Skip to main content

Measuring the effectiveness of an automated text messaging active surveillance system for COVID-19 in the south of Ireland, March to April 2020

Our take —

Active surveillance using text-messaging may be an effective and efficient way to monitor exposed asymptomatic contacts to COVID-19. This study assessed the potential utility of such an automated text-messaging monitoring system in Ireland. While the overall response rate to daily texts was high (~80%) among those who opted for text monitoring, there was a substantial proportion of contacts who failed to respond, resulting in increased burden to local public health departments.

Study design

Prospective Cohort

Study population and setting

Authors investigated the effectiveness of a text-based active surveillance system in the Cork/Kerry region of Ireland by following 1,336 asymptomatic close contacts (defined as >15 minutes of close face-to-face exposure) of confirmed cases with automated daily text messages from March 8, 2020 to April 23, 2020. Asymptomatic contacts were asked to respond (with y/n) to a daily text inquiring about the presence of main COVID-19 symptoms for 14-days. A “yes” response, clinical inquiries, or a continued non-response after a follow-up text led to a clinical follow-up.

Summary of Main Findings

12,421 texts were sent in total over the 14-day period. Of the 1,336 contacts in the study, 14.4% (n=192) needed a clinical follow-up due to the presence of symptoms or clinical inquiries. The daily texts response rate was high, with 82.9% (n=10,300) of texts in the overall 14-day study period receiving an answer. However, a substantial portion, 17.1% (n=2,121) of texts did not provide a response, and needed to be followed-up by the clinical and administrative staff. A practical barrier to implementation included additional financial burden to those who used pay-as-you-go phones.

Study Strengths

This study conducted active surveillance of asymptomatic contacts of confirmed COVID-19 cases using an automated text-messaging system.

Limitations

There were no data reported on the proportion of asymptomatic contacts who opted out of automated text message monitoring in favor of monitoring through phone calls. The main symptoms assessed, fever, cough or shortness of breath, do not cover the increasing breadth of symptoms associated with COVID-19. Among those with clinical queries or “yes” responses (n=192), authors indicated that 72 did not meet testing criteria, but specifics as to which symptoms the study considered to be non-COVID-19 related are lacking. Authors acknowledged that testing eligibility criteria shifted during the study period, introducing additional limitations. The study did not investigate the reasoning behind non-responses, nor did authors specify how many non-responses were symptomatic upon clinical follow-up. Authors also did not report whether participants (if any) dropped out of the monitoring system. Aggregated data makes it impossible to differentiate if a small number of participants contributed to daily non-responses.

Value added

This study contributes to a limited existing literature on the effectiveness and utilization of active surveillance methods (e.g., text-messaging) to control COVID-19 spread.

This review was posted on: 18 July 2020