Digital Applications to Monitor Novel Coronavirus Disease and Response in Colombia - Syndromic and Vaccination Surveillance

The investigators deploy mobile phone surveys using Interactive Voice Response (IVR) in a random sample of mobile phone numbers (random digit dialing or RDD) in Colombia.

Adult individuals who have not been fully vaccinated against COVID-19 (defined as those who 1) did not get any vaccine; 2) did not get the complete vaccination scheme 3) did not get one booster when less than 50 years old; 4) did not get two boosters when older than 50 years old), and who consent to be recontacted are invited to participate in the messaging trial. After consent, participants are randomized in three different treatment arms or one control arm. Participants in each treatment arm are presented with different messages to improve COVID-19 vaccination. Participants in the arm with factual messages listen to actual data and scientifical information on COVID-19 vaccination. Those in the narrative arm listen to messages appealing to emotions drawing on the COVID-19 vaccination experiences that were explored in a qualitative study conducted earlier. The mixed arm combines both factual and narrative messages. Finally, the control arm does not deliver any message and will only ask for whether the participant got vaccinated in the last two weeks. Individuals in each of the three treatment arms receive seven different types of messages. Each message is delivered twice every week for seven weeks and their vaccination status is inquired in the same calls to assess whether the participants received their vaccine after the last call. IN all arms, respondents are asked a simple question to test the attention they pay to the information provided in the message. The three treatment arms are compared against the control to assess the probability that any of the arms is significantly better than the control. With this aim, the minimum target sample size for each messaging group is N=173 which allows for the detection of a difference between groups of at least 15% with a double-tailed distribution at alpha=0.05 and power =0.8.

The investigators launched a first cohort on September 21, 2022 over sampling both the treatment and control arms to reduce the risk of attrition and aiming at obtaining 250 recruited participants that agree to participate. Due to high attrition, a second cohort was deployed on October 25, 2022 following the same procedure as for the first cohort, this time with a target of 400 recruited respondents with the aim of achieving the required sample size.

The investigators assess the differential likelihood of vaccinating during the trial for individuals participating in the different arms, including the control.