[Proposal] A Free Online Intervention to Reduce Depression and Anxiety: A Randomized Controlled Trial with Prolific Participants

Motivation

Depression and anxiety are leading causes of disability worldwide, affecting around 20% and 25% of people in the U.S. at some point in their life. Although effective treatments exist, most people do not access any treatment. Evidence suggests that treatments delivered via the internet can help people manage depression and anxiety.

Participants in online panels like Prolific are at increased risk for depression and anxiety. For example, the rate of depression among Amazon Mechanical Turk workers is estimated to be 1.6-3.6 times higher than in the general population (Ophir et al., 2020). Similarly, in our own work, we found that 42% of participants in a recent Qualtrics sample had been diagnosed with depression. Online interventions may be especially suitable for expanding access to support for communities of online panel participants.

Over the last two years, our team has been working on the development and evaluation of highly scalable online interventions to promote wellbeing globally. Our efforts have led to the development of the Common Elements Toolbox (COMET), a freely available online intervention that lasts 45-60 minutes (Wasil et al., 2021). The intervention includes a set of online modules from the most well-studied mental health and wellbeing interventions. Prior meta-analytic reviews of RCTs suggest that single-session interventions (SSIs) like COMET can effectively reduce depression and anxiety for youth and adolescents (Schleider & Weisz, 2017). Additionally, in two RCTs with college students in the US and India, COMET has been shown to reduce depressive symptoms and improve subjective well-being (Wasil et al., in prep). We are encouraged by these initial findings, and we now wish to examine if an adapted version of COMET can improve mental health and wellness for adults recruited via Prolific.

We propose a randomized controlled trial (RCT) of COMET, a free, brief, and scalable online mental health intervention vs. a waiting list control. This RCT will help us understand if COMET can be an effective and scalable intervention for adults recruited through Prolific. In addition to measuring the intervention’s effects on depression, we will measure ratings of its acceptability and appropriateness.

Additionally, we aim to understand for whom COMET is effective. To do this, we will apply methods from precision mental health to identify participant-level factors that predict response to the intervention (see DeRubeis et al., 2014; Lorenzo-Luaces et al., 2017).

In summary, our study aims at answering three questions:

  • I) Does a brief online intervention reduce symptoms of depression and anxiety among Prolific participants?
  • II) Is the intervention considered acceptable and appropriate?
  • III) For whom is the intervention most effective?

Methodology

Experimental Design

We plan to conduct an RCT via Prolific. Adult participants will be randomized to the intervention condition (see below) or a waitlist control. The primary outcome variables for the study will be participants’ symptoms of depression and anxiety (measured by the K6). Secondary outcomes include subjective well-being (measured by the WHO-5 Wellbeing Index), flourishing (measured by a modified version of the PERMA-Profiler), and positive and negative affect (measured by the Positive and Negative Affect Schedule; PANAS). Participants in the intervention condition will also be asked to provide ratings of the intervention’s acceptability and appropriateness, as well as to offer feedback that helps us further improve the intervention.**

Intervention: The Common Elements Toolbox

COMET lasts about 45-60 minutes. Participants complete the intervention on a computer, anywhere around the world. COMETincorporates four interventions: behavioral activation, (Dimidjian et al., 2006; Patel et al., 2017), cognitive restructuring (e.g., Beck et al., 1979). gratitude (Emmons & Stern; 2013; Seligman et al., 2005), and self-compassion (Gilbert, 2009; Neff et al., 2007).

The intervention is informed by previous research on single-session interventions (Schleider & Weisz, 2018; Schleider et al., 2020) and a review of commercially available smartphone-based interventions (Wasil et al., 2019). We have published details about the development of the intervention, as well as its content (Wasil et al., 2021).

Waitlist control

Participants in the waitlist control condition will complete outcome measures. They will receive access to COMET six months after they enroll in the study.

Precision mental health models

We will examine if demographic variables or clinical variables, which we selected from a review of replicated predictors by Kessler et al. (2017), predict who is most likely to benefit from the intervention. To do this, we will apply state-of-the-art methods in precision mental health (see DeRubeis et al., 2014; Lorenzo-Luaces et al., 2017).

Sample size and costs

Given the brevity and scalability of our intervention, we are powering to detect ‚Äúsmall‚ÄĚ effect sizes. We performed a power analysis using G*Power. Assuming a standardized mean difference of d = 0.2, an alpha level of 0.05, and a power of 0.9, we will require 1,054 participants. We plan to compensate participants at a rate of ¬£7.50 per hour. Assuming a total duration of 90 minutes (including the initial survey and the follow-up surveys), this corresponds to a total participant cost of ¬£7,905. Including the 33% service fee, the total cost of the study will be ¬£10,513.65. If we are able to obtain support from Prolific, we will be able to cover the remaining ¬£513.65 through institutional support.

Commitment to Open Science

Before data collection begins, we will register our study via AsPredicted (https://aspredicted.org/) and clinicaltrials.gov. We will also make a de-identified dataset and our R code available via the Open Science Framework (OSF; https://osf.io/). We also aim to publish the findings of the study in an open access journal. Additionally, if the intervention is effective, we will also make the intervention (and all intervention materials) freely available via OSF.

Broader Impact

Through this research, we will test an evidence-based intervention that can be delivered freely and briefly. The intervention could be one important tool in reducing the rate of common mental health problems and improving wellbeing among Prolific participants. More broadly, the data and feedback from this project will help us refine this intervention to promote mental health and wellbeing worldwide.