the effect of early compliance for late compliance, and finally the early occurrence of two successive weeks with non-compliance. Result Compliance was excellent across all five studies, and only influenced somewhat by age, sex and pain-level. The factor “study” remained significant in the final model thus the observed differences may be a result of the conditions studied but does not seem to be attributable to severity or development of these conditions. Number and frequency of questions did not influence compliance, nor did study duration. Conclusions Compliance was excellent in the included studies and was not affected by population factors. However, differences in compliance were observed that cannot be easily explained and warrant further investigation. In particular, the nature of the variables or the management of the study are potential areas for further investigations.

Keywords