The Impact of Digital Mental Health Services on Loneliness and Mental Health: Results from a Prospective, Observational Study.

BACKGROUND: Loneliness has increased since the COVID-19 pandemic and negatively impacts mental health. This study examined relationships between loneliness and mental health among adults using a digital mental health platform. METHODS: A purposive sample of 919 participants (97% response rate) who were newly enrolled in the platform completed a survey on loneliness, depression, anxiety, well-being, stress, social support, and comorbidities at baseline and 3 months. Platform engagement was tracked during this period. We examined baseline differences between lonely and non-lonely participants; associations between loneliness, mental health symptoms, and comorbidities; and changes in loneliness and mental health through engagement in any form of care. RESULTS: At baseline, 57.8% of the sample were categorized as lonely. Loneliness was associated with younger age, fewer years of education, and the presence of a comorbidity (p values < .05). Baseline loneliness was associated with greater depression, anxiety, and stress and lower well-being and social support (ps < .001). The percentage of lonely participants decreased at follow-up (57.6% to 52.9%, p = .03). Those who improved in loneliness improved in mental health symptoms, well-being, and social support (ps < .001). Lonely participants who engaged in any form of care reported a greater reduction in loneliness than those who did not engage (p = .04). CONCLUSIONS: This study confirms previous findings of the high prevalence of loneliness among adults and risk factors for increased loneliness. Findings highlight the potential of digital platforms to reach lonely individuals and alleviate loneliness through remote mental health support.

Association between the timing of childhood adversity and epigenetic patterns across childhood and adolescence: findings from the Avon Longitudinal Study of Parents and Children (ALSPAC) prospective cohort.

BACKGROUND: Childhood adversity is a potent determinant of health across development and is associated with altered DNA methylation signatures, which might be more common in children exposed during sensitive periods in development. However, it remains unclear whether adversity has persistent epigenetic associations across childhood and adolescence. We aimed to examine the relationship between time-varying adversity (defined through sensitive period, accumulation of risk, and recency life course hypotheses) and genome-wide DNA methylation, measured three times from birth to adolescence, using data from a prospective, longitudinal cohort study. METHODS: We first investigated the relationship between the timing of exposure to childhood adversity between birth and 11 years and blood DNA methylation at age 15 years in the Avon Longitudinal Study of Parents and Children (ALSPAC) prospective cohort study. Our analytic sample included ALSPAC participants with DNA methylation data and complete childhood adversity data between birth and 11 years. We analysed seven types of adversity (caregiver physical or emotional abuse, sexual or physical abuse [by anyone], maternal psychopathology, one-adult households, family instability, financial hardship, and neighbourhood disadvantage) reported by mothers five to eight times between birth and 11 years. We used the structured life course modelling approach (SLCMA) to identify time-varying associations between childhood adversity and adolescent DNA methylation. Top loci were identified using an R2 threshold of 0·035 (ie, ≥3·5% of DNA methylation variance explained by adversity). We attempted to replicate these associations using data from the Raine Study and Future of Families and Child Wellbeing Study (FFCWS). We also assessed the persistence of adversity-DNA methylation associations we previously identified from age 7 blood DNA methylation into adolescence and the influence of adversity on DNA methylation trajectories from ages 0-15 years. FINDINGS: Of 13 988 children in the ALSPAC cohort, 609-665 children (311-337 [50-51%] boys and 298-332 [49-50%] girls) had complete data available for at least one of the seven childhood adversities and DNA methylation at 15 years. Exposure to adversity was associated with differences in DNA methylation at 15 years for 41 loci (R2 ≥0·035). Sensitive periods were the most often selected life course hypothesis by the SLCMA. 20 (49%) of 41 loci were associated with adversities occurring between age 3 and 5 years. Exposure to one-adult households was associated with differences in DNA methylation at 20 [49%] of 41 loci, exposure to financial hardship was associated with changes at nine (22%) loci, and physical or sexual abuse was associated with changes at four (10%) loci. We replicated the direction of associations for 18 (90%) of 20 loci associated with exposure to one-adult household using adolescent blood DNA methylation from the Raine Study and 18 (64%) of 28 loci using saliva DNA methylation from the FFCWS. The directions of effects for 11 one-adult household loci were replicated in both cohorts. Differences in DNA methylation at 15 years were not present at 7 years and differences identified at 7 years were no longer apparent by 15 years. We also identified six distinct DNA methylation trajectories from these patterns of stability and persistence. INTERPRETATION: These findings highlight the time-varying effect of childhood adversity on DNA methylation profiles across development, which might link exposure to adversity to potential adverse health outcomes in children and adolescents. If replicated, these epigenetic signatures could ultimately serve as biological indicators or early warning signs of initiated disease processes, helping identify people at greater risk for the adverse health consequences of childhood adversity. FUNDING: Canadian Institutes of Health Research, Cohort and Longitudinal Studies Enhancement Resources, EU's Horizon 2020, US National Institute of Mental Health.

Socioeconomic changes predict genome-wide DNA methylation in childhood.

Childhood socioeconomic position (SEP) is a major determinant of health and well-being across the entire life course. To effectively prevent and reduce health risks related to SEP, it is critical to better understand when and under what circumstances socioeconomic adversity shapes biological processes. DNA methylation (DNAm) is one such mechanism for how early life adversity 'gets under the skin'. In this study, we evaluated the dynamic relationship between SEP and DNAm across childhood using data from 946 mother-child pairs in the Avon Longitudinal Study of Parents and Children. We assessed six SEP indicators spanning financial, occupational and residential domains during very early childhood (ages 0-2), early childhood (ages 3-5) and middle childhood (ages 6-7). Epigenome-wide DNAm was measured at 412 956 cytosine-guanines (CpGs) from peripheral blood at age 7. Using an innovative two-stage structured life-course modeling approach, we tested three life-course hypotheses for how SEP shapes DNAm profiles-accumulation, sensitive period and mobility. We showed that changes in the socioeconomic environment were associated with the greatest differences in DNAm, and that middle childhood may be a potential sensitive period when socioeconomic instability is especially important in shaping DNAm. Top SEP-related DNAm CpGs were overrepresented in genes involved in pathways important for neural development, immune function and metabolic processes. Our findings highlight the importance of socioeconomic stability during childhood and if replicated, may emphasize the need for public programs to help children and families experiencing socioeconomic instability and other forms of socioeconomic adversity.

Testing lifecourse theories characterising associations between maternal depression and offspring depression in emerging adulthood: the Avon Longitudinal Study of Parents and Children.

BACKGROUND: Maternal depression is a major determinant of offspring mental health. Yet, little is understood about how the duration and timing of maternal depression shapes youth risk for depressive symptoms, which if understood could inform when best to intervene. This study aimed to determine how the timing and duration of maternal depression was related to offspring depression in emerging adulthood, and if these associations varied by sex. METHODS: We analysed data from the Avon Longitudinal Study of Parents and Children (a prenatal cohort in the Avon area of England, 1991-2003), n = 3,301. We applied the structured lifecourse modelling approach to maternal depression (assessed at 13 points from prenatal period to adolescence) and emerging adult depressive symptoms (age 21). Lifecourse models assessed were accumulation (sum of timepoints when maternal depression was reported), sensitive periods (each period assessed as one during which maternal depression has a stronger effect) and instability (frequent fluctuations in maternal depression). RESULTS: Female adolescents (n = 2,132) had higher SMFQ scores (mean = 6.15, SD = 5.90) than males (n = 1,169, mean = 4.87, SD = 4.82). Maternal depression was most common in the infancy period (21.2% males; 21.4% females). For males, accumulation was the most appropriate lifecourse model; for each additional period of maternal depression, depressive symptoms in emerging adulthood increased by 0.11 (95% CI: 0.07, 0.15, one-sided p value ≤ .001). For females, exposure to maternal depression was associated with increasing depressive symptoms in emerging adulthood, with the largest effect in mid-childhood (increase of 0.27 units, 95% CI 0.03-0.50, p = .015 for difference between mid-childhood and other time-periods) and a smaller, equal effect at all other time-periods (increase of 0.07 units per time-period, 95% CI: 0.03-0.12, p = .002). CONCLUSIONS: This study highlights the importance of ongoing maternal depression for the development of depression in offspring through to emerging adulthood. Because long-term exposure to maternal depression was particularly important, early interventions are warranted.

Understanding Components of Therapeutic Alliance and Well-Being from Use of a Global Digital Mental Health Benefit During the COVID-19 Pandemic: Longitudinal Observational Study.

Digital mental health services leverage technology to increase access to care, yet less is known about the quality of therapeutic relationships in a virtual setting. This study examined components of therapeutic alliance (a mechanism underlying successful treatment) and its association with beneficial treatment outcomes in a real-world, virtual setting. The objective is to examine (1) participant ratings of components of therapeutic alliance with providers in a virtual setting, (2) changes in subjective well-being and depressive symptoms among participants who began care with elevated depressive symptoms, and (3) the association between components of alliance and changes in participants' well-being. Adults (N = 3,087, M age = 36 ± 9 years, 54% female) across the world with access to digital mental health benefits who engaged in videoconference sessions with a licensed therapist (18%, 555/3,087), certified coach (65%, 2,003/3,087), or both (17%, 529/3,087) between Sept. 29, 2020 and Oct. 12, 21. Participants completed 2 adapted items from the Working Alliance Inventory (goals and bonds subscales) after each session, and ratings were averaged across visits (Cronbach's ɑ = .72). Participants' World Health Organization-Five (WHO-5) Well-Being Index scores at the start and end of the study period were used to measure changes in subjective well-being. Descriptive and inferential statistics were conducted to examine average alliance ratings across demographics and utilization types and the association between alliance and well-being. The median adapted therapeutic alliance score was 4.8 (range: 1-5) and did not differ by age, country, or baseline well-being (Ps > .07). Females reported higher components of alliance than males (4.88 vs. 4.67, P = .01). Participants utilizing telecoaching reported higher components of alliance than those utilizing teletherapy or both telecoaching and teletherapy (4.83 v. 4.75, P = .004), though effect sizes were negligible. Among those with elevated baseline depressive symptoms (n = 835), participants reported an average WHO-5 increase of 15.42 points (95% CI 14.19-16.65, P < .001, Cohen d = 1.06) with 58% (485/835) reporting clinical recovery and 57% (481/835) reporting clinical improvement in depressive symptoms. Higher components of therapeutic alliance scores predicted greater well-being at follow-up (b = 2.04, 95% CI 0.09-3.99, P = .04) after controlling for age, sex, baseline WHO-5, and number of days in care (R 2 = .06, P < .001). Exploratory analyses indicated this association did not differ by utilization type, baseline well-being, or session utilization (Ps > .34). People with access to one-on-one videoconferencing care via a digital mental health benefit formed a strong bond and sense of alignment on goals with both coaches and therapists. Higher components of alliance scores were associated with improvements in subjective well-being among participants who began care with elevated depressive symptoms, providing evidence that a positive bond and goal alignment with a provider are two of many factors influencing virtual care outcomes. Continued focus on the quality of therapeutic relationships will ensure digital mental health services are patient-tailored as these platforms expand equitable access to evidence-based care.

Statistical Modeling of Sensitive Period Effects Using the Structured Life Course Modeling Approach (SLCMA).

Sensitive periods are times during development when life experiences can have a greater impact on outcomes than at other periods during the life course. However, a dearth of sophisticated methods for studying time-dependent exposure-outcome relationships means that sensitive periods are often overlooked in research studies in favor of more simplistic and easier-to-use hypotheses such as ever being exposed, or the effect of an exposure accumulated over time. The structured life course modeling approach (SLCMA; pronounced "slick-mah") allows researchers to model complex life course hypotheses, such as sensitive periods, to determine which hypothesis best explains the amount of variation between a repeated exposure and an outcome. The SLCMA makes use of the least angle regression (LARS) variable selection technique, a type of least absolute shrinkage and selection operator (LASSO) estimation procedure, to yield a parsimonious model for the exposure-outcome relationship of interest. The results of the LARS procedure are complemented with a post-selection inference method, called selective inference, which provides unbiased effect estimates, confidence intervals, and p-values for the final explanatory model. In this chapter, we provide a brief overview of the genesis of this sensitive period modeling approach and provide a didactic step-by-step user's guide to implement the SLCMA in sensitive- period research. R code to complete the SLCMA is available on our GitHub page at: https://github.com/thedunnlab/SLCMA-pipeline .

Updates to data versions and analytic methods influence the reproducibility of results from epigenome-wide association studies.

Biomedical research has grown increasingly cooperative through the sharing of consortia-level epigenetic data. Since consortia preprocess data prior to distribution, new processing pipelines can lead to different versions of the same dataset. Similarly, analytic frameworks evolve to incorporate cutting-edge methods and best practices. However, it remains unknown how different data and analytic versions alter the results of epigenome-wide analyses, which could influence the replicability of epigenetic associations. Thus, we assessed the impact of these changes using data from the Avon Longitudinal Study of Parents and Children (ALSPAC) cohort. We analysed DNA methylation from two data versions, processed using separate preprocessing and analytic pipelines, examining associations between seven childhood adversities or prenatal smoking exposure and DNA methylation at age 7. We performed two sets of analyses: (1) epigenome-wide association studies (EWAS); (2) Structured Life Course Modelling Approach (SLCMA), a two-stage method that models time-dependent effects. SLCMA results were also compared across two analytic versions. Data version changes impacted both EWAS and SLCMA analyses, yielding different associations at conventional p-value thresholds. However, the magnitude and direction of associations was generally consistent between data versions, regardless of p-values. Differences were especially apparent in analyses of childhood adversity, while smoking associations were more consistent using significance thresholds. SLCMA analytic versions similarly altered top associations, but time-dependent effects remained concordant. Alterations to data and analytic versions influenced the results of epigenome-wide analyses. Our findings highlight that magnitude and direction are better measures for replication and stability than p-value thresholds.

Childhood Emotional Neglect and Adolescent Depression: Assessing the Protective Role of Peer Social Support in a Longitudinal Birth Cohort.

Introduction: Childhood adversities have been shown to increase psychopathology risk, including depression. However, the specific impact of childhood emotional neglect on later depression has been understudied. Moreover, few studies have investigated relational protective factors that may offset the risk of depression for children who experienced emotional neglect. Analyzing data (n = 3,265) from the Avon Longitudinal Study of Parents and Children (ALSPAC) study, a longitudinal birth cohort of children born to pregnant women residing in Avon, UK from 1990 to 1992, we assessed the prospective relationship between childhood emotional neglect and depressive symptoms in late adolescence, and tested whether peer social support in mid-adolescence moderates this relationship. Methods: Childhood emotional neglect, defined as the absence of parental attention and support, was measured across seven assessments from age 8 to 17.5. Peer social support was measured at age 15. Depressive symptoms were measured at age 18. We analyzed the associations between emotional neglect and depressive symptoms, and between peer support and depressive symptoms, and also tested interactive effects of peer support on the association between emotional neglect and depressive symptoms. Results: Higher levels of emotional neglect were associated with increased depressive symptoms at 18. Conversely, strong peer social support was associated with reduced depressive symptoms, though no significant interaction with emotional neglect was detected. Conclusion: Although childhood emotional neglect is a risk factor for later depression, our results suggest that strong peer social support at age 15 may generally reduce the risk of depressive symptoms by the time children reach late adolescence. Fostering strong peer support in youth may help offset depression risk for all youth, even among those who have experienced emotional neglect.