Artificial intelligence applications in social media for depression screening: A systematic review protocol for content validity processes.

BACKGROUND: The popularization of social media has led to the coalescing of user groups around mental health conditions; in particular, depression. Social media offers a rich environment for contextualizing and predicting users' self-reported burden of depression. Modern artificial intelligence (AI) methods are commonly employed in analyzing user-generated sentiment on social media. In the forthcoming systematic review, we will examine the content validity of these computer-based health surveillance models with respect to standard diagnostic frameworks. Drawing from a clinical perspective, we will attempt to establish a normative judgment about the strengths of these modern AI applications in the detection of depression. METHODS: We will perform a systematic review of English and German language publications from 2010 to 2020 in PubMed, APA PsychInfo, Science Direct, EMBASE Psych, Google Scholar, and Web of Science. The inclusion criteria span cohort, case-control, cross-sectional studies, randomized controlled studies, in addition to reports on conference proceedings. The systematic review will exclude some gray source materials, specifically editorials, newspaper articles, and blog posts. Our primary outcome is self-reported depression, as expressed on social media. Secondary outcomes will be the types of AI methods used for social media depression screen, and the clinical validation procedures accompanying these methods. In a second step, we will utilize the evidence-strengthening Population, Intervention, Comparison, Outcomes, Study type (PICOS) tool to refine our inclusion and exclusion criteria. Following the independent assessment of the evidence sources by two authors for the risk of bias, the data extraction process will culminate in a thematic synthesis of reviewed studies. DISCUSSION: We present the protocol for a systematic review which will consider all existing literature from peer reviewed publication sources relevant to the primary and secondary outcomes. The completed review will discuss depression as a self-reported health outcome in social media material. We will examine the computational methods, including AI and machine learning techniques which are commonly used for online depression surveillance. Furthermore, we will focus on standard clinical assessments, as indicating content validity, in the design of the algorithms. The methodological quality of the clinical construct of the algorithms will be evaluated with the COnsensus-based Standards for the selection of health status Measurement Instruments (COSMIN) framework. We conclude the study with a normative judgment about the current application of AI to screen for depression on social media. SYSTEMATIC REVIEW REGISTRATION: International Prospective Register of Systematic Reviews PROSPERO (registration number CRD42020187874).

Prevalence of High Resilience in Old Age and Association with Perceived Threat of COVID-19-Results from a Representative Survey.

Little is known about resilience in old age and its manifestation during the COVID-19 pandemic. This study aims to estimate the prevalence of high resilience in the German old age population. We further examine the socio-demographic correlates and whether high resilience reflects on older adults' perception of the threat posed by COVID-19. The data were derived from a representative telephone survey of n = 1005 older adults (≥65 years) during the first COVID-19 lockdown. Assessments included socio-demographic variables, the perceived threat of COVID-19, and high resilience (Brief Resilience Scale; cutoff: ≥4.31). The association between high resilience and threat from COVID-19 was analyzed using ordinal logistic regression. The study sample had a mean age (SD) of 75.5 (7.1) years, and n = 566 (56.3%) were female. The estimated prevalence of high resilience was 18.7% (95% CI = [16.3; 21.2]). High resilience was more prevalent in the younger age group and participants with higher education levels. High resilience was significantly associated with a lower perception of threat from COVID-19. The results of the representative survey in the German old age population showed that one out of five adults aged 65 years and older had high resilience. Older adults with high resilience tended to feel less threatened by COVID-19. Further research on resilience in old age is needed to support vulnerable groups in the context of care.

Social Isolation and Loneliness during COVID-19 Lockdown: Associations with Depressive Symptoms in the German Old-Age Population.

Lockdowns during the COVID-19 pandemic increase the risk of social isolation and loneliness, which may affect mental wellbeing. Therefore, we aimed to investigate associations between social isolation and loneliness with depressive symptoms in the German old-age population during the first COVID-19 lockdown. A representative sample of randomly selected individuals at least 65 years old (n = 1005) participated in a computer-assisted standardized telephone interview in April 2020. Sociodemographic data, aspects of the personal life situation, attitudes towards COVID-19 and standardized screening measures on loneliness (UCLA 3-item loneliness scale), depression (Brief Symptom Inventory/BSI-18), and resilience (Brief Resilience Scale/BRS) were assessed. Associations were inspected using multivariate regression models. Being lonely, but not isolated (ß = 0.276; p < 0.001) and being both isolated and lonely (ß = 0.136; p < 0.001) were associated with higher depressive symptoms. Being isolated, but not lonely was not associated with depressive symptoms. Thus, the subjective emotional evaluation, i.e., feeling lonely, of the social situation during lockdown seems more relevant than the objective state, i.e., being isolated. Normal (ß = -0.203; p < 0.001) and high resilience (ß = -0.308; p < 0.001) were associated with lower depressive symptoms across groups. Therefore, strengthening coping skills may be a support strategy during lockdowns, especially for lonely older individuals.

Social isolation, mental health, and use of digital interventions in youth during the COVID-19 pandemic: A nationally representative survey.

BACKGROUND: Public health measures to curb SARS-CoV-2 transmission rates may have negative psychosocial consequences in youth. Digital interventions may help to mitigate these effects. We investigated the associations between social isolation, COVID-19-related cognitive preoccupation, worries, and anxiety, objective social risk indicators, and psychological distress, as well as use of, and attitude toward, mobile health (mHealth) interventions in youth. METHODS: Data were collected as part of the "Mental Health And Innovation During COVID-19 Survey"-a cross-sectional panel study including a representative sample of individuals aged 16-25 years (N = 666; Mage = 21.3; assessment period: May 5, 2020 to May 16, 2020). RESULTS: Overall, 38% of youth met criteria for moderate or severe psychological distress. Social isolation worries and anxiety, and objective risk indicators were associated with psychological distress, with evidence of dose-response relationships for some of these associations. For instance, psychological distress was progressively more likely to occur as levels of social isolation increased (reporting "never" as reference group: "occasionally": adjusted odds ratio [aOR] 9.1, 95% confidence interval [CI] 4.3-19.1, p < 0.001; "often": aOR 22.2, CI 9.8-50.2, p < 0.001; "very often": aOR 42.3, CI 14.1-126.8, p < 0.001). There was evidence that psychological distress, worries, and anxiety were associated with a positive attitude toward using mHealth interventions, whereas psychological distress, worries, and anxiety were associated with actual use. CONCLUSIONS: Public health measures during pandemics may be associated with poor mental health outcomes in youth. Evidence-based digital interventions may help mitigate the negative psychosocial impact without risk of viral infection given there is an objective need and subjective demand.

Evidence Synthesis of Digital Interventions to Mitigate the Negative Impact of the COVID-19 Pandemic on Public Mental Health: Rapid Meta-review.

BACKGROUND: Accumulating evidence suggests the COVID-19 pandemic has negative effects on public mental health. Digital interventions that have been developed and evaluated in recent years may be used to mitigate the negative consequences of the pandemic. However, evidence-based recommendations on the use of existing telemedicine and internet-based (eHealth) and app-based mobile health (mHealth) interventions are lacking. OBJECTIVE: The aim of this study was to investigate the theoretical and empirical base, user perspective, safety, effectiveness, and cost-effectiveness of digital interventions related to public mental health provision (ie, mental health promotion, prevention, and treatment of mental disorders) that may help to reduce the consequences of the COVID-19 pandemic. METHODS: A rapid meta-review was conducted. The MEDLINE, PsycINFO, and CENTRAL databases were searched on May 11, 2020. Study inclusion criteria were broad and considered systematic reviews and meta-analyses that investigated digital tools for health promotion, prevention, or treatment of mental health conditions and determinants likely affected by the COVID-19 pandemic. RESULTS: Overall, 815 peer-reviewed systematic reviews and meta-analyses were identified, of which 83 met the inclusion criteria. Our findings suggest that there is good evidence on the usability, safety, acceptance/satisfaction, and effectiveness of eHealth interventions. Evidence on mHealth apps is promising, especially if social components (eg, blended care) and strategies to promote adherence are incorporated. Although most digital interventions focus on the prevention or treatment of mental disorders, there is some evidence on mental health promotion. However, evidence on process quality, cost-effectiveness, and long-term effects is very limited. CONCLUSIONS: There is evidence that digital interventions are particularly suited to mitigating psychosocial consequences at the population level. In times of physical distancing, quarantine, and restrictions on social contacts, decision makers should develop digital strategies for continued mental health care and invest time and efforts in the development and implementation of mental health promotion and prevention programs.

Mental wellbeing in the German old age population largely unaltered during COVID-19 lockdown: results of a representative survey.

BACKGROUND: Older individuals are at increased risk of a severe and lethal course of COVID-19. They have typically been advised to practice particularly restrictive social distancing ('cocooning'), which has sparked much debate on the consequences for their mental wellbeing. We aimed to provide evidence by conducting a representative survey among the German old population during COVID-19 lockdown. METHODS: A computer-assisted standardized telephone interview was conducted in a randomly selected and representative sample of the German old age population (n = 1005; age ≥ 65 years) during the first lockdown in April 2020. Assessments included sociodemographic factors, aspects of the personal life situation during lockdown, attitudes towards COVID-19, and standardized screening measures on depression, anxiety, somatization, overall psychological distress (Brief Symptom Inventory/BSI-18) and loneliness (UCLA 3-item loneliness scale). Sampling-weighted descriptive statistics and multiple multivariable regression analyses were conducted. RESULTS: Participants were M = 75.5 (SD = 7.1) years old; 56.3% were women. At data collection, COVID-19 lockdown had been in force for M = 28.0 (SD = 4.8) days. Overall, older individuals were worried about COVID-19, but supportive of the lockdown. Mean BSI-18 scores were 1.4 for depression, 1.6 for anxiety and 2.2 for somatization as well as 5.1 for global psychological distress. These figures did not indicate worse mental wellbeing, given normative values established by studies before the pandemic (2.0, 1.6, 2.4, 6.0, respectively). The prevalence of loneliness was 13.1%, which also fell within a range of estimates reported by studies before the pandemic. There were only few significant associations of aspects of the personal life situation during lockdown and attitudes towards COVID-19 with mental wellbeing. Resilience explained a large amount of variance. CONCLUSIONS: In the short-term, the mental wellbeing of the German old age population was largely unaltered during COVID-19 lockdown, suggesting resilience against the challenging pandemic situation. Our results refute common ageist stereotypes of "the weak and vulnerable older adults" that were present during the pandemic. Long-term observations are needed to provide robust evidence.