Predicting recurrence of depression using lifelog data: an explanatory feasibility study with a panel VAR approach

The role of the circadian system in the etiology of depression

Circadian rhythms have evolved in almost all organisms enabling them to anticipate alternating changes in the environment. As a consequence, the circadian clock controls a broad range of bodily functions including appetite, sleep, activity and cortisol levels. The circadian clock synchronizes itself to the external world mainly by environmental light cues and can be disturbed by a variety of factors, including shift-work, jet-lag, stress, ageing and artificial light at night. Interestingly, mood has also been shown to follow a diurnal rhythm. Moreover, circadian disruption has been associated with various mood disorders and patients suffering from depression have irregular biological rhythms in sleep, appetite, activity and cortisol levels suggesting that circadian rhythmicity is crucially involved in the etiology and pathophysiology of depression. The aim of the present review is to give an overview and discuss recent findings in both humans and rodents linking a disturbed circadian rhythm to depression. Understanding the relation between a disturbed circadian rhythm and the etiology of depression may lead to novel therapeutic and preventative strategies.

Characteristics of Mobile Health Platforms for Depression and Anxiety: Content Analysis Through a Systematic Review of the Literature and Systematic Search of Two App Stores

Background

Mobile health (mHealth) platforms show promise in the management of mental health conditions such as anxiety and depression. This has resulted in an abundance of mHealth platforms available for research or commercial use.

Objective

The objective of this review is to characterize the current state of mHealth platforms designed for anxiety or depression that are available for research, commercial use, or both.

Methods

A systematic review was conducted using a two-pronged approach: searching relevant literature with prespecified search terms to identify platforms in published research and simultaneously searching 2 major app stores—Google Play Store and Apple App Store—to identify commercially available platforms. Key characteristics of the mHealth platforms were synthesized, such as platform name, targeted condition, targeted group, purpose, technology type, intervention type, commercial availability, and regulatory information.

Results

The literature and app store searches yielded 169 and 179 mHealth platforms, respectively. Most platforms developed for research purposes were designed for depression (116/169, 68.6%), whereas the app store search reported a higher number of platforms developed for anxiety (Android: 58/179, 32.4%; iOS: 27/179, 15.1%). The most common purpose of platforms in both searches was treatment (literature search: 122/169, 72.2%; app store search: 129/179, 72.1%). With regard to the types of intervention, cognitive behavioral therapy and referral to care or counseling emerged as the most popular options offered by the platforms identified in the literature and app store searches, respectively. Most platforms from both searches did not have a specific target age group. In addition, most platforms found in app stores lacked clinical and real-world evidence, and a small number of platforms found in the published research were available commercially.

Conclusions

A considerable number of mHealth platforms designed for anxiety or depression are available for research, commercial use, or both. The characteristics of these mHealth platforms greatly vary. Future efforts should focus on assessing the quality—utility, safety, and effectiveness—of the existing platforms and providing developers, from both commercial and research sectors, a reporting guideline for their platform description and a regulatory framework to facilitate the development, validation, and deployment of effective mHealth platforms.

Prodromal Symptoms in Depression: A Systematic Review

INTRODUCTION

Appraisal of prodromal symptoms of unipolar depression may complement the traditional cross-sectional approach and provide a longitudinal perspective, according to a staging model of the illness.

OBJECTIVE

To provide an updated systematic review of clinical studies concerned with prodromal symptoms of unipolar depression, according to PRISMA guidelines.

METHODS

Keyword searches were conducted in PubMed, Scopus, and Web of Science. Longitudinal studies on prodromal symptoms and signs in adult patients primarily diagnosed with unipolar depression were selected. Findings were examined separately according to study design (i.e., retrospective or prospective).

RESULTS

Twenty-five studies met the criteria for inclusion in this systematic review. Findings indicate that a distinct prodromal symptomatology - commonly characterized by anxiety, tension, irritability, and somatic complaints - exists before the onset of unipolar depression. The duration of the prodromal phase was highly variable across studies, ranging from less than a month to several years. Prodromal symptoms profile and duration were consistent within individuals across depressive episodes. There was a close relationship between prodromal and residual symptoms of the same depressive episode.

CONCLUSIONS

The present systematic review addresses an important, and yet relatively neglected, clinical issue that deserves further investigation and may be of immediate practical value. The findings provide challenging insights into the pathogenesis and course of unipolar depression, which may result in more timely and effective treatment of recurrences. The definition of a prodromal phase in depression would benefit from the joint use of symptom identification, biomarkers, and neuroimaging.

Leveraging Longitudinal Lifelog Data Using Survival Models for Predicting Risk of Relapse among Patients with Depression in Remission

Managing depression relapse is a challenge given factors such as inconsistent follow-up and cumbersome psychological distress evaluation methods which leaves patients with a high risk of relapse to leave their symptoms untreated. In an attempt to bridge this gap, we proposed an approach on the use of personal longitudinal lifelog activity data gathered from individual smartphones of patients in remission and maintenance therapy (N=87) to predict their risk of depression relapse. Through the use of survival models, we modeled the activity data as covariates to predict survival curves to determine if patients are at risk of relapse. We compared three models: CoxPH, Random Survival Forests, and DeepSurv, and found that DeepSurv performed the best in terms of Concordance Index and Brier Score. Our results show the possibility of utilizing lifelog data as a means of predicting the onset of relapse and towards building eventual tools for a more coherent patient evaluation and intervention system.

Depression during the COVID-19 pandemic amongst residents of homeless shelters in France

Background

Accumulating evidence suggests that the COVID-19 pandemic has negatively affected global mental health and well-being. However, the impact amongst homeless persons has not been fully evaluated. The ECHO study reports factors associated with depression amongst the homeless population living in shelters in France during the spring of 2020.

Methods

Interview data were collected from 527 participants living in temporary and/or emergency accommodation following France's first lockdown (02/05/20 – 07/06/20), in the metropolitan regions of Paris (74%), Lyon (19%) and Strasbourg (7%). Interviews were conducted in French, English, or with interpreters (33% of participants, ∼20 languages). Presence of depression was ascertained using the Patient Health Questionnaire (PHQ-9).

Results

Amongst ECHO study participants, 30% had symptoms of moderate to severe depression (PHQ-9 ≥ 10). Multivariate analysis revealed depression to be associated with being female (aOR: 2.15; CI: 1.26–3.69), single (aOR: 1.60; CI: 1.01–2.52), chronically ill (aOR: 2.32; CI: 1.43: 3.78), facing food insecurity (aOR: 2.12; CI: 1.40–3.22) and participants’ region of origin. Persons born African and Eastern Mediterranean regions showed higher levels of depression (30–33% of participants) than those migrating from other European countries (14%). Reduced rates of depression were observed amongst participants aged 30–49 (aOR: 0.60; CI: 0.38–0.95) and over 50 (aOR: 0.28; CI: 0.13–0.64), compared to 18–29-year-olds.

Limitations

These data are cross-sectional, only providing information on a given moment in time.

Conclusions

Our results indicate high levels of depression amongst homeless persons during the COVID-19 pandemic. Predicted future instability and economic repercussions could particularly impact the mental health of this vulnerable group.

The Impact of COVID-19 on Informal Caregiving and Care Receiving Across Europe During the First Phase of the Pandemic

Purpose: We analyzed the effects of COVID-19 as well as its accompanying epidemiological control measures on health-related outcomes (physical and mental health) and unmet care needs of both caregivers and care recipients across Europe and Israel by taking into account country differences. Methods: We applied comparisons of adjusted predictions, controlling for a large set of relevant respondent characteristics, to investigate changes in the physical and mental health of caregivers and care recipients due to COVID-19. Furthermore, multilevel regression models were used to analyze the effect of individual and contextual indicators on the probability of reporting difficulties in receiving care. For the analyses, we used data from 26 countries with 51,983 respondents over 50 years based on the eighth wave of the Survey of Health, Aging and Retirement in Europe (SHARE), which had to be suspended in March 2020, and the SHARE Corona Survey fielded from June to August 2020. Results: During the first phase of the pandemic in spring/summer 2020, the frequency of providing personal care to parents increased in almost all European countries, while care to children, in turn, decreased. Parental caregivers who increased the frequency of providing personal care reported significantly more mental health strains, that is, feeling sad/depressed and anxious/nervous more often since the outbreak of the pandemic. With respect to receiving care, about one out of five care recipients had difficulty in obtaining adequate care from outside the household during the pandemic. The perception of unmet care needs was significantly associated with country differences regarding the duration of the stay-at-home orders. In contrast, the number of confirmed deaths did not have a significant effect on perceiving difficulties related to receiving care. Conclusions: Our findings show the extent of the burden to which caregivers and care recipients were exposed with respect to the unintended consequences of COVID-19-related epidemiological control measures. There is a great need within this population for interventions, which effectively reduce the burden as well as the symptoms of anxiety or depression for caregivers as well as care recipients. This should be recognized by (health) policymakers and social organizations.

Assessing recurrence of depression using a zero-inflated negative binomial model: A secondary analysis of lifelog data

When studying recurrence of depression, researchers should pay attention to cases where physicians' assessment corresponds to the patients' perception. However, they should also focus on potential signs of recurrence when the recurrence is suspected by the physicians but not the patients (false-negative zeros). Because false negatives can delay diagnosis and treatment, we aimed to investigate "sitting idly" as a predictor influencing no alert sign of recurrence and estimated the counts of recurrence of depression. A smartphone application and a wearable device were used to collect lifelog data from 89 remitted depressive patients over one year. Recurrent depression was defined using the Japanese version of the Kessler Psychological Distress Scale and Patient Health Questionnaire-9 scores. Estimates of the population-averaged parameters indicated that daily hours of sitting idly increased the chances of recurrent depression occurring two to four weeks later. Exposure to daily ultraviolet light reduced depression relapse. Although long sleep was a determinant of zero outcome of the recurrence of depression after two to four weeks, daily hours of sitting idly can negate it. Thus, daily hours of sitting idly could reduce overdispersion of the recurrence of depression, and we could measure recurrent depression accurately by considering changes in sitting idly.

Prognostic models for predicting relapse or recurrence of major depressive disorder in adults

BACKGROUND

Relapse (the re-emergence of depressive symptoms after some level of improvement but preceding recovery) and recurrence (onset of a new depressive episode after recovery) are common in depression, lead to worse outcomes and quality of life for patients and exert a high economic cost on society. Outcomes can be predicted by using multivariable prognostic models, which use information about several predictors to produce an individualised risk estimate. The ability to accurately predict relapse or recurrence while patients are well (in remission) would allow the identification of high-risk individuals and may improve overall treatment outcomes for patients by enabling more efficient allocation of interventions to prevent relapse and recurrence.

OBJECTIVES

To summarise the predictive performance of prognostic models developed to predict the risk of relapse, recurrence, sustained remission or recovery in adults with major depressive disorder who meet criteria for remission or recovery.

SEARCH METHODS

We searched the Cochrane Library (current issue); Ovid MEDLINE (1946 onwards); Ovid Embase (1980 onwards); Ovid PsycINFO (1806 onwards); and Web of Science (1900 onwards) up to May 2020. We also searched sources of grey literature, screened the reference lists of included studies and performed a forward citation search. There were no restrictions applied to the searches by date, language or publication status .

SELECTION CRITERIA

We included development and external validation (testing model performance in data separate from the development data) studies of any multivariable prognostic models (including two or more predictors) to predict relapse, recurrence, sustained remission, or recovery in adults (aged 18 years and over) with remitted depression, in any clinical setting. We included all study designs and accepted all definitions of relapse, recurrence and other related outcomes. We did not specify a comparator prognostic model.

DATA COLLECTION AND ANALYSIS

Two review authors independently screened references; extracted data (using a template based on the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS)); and assessed risks of bias of included studies (using the Prediction model Risk Of Bias ASsessment Tool (PROBAST)). We referred any disagreements to a third independent review author. Where we found sufficient (10 or more) external validation studies of an individual model, we planned to perform a meta-analysis of its predictive performance, specifically with respect to its calibration (how well the predicted probabilities match the observed proportions of individuals that experience the outcome) and discrimination (the ability of the model to differentiate between those with and without the outcome). Recommendations could not be qualified using the GRADE system, as guidance is not yet available for prognostic model reviews.

MAIN RESULTS

We identified 11 eligible prognostic model studies (10 unique prognostic models). Seven were model development studies; three were model development and external validation studies; and one was an external validation-only study. Multiple estimates of performance measures were not available for any of the models and, meta-analysis was therefore not possible. Ten out of the 11 included studies were assessed as being at high overall risk of bias. Common weaknesses included insufficient sample size, inappropriate handling of missing data and lack of information about discrimination and calibration. One paper (Klein 2018) was at low overall risk of bias and presented a prognostic model including the following predictors: number of previous depressive episodes, residual depressive symptoms and severity of the last depressive episode. The external predictive performance of this model was poor (C-statistic 0.59; calibration slope 0.56; confidence intervals not reported). None of the identified studies examined the clinical utility (net benefit) of the developed model.

AUTHORS' CONCLUSIONS

Of the 10 prognostic models identified (across 11 studies), only four underwent external validation. Most of the studies (n = 10) were assessed as being at high overall risk of bias, and the one study that was at low risk of bias presented a model with poor predictive performance. There is a need for improved prognostic research in this clinical area, with future studies conforming to current best practice recommendations for prognostic model development/validation and reporting findings in line with the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement.