Prediction of well-being and insight into work-life integration among physicians using machine learning approach.
Ontology highlight
ABSTRACT:
SUBMITTER: Nishi M
PROVIDER: S-EPMC8282024 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:ImportancePoor work-life integration (WLI) occurs when career and personal responsibilities come in conflict and may contribute to the ongoing high rates of physician burnout. The characteristics associated with WLI are poorly understood.ObjectiveTo identify personal and professional factors associated with WLI in physicians and identify factors that modify the association between gender and WLI.Design, setting, and participantsThis cross-sectional study was based on electronic and paper surveys administered October 2017 to March 2018 at private, academic, military, and veteran's practices across the US. It used a population-based sample of US physicians across all medical specialties. Data analysis was performed from November 2019 to July 2020.Main outcomes and measuresWLI was assessed using an 8-item scale (0-100 point scale, with higher scores indicating favorable WLI), alongside personal and professional factors. Multivariable linear regressions evaluated independent associations with WLI as well as factors that modify the association between gender and WLI.ResultsOf 5197 physicians completing surveys, 4370 provided complete responses. Of the physicians who provided complete responses, 2719 were men, 3491 were White/Caucasian (80.8%), 3560 were married (82.4%), and the mean (SD) age was 52.3 (12.0) years. The mean (SD) WLI score was 55 (23). Women reported lower (worse) mean (SD) WLI scores than men overall (52 [22] vs 57 [23]; mean difference, -5 [-0.2 SDs]; P < .001). In multivariable regression, lower WLI was independently associated with being a woman (linear regression coefficient, -6; SE, 0.7; P < .001) as well as being aged 35 years or older (eg, aged 35 to 44 years: linear regression coefficient, -7; SE, 1.4; P < .001), single (linear regression coefficient, -3 vs married; SE, 1.1; P = .003), working more hours (eg, 50 to 59 hours per week vs less than 40 hours per week: linear regression coefficient, -9; SE, 1.0; P < .001) and call nights (linear regression coefficient, -1 for each call night per week; SE, 0.2; P < .001), and being in emergency medicine (linear regression coefficient, -18; SE, 1.6, P < .001), urology (linear regression coefficient, -11; SE, 4.0; P = .009), general surgery (linear regression coefficient, -4; SE, 2.0; P = .04), anesthesiology (linear regression coefficient, -4; SE, 1.7; P = .03), or family medicine (linear regression coefficient, -3; SE, 1.4; P = .04) (reference category, internal medicine subspecialties). In interaction modeling, physician age, youngest child's age, and hours worked per week modified the associations between gender and WLI, such that the largest gender disparities were observed in physicians who were aged 45 to 54 years (estimated WLI score for women, 49; 95% CI, 47-51; estimated WLI score for men, 57, 95% CI, 55-59; P < .001), had youngest child aged 23 years or older (estimated WLI score for women, 51; 95% CI, 48-54; estimated WLI score for men, 60; 95% CI, 58-62; P < .001), and were working less than 40 hours per week (estimated WLI score for women, 61; 95% CI, 59-63; estimated WLI score for men; 70; 95% CI, 68-72; P < .001).Conclusions and relevanceThis study found that lower WLI was reported by physicians who are women, single, aged 35 years or older, and who work more hours and call nights. These findings suggest that systemic change is needed to improve WLI among physicians.
| S-EPMC8160595 | biostudies-literature
Project description:Gene expression profiles were generated from 199 primary breast cancer patients. Samples 1-176 were used in another study, GEO Series GSE22820, and form the training data set in this study. Sample numbers 200-222 form a validation set. This data is used to model a machine learning classifier for Estrogen Receptor Status. RNA was isolated from 199 primary breast cancer patients. A machine learning classifier was built to predict ER status using only three gene features.
2013-01-01 | E-GEOD-29210 | biostudies-arrayexpress
Project description:Analyses of physician well-being typically rely on small and unrepresentative samples. In April 2011, the UK Office for National Statistics incorporated subjective well-being metrics (SWB) into the Annual Population Survey (APS), a well-established survey. This survey includes variables from the labor market, making APS an ideal source for measuring the association between work hours and SWB metrics and comparing among different professionals. Using APS data from 2011/12 to 2014/15, this study examined the association between SWB levels and work hours using multiple linear models for physicians (primary care physicians and hospital doctors), lawyers, and accountants. Of the 11,810 observations, physicians were more satisfied, happier, and less anxious; females were more stressed (10.7%); and age was negatively associated with happiness and satisfaction. Incorporating information on preferences to work more hours (underemployment) did not affect physicians' but worsened the well-being of other professionals (lawyers and accountants). Surveyed physicians were less anxious, happier, and more satisfied than lawyers or accountants before Covid. Although the total work hours did not alter the SWB metrics, overtime hours for other professionals did. Increasing the working hours of underemployed physicians (with appropriate compensation) could be a relatively inexpensive solution to tackle the shortage of health workers in the short run.
| S-EPMC10723739 | biostudies-literature
Project description:IntroductionIntegrating work and home domains is a constant challenge for medical professionals. Only half of physicians report positive work-life satisfaction, implying that negotiating the inherent conflicts between work and home may not be intuitive. Early teaching of skills focused on professional sustainability may best prepare physicians to navigate conflicts between work and home domains.MethodsThis interactive workshop targets trainees and junior faculty. It aims to highlight the current state of physician career satisfaction, to bring awareness to the risk of physician burnout, and to apply strategies that promote work-life integration as a lifelong practice for sustained career satisfaction. It includes a detailed presentation with structured resources to reinforce skill development.ResultsThis workshop was delivered five times to trainees and junior faculty. Workshop evaluations (n = 50) revealed that all participants believed the information presented was useful, addressed competencies relevant to their training, and increased their knowledge about how to create better work-life integration; all anticipated improvement in their professional work. They all recommended this program to a colleague.DiscussionThis workshop offers an effective way to teach a skill set that enhances physicians' abilities to negotiate conflicting work and life domain boundaries. Our results indicate that learners intend to apply newly acquired strategies for work-life integration so as to improve career satisfaction and wellness. Such skill sets may mitigate physician burnout and promote career sustainability, both critical issues with far-reaching implications for the delivery of safe, high-quality health care at the provider and system levels.
| S-EPMC6374746 | biostudies-literature
Project description:CRISPR interference (CRISPRi), the targeting of a catalytically dead Cas protein to block transcription, is the leading technique to silence gene expression in bacteria. However, design rules for CRISPRi remain poorly defined, limiting predictable design for gene interrogation, pathway manipulation, and high-throughput screens. Here we develop a best-in-class prediction algorithm for guide silencing efficiency by systematically investigating factors influencing guide depletion in multiple genome-wide essentiality screens, with the surprising discovery that gene-specific features such as transcriptional activity substantially impact prediction of guide activity. Accounting for these features as part of algorithm development allowed us to develop a mixed-effect random forest regression model that provides better estimates of guide efficiency than existing methods, as demonstrated in an independent saturating screen. We further applied methods from explainable AI to extract interpretable design rules from the model, such as sequence preferences in the vicinity of the PAM distinct from those previously described for genome engineering applications. Our approach provides a blueprint for the development of predictive models for CRISPR technologies where only indirect measurements of guide activity are available.
2023-12-19 | GSE196911 | GEO
Project description:In response to the outbreak of the COVID-19 pandemic, educational institutions around the world were forced into lockdown in order to contain the spread of the virus. To ensure continuous provision of education, most transitioned to emergency remote instruction. This has been particularly the case in higher education (HE) institutions. The circumstances of the pandemic have brought unprecedented psychological pressure on the population, in the case of educators and students exacerbated by the transition to a mode of instruction that was completely novel to the majority. The present study examines how college and university instructors dealt with teaching online in these unparalleled circumstances, with a focus on how factors connected with their daily lives and livelihoods influenced their well-being. Between April and September 2020, a comprehensive online survey was filled out by 804 HE instructors from 92 countries. We explore how sociodemographic variables such as gender, age, relationship status, living conditions, and length of professional experience non-trivially affect situational anxiety, work-life synergy, coping, and productivity. The results contribute to a better understanding of the impact of the pandemic and emergency remote instruction on college and university instructors' well-being by explaining the mechanisms mediating the relationship between individual, contextual, and affective variables. It may provide helpful guidelines for college and university administrators as well as teachers themselves as to how help alleviate the adverse effects of the continuing pandemic and possible similar disruptions leading to school closures on coping and well-being.
| S-EPMC8226323 | biostudies-literature
Project description:Background: Recent evidence suggests that integration of multi-modal data improves performance in machine learning prediction of depression treatment outcomes. Here, we compared the predictive performance of three machine learning classifiers using differing combinations of sociodemographic characteristics, baseline clinical self-reports, cognitive tests, and structural magnetic resonance imaging (MRI) features to predict treatment outcomes in late-life depression (LLD). Methods: Data were combined from two clinical trials conducted with depressed adults aged 60 and older, including response to escitalopram (N = 32, NCT01902004) and Tai Chi (N = 35, NCT02460666). Remission was defined as a score of 6 or less on the 24-item Hamilton Rating Scale for Depression (HAMD) at the end of 24 weeks of treatment. Features subsets were constructed from baseline sociodemographic and clinical features, gray matter volumes (GMVs), or both. Three classification algorithms were compared: (1) Support Vector Machine-Radial Bias Function (SVMRBF), (2) Random Forest (RF), and (3) Logistic Regression (LR). A repeated 5-fold cross-validation approach with a wrapper-based feature selection method was used for model fitting. Model performance metrics included Area under the ROC Curve (AUC) and Matthews correlation coefficient (MCC). Cross-validated performance significance was tested by permutation analysis. Classifiers were compared by Cochran's Q and post-hoc pairwise comparisons using McNemar's Chi-Square test with Bonferroni correction. Results: For the RF and SVMRBF algorithms, the combined feature set outperformed the clinical and GMV feature sets with a final cross-validated AUC of 0.83 ± 0.11 and 0.80 ± 0.11, respectively. Both classifiers passed permutation analysis. The LR algorithm performed best using GMV features alone (AUC 0.79 ± 0.14) but failed to pass permutation analysis using any feature set. Performance of the three classifiers differed significantly for all three features sets. Important predictive features of treatment response included anterior and posterior cingulate volumes, depression characteristics, and self-reported health-related quality scores. Conclusion: This preliminary exploration into the use of ML and multi-modal data to identify predictors of general treatment response in LLD indicates that integration of clinical and structural MRI features significantly increases predictive capability. Identified features are among those previously implicated in geriatric depression, encouraging future work in this arena.
| S-EPMC8563624 | biostudies-literature
Project description:Gene expression profiles were generated from 199 primary breast cancer patients. Samples 1-176 were used in another study, GEO Series GSE22820, and form the training data set in this study. Sample numbers 200-222 form a validation set. This data is used to model a machine learning classifier for Estrogen Receptor Status.
2013-01-01 | GSE29210 | GEO
Project description:Alongside academic learning, there is increasing recognition that educational systems must also cater to students' well-being. This study examines the key factors that predict adolescent students' subjective well-being, indexed by life satisfaction, positive affect, and negative affect. Data from 522,836 secondary school students from 71 countries/regions across eight different cultural contexts were analyzed. Underpinned by Bronfenbrenner's bioecological theory, both machine learning (i.e., light gradient-boosting machine) and conventional statistics (i.e., hierarchical linear modeling) were used to examine the roles of person, process, and context factors. Among the multiple predictors examined, school belonging and sense of meaning emerged as the common predictors of the various well-being dimensions. Different well-being dimensions also had distinct predictors. Life satisfaction was best predicted by a sense of meaning, school belonging, parental support, fear of failure, and GDP per capita. Positive affect was most strongly predicted by resilience, sense of meaning, school belonging, parental support, and GDP per capita. Negative affect was most strongly predicted by fear of failure, gender, being bullied, school belonging, and sense of meaning. There was a remarkable level of cross-cultural similarity in terms of the top predictors of well-being across the globe. Theoretical and practical implications are discussed.
| S-EPMC11006657 | biostudies-literature
Project description:BackgroundAlthough participation in paid work improves women's quality of life and well-being, the health benefits decline for women with young children. Implementing family-friendly work conditions is one strategy for improving working women's well-being, especially those with competing unpaid work responsibilities.ObjectiveThis study investigated the extent to which accessibility and use of 11 specific family-friendly work conditions were associated with physical health, anxiety and depression in Malaysian women with young children.DesignA cross-sectional design using a retrospective self-complete, anonymous, online survey was conducted between March and October 2021.MethodsWomen with a child aged 5 years or less (N = 190) completed an online survey measuring their exposure (availability and use) to 11 specific family-friendly work conditions, and their physical health, anxiety, and depression. The sample included women who were currently and recently working and with both formal and informal employment.ResultsAfter accounting for potential confounders, women who used paid maternity leave have a lower likelihood of having anxiety symptoms.ConclusionFuture research is needed to extend the findings from this study by over-sampling women who are informally employed and not currently working. Policy creation and development processes, including research and decision-making, should be led by and inclusive of women. For example, research funding could be allocated to 'lived experience' research that privileges the co-design of research with consumers. Based on these findings, the extent to which family-friendly work conditions fulfill their intent to improve the well-being for working women requires further critique.
| S-EPMC10908238 | biostudies-literature