Shorter communication
Behavior change through automated e-mails: Mediation analysis of self-help strategy use for depressive symptoms

https://doi.org/10.1016/j.brat.2012.11.002Get rights and content

Abstract

Objective

To evaluate whether automated e-mails promoting effective self-help strategies for depressive symptoms were effective in changing self-help behavior, and whether this improved depression outcomes.

Method

568 adults with sub-threshold depression participated in a randomized controlled trial and provided complete data. A series of 12 e-mails promoting the use of evidence-based self-help strategies was compared with e-mails providing non-directive depression information. Depression symptoms were assessed with the Patient Health Questionnaire depression scale (PHQ-9) and use of self-help strategies was assessed at baseline and post-intervention. We hypothesized that those receiving the self-help e-mails would increase their use of evidence-based self-help and this would be associated with improvements in depression. Mediation analyses were conducted using a non-parametric bootstrapping procedure.

Results

Total use of the self-help strategies promoted in the e-mails significantly mediated the effect of the intervention on depressive symptoms (B = −0.75, SE = 0.16, 95% CI: −1.06 to −0.48). The direct effect of the intervention on depressive symptoms was much smaller and not significant when the mediation path was included. The majority of the individual strategies also had a significant indirect effect on depressive symptoms.

Conclusions

In adults with sub-threshold depression, automated e-mails based on behavior change principles can successfully increase use of self-help strategies, leading to a reduction in depressive symptoms.

Highlights

► Adults with depressive symptoms received emails with coping advice or information. ► The coping advice emails increased use of effective self-help techniques. ► Increased use of self-help techniques led to a reduction in depression symptoms. ► Messages about effective self-help can change behavior and improve mild depression.

Section snippets

Participants

Participants were 568 adults with sub-threshold depression who completed the trial and for whom baseline and post-intervention data were available. Sub-threshold depression was defined as 2 to 4 symptoms of depression experienced more than half the days or nearly every day for two or more weeks, which had affected work, home, or social functioning. Participants were recruited online to the study website (www.moodmemos.com) from a variety of internet-based sources including Google advertising,

Results

Table 1 presents the means and standard deviations of depressive symptoms, frequency of use of all self-help strategies combined and frequency of use of each strategy individually at baseline and post-intervention. Mean scores on the PHQ-9 at baseline were in the mild to moderate severity range (Kroenke et al., 2010). All self-help strategies had potential for improvement in mean usage.

The total strategy use score was examined first as a potential mediator, to test the overall effect of the use

Discussion

The central precept of the Mood Memos study was that the active e-mails would lead to an increase in usage of the promoted strategies, relative to the control group, and that this increase would improve depressive symptoms. This hypothesis was supported for use of the strategies overall. The total use of the promoted self-help strategies completely mediated the effect of the intervention on depression symptoms. This finding provides strong support for the premise that the intervention worked as

Acknowledgments

This research was supported by a National Health and Medical Research Council Postgraduate Scholarship (567056), a Sidney Myer Health Scholarship, and an NHMRC Australia Fellowship. Orygen Youth Health Research Centre receives funding from the Colonial Foundation. The funding sources had no role in the study design, collection, analysis or interpretation of data, in the writing of the report, and in the decision to submit the paper for publication. Thanks to the working group of Sarah

References (28)

  • J.P. Fry et al.

    Periodic prompts and reminders in health promotion and health behavior interventions: systematic review

    Journal of Medical Internet Research

    (2009)
  • J.L. Horowitz et al.

    The prevention of depressive symptoms in children and adolescents: a meta-analytic review

    Journal of Consulting and Clinical Psychology

    (2006)
  • N.E. Hundt et al.

    The relationship between use of CBT skills and depression treatment outcome: a theoretical and methodological review of the literature

    Behavior Therapy

    (2012)
  • J. Jones et al.

    Qualitative research: consensus methods for medical and health services research

    British Medical Journal (Clinical Research Edition)

    (1995)
  • Cited by (18)

    • Mediators and mechanisms of change in internet- and mobile-based interventions for depression: A systematic review

      2021, Clinical Psychology Review
      Citation Excerpt :

      Finally, a total of 26 RCTs met the eligibility criteria and were included in this review. Half of these studies examined samples with clinically depressed patients (n = 2769) (Choi et al., 2014; Clarke et al., 2014; Forand et al., 2018; Hirsch et al., 2018; Lamers et al., 2015; Newby, Williams, & Andrews, 2014; Pictet, Jermann, & Ceschi, 2016; Pots, Trompetter, Schreurs, & Bohlmeijer, 2016; Seeley et al., 2019; Terides et al., 2017; van der Zanden, Galindo-Garre, Curie, Kramer, & Cuijpers, 2014; Warmerdam, van Straten, Jongsma, Twisk, & Cuijpers, 2010; Williams, Blackwell, Mackenzie, Holmes, & Andrews, 2013); whereas the other 13 studies focused on samples in which depression symptoms were assessed without applying a clinical cut-off score (i.e., “non-clinical”), comprising mixed samples with other physical and mental impairments, or without pathology (n = 4051) (Auyeung & Mo, 2018; Bakker, Kazantzis, Rickwood, & Rickard, 2018; Heckendorf, Lehr, Ebert, & Freund, 2019; Kauer et al., 2012; Levin, Haeger, Pierce, & Twohig, 2017; Ma, She, Siu, Zeng, & Liu, 2018; Morgan, Mackinnon, & Jorm, 2013; Morris, Schueller, & Picard, 2015; Querstret, Cropley, & Fife-Schaw, 2018; van Luenen, Kraaij, Spinhoven, Wilderjans, & Garnefski, 2019; Viskovich & Pakenham, 2018; Viskovich & Pakenham, 2019; Willems et al., 2017). Table 1 provides a detailed overview on study characteristics and extracted data.

    • Regression-based statistical mediation and moderation analysis in clinical research: Observations, recommendations, and implementation

      2017, Behaviour Research and Therapy
      Citation Excerpt :

      Only one test of mediation is required, and we recommend the use of a bootstrap confidence interval or another method that makes no assumption about the shape of the sampling distribution of ab. Researchers frequently attempt to label a mediation process as partial or complete, language frequently used in mediation analyses reported in BRaT (e.g., Deacon, Kemp, & Dixon et al., 2013; Manuel & Wade, 2013; Morgan, MacKinnon, & Jorm, 2013; Staring et al., 2016; Vincent & Walsh, 2013). Partial mediation refers to a pattern of findings where mediation is established in the presence of significant total effect of X and the direct effect of X (c') is statistically different from zero.

    • Posttraumatic Trigeminal Nerve Neuropathy

      2015, Nerves and Nerve Injuries: Pain, Treatment, Injury, Disease and Future Directions: Vol 2
    View all citing articles on Scopus
    View full text