Associations between smartphone keystroke metadata and mental health symptoms in adolescents: findings from the future proofing study
- Braund, Taylor, O'Dea, Bridianne, Bal, Debopriyo, Maston, Kate, Larsen, Mark, Werner-Seidler, Aliza, Tillman, Gabriel, Christensen, Helen
- Authors: Braund, Taylor , O'Dea, Bridianne , Bal, Debopriyo , Maston, Kate , Larsen, Mark , Werner-Seidler, Aliza , Tillman, Gabriel , Christensen, Helen
- Date: 2023
- Type: Text , Journal article
- Relation: JMIR Mental Health Vol. 10, no. (2023), p.
- Full Text:
- Reviewed:
- Description: Background: Mental disorders are prevalent during adolescence. Among the digital phenotypes currently being developed to monitor mental health symptoms, typing behavior is one promising candidate. However, few studies have directly assessed associations between typing behavior and mental health symptom severity, and whether these relationships differs between genders. Objective: In a cross-sectional analysis of a large cohort, we tested whether various features of typing behavior derived from keystroke metadata were associated with mental health symptoms and whether these relationships differed between genders. Methods: A total of 934 adolescents from the Future Proofing study undertook 2 typing tasks on their smartphones through the Future Proofing app. Common keystroke timing and frequency features were extracted across tasks. Mental health symptoms were assessed using the Patient Health Questionnaire-Adolescent version, the Children's Anxiety Scale-Short Form, the Distress Questionnaire 5, and the Insomnia Severity Index. Bivariate correlations were used to test whether keystroke features were associated with mental health symptoms. The false discovery rates of P values were adjusted to q values. Machine learning models were trained and tested using independent samples (ie, 80% train 20% test) to identify whether keystroke features could be combined to predict mental health symptoms. Results: Keystroke timing features showed a weak negative association with mental health symptoms across participants. When split by gender, females showed weak negative relationships between keystroke timing features and mental health symptoms, and weak positive relationships between keystroke frequency features and mental health symptoms. The opposite relationships were found for males (except for dwell). Machine learning models using keystroke features alone did not predict mental health symptoms. Conclusions: Increased mental health symptoms are weakly associated with faster typing, with important gender differences. Keystroke metadata should be collected longitudinally and combined with other digital phenotypes to enhance their clinical relevance. ©Taylor A Braund, Bridianne O'Dea, Debopriyo Bal, Kate Maston, Mark Larsen, Aliza Werner-Seidler, Gabriel Tillman, Helen Christensen.
- Authors: Braund, Taylor , O'Dea, Bridianne , Bal, Debopriyo , Maston, Kate , Larsen, Mark , Werner-Seidler, Aliza , Tillman, Gabriel , Christensen, Helen
- Date: 2023
- Type: Text , Journal article
- Relation: JMIR Mental Health Vol. 10, no. (2023), p.
- Full Text:
- Reviewed:
- Description: Background: Mental disorders are prevalent during adolescence. Among the digital phenotypes currently being developed to monitor mental health symptoms, typing behavior is one promising candidate. However, few studies have directly assessed associations between typing behavior and mental health symptom severity, and whether these relationships differs between genders. Objective: In a cross-sectional analysis of a large cohort, we tested whether various features of typing behavior derived from keystroke metadata were associated with mental health symptoms and whether these relationships differed between genders. Methods: A total of 934 adolescents from the Future Proofing study undertook 2 typing tasks on their smartphones through the Future Proofing app. Common keystroke timing and frequency features were extracted across tasks. Mental health symptoms were assessed using the Patient Health Questionnaire-Adolescent version, the Children's Anxiety Scale-Short Form, the Distress Questionnaire 5, and the Insomnia Severity Index. Bivariate correlations were used to test whether keystroke features were associated with mental health symptoms. The false discovery rates of P values were adjusted to q values. Machine learning models were trained and tested using independent samples (ie, 80% train 20% test) to identify whether keystroke features could be combined to predict mental health symptoms. Results: Keystroke timing features showed a weak negative association with mental health symptoms across participants. When split by gender, females showed weak negative relationships between keystroke timing features and mental health symptoms, and weak positive relationships between keystroke frequency features and mental health symptoms. The opposite relationships were found for males (except for dwell). Machine learning models using keystroke features alone did not predict mental health symptoms. Conclusions: Increased mental health symptoms are weakly associated with faster typing, with important gender differences. Keystroke metadata should be collected longitudinally and combined with other digital phenotypes to enhance their clinical relevance. ©Taylor A Braund, Bridianne O'Dea, Debopriyo Bal, Kate Maston, Mark Larsen, Aliza Werner-Seidler, Gabriel Tillman, Helen Christensen.
Efficacy of a digital mental health biopsychosocial transdiagnostic intervention with or without therapist assistance for adults with anxiety and depression : adaptive randomized controlled trial
- Andrews, Brooke, Klein, Britt, Nguyen, Huy, Corboy, Denise, McLaren, Suzanne, Watson, Shaun
- Authors: Andrews, Brooke , Klein, Britt , Nguyen, Huy , Corboy, Denise , McLaren, Suzanne , Watson, Shaun
- Date: 2023
- Type: Text , Journal article
- Relation: Journal of Medical Internet Research Vol. 25, no. (2023), p.
- Full Text:
- Reviewed:
- Description: Background: Digital mental health (DMH) interventions incorporating elements that adapt to the evolving needs of consumers have the potential to further our understanding of the optimal intensity of therapist assistance and inform stepped-care models. Objective: The primary objective was to compare the efficacy of a transdiagnostic biopsychosocial DMH program, with or without therapist assistance for adults with subthreshold symptoms or a diagnosis of anxiety or depression. Methods: In a randomized adaptive clinical trial design, all participants had access to the DMH program, with eligibility to have their program augmented with therapist assistance determined by program engagement or symptom severity. Participants who met stepped-care criteria were randomized to have their treatment program augmented with either low-intensity (10 min/week of video chat support for 7 weeks) or high-intensity (50 min/week of video chat support for 7 weeks) therapist assistance. A total of 103 participants (mean age 34.17, SD 10.50 years) were assessed before (week 0), during (weeks 3 and 6), and after the intervention (week 9) and at the 3-month follow-up (week 21). The effects of 3 treatment conditions (DMH program only, DMH program+low-intensity therapist assistance, and DMH program+high-intensity therapist assistance) on changes in the 2 primary outcomes of anxiety (7-item Generalized Anxiety Disorder Scale [GAD-7]) and depression (9-item Patient Health Questionnaire [PHQ-9]) were assessed using the Cohen d, reliable change index, and mixed-effects linear regression analyses. Results: There were no substantial differences in the outcome measures among intervention conditions. However, there were significant time effect changes in most outcomes over time. All 3 intervention conditions demonstrated strong and significant treatment effect changes in GAD-7 and PHQ-9 scores, with absolute Cohen d values ranging from 0.82 to 1.79 (all P<.05). The mixed-effects models revealed that, in the Life Flex program–only condition at week 3, mean GAD-7 and PHQ-9 scores significantly decreased from baseline by 3.54 and 4.38 (all P<.001), respectively. At weeks 6, 9, and 21, GAD-7 and PHQ-9 scores significantly decreased from baseline by at least 6 and 7 points (all P<.001), respectively. Nonresponders at week 3 who were stepped up to therapist assistance increased program engagement and treatment response. At the postintervention time point and 3-month follow-up, 67% (44/65) and 69% (34/49) of the participants, respectively, no longer met diagnostic criteria for anxiety or depression. Conclusions: The findings highlight that early detection of low engagement and non–treatment response presents an opportunity to effectively intervene by incorporating an adaptive design. Although the study findings indicate that therapist assistance was no more effective than the DMH intervention program alone for reducing symptoms of anxiety or depression, the data highlight the potential influence of participant selection bias and participant preferences within stepped-care treatment models. ©Brooke Andrews, Britt Klein, Huy Van Nguyen, Denise Corboy, Suzanne McLaren, Shaun Watson.
- Authors: Andrews, Brooke , Klein, Britt , Nguyen, Huy , Corboy, Denise , McLaren, Suzanne , Watson, Shaun
- Date: 2023
- Type: Text , Journal article
- Relation: Journal of Medical Internet Research Vol. 25, no. (2023), p.
- Full Text:
- Reviewed:
- Description: Background: Digital mental health (DMH) interventions incorporating elements that adapt to the evolving needs of consumers have the potential to further our understanding of the optimal intensity of therapist assistance and inform stepped-care models. Objective: The primary objective was to compare the efficacy of a transdiagnostic biopsychosocial DMH program, with or without therapist assistance for adults with subthreshold symptoms or a diagnosis of anxiety or depression. Methods: In a randomized adaptive clinical trial design, all participants had access to the DMH program, with eligibility to have their program augmented with therapist assistance determined by program engagement or symptom severity. Participants who met stepped-care criteria were randomized to have their treatment program augmented with either low-intensity (10 min/week of video chat support for 7 weeks) or high-intensity (50 min/week of video chat support for 7 weeks) therapist assistance. A total of 103 participants (mean age 34.17, SD 10.50 years) were assessed before (week 0), during (weeks 3 and 6), and after the intervention (week 9) and at the 3-month follow-up (week 21). The effects of 3 treatment conditions (DMH program only, DMH program+low-intensity therapist assistance, and DMH program+high-intensity therapist assistance) on changes in the 2 primary outcomes of anxiety (7-item Generalized Anxiety Disorder Scale [GAD-7]) and depression (9-item Patient Health Questionnaire [PHQ-9]) were assessed using the Cohen d, reliable change index, and mixed-effects linear regression analyses. Results: There were no substantial differences in the outcome measures among intervention conditions. However, there were significant time effect changes in most outcomes over time. All 3 intervention conditions demonstrated strong and significant treatment effect changes in GAD-7 and PHQ-9 scores, with absolute Cohen d values ranging from 0.82 to 1.79 (all P<.05). The mixed-effects models revealed that, in the Life Flex program–only condition at week 3, mean GAD-7 and PHQ-9 scores significantly decreased from baseline by 3.54 and 4.38 (all P<.001), respectively. At weeks 6, 9, and 21, GAD-7 and PHQ-9 scores significantly decreased from baseline by at least 6 and 7 points (all P<.001), respectively. Nonresponders at week 3 who were stepped up to therapist assistance increased program engagement and treatment response. At the postintervention time point and 3-month follow-up, 67% (44/65) and 69% (34/49) of the participants, respectively, no longer met diagnostic criteria for anxiety or depression. Conclusions: The findings highlight that early detection of low engagement and non–treatment response presents an opportunity to effectively intervene by incorporating an adaptive design. Although the study findings indicate that therapist assistance was no more effective than the DMH intervention program alone for reducing symptoms of anxiety or depression, the data highlight the potential influence of participant selection bias and participant preferences within stepped-care treatment models. ©Brooke Andrews, Britt Klein, Huy Van Nguyen, Denise Corboy, Suzanne McLaren, Shaun Watson.
Evaluation of various support intensities of digital mental health treatment for reducing anxiety and depression in adults : protocol for a mixed methods, adaptive, randomized clinical trial
- Andrews, Brooke, Klein, Britt, McLaren, Suzanne, Watson, Shaun, Corboy, Denise
- Authors: Andrews, Brooke , Klein, Britt , McLaren, Suzanne , Watson, Shaun , Corboy, Denise
- Date: 2023
- Type: Text , Journal article
- Relation: JMIR Research Protocols Vol. 12, no. (2023), p.
- Full Text:
- Reviewed:
- Description: Background: Anxiety and depression are leading causes of disease worldwide, requiring timely access to evidence-based treatment. Digital mental health (dMH) interventions increase accessibility to evidence-based psychological services delivered in a variety of web-based formats (eg, self-help and therapist-assisted interventions). Robust and rigorous studies of adaptive web-based intervention designs are scarce. No identified randomized clinical trial has investigated the efficacy of a 2-stage adaptive design, whereby the program-only condition or no support dMH treatment program is augmented by either low or high therapist assistance, if a participant does not improve or engage in the program-only condition. Objective: The primary objective is to assess whether low or high therapist-assisted support delivered via video chat is more effective in reducing anxiety and depressive symptoms compared with a dMH program–only condition. The secondary objective is to evaluate the role of motivation; self-efficacy; and preferences in participant engagement, adherence, and clinical outcomes (anxiety and depression symptoms) among the 3 treatment conditions (program only, low-intensity therapist assistance, and high-intensity therapist assistance). A mixed methods analysis of factors affecting participant attrition, participant reasons for nonengagement and withdrawal, and therapist training and implementation of dMH interventions will be completed. Qualitative data regarding participant and therapist experiences and satisfaction with video chat assessment and treatment will also be analyzed. Methods: Australian adults (N=137) with symptoms or a diagnosis of anxiety or depression will be screened for eligibility and given access to the 8-module Life Flex dMH treatment program. On day 15, participants who meet the augmentation criteria will be stepped up via block randomization to receive therapist assistance delivered via video chat for either 10 minutes (low intensity) or 50 minutes (high intensity) per week. This adaptive trial will implement a mixed methods design, with outcomes assessed before the intervention (week 0), during the intervention (weeks 3 and 6), after the intervention (week 9), and at the 3-month follow-up (week 21). Results: The primary outcome measures are for anxiety (Generalized Anxiety Disorder–7) and depression severity (Patient Health Questionnaire–9). Measures of working alliance, health status, health resources, preferences, self-efficacy, and motivation will be used for secondary outcomes. Qualitative methods will be used to explore participant and therapist experiences of video chat assessment and treatment, participant reasons for withdrawal and nonengagement, and therapist training and implementation experiences. Data collection commenced in November 2020 and was completed at the end of March 2022. Conclusions: This is the first mixed methods adaptive trial to explore the comparative efficacy of different intensity levels of self-help and a therapist-assisted dMH intervention program delivered via video chat for adults with anxiety or depression. Anticipated results may have implications for the implementation of dMH interventions. © Brooke Andrews, Britt Klein, Suzanne McLaren, Shaun Watson, Denise Corboy. Originally published in JMIR Research Protocols (https://www.researchprotocols.org), 28.04.2023.
- Authors: Andrews, Brooke , Klein, Britt , McLaren, Suzanne , Watson, Shaun , Corboy, Denise
- Date: 2023
- Type: Text , Journal article
- Relation: JMIR Research Protocols Vol. 12, no. (2023), p.
- Full Text:
- Reviewed:
- Description: Background: Anxiety and depression are leading causes of disease worldwide, requiring timely access to evidence-based treatment. Digital mental health (dMH) interventions increase accessibility to evidence-based psychological services delivered in a variety of web-based formats (eg, self-help and therapist-assisted interventions). Robust and rigorous studies of adaptive web-based intervention designs are scarce. No identified randomized clinical trial has investigated the efficacy of a 2-stage adaptive design, whereby the program-only condition or no support dMH treatment program is augmented by either low or high therapist assistance, if a participant does not improve or engage in the program-only condition. Objective: The primary objective is to assess whether low or high therapist-assisted support delivered via video chat is more effective in reducing anxiety and depressive symptoms compared with a dMH program–only condition. The secondary objective is to evaluate the role of motivation; self-efficacy; and preferences in participant engagement, adherence, and clinical outcomes (anxiety and depression symptoms) among the 3 treatment conditions (program only, low-intensity therapist assistance, and high-intensity therapist assistance). A mixed methods analysis of factors affecting participant attrition, participant reasons for nonengagement and withdrawal, and therapist training and implementation of dMH interventions will be completed. Qualitative data regarding participant and therapist experiences and satisfaction with video chat assessment and treatment will also be analyzed. Methods: Australian adults (N=137) with symptoms or a diagnosis of anxiety or depression will be screened for eligibility and given access to the 8-module Life Flex dMH treatment program. On day 15, participants who meet the augmentation criteria will be stepped up via block randomization to receive therapist assistance delivered via video chat for either 10 minutes (low intensity) or 50 minutes (high intensity) per week. This adaptive trial will implement a mixed methods design, with outcomes assessed before the intervention (week 0), during the intervention (weeks 3 and 6), after the intervention (week 9), and at the 3-month follow-up (week 21). Results: The primary outcome measures are for anxiety (Generalized Anxiety Disorder–7) and depression severity (Patient Health Questionnaire–9). Measures of working alliance, health status, health resources, preferences, self-efficacy, and motivation will be used for secondary outcomes. Qualitative methods will be used to explore participant and therapist experiences of video chat assessment and treatment, participant reasons for withdrawal and nonengagement, and therapist training and implementation experiences. Data collection commenced in November 2020 and was completed at the end of March 2022. Conclusions: This is the first mixed methods adaptive trial to explore the comparative efficacy of different intensity levels of self-help and a therapist-assisted dMH intervention program delivered via video chat for adults with anxiety or depression. Anticipated results may have implications for the implementation of dMH interventions. © Brooke Andrews, Britt Klein, Suzanne McLaren, Shaun Watson, Denise Corboy. Originally published in JMIR Research Protocols (https://www.researchprotocols.org), 28.04.2023.
- «
- ‹
- 1
- ›
- »