Comparison of contact diaries and wearable sensors in measuring social contact patterns
This project aimed at investigating unresolved methodological issues of collection of data on social interactions, which is done in network epidemiology and needed in modelling infectious disease dynamics to understand epidemics.
Specifically, we assessed the measurement error of the two most relevant methods of contact recording - paper diaries and wearable proximity sensors - by comparing them. Both were applied concurrently to the same population during a scientific conference in 2014. In addition, we investigated the acceptability of both approaches. Seventy-six conference attendees wore radio-frequency identification (RFID) sensors during one day while at the same time recording their interactions with other study participants in a contact paper-diary.
Based on the diaries, this resulted in 199 contacts, of which 130 were noted by both individuals, whereas the sensors recorded 316 contacts. Comparing both measurements, the following probabilities that a sensor-recorded contact was reported in a diary were estimated: 72% for <5 min contact duration, 86% for 5-15 min, 89% for 15-60 min, and 94% for >60 min. Thus, sensor-measured and self-reported contacts intersected largely. However, both set were not completely part of the other.
Concerning acceptability of the diary, 20% of respondents regarded it as too much work and 25% had difficulties in remembering contacts; concerning the sensor measurement, 93% were comfortable with it.
We concluded that both reporting and recording are not done completely. This is especially true for reporting short interactions <5 min. Both approaches detect partly different contact types while being acceptable to study participants. Therefore, both methods can be used in future network studies if their limitations are kept in mind and their incompleteness is approximately corrected.
Publikation zum Projekt
Smieszek T*/Castell S*, Barrat A, Cattuto C, White PJ, Krause G. Contact diaries versus wearable proximity sensors in measuring contact patterns at a conference: method comparison and participants' attitudes. BMC Infect Dis. 2016;16(1):341. [*equal contribution]
- NIHR Health Protection Research Unit in Modelling Methodology and MRC Outbreak Centre for Outbreak Analysis and Modelling, Department of Infectious Disease Epidemiology, School of Public Health, Imperial College London, London, UK.
- Modelling and Economics Unit, Centre for Infectious Disease Surveillance and Control, Public Health England, London, UK.
- Aix Marseille Université, Université de Toulon, CNRS, CPT, UMR 7332, Marseille 13288, France.
- Data Science Laboratory, ISI Foundation, Torino, Italy.
- Hannover Medical School, Hannover, Germany.
- Epidemiologie - Prof. Dr. Gérard Krause
Geldgeber / Förderer
HZI - Helmholtz-Zentrum für Infektionsforschung