Comparing a ses-sensitive and an all-ses implementation strategy to improve participation rates of patients with a lower socioeconomic background in a web-based intervention for depressive complaints: A cluster randomised trial in primary care

Background
Depression is a major public health concern, which is most pronounced in population segments with a lower social-economic status (SES). E-health interventions for depressive complaints are proven to be effective, but their reach needs to be improved, especially among people with a lower socioeconomic status (SES). Implementing e-health interventions in the primary care setting with SES-sensitive guidance from General Practice nurses (GP nurses) may be a useful strategy to increase the reach of e-health in lower SES groups. We implemented an evidence-based online intervention that targets depressive complaints in primary care.

Methods
A pragmatic cluster-randomised trial was conducted in two parallel groups where a SES sensitive (SES-sens) implementation strategy with additional face-to-face guidance by GP nurses was compared to an all-SES implementation strategy. The primary outcome was the percentage of lower SES participants in either condition. Participation was defined as completing at least 1 face-to-face session and 2 online exercises. Participation rates were evaluated using logistic mixed modelling.

Results
In both conditions, the participation rates of lower SES participants were quite high, but were notably lower in the SES-sens implementation condition (44%) than in the all-SES implementation condition (58%). This unexpected outcome remained statistically significant even after adjusting for potential confounders between the conditions (Odds Ratio 0.43, 95%-CI 0.22 to 0.81). Less guidance was provided by the GP nurses in the SES-sens group, contrary to the implementation instructions.

Conclusions
From a public health point of view, it is good news that a substantial number of primary care patients with a lower SES level used the implemented e-health intervention. It is also positive that an all-SES implementation strategy performed well, and even outperformed a SES-sensitive strategy. However, this was an unexpected finding, warranting further research into tailoring implementation strategies of e-health interventions towards specific target groups in the primary care setting.