Abstract
Background: Internet-delivered health behaviour change interventions have potential to reach a large audience cost-effectively at a convenient time and location. However, the public health impact of these interventions in practice settings may be limited if sufficient people are not recruited. Moreover, males, younger adults, and those with lower income or education are often under-represented participants.
Objective: This translational research evaluates the impact and cost-effectiveness of four recruitment strategies on participant registration and initial use (Reach) of the 5+YourWay® Coach service. This Internet-delivered health service offers adults free computer-tailored communications to improve fruit and vegetable consumption.
Methods: An observational comparative approach was used to evaluate recruitment impact on registration rates and Reach. Firstly, six Otago general practices agreed to offer the 5+YourWay® Coach service to eligible patients (n=9228) for one year during consultations. Secondly, the service was featured in the email newsletter of an existing Internet-based physical activity campaign for Dunedin adults (n=816 subscribers). Thirdly, Facebook advertisements were shown to 5155 Dunedin adults with low or unspecified educational attainment, and finally, flyers were included in 265 Dunedin food bank parcels. At registration and the first session, the number, proportion, and characteristics of participants were compared between strategies.
Results: In participating general practices, 27% of clinical staff adopted the service, but only 2.7% of eligible patients were offered it (n=249). Of those, 52% registered, and 22% began the first session (Reach). In comparison, only 10% of email newsletter subscribers registered, and Reach was 7.5%; but attrition was lower resulting in more participants completing the first session than general practices. Registration rates and Reach were low for strategies targeting low-socio-economic individuals (Facebook advertisements: 0.06% and 0.02%, respectively; food bank flyer: 0.4% for both). The email newsletter to campaign subscribers was the most cost-effective strategy (zero invitation cost for 24 targeted participants). Food bank flyers cost $11.43 per targeted participant, general practice invitations $196.93 and Facebook advertisements $256.80. In general, most participants accessing the service were female (79%) and of European descent (90%); 44% were university qualified. On average participants were middle class, and few had characteristics consistent with being socio-economically disadvantaged. General practices recruited a significantly higher proportion of males (p < 0.001) and older people (p < 0.001) than the email newsletter.
Conclusion: Results indicate that a promoted email newsletter can be both successful and cost-effective when sent from a trusted source to an interested audience. Recruitment through general practices has potential for high Reach, especially in recruiting males over the age of 45 years, but to be successful, more patients need to be offered the service. All strategies, even targeted ones, were ineffective in recruiting low socio-economic individuals, but the reasons for this are unknown and require further investigation. If the free service fails to attract a greater proportion of participants from lower socio-economic groups, then it may contribute to increasing rather than decreasing socio-economic disparities in fruit and vegetable consumption. Nevertheless, this free online nutrition service is likely to fill an important gap in personalised, nutrition wellness services, especially in rural communities with limited access to health services.