A protocol for a multicentre, parallel-group, pragmatic randomised controlled trial to evaluate the NEVERMIND system in preventing and treating depression in patients with severe somatic conditions

Abstract

Background: Depressive symptoms are common in individuals suffering from severe somatic conditions. There is a lack of interventions and evidence-based interventions aiming to reduce depressive symptoms in patients with severe somatic conditions. The aim of the NEVERMIND project is to address these issues and provide evidence by testing the NEVERMIND system, designed to reduce and prevent depressive symptoms in comparison to treatment as usual. Methods: The NEVERMIND study is a parallel-groups, pragmatic randomised controlled trial to assess the effectiveness of the NEVERMIND system in reducing depressive symptoms among individuals with severe somatic conditions. The NEVERMIND system comprises a smart shirt and a user interface, in the form of a mobile application. The system is a real-time decision support system, aiming to predict the severity and onset of depressive symptoms by modelling the well-being condition of patients based on physiological data, body movement, and the recurrence of social interactions. The study includes 330 patients who have a diagnosis of myocardial infarction, breast cancer, prostate cancer, kidney failure, or lower limb amputation. Participants are randomised in blocks of ten to either the NEVERMIND intervention or treatment as usual as the control group. Clinical interviews and structured questionnaires are administered at baseline, at 12 weeks, and 24 weeks to assess whether the NEVERMIND system is superior to treatment as usual. The endpoint of primary interest is Beck Depression Inventory II (BDI-II) at 12 weeks defined as (i) the severity of depressive symptoms as measured by the BDI-II. Secondary outcomes include prevention of the onset of depressive symptoms, changes in quality of life, perceived stigma, and self-efficacy. Discussion: There is a lack of evidence-based interventions aiming to reduce and prevent depressive symptoms in patients with severe somatic conditions. If the NEVERMIND system is effective, it will provide healthcare systems with a novel and innovative method to attend to depressive symptoms in patients with severe somatic conditions.

The NEVERMIND e-health system in the treatment of depressive symptoms among patients with severe somatic conditions: A multicentre, pragmatic randomised controlled trial

“…This procedure, along with the statistical analysis plan, have been described in detail in the previously published protocol of this study. 21 The trial has been registered in the German Clinical Trials Register (DRKS00013391). This study did not have a data monitoring committee, as each local principal investigator had the responsibility to oversee their centre.…”

“… 20 More detailed information about the NEVERMIND system is included in the protocol of the study. 21 …”

Well-being Forecasting using a Parametric Transfer-Learning method based on the Fisher Divergence and Hamiltonian Monte Carlo

“…For the purposes of this work, we used data collected over 112 participants in the context of the NEVERMIND randomised controlled trial [11]. This dataset consists of participants aged 18 or older, who have received a diagnosis of a severe somatic disease, including myocardial infarction, breast cancer, prostate cancer, kidney failure and lower limb amputation.…”

“…The above conundrum was one that was also faced in the NEVERMIND project [10]. NEVERMIND is an EU-funded project, which includes a randomisedcontrolled trial, and is tasked with helping individuals at risk of developing depressive symptoms following a primary clinical event such as cancer, myocardial infarction, amputation, and kidney failure [11]. It does so by providing effective, smartphone/wearable-based self-management tools, which in turn complement and guide the patients' clinical support plan.…”

“…The main contribution of this paper is the following: we propose a personalised Bayesian inference method making use of TL in the context of Hamiltonian Monte Carlo sampling, which allows a population prior to be directly represented in the sampling process through the use of the Fisher divergence. We demonstrate the effectiveness of the above technique in the context of personalised prediction of self-reported well-being scores, using data from the NEVERMIND project [10,11]. Our method allows for a seamless transition from generalised to highly personalised models, as data become gradually available.…”