Differences between online trial participants who have used statutory mental health services and who have not: analysis of baseline data from two pragmatic trials of a digital health intervention

Background:Digital health interventions have become an established part of mental health service provision internationally. Regulators have positioned the best-practice standard of evidence as an interventional study with a comparator reflective of standard care, often operationalised as a pragmatic trial. Digital health interventions can extend healthcare provision to those not currently using mental health services, and hence for external validity trials might openly recruit a mixture of people who have and have not used current mental health services. Prior research has demonstrated phenomenological differences in mental health experiences between these groups. Some differences between service users and non-service users might influence the change created by digital health interventions, and hence research should systematically examine these differences so as to inform intervention development and evaluation work. This paper presents an analysis of baseline data collected in the NEON Trial (ISRCTN11152837) and NEON-O Trial (ISRCTN63197153). These were pragmatic trials of a digital health intervention which openly recruited people who had and had not used specialist mental health services. All participants were experiencing mental health distress. NEON Trial participants had experienced psychosis in the previous five years. Objective:To identify differences in baseline sociodemographic and clinical characteristics associated with specialist mental health service use, for NEON Trial and NEON-O Trial participants. Methods:For both trials, hypothesis testing (with P-value adjustment for multiple comparison) was used to compare baseline sociodemographic and clinical characteristics of participants in the intention to treat sample who had and had not used specialist mental health services. The significance threshold was P=.05. Results:Compared to non-service users (n=124), NEON Trial specialist service users (n=609) were more likely to be female (adjusted P<.001), older (adjusted P<.001) and White British (adjusted P<.001), with lower quality of life (adjusted P=.001) and lower health status (adjusted P=.03). There were differences in geographical distribution (adjusted P=.03), employment (adjusted P=.009, more unemployment), current mental health problem (adjusted P=.009, more psychosis and personal disorders) and recovery status (adjusted P=.009, more recovered). Compared to non-service users (n=399), NEON-O Trial specialist service users (n=614) had differences in employment (adjusted P=.02, more unemployment) and current mental health problem (adjusted P=.009, more personality disorders), with lower quality of life (adjusted P=.004), more distress (adjusted P<.001), less hope (adjusted P<.001), less empowerment (adjusted P<.001), less meaning in life (adjusted P<.001) and lower health status (adjusted P<.001). Conclusions:Specialist mental health service use was associated with numerous demographic and clinical differences in baseline characteristics. Change models for complex intervention should attend to service use history. Investigators for trials recruiting openly should consider randomised stratification on service use history. Evaluations of interventions should attend to in-study changes in service use. Clinical Trial: International Standard Randomized Controlled Trial Number (ISRCTN) 11152837; http://www.controlled-trials.com/ISRCTN11152837 and 63197153; http://www.controlled-trials.com/ISRCTN63197153