Comparison of response rates on invitation mode of a web-based survey on influenza vaccine adverse events among healthcare workers: a pilot study.

BACKGROUND

Web-based surveys have become increasingly popular but response rates are low and may be prone to selection bias. How people are invited to participate may impact response rates and needs further study as previous evidence is contradictory. The purpose of this study was to determine whether response to a web-based survey of healthcare workers would be higher with a posted or an emailed invitation. We also report results of the pilot study, which aims to estimate the percentage of adults vaccinated against influenza who report recurrent systemic adverse events (the same systemic adverse event occurring successively following receipt of influenza vaccines).

METHODS

The pilot study was conducted in November 2016 in Toronto, Canada. Members of a registry of adults (18 years and older and predominantly healthcare workers) who volunteered to receive information regarding future studies about influenza were randomly assigned to receive either an email or postal invitation to complete a web-based survey regarding influenza vaccinations. Non-respondents received one reminder using the same mode of contact as their original invitation.

RESULTS

The overall response rate was higher for those sent the invitation by email (34.8%) than by post (25.8%; p < 0.001) and for older versus younger participants (p < 0.001). Of those who responded, 387/401 had been vaccinated against influenza at least once since adulthood. Of those responding to the question, 70/386 (18.1%) reported a systemic adverse event after their most recent influenza vaccine including 22 (5.7%) who reported a recurring systemic event. Systemic adverse events were reported more often by males 18-49 years old than by other groups (p = 0.01). Recurrent systemic adverse events were similar by age and sex with muscle ache being the most commonly reported recurrent reaction. More respondents who reported only a local adverse event (93.1%) planned to be vaccinated again next year than those with a systemic adverse event (69.7%; p = 0.04).

CONCLUSIONS

In this convenience sample of registry volunteers, response rates were generally low, but were higher for the emailed than posted invitations and for older than younger adults.