Sustainable Upscaling of Depression Prevention

Sustainable Upscaling of Depression Prevention, Finding the Optimal Balance Between Investment and Benefit (SPRINT)

Research shows that online unguided self-help interventions focused on psycho-education, skills training and lifestyle can prevent mild mood complaints from turning into a full-blown depression. These encouraging results are found even though the adherence to these types of interventions is generally low. With this project, the investigators examine whether adherence to online unguided self-help interventions can be increased by additional motivational guidance elements and as such further improve the effectiveness. This is examined by adding three additional components to the intervention: 1) A coach who provides online feedback once a week to provide support. 2) Mobile application to monitor mood and related factors and to receive automated personalized messages, 3) Content based on the principles of motivational interviewing. A secondary aim is to compare the additional effects of the individual components against the additional costs.

Given the substantial prevalence rate of Major Depression and its extreme burden among the general population, depression prevention is a high priority on the Dutch public health agenda. The aim of the Depression Prevention Program of the Dutch Ministry of Health, Welfare and Sport (Meerjarenprogramma (MJP, VWS 2017) entails a decrease in major depression prevalence of 30% by the year 2030. One solution to the problem is to offer online self-help interventions focusing on psycho-education, skills-training and lifestyle with the aim to improve mood. These interventions have proven to be effective and can prevent mood problems to sustain and/or worsen (van Zoonen et al., 2014). Self-help interventions are easily accessible and acceptable, and they can reach a population at low costs and on a large scale (Riper et al. 2010). Still, while online self-help interventions can be effective (Karyotaki et al., 2017), engagement barriers exist, adherence rates are generally low, and integration into daily life routines is difficult to achieve (Karyotaki et al., 2015), which may jeopardize the potential population health impact of these interventions. From this perspective there is a clear optimization need of evidence-based online self-help interventions to increase their impact on the general population. One way to increase adherence and engagement, and subsequently the effectiveness of such interventions, is to administer the intervention with the help of (motivational) guidance elements. Guided interventions are known to increase adherence, engagement and effectiveness of interventions and can be operationalized in various ways (Mohr, Cuijpers & Lehman, 2011; Kelders, 2017). Examples for types of guidance are human coaches, computerized coaches, chat support functions, personalized messages, and many more. While those motivational guidance elements can help the self-help interventions effectiveness, they come with higher costs as they need, for example, an infrastructure of therapists or coaches. It is therefore of high value to find the optimal balance between the effectiveness of the intervention and the necessary support components to establish a product with the potential to be implemented at scale. The first objective of this study is to examine whether the effectiveness of an online self-help intervention ("Moodbuster Life") for adults who want to improve their mood can be optimized by three different motivational guidance components. The motivational components are: 1) A coach who provides online feedback once a week to provide support. 2) Mobile application to monitor mood and related factors and to receive automated personalized messages, 3) Content based on the principles of motivational interviewing. A secondary aim is to compare the additional effects of one component against additional costs defined as extra time investment (in the platform and beyond) and financial costs (service costs, costs incurred by participants).

All participants receive access to the Moodbuster Life intervention with no or a maximum of 3 additional components. The investigators will use a full-factorial design to study the additional effect of each component either alone or in interaction. Participants are randomized to one of all possible combinations of the 3 components. In our study, with 3 additional components, this means that there are 8 conditions (2x2x2).

Study participants cannot be blinded to the allocation scheme as the participants will know what components are added to their intervention.

All participants get access to the Moodbuster Life intervention. Moodbuster Life is an online self-help intervention that contains 5 web-based modules focusing on lifestyle and coping: psycho-education, behavioral activation, physical activity, problem-solving, and worrying. All participants start with module 1, psycho-education. Next, participants can choose what module they wish to continue with. All modules take about 45 minutes to complete and contain text, exercises, video clips and preparing the home-work assignments. Executing the home-work assignments may take 20 minutes each week.

The participants randomized to receive this component will receive access to a mobile application. The aim of this app is two-folded, (1) used for diary ratings, (2) sending out personalized automated messages. First, the participants will rate their mood, sleep and related factors on a daily basis. The participants are prompted to rate the diary ratings three times a day (morning, afternoon, evening). Moreover, the application graphically pictures progression over time. Second, the application will send personalized automated messages. The content of the messages is informative, affirmative or encouraging. The investigators will use reinforcement learning (RL) to find so-called policies that show best long-term engagement and most sustained improvement of participants' mood. To drive choices, the investigators will use the data mentioned in the advising for the modules as well as behavioral data (mood ratings), data across all users is exploited.

A coach will provide support once per week at a scheduled time to participants who are allocated to receive support. The coaches are psychologists who are not part of the research team. The support will be provided via the Moodbuster Life messaging system and is focused on helping the participant work through the modules, showing empathy and motivating the participants to continue with the modules. The coaching is not aimed at developing a patient-therapist relationship.

Participants who are randomized to this component, receive access to extra content that is based on the principles of motivational interviewing. This includes an extended first module that contains psychoeducation on the importance of motivations and on how persons can motivate themselves to engage with the interventions. Participants are asked about their life goals (long term) and intervention goals (short time) and are guided in how they should formulate these goals to increase the chance of success. Moreover, in each of the 4 modules a short exercise aimed at increasing motivation is included.

Mood is assessed with the Center for Epidemiological Studies Depression Scale (CES-D). The total score ranges from 0 to 60, with a lower score indicating better mood. The CES-D is assessed at baseline and then again after 6 weeks.

Adherence to the intervention is measured with "meta-data". That is, number of logins, duration on the platform, visiting pages, completion of homework assignments (yes/no). Participants are advised to use the intervention for 5 weeks.

Anxiety symptoms are measured with the 7-item anxiety subscale of the Hospital Anxiety and Depression Scale (HADS; with a total score ranging from 0 to 21, where higher scores indicate higher anxiety levels).

Problem solving skills are measured with 6-items (total score ranging from 6 to 36, with higher scores representing better problem solving skills). These 6 items are the six highest loading items of the Approach Avoidance Style subscale of the Problem-Solving Inventory (PSI), which in turn represent the problem solving subscale of the Cognitive Behavioral Therapy Skills scale (CBT-Skills).

Levels of behavioral activation are measured with the 9-item Behavioral Activation for Depression Scale - Short Form (BADS-SF; with a total range ranging from 0 to 54, with high scores representing higher activation)

To assess worrying, the abbreviated Penn State Worry Questionnaire (PSWQ) is administered. This 11-item questionnaire has total scores of 11 to 55, with higher scores indicating more worrying.

Information about levels of physical activity is gathered with the 7-item International Physical Activity Questionnaire - Short Form (IPAQ - SF). The scoring of the IPAQ is based on a metric called MET (multiples of the resting metabolic rate) minutes. MET minutes represent the amount of energy expended carrying out a physical activity. With higher scores indicating more vigorous physical activity.

Motivation for following the self-help intervention is measured with the 8-item Short Motivation Feedback List (SMFL; with total scores ranging from 0 to 80, where higher scores reflect higher levels of motivation). There are two different versions, of which the pre-intervention version will be assessed at baseline (t0) and the post-intervention one after 6 weeks (t1).

Satisfaction with the intervention will be assessed with the Client Satisfaction Questionnaire for internet-based interventions (CSQ-I). The total score of this 8-item questionnaire ranges from 8 to 32, with higher scores indicating higher levels of participant satisfaction.

Past intervention engagement will be measured with the Twente Engagement with eHealth Technologies Scale (TWEETS) at t1. The total score of this 9-item questionnaire ranges from 0 to 36, with higher scores indicating higher levels of engagement.

Technical alliance will be assessed with the Technical Alliance Inventory (TAI) at past-intervention. The total score of this 7-item questionnaire ranges from 7 to 84, with higher scores indicating higher levels of technical alliance.

Costs will be assessed on two levels: (1) costs of administering the component (service costs, monitored with administrative means) and (2) user's costs of executing the component (participant level) will be estimated

Time investment is measured in two ways at participant level: (1) Log-file analysis of the use of the online platform and (2) the time investment each user spends 'outside' the platform will be estimated.

Inclusion Criteria: Aged 18 years or older Mild to moderate depression as defined by a score between 5 and 15 on the Patient Health Questionnaire - 9 (PHQ-9) Adequate written proficiency in the Dutch language Have a valid email address and computer with internet access In possession of a smartphone Exclusion Criteria: Current risk for suicide according to the PHQ-9 questionnaire (question 9, score of 1 or higher) Currently receiving psychological treatment for depression or another psychiatric disorder in primary or specialized mental health care Currently having a psychiatric disorder

Department of Clinical, Neuro and Developmental Psychology, Vrije Universiteit, 1081BT Amsterdam, The Netherlands

Department of Clinical, Neuro and Developmental Psychology, Vrije Universiteit, 1081BT Amsterdam, The Netherlands

Individual participant data (anonymised and encrypted) will be available on request following a standardized data accession form.

Accession will be granted through a standardized accession form following review by the project's PI.

Buntrock C, Ebert DD, Lehr D, Smit F, Riper H, Berking M, Cuijpers P. Effect of a Web-Based Guided Self-help Intervention for Prevention of Major Depression in Adults With Subthreshold Depression: A Randomized Clinical Trial. JAMA. 2016 May 3;315(17):1854-63. doi: 10.1001/jama.2016.4326.

Simon GE, VonKorff M, Rutter C, Wagner E. Randomised trial of monitoring, feedback, and management of care by telephone to improve treatment of depression in primary care. BMJ. 2000 Feb 26;320(7234):550-4. doi: 10.1136/bmj.320.7234.550.

Batterham PJ, Calear AL. Preferences for Internet-Based Mental Health Interventions in an Adult Online Sample: Findings From an Online Community Survey. JMIR Ment Health. 2017 Jun 30;4(2):e26. doi: 10.2196/mental.7722.

Buntrock C, Ebert D, Lehr D, Riper H, Smit F, Cuijpers P, Berking M. Effectiveness of a web-based cognitive behavioural intervention for subthreshold depression: pragmatic randomised controlled trial. Psychother Psychosom. 2015;84(6):348-58. doi: 10.1159/000438673. Epub 2015 Sep 24.

Hassouni, A. E., Hoogendoorn, M., van Otterlo, M., Eiben, A. E., Muhonen, V., & Barbaro, E. (2018). A clustering-based reinforcement learning approach for tailored personalization of e-Health interventions. arXiv preprint arXiv:1804.03592.

Karyotaki E, Kleiboer A, Smit F, Turner DT, Pastor AM, Andersson G, Berger T, Botella C, Breton JM, Carlbring P, Christensen H, de Graaf E, Griffiths K, Donker T, Farrer L, Huibers MJ, Lenndin J, Mackinnon A, Meyer B, Moritz S, Riper H, Spek V, Vernmark K, Cuijpers P. Predictors of treatment dropout in self-guided web-based interventions for depression: an 'individual patient data' meta-analysis. Psychol Med. 2015 Oct;45(13):2717-26. doi: 10.1017/S0033291715000665. Epub 2015 Apr 17.

Karyotaki E, Riper H, Twisk J, Hoogendoorn A, Kleiboer A, Mira A, Mackinnon A, Meyer B, Botella C, Littlewood E, Andersson G, Christensen H, Klein JP, Schroder J, Breton-Lopez J, Scheider J, Griffiths K, Farrer L, Huibers MJ, Phillips R, Gilbody S, Moritz S, Berger T, Pop V, Spek V, Cuijpers P. Efficacy of Self-guided Internet-Based Cognitive Behavioral Therapy in the Treatment of Depressive Symptoms: A Meta-analysis of Individual Participant Data. JAMA Psychiatry. 2017 Apr 1;74(4):351-359. doi: 10.1001/jamapsychiatry.2017.0044.

Kelders, S. M. (2015, June). Involvement as a working mechanism for persuasive technology. In International Conference on Persuasive Technology (pp. 3-14). Springer, Cham.

Kranzler HR, McKay JR. Personalized treatment of alcohol dependence. Curr Psychiatry Rep. 2012 Oct;14(5):486-93. doi: 10.1007/s11920-012-0296-5.

Mohr DC, Cuijpers P, Lehman K. Supportive accountability: a model for providing human support to enhance adherence to eHealth interventions. J Med Internet Res. 2011 Mar 10;13(1):e30. doi: 10.2196/jmir.1602.

Riper H, Andersson G, Christensen H, Cuijpers P, Lange A, Eysenbach G. Theme issue on e-mental health: a growing field in internet research. J Med Internet Res. 2010 Dec 19;12(5):e74. doi: 10.2196/jmir.1713.

van Zoonen K, Buntrock C, Ebert DD, Smit F, Reynolds CF 3rd, Beekman AT, Cuijpers P. Preventing the onset of major depressive disorder: a meta-analytic review of psychological interventions. Int J Epidemiol. 2014 Apr;43(2):318-29. doi: 10.1093/ije/dyt175.

Warmerdam L, Riper H, Klein M, van den Ven P, Rocha A, Ricardo Henriques M, Tousset E, Silva H, Andersson G, Cuijpers P. Innovative ICT solutions to improve treatment outcomes for depression: the ICT4Depression project. Stud Health Technol Inform. 2012;181:339-43.