An Explainable Machine Learning Approach for COVID-19’s Impact on Mood States of Children and Adolescents during the First Lockdown in Greece

The global spread of COVID-19 led the World Health Organization to declare a pandemic on 11 March 2020. To decelerate this spread, countries have taken strict measures that have affected the lifestyles and economies. Various studies have focused on the identification of COVID-19’s impact on the mental health of children and adolescents via traditional statistical approaches. However, a machine learning methodology must be developed to explain the main factors that contribute to the changes in the mood state of children and adolescents during the first lockdown. Therefore, in this study an explainable machine learning pipeline is presented focusing on children and adolescents in Greece, where a strict lockdown was imposed. The target group consists of children and adolescents, recruited from children and adolescent mental health services, who present mental health problems diagnosed before the pandemic. The proposed methodology imposes: (i) data collection via questionnaires; (ii) a clustering process to identify the groups of subjects with amelioration, deterioration and stability to their mood state; (iii) a feature selection process to identify the most informative features that contribute to mood state prediction; (iv) a decision-making process based on an experimental evaluation among classifiers; (v) calibration of the best-performing model; and (vi) a post hoc interpretation of the features’ impact on the best-performing model. The results showed that a blend of heterogeneous features from almost all feature categories is necessary to increase our understanding regarding the effect of the COVID-19 pandemic on the mood state of children and adolescents.

Functional status, mood state, and physical activity among women with post-acute COVID-19 syndrome

Objectives: While organ-specific pathophysiology has been well-described in SARS-CoV-2 infection, less is known about the attendant effects on functional status, mood state and leisure-time physical activity (PA) in post-acute COVID-19 syndrome. Methods: A case-control design was employed to recruit 32 women (n = 17 SARS-CoV-2; n = 15 controls) matched on age (54 +/- 12 years), body mass index (27 +/- 6 kg/m2), smoking status, and history of cardiopulmonary disease. Participants completed a series of assessments including the Modified Pulmonary Functional Status and Dyspnea Questionnaire (PFSDQ-M), Profile of Mood States (POMS), and Godin-Shephard Leisure-Time PA. Results: SARS-CoV-2 participants exhibited poorer functional status (p = 0.008) and reduced leisure-time PA (p = 0.004) compared to controls. Significant between-group differences were also detected for the POMS total mood disturbance with sub-scale analyses revealing elevated tension, confusion, and lower vigor among SARS-CoV-2 participants (all p-values < 0.05). The number of SARS-CoV-2 symptoms (e.g.,loss of taste / smell, muscle aches etc.) were associated (r = 0.620, p = 0.008) with confusion. Conclusion: The sequela of persistent SARS-CoV-2 symptoms elicit clear disturbances in functional status, mood state, and leisure-time PA among women with post-acute COVID-19 syndrome.

Real-world clinical predictors of manic/hypomanic episodes among outpatients with bipolar disorder

Background Bipolar disorder is a mental illness in which manic and depressive states are repeated, causing psychosocial dysfunction. Manic/hypomanic episodes cause problems with interpersonal, social and financial activities, but there is limited evidence regarding the predictors of manic/hypomanic episodes in real-world clinical practice. Methods The multicenter treatment survey on bipolar disorder (MUSUBI) in Japanese psychiatric clinics was administered in an observational study that was conducted to accumulate evidence regarding bipolar disorder in real-world clinical practice. Psychiatrists were asked to complete a questionnaire about patients with bipolar disorder who visited 176 member clinics of the Japanese Association of Neuro-Psychiatric Clinics by conducting a retrospective medical record survey. Our study extracted baseline patient characteristics from September to October 2016, including comorbidities, mental status, duration of treatment, Global Assessment of Functioning (GAF) score, and pharmacological treatment details. We investigated the presence or absence of manic/hypomanic episodes over the course of one year from baseline to September-October 2017. Results In total, 2231 participants were included in our study, 29.1% of whom had manic/hypomanic episodes over the course of one year from baseline. Binomial logistic regression analysis revealed that the presence of manic/hypomanic episodes was correlated with lower baseline GAF scores, rapid cycling, personality disorder, bipolar I disorder, and a mood state with manic or mixed features. Substance abuse was also a risk factor for manic episodes. There was no significant association between a baseline antidepressant prescription and manic/hypomanic episodes. Conclusions In Japan, 29.1% of outpatients with bipolar disorder had manic/hypomanic episodes over the course of one year. Our study suggested that a low GAF score, rapid cycling, personality disorder, bipolar I disorder, substance abuse, and baseline mood state could be predictors of manic/hypomanic episodes. Based on our findings, an antidepressant prescription is not a predictor of manic/hypomanic episodes.

Positive and negative mood states do not influence cross-modal auditory distraction in the serial-recall paradigm

The aim of this study was to examine whether positive and negative mood states affect auditory distraction in a serial-recall task. The duplex-mechanism account differentiates two types of auditory distraction. The changing-state effect is postulated to be rooted in interference-by-process and to be automatic. The auditory-deviant effect is attributed to attentional capture by the deviant distractors. Only the auditory-deviant effect, but not the changing-state effect, should be influenced by emotional mood states according to the duplex-mechanism account. Four experiments were conducted to test how auditory distraction is affected by emotional mood states. Mood was induced by autobiographical recall (Experiments 1 and 2) or the presentation of emotional pictures (Experiments 3 and 4). Even though the manipulations were successful in inducing changes in mood, neither positive mood (Experiments 1 and 3) nor negative mood (Experiments 2 and 4) had any effect on distraction despite large samples sizes (N = 851 in total). The results thus are not in line with the hypothesis that auditory distraction is affected by changes in mood state. The results support an automatic-capture account according to which the auditory-deviant effect and the changing-state effect are mainly stimulus-driven effects that are rooted in the automatic processing of the to-be-ignored auditory stream.

Assessing changes in mood state in university students following short-term study abroad

Short-term study-abroad (STSA) programs provide a more accessible alternative for students who would otherwise not consider engaging in academic activities overseas. Though improvements in the levels of intercultural sensitivity and general academic aspects attained by STSA programs have been previously examined, much less is known regarding the impact such programs have in the mood of students. Here, we examined changes in mood state associated with participation in an STSA program in a group of Japanese university students. Mood states were assessed using the Profile of Mood States (POMS), the Satisfaction With Life Scale (SWLS), and the Gratitude Questionnaire (GQ-6). Results indicated that the POMS mean scores of Vigor-Activity and SWLS peaked at the time immediately following participation in the STSA program; moreover, the same scores were found to be at comparable levels even one month after the end of the program. These results indicate that participation in STSA programs can positively influence the mood state of university students, suggesting that the benefits associated with participation in such programs extend beyond typically reported improvements in the academic domain.

An Explainable Machine Learning Approach for COVID-19 Impact on Mood State of Children and Adolescents during the 1st Lockdown in Greece

The global spread of COVID-19 led the World Health Organization to declare a pandemic on 11 March 2020. To decelerate this spread, countries have taken strict measures that affected the lifestyle and economy. Various studies have been focused on the identification of COVID-19 impact to mental health of children and adolescents via traditional statistical approaches. However, a machine learning methodology must be developed to explain the main factors that contribute to the change of mood state of children and adolescents during the first lockdown. Therefore, to this study an explainable machine learning pipeline is presented focusing on children and adolescents in Greece, where a strict lockdown was imposed. The target group consists of children and adolescents, recruited from children and adolescent mental health services, who present mental health problems diagnosed before the pandemic. The proposed methodology imposes: (i) data collection via questionnaires; (ii) a clustering process to identify the groups of subjects with amelioration, deterioration and stability to their mood state; (iii) a feature selection process to identify the most informative features that contribute to mood state prediction; (iv) a decision-making process based on an experimental evaluation among classifiers; (v) calibration of the best performing model and (v) a post-hoc interpretation of the features&rsquo; impact on the best performing model. The results showed that a blend of heterogeneous features from almost all feature categories is necessary to increase our understanding regarding the effect of COVID-19 pandemic on the mood state of children and adolescents.

“The Competitive Season and Off-Season”: Preliminary Research concerning the Sport-Specific Performance, Stress, and Sleep in Elite Male Adolescent Basketball Athletes

Background: Through scholastic sports programs, adolescent athletes compete to represent their communities. However, few studies investigate the changes in physiological and mental profiles during varied sport periodization among this population. Therefore, the purpose of this study was to compare the changes in sports performance and stress-related biomarkers between the competitive season (CS) and off-season (OS) in elite adolescent basketball players. Method: Nine elite Division I male basketball players (age: 15–18 years. old) participated in this study. Basketball-specific performance, salivary dehydroepiandrosterone sulfate (DHEA-S)/cortisol levels, mood state, and sleep quality were all accessed during the CS and OS periods. Results: The training load during OS was 26.0% lower than CS (p = 0.001). Muscle mass, aerobic capacity, 10 m sprint, and Abalakov jump (AJ) power during OS were greater than that during CS (+2.2–9.8%, p < 0.05), but planned agility was greater during CS (p = 0.003). The salivary DHEA-S/cortisol was greater during CS than during OS (p = 0.039). The overall mood state and sleep quality did not differ between periods, but the POMS-tension was higher during CS (p = 0.005). Conclusion: The present study demonstrates that muscle mass, aerobic capacity, peak AJ power, and 10 m sprint performance, but not planned agility, were greater during OS compared to CS among elite adolescent basketball players. Furthermore, the stress-related responses reflected by the D/C ratio and mood tension were relatively lower during the OS in these athletes. Thus, this study suggests that coaches and sport science professionals should closely monitor athletes’ training states across varied training/competition periods to better react to modifying training or recovery plans.

Message Sidedness in Health Claims: Roles of Mood State, Product Involvement, and Self-Rated Health Status

Most of the previous studies with respect to message sidedness mainly focus on the effect of message sidedness in advertising on behavior of consumers and it is unknown how consumers respond to different message sidedness when a one-sided or two-sided message in claims shown on the package of a healthy food product. This study explores the underlying mechanisms how consumers respond to different message sidedness in claims. The results indicate that two-sided messages in claims are more persuasive than one-sided messages because they pass the “sufficiency threshold.” In addition, the results of this article show that mood state, product involvement, and self-rated health of individuals moderate the relationship between message sidedness in claims and product evaluation.

Relationshp Between Sleep Quality, Mood State, and Performance of Elite Air-Rifle Shooters

Background To evaluate the impact of pre-competition sleep quality on the mood and performance of elite air-rifle shooters. Methods This study included 23 elite air-rifle shooters who participated in an air-rifle shooting-competition from April 2019 to October 2019. Sleep time, sleep efficiency, sleep latency, and wake-up time after sleep onset were monitored using actigraphy. The Pittsburgh sleep quality index and Profile of Mood State were used to assess sleep quality. Competitive State Anxiety Inventory-2 was used to evaluate mood state. Results The average time to fall asleep, sleep time, sleep efficiency, and subjective sleep quality were 20.6 ± 14.9 min, 7.0 ± 0.8 h, 85.9 ± 5.3%, and 5.2 ± 2.2, respectively. Sleep quality decreased as the competition progressed. Pre-competition sleep time in female athletes was significantly higher than that on the competition day (P = 0.05). Pre-competition sleep latency was significantly longer in women than in men (P = 0.021). During training and pre-competition, the tension, fatigue, depression, and emotional disturbance were significantly lower in athletes with good sleep quality than in athletes with poor sleep quality. Athletes with good sleep quality had significantly more energy. The PSQI total score was positively correlated with positive emotion, TMD, cognitive anxiety, and somatic anxiety POMS scores, and negatively correlated with energy and self-confidence scores. Race scores and depression and somatic anxiety scores were negatively correlated. Conclusion Poor sleep quality negatively impacted the mood of athletes; however, sleep indices and competition performance of athletes during competitions were not significantly correlated.