Harnessing telemedicine for suicide prevention: a clinical trial

Methods

Summary

MindMe is currently a beta-stage application that is currently available only on the Android platform. The app allows users to access personalized therapeutic exercises, videos and strategies that have been preset by their therapist in crisis situations. It also encourages daily reflection on negative emotions that are being experienced, and rating of the intensity of these emotions, providing the therapist with the ability to compile this information, allowing the therapist to harness the power of our customized analytics platform to make data-driven clinical decisions.

Our plan is to develop the capabilities of the application, and expand its availability to multiple smartphone platforms so that any mobile device can launch MindMe, moving it out of the Beta phase of development and having it ready to power the clinical trial. Alongside further software development, we will also be using funds from this campaign to organize a large-scale clinical trial. We have encouraging pilot results to suggest that MindMe can make a difference in the lives of individuals with suicidal thoughts and ideation.

However, in order to conduct a large clinical trial, certain expensive needs must be met - we must recruit skilled clinical researchers to organize the study, and seek administrative approval from hospitals to begin trials. Clinicians must be trained to use MindMe, and then paid for their time spent recruiting and treating eligible research participants. Assessing therapists must be employed to measure the effects of MindMe so that any clinical effects from using the app can be quantified. Statisticians must be employed to collate and analyze the data, and then present findings in a way that is acceptable for publication.

Challenges

Our research plan has been carefully designed because we are dealing with such a high risk population. A major challenge associated with research of this nature is that the stakes are so high: if there is a negative response to an intervention, the results can be tragic. Our research will be conducted whilst adhering to the highest ethical and research practice standards in order collect data that can demonstrate a high degree of clinical efficacy. We are fortunate to be working closely with researchers who have been conducting responsible clinical research on high-risk, depressed individuals for decades. Our project proposal has been approved by the Institutional Review Board (IRB) of Weill Cornell Medical College, and has been judged to be safe and ethical to deploy into a clinical environment. Each research participant will provide informed consent before engaging in the study.

Pre Analysis Plan

Study Design

We will be conducting a randomized controlled trial to show the efficacy of MindMe in individuals with a diagnosis of unipolar non-psychotic Major Depressive Disorder. A randomized controlled trial is a gold standard approach to determining the clinical efficacy of an approach in clinical research. Research participants will be assessed for eligibility in the study using DSM-IV criteria: blinded assessors will conduct a Montgomery-Åsberg Depression Rating Scale (MADRS). After a description of the study, potential participants with capacity to consent (assessed with Cornell Scale for Capacity to Consent) will be required to consent forms and will be randomly assigned to one of two groups: (1) Problem Adaptation Therapy (PATH; a psychosocial intervention that has been shown to reduce depression) with MindMe for adjunct assistance outside of therapy (the INTERVENTION group), (2) PATH alone (the CONTROL group). Staff who are assessing research outcomes will be unaware of participants’ randomization status and the study hypotheses.

Assessment and Instruments

We will suicidal ideation at entry (baseline), week 4, week 8, and week 12 with Montgomery-Åsberg Depression Rating Scale– Suicide Item (MADRS-SI), rated on a scale of 0-6 with the following anchor points; 0=No suicidal ideation; 1=Occasional thoughts that life is not worth living; 2= Weary of life, fleeting suicidal thoughts; 3= Wish were dead, occasional suicidal thoughts but suicide may not be considered a solution; 4= Probably better off dead; suicidal thoughts are common and suicide is considered as a possible solution, but without specific plans or intentions. 5= Suicide intent but vague plan; 6=Explicit plans for suicide when there is an opportunity; active preparations for suicide. Depression and disability will also be quantified at all assessment points (baseline, week 4, week 8, and week 12) by the MADRS total score and the World Health Organization Disability Assessment Schedule II (WHODAS II)-12 item form. Emotion dysregulation (i.e. MADRS Dysregulated Emotions) will be quantified by summation of MADRS items that capture emotional states, i.e. sadness (MADRS 1 & MADRS 2), anxiety and worry (MADRS 3), anhedonia/lack of pleasure or interest in activities (MADRS 8), and feelings of guilt, self-reproach, and hopelessness (MADRS 9).

Statistical Analysis

We plan to analyze data using the Wilcoxon-Mann-Whitney test (for our continuous variables) and the Chi-square test or Fisher’s exact (for our categorical variables) to test differences between participants in the intervention group. In all analyses we will use used mixed effects models for longitudinal data. To test whether the intervention group have greater improvements in emotion regulation (or greater reduction in emotion dysregulation) over 12 weeks of treatment, will use a generalized linear mixed-effects model for longitudinal data. These gold-standard statistical analysis techniques will be conducted in MATLAB and will allow us to determine whether our application can significantly decrease symptoms of suicidal thought and ideation from at-risk individuals.