Evaluating digital public health campaigns

Abstract

Health campaigns have been an important component of initiatives to improve population health behaviours. There has been a substantial increase in the use of digital channels by campaigns in recent years, but this has not been mirrored with the same level of research and evaluation. Often the evaluations of digital health campaigns continue to use the same evaluation approach that has been used for traditional broadcast media, with little consideration given to how online engagement metrics are relevant for health campaign evaluations. The purpose of my thesis is to explore how health campaigns that use digital channels should be evaluated. The first part of my thesis explored how we currently evaluate digital health campaigns by conducting a systematic literature review. Utilising the evaluation approach identified in the literature, I evaluated three campaigns that used digital channels (Movements Matter, Still Six Lives, and Shisha No Thanks). These evaluations suggested that online engagement is not necessarily a linear step between reach and behaviours. This led to my second research question examining how digitalspecific measures, particularly engagement, should be understood in relation to the overall campaign evaluation. I conducted a further two studies (a content analysis of Shisha No Thanks Facebook comments, and focus groups about online engagement for a Healthy Lunch Box campaign) and found that digital engagement actions did not always represent a precursor to making the behaviour change advised by the campaign. Bringing together the findings of these studies, I outline six evaluation areas to be considered or adapted for digital health campaigns, including a more nuanced approach to understanding engagement metrics. The insights of this thesis will enable researchers and practitioners to conduct more appropriate campaign evaluations, and contribute to the evidence on how we can effectively use digital health campaigns to improve population health.