The Self-help Online Against Suicidal Thoughts (SOS) Trial

The Self-help Online Against Suicidal Thoughts (SOS) Trial - a Danish Randomized Wait-list Controlled Trial for People With Suicidal Thoughts

The objective of the SOS-trial is to examine if an online self-help intervention is effective in reducing suicidal thoughts among people at risk of suicide. The SOS-trial is a randomized, wait-list controlled trial with 1:1 allocation ratio. A total of 438 people with suicidal thoughts will be allocated to the intervention condition (N=219) or the control condition (N=219). The intervention condition consists of a six-week internet-based self-help therapy intervention. The control condition consists of a waiting list assignment for 32 weeks. The primary hypothesis is that the intervention is superior to the control condition in reducing suicidal thoughts at post-test (6 weeks). The SOS-trial is a partial replication of a previously conducted Dutch trial.

Suicide accounts for over 800,000 deaths each year and suicidal thoughts are common and cause distress for millions of people over the world. With this study, the investigators wish to examine the effectiveness of an online self-help intervention in reducing suicidal thoughts among people at risk of suicide. The participants will be recruited through the Danish online and telephone service, the Lifeline's website. Persons who consent to participate in the study will be randomized to the intervention or control condition using a centralized and computer-based algorithm that is stratified by sex and level of suicidal ideation. The intervention consists of an online self-help program with six modules that each takes a week to complete. Every module has a theoretical section, a compulsory exercise, and several optional exercises. The participants will be encouraged to spend a minimum of 30 minutes every day with the self-help program. The participants in the control condition will have access to the website but not the intervention modules. Measurements with self-report questionnaires will be carried out at baseline and after 2, 4, 6, and 32 weeks. The participants in the control condition, will after 6 months on the waiting list mode assessment have access to the intervention. These participants will further be assessed 3 times, i.e. after 34, 36 and 38 weeks. The null-hypothesis assumption of no difference between the intervention and control condition with respect to suicidal thoughts and the secondary outcome measures will be tested. Mixed models will be used to assess changes over time in suicidal thoughts and level of depression, with the measurements at baseline, 2, 4, and 6 weeks (post-test) as well as the long-term effect at 32 weeks (6 months after post-test). Changes in secondary outcomes will, apart from the participant's depression level, be measured with ANCOVA using multiple imputations. The investigators will use Cohen's d test to determine the difference between the intervention and control condition. If the true difference in the experimental and control mean is 0.3 in the Danish trial and a pooled standard deviation at post-test at 1 is expected, then 350 participants in each condition is needed in order to reject the null hypothesis that there is no difference between the intervention and control condition at post-test on suicidal thoughts with probability (power) 0.8. The type I error probability associated with this test of the null hypothesis is 0.05. By accounting for a drop-out rate of 20%, a total of 438 participants will be needed in the SOS-trial. The participants constitute a vulnerable group, and a safety procedure has therefore been developed. Participants will only be included if they provide their telephone number and the number of a contact person upon enrollment. During the first 6 weeks of the trial, the Beck Scale for Suicide Ideation will be administered biweekly; if a participant in either the intervention or control condition scores ≥ 27 on the scale, he or she will receive a telephone call from the trial manager. If the participant cannot be reached during three attempts over three days, the participant's contact person will be contacted. The same procedure will be used out if a participant ceases to follow the intervention without notifying the trial manager.

The internet-based self-help program consist of 6 modules that each take a week to complete. Every module consists of a theoretical section, a compulsory exercise, and several optional exercises: Participants are encouraged to spend a minimum of 30 minutes every day with the self-help program. On the website, the participants in the intervention condition may also access an "Acute Help" page where information on help options are listed and a "My Profile" page where changes to login and contact details may be made, also, the informed consent form is found here. Participants may furthermore access a message system where they can write questions to the trial manager.

Changes in suicidal thoughts from baseline to 2, 4, and 6 weeks measured with the Beck Scale for Suicide Ideation

Linear mixed models with repeated measurements and an unstructured covariance matrix, will use measurements at baseline, 2, 4, and 6 weeks (post-test) with the purpose of modeling changes over time in suicidal thoughts.

Changes in suicidal thoughts from baseline to post-test (6 weeks) and follow-up (32 weeks) measured with the Beck Scale for Suicide Ideation

Mixed models will be used to assess the long-term effect with measurements at baseline, 6 weeks (post-test), and 32 weeks (6 months after post-test).

Changes in suicidal thoughts from follow-up (32 weeks) to 34, 36, and 38 weeks in suicidal thoughts measured with the Beck Scale for Suicide Ideation (only for participants in the control condition who decides to start the self-help intervention)

Mixed models will use measurements at follow-up (32 weeks), 34, 36, and 38 weeks with the purpose of modeling changes over time in suicidal thoughts for those participants in the control condition who decides to start the self-help intervention.

Changes in suicidal thoughts from baseline to post-test (6 weeks) measured with the Suicidal Ideation Attributes Scale

Changes in suicidal thoughts from post-test (6 weeks) to follow-up (32 weeks) measured with the Suicidal Ideation Attributes Scale

Changes in depression level from baseline to 2, 4, and 6 weeks measured with the Six-item Hamilton Depression Scale

Mixed models with repeated measurements and an unstructured covariance matrix, will use measurements at baseline, 2, 4, and 6 weeks (post-test) with the purpose of modeling changes over time in depression level.

Changes in depression level from post-test (6 weeks) to follow-up (32 weeks) measured with the Six-item Hamilton Depression Scale

Changes in hopelessness from baseline to post-test (6 weeks) measured with the Beck Hopelessness Scale

Changes in hopelessness from post-test (6 weeks) to follow-up (32 weeks) measured with the Beck Hopelessness Scale

Changes in worrying from baseline to post-test (6 weeks) measured with the Penn State Worry Questionnaire Past Week

Changes in worrying from post-test (6 weeks) to follow-up (32 weeks) measured with the Penn State Worry Questionnaire Past Week

Changes in quality of life from baseline to post-test (6 weeks) measured with the WHO-Five Well-being Index

Changes in quality of life from post-test (6 weeks) to follow-up (32 weeks) measured with the WHO-Five Well-being Index

Episodes of deliberate self-harm the last six weeks measured with data from the Danish National Patient Register.

Episodes of deliberate self-harm the last six months measured with data from the Danish National Patient Register

Episodes of deliberate self-harm the last six weeks measured with data from the Danish National Patient Register

Death by suicide measured with the data extract from the Danish Cause of Death Register; (ICD-10: X60-X84)

Death by suicide measured with the data extract from the Danish Cause of Death Register; (ICD-10: X60-X84)

Death by suicide measured with the data extract from the Danish Cause of Death Register; (ICD-10: X60-X84)

Health care utilization from baseline to post-test (6 weeks) measured with The National Patient Register

Health care utilization from post-test (6 weeks) to follow-up (32 weeks) measured with The National Patient Register

Health care utilization from baseline to post-test (6 weeks) measured with National Health Service Register

Health care utilization from post-test (6 weeks) to follow-up (32 weeks) measured with National Health Service Register

Health care utilization from baseline to post-test (6 weeks) measured with the National Prescription Registry

Health care utilization from post-test (6 weeks) to follow-up (32 weeks) measured with the National Prescription Registry

Health care utilization from baseline to post-test (6 weeks) measured with the Psychiatric Central Research Register

Health care utilization from post-test (6 weeks) to follow-up (32 weeks) measured with the Psychiatric Central Research Register

Production loss from baseline to post-test (6 weeks) measured with the Trimbos and iMTA questionnaire for Costs associated with Psychiatric Illness

Production loss from post-test (6 weeks) to follow-up (32 weeks) measured with the Trimbos and iMTA questionnaire for Costs associated with Psychiatric Illness

Demographic questions measured with questions from the Trimbos and iMTA questionnaire for Costs associated with Psychiatric Illness

Evaluation and utility of the intervention measured with the Internet Evaluation and Utility Questionnaire

Evaluation and utility of the intervention measured with the Internet Evaluation and Utility Questionnaire

Inclusion Criteria: Age above 18 years old Have a personal code card (NemID) Sufficient command of the Danish language Exclusion Criteria: No suicidal thoughts (defined as a score below 3 on the Beck Scale for Suicide Ideation).

A de-identified dataset generated during the study and the statistical code used for the analysis will be available upon request, once the data has been collected and analyzed by the research group.

Muhlmann C, Madsen T, Hjorthoj C, Forman JL, Kerkhof AJFM, Nordentoft M, Erlangsen A. Effectiveness of an Internet-Based Self-help Therapy Program for Suicidal Ideation With Follow-up at 6 Months: Results of a Randomized Controlled Trial. J Clin Psychiatry. 2021 Aug 31;82(5):20m13803. doi: 10.4088/JCP.20m13803.

Muhlmann C, Madsen T, Hjorthoj C, Kerkhof A, Nordentoft M, Erlangsen A. The Self-help Online against Suicidal thoughts (SOS) trial: study protocol for a randomized controlled trial. Trials. 2017 Jan 28;18(1):45. doi: 10.1186/s13063-017-1794-x.