This study aimed to determine whether both subjective sleep quality and sleep duration are directly associated with quality of life (QOL), as well as indirectly associated with QOL through insomnia symptoms. Individuals aged 20–69 years without mental illness (n = 9305) were enrolled in this web-based cross-sectional survey. The Short Form-8 was used to assess physical and mental QOL. We used the Pittsburgh Sleep Quality Index (PSQI) and extracted items related to subjective sleep quality and sleep duration. Insomnia symptoms were also extracted from the PSQI. The hypothesized models were tested using structural equation modeling. Worse sleep quality, but not shorter sleep duration, was related to worse physical QOL. Both worse sleep quality and shorter sleep duration were related to worse mental QOL. Insomnia symptoms mediated these relationships. Subgroup analyses revealed a U-shaped relationship between sleep duration and physical/mental QOL. However, the relationship between sleep quality and physical/mental QOL was consistent regardless of sleep duration. The results suggest that subjective sleep quality has a more coherent association with QOL than subjective sleep duration. Because of its high feasibility, a questionnaire on overall sleep quality could be a useful indicator in future epidemiological studies of strategies for improving QOL.

Background: The screening process for systematic reviews is resource-intensive. Although previous machine learning solutions have reported reductions in workload, they risked excluding relevant papers. Objective: We evaluated the performance of a 3-layer screening method using GPT-3.5 and GPT-4 to streamline the title and abstract-screening process for systematic reviews. Our goal is to develop a screening method that maximizes sensitivity for identifying relevant records. Methods: We conducted screenings on 2 of our previous systematic reviews related to the treatment of bipolar disorder, with 1381 records from the first review and 3146 from the second. Screenings were conducted using GPT-3.5 (gpt-3.5-turbo-0125) and GPT-4 (gpt-4-0125-preview) across three layers: (1) research design, (2) target patients, and (3) interventions and controls. The 3-layer screening was conducted using prompts tailored to each study. During this process, information extraction according to each study’s inclusion criteria and optimization for screening were carried out using a GPT-4–based flow without manual adjustments. Records were evaluated at each layer, and those meeting the inclusion criteria at all layers were subsequently judged as included. Results: On each layer, both GPT-3.5 and GPT-4 were able to process about 110 records per minute, and the total time required for screening the first and second studies was approximately 1 hour and 2 hours, respectively. In the first study, the sensitivities/specificities of the GPT-3.5 and GPT-4 were 0.900/0.709 and 0.806/0.996, respectively. Both screenings by GPT-3.5 and GPT-4 judged all 6 records used for the meta-analysis as included. In the second study, the sensitivities/specificities of the GPT-3.5 and GPT-4 were 0.958/0.116 and 0.875/0.855, respectively. The sensitivities for the relevant records align with those of human evaluators: 0.867-1.000 for the first study and 0.776-0.979 for the second study. Both screenings by GPT-3.5 and GPT-4 judged all 9 records used for the meta-analysis as included. After accounting for justifiably excluded records by GPT-4, the sensitivities/specificities of the GPT-4 screening were 0.962/0.996 in the first study and 0.943/0.855 in the second study. Further investigation indicated that the cases incorrectly excluded by GPT-3.5 were due to a lack of domain knowledge, while the cases incorrectly excluded by GPT-4 were due to misinterpretations of the inclusion criteria. Conclusions: Our 3-layer screening method with GPT-4 demonstrated acceptable level of sensitivity and specificity that supports its practical application in systematic review screenings. Future research should aim to generalize this approach and explore its effectiveness in diverse settings, both medical and nonmedical, to fully establish its use and operational feasibility.

This study investigated the difference in the severity of mental distress and factors contributing to mental distress in frontline and non-frontline healthcare professionals during the coronavirus disease (COVID-19) pandemic. A cross-sectional web-based survey of medical staff collected by snow-ball sampling was performed in Japan in October 2020 using the Kessler Psychological Distress Scale (K6) as an outcome measure for mental distress. Originally developed items asking about the degree of change in psychological and physical burdens, COVID-19-related fear, and experience of discrimination were obtained. The median score of the K6 was 7 in the frontline staff group (n = 86) and 6 in the non-frontline staff group (n = 504), without a statistically significant difference. Multiple regression analyses showed that among the participants, an increase in psychological burden and COVID-19-related fear was significantly associated with mental distress in both groups. Experience of discrimination was significantly associated with mental distress only in the frontline staff group. However, an increase in physical burden was significantly associated with mental distress only in the non-frontline staff group. The results indicate that the factors contributing to mental distress between frontline and non-frontline staff can be different, although the severity of mental distress is comparable between them.

Aim: Pharmacological treatments recommended for bipolar depression are inconsistent across guidelines. We compared the efficacy and safety of antipsychotics and mood stabilizers for bipolar depression. Methods: A systemic review and meta-analysis of randomized controlled trials comparing antipsychotics and mood stabilizers for bipolar depression was conducted based on a literature search of major electronic databases. Results: Three studies comparing quetiapine with lithium were identified and analyzed; no other antipsychotic-mood stabilizer combinations were found. The meta-analysis revealed no significant differences between quetiapine and lithium for the following outcomes: (1) remission from depressive episodes (risk ratio [RR]: 1.80, 95% CI: 0.51-6.40, P = 0.36), (2) changes in depressive symptom (standardized mean difference: -0.22, 95% CI: -0.52-0.08, P = 0.15), (3) changes in social function (standardized mean difference: -0.00, 95% CI: -0.19-0.18, P = 0.98), (4) suicide-related events (odds ratio [OR]: 2.35, 95% CI: 0.40-13.65, P = 0.34), (5) severe adverse events (OR: 1.63, 95% CI: 0.51-5.20, P = 0.41), (6) dropouts due to adverse events (RR: 1.19, 95% CI: 0.76-1.87, P = 0.45, 7) dropout for any reasons (RR: 0.95, 95% CI: 0.74-1.22, P = 0.70). Conclusion: Although this study found no differences in the efficacy and safety of quetiapine and lithium for bipolar depression, a comprehensive comparison of antipsychotics and mood stabilizers was not performed. Further studies are needed to clarify which of these, not just quetiapine and lithium, is more useful for bipolar depression.

Health anxiety (HA), defined as excessive worry about having a serious medical condition, may affect preventive behaviors during the coronavirus disease 2019 (COVID-19) pandemic. We examined the distinct role of two dimensions of HA-perceived likelihood (probability dimension) and awfulness of illness (awfulness dimension)-in self-protection, as reflected in preventive behaviors during the pandemic. Participants comprised 657 healthcare workers. Data were collected between February 24 and 26, 2021. The Short Health Anxiety Inventory determined the HA dimensions. Adherence to the government's recommendations for COVID-19 preventive behaviors was self-rated. An independent association between each HA dimension and participants' adherence to the recommendations was examined using multivariable regression. Within the analyzed sample of 560 subjects, severe HA was observed in 9.1 %. The more the participants felt awful, the less frequently they engaged in the recommended preventive behaviors (adjusted odds ratio = 0.993, 95 % confidence interval: 0.989, 0.998, p = 0.003) regardless of their profession, working position, psychological distress, sleep disturbance, and current physical diseases. However, the probability dimension was not associated with their preventive behaviors. The awfulness dimension of HA could be a more sensitive marker of preventive behaviors than the probability dimension. Paying particular attention to the awfulness dimension may help optimize self-protection strategies during the COVID-19 pandemic. A two-dimensional understanding of HA may be useful for the maintenance of the healthcare system and public health as well as healthcare workers' own health.