A Mobile Self-control Training App to Improve Self-control and Physical Activity in People With Severe Mental Illness: Protocol for 2 Single-Case Experimental Designs

Abstract

Background Lack of physical activity is a common issue with detrimental consequences for the health of people with severe mental illness (SMI). Existing physical activity interventions show suboptimal effects as they require substantial cognitive skills, including goal setting and writing, whereas cognitive deficits are common in this population. To bolster the effectiveness of physical activity interventions, self-control training (SCT), in which users practice the ability to override unwanted thoughts and behaviors, can be used in addition. Recent research has demonstrated the initial effectiveness of a mobile SCT app, but this has not been studied in psychiatric clinical practice. Objective This study aims to evaluate to what extent adding a mobile SCT app designed for and with people with SMI to a mobile lifestyle intervention aimed at increasing physical activity increases physical activity and self-control levels. Methods A mixed methods approach incorporating 2 single-case experimental designs (SCEDs) and qualitative interviews was used to evaluate and optimize SCT. Overall, 12 participants with SMI will be recruited from 2 organizations offering outpatient and inpatient care to people with SMI. Each experiment will include 6 patients. SCED I is a concurrent multiple-baseline design across participants that explores initial effectiveness and optimal intervention duration. Using accelerometry and experience sampling questionnaires, participants’ physical activity and self-control will be monitored for ≥5 days from baseline, followed by the sequential introduction of Google Fit, the physical activity intervention, for 7 days and the addition of SCIPP: Self-Control Intervention App for 28 days. SCED II is an introduction/withdrawal design in which optimized SCT will be introduced and withdrawn to validate the findings from SCED I. In both experiments, the daily average of total activity counts per hour and the state level of self-control will serve as the primary and secondary outcome measures. Data will be analyzed using visual analysis and piecewise linear regression models. Results The study was designated as not subject to the Dutch Medical Research Involving Human Subjects Act by the Medical Research Ethical Committee Oost-Nederland and approved by the Ethics Committee/domain Humanities and Social Sciences of the Faculty of Behavioural, Management, and Social Sciences at the University of Twente. Participant recruitment started in January 2022, and we expect to publish the results in early 2023. Conclusions The mobile SCT app is expected to be feasible and effective. It is self-paced and scalable and can increase patient motivation, making it a suitable intervention for people with SMI. SCED is a relatively novel yet promising method for gaining insights into whether and how mobile apps work that can handle heterogeneous samples and makes it possible to involve a diverse population with SMI without having to include a large number of participants. International Registered Report Identifier (IRRID) PRR1-10.2196/37727

A Protocol for a feasibility study of Cognitive Bias Modification training (IVY 2.0) countering fatigue in people with breast cancer

“…PLM is an analysis method suited for time-series data that are segmented into phases (47,48). The dataset is split at certain break points (the phases) and regression parameters (intercept, trend, level, and slope) are calculated separately for each phase, and compared with each other (47,48).…”

“…PLM is an analysis method suited for time-series data that are segmented into phases (47,48). The dataset is split at certain break points (the phases) and regression parameters (intercept, trend, level, and slope) are calculated separately for each phase, and compared with each other (47,48). In this model, the intercept refers to the outcome at the start of the study, the trend effect refers to the trendline in one phase, the level effect refers to the instant and constant effect of the intervention, and the slope effect is an indication of the effect of the intervention by comparing the trendline (trend) at baseline to the trendline during the intervention (47,48).…”

“…The dataset is split at certain break points (the phases) and regression parameters (intercept, trend, level, and slope) are calculated separately for each phase, and compared with each other (47,48). In this model, the intercept refers to the outcome at the start of the study, the trend effect refers to the trendline in one phase, the level effect refers to the instant and constant effect of the intervention, and the slope effect is an indication of the effect of the intervention by comparing the trendline (trend) at baseline to the trendline during the intervention (47,48). Averages for each measurement will be investigated with the PLM, yielding beta weights for trend, level, and slope effects, as well as a P value (with P < .05 as cut-off for signi cance) and R 2 effect sizes (47).…”