BackgroundAtheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study.MethodsThe study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009–2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators.ResultsAfter the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001).ConclusionThe systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin.
There is a wealth of research that has highlighted the relationship between personality and eating disorders. It has been suggested that understanding how subclinical disordered eating behaviours are uniquely associated with personality can help to improve the conceptualization of individuals with eating disorders. This study aimed to explore how the facets of the Five‐Factor Model (FFM) predicted restrictive eating, binge eating, purging, chewing and spitting, excessive exercising and muscle building among males and females. An online survey assessing disordered eating behaviours, FFM and general psychopathology was completed by 394 females and 167 males aged between 16 and 30 years. Simultaneous equations path models were systematically generated for each disordered eating behaviour to identify how the FFM facets, body dissatisfaction and age predicted behaviour. The results indicated that each of the six disordered behaviours were predicted by a unique pattern of thinking, feeling and behaving. Considerable differences between males and females were found for each path model, suggesting differences between males and females in the personality traits that drive disordered eating behaviours. It was concluded that it is important to take personality into account when treating males and females who engage in disordered eating behaviours.
Objectives: This mixed method study assessed the psychological impacts of six weeks of exposure to smoke and ash from the 2014 Hazelwood mine fire in the Latrobe Valley in Victoria, Australia.The quantitative component compared residents from the most exposed community (Morwell) with those from a similar, but minimally-exposed, control community (Sale). Qualities of the experience were examined in interviews with Morwell residents.
Methods:A cross-sectional survey involved 3,091 Morwell and 960 Sale adults with multiple psychological measures was complimented by 26 interviews with Morwell residents.Results: Morwell residents scored significantly higher than Sale residents on the primary outcome measure, the Impact of Event Scale -Revised (difference = 6.53; 95%CI: 5.37, 7.35, p<0.001), which measured the posttraumatic stress symptoms of intrusive rumination, hyperarousal and avoidance behaviour. Morwell residents also scored significantly higher on the Kessler 10-item general distress scale (difference = 1.69; 95%CI: 1.05, 2.33, p<0.001). On average, the Hazelwood mine fire continued to generate moderate levels of participant distress more than two years after the event, however this ranged from no impact to more severe distress. This range of impact was also evident in the qualitative interviews, where intrusive thoughts were the most frequently reported symptom of posttraumatic stress. The interviews highlighted the increased vulnerability of people with pre-existing mental health concerns.
Conclusions:The finding that moderate distress was apparent in the community several years after an extended community-wide pollution event highlights the need for improving response to such events, including providing support to more vulnerable subgroups.
Background
A major problem in quantifying symptoms of schizophrenia is establishing a reliable distinction between enduring and dynamic aspects of psychopathology. This is critical for accurate diagnosis, monitoring and evaluating treatment effects in both clinical practice and trials.
Materials and methods
We applied Generalizability Theory, a robust novel method to distinguish between dynamic and stable aspects of schizophrenia symptoms in the widely used Positive and Negative Symptom Scale (PANSS) using a longitudinal measurement design. The sample included 107 patients with chronic schizophrenia assessed using the PANSS at five time points over a 24‐week period during a multi‐site clinical trial of N‐Acetylcysteine as an add‐on to maintenance medication for the treatment of chronic schizophrenia.
Results
The original PANSS and its three subscales demonstrated good reliability and generalizability of scores (G = 0.77‐0.93) across sample population and occasions making them suitable for assessment of psychosis risks and long‐lasting change following a treatment, while subscales of the five‐factor models appeared less reliable. The most enduring symptoms represented by the PANSS were poor attention, delusions, blunted affect and poor rapport. More dynamic symptoms with 40%‐50% of variance explained by patient transient state including grandiosity, preoccupation, somatic concerns, guilt feeling and hallucinatory behaviour.
Conclusions
Identified dynamic symptoms are more amendable to change and should be the primary target of interventions aiming at effectively treating schizophrenia. Separating out the dynamic symptoms would increase assay sensitivity in trials, reduce the signal to noise ratio and increase the potential to detect the effects of novel therapies in clinical trials.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.