A cohort study on the predictive capability of body composition for diabetes mellitus using machine learning.

Purpose: We applied machine learning to study associations between regional body fat distribution and diabetes mellitus in a population of community adults in order to investigate the predictive capability. We retrospectively analyzed a subset of data from the published Fasa cohort study using individual standard classifiers as well as ensemble learning algorithms. Methods: We measured segmental body composition using the Tanita Analyzer BC-418 MA (Tanita Corp, Japan). The following features were input to our machine learning model: fat-free mass, fat percentage, basal metabolic rate, total body water, right arm fat-free mass, right leg fat-free mass, trunk fat-free mass, trunk fat percentage, sex, age, right leg fat percentage, and right arm fat percentage. We performed classification into diabetes vs. no diabetes classes using linear support vector machine, decision tree, stochastic gradient descent, logistic regression, Gaussian naïve Bayes, k-nearest neighbors (k = 3 and k = 4), and multi-layer perceptron, as well as ensemble learning using random forest, gradient boosting, adaptive boosting, XGBoost, and ensemble voting classifiers with Top3 and Top4 algorithms. 4661 subjects (mean age 47.64 ± 9.37 years, range 35 to 70 years; 2155 male, 2506 female) were analyzed and stratified into 571 and 4090 subjects with and without a self-declared history of diabetes, respectively. Results: Age, fat mass, and fat percentages in the legs, arms, and trunk were positively associated with diabetes; fat-free mass in the legs, arms, and trunk, were negatively associated. Using XGBoost, our model attained the best excellent accuracy, precision, recall, and F1-score of 89.96%, 90.20%, 89.65%, and 89.91%, respectively. Conclusions: Our machine learning model showed that regional body fat compositions were predictive of diabetes status.

Clinical learning evaluation questionnaire: a reliable and valid tool for the evaluation of clinical education by educators and students.

BACKGROUND: The clinical learning evaluation questionnaire can be used in the clinical trial period of medical students to measure the effectiveness of the clinical learning environment. The purpose of this study was to collect validity evidence of an adapted questionnaire to measure the transcultural adaptation of a Persian version of CLEQ. METHODS: A total of 200 questionnaires were completed by students who were at the end of their clinical rotation. The study instrument was the latest version of the CLEQ consists of 18 Items in four dimensions. The CLEQ was translated into Persian language through a four-step process of forward and backward translation. Data analysis was performed on two softwares, SPSS, version 22 and Lisrel, version 8.8. RESULTS: The results showed that the 18-question CLEQ could be applied to the Persian translation of the tool. The response process evidence of the Persian questionnaire was established through feedback from 15 students in the sample group. The content validity index (CVI) for the items were between 0.8 and 0.9, and the content validity ratio (CVR) for the entire questionnaire was 0.9. The 4-factor feature of CLEQ was good model fit. The internal consistency analysis indicated that the Cronbach's alpha values for all items of the 18-item questionnaire were equal to 0.87 and for the subscales were 0.68 to 0.79. CONCLUSION: The Persian translation of the 4-factor CLEQ has sufficient validity evidence to measure the transcultural adaptability of clinical education activities by instructors and students. The validity evidence are content, response process and internal structure. We recommend that the English 6-factor and 6-factor versions of CLEQ be tested on medical students at multiple foreign academic institutions to assess their efficiency.

General Anxiety Disorder and Primary Hypertension: Is There a Link?

ABSTRACT: Chronic forms of morbidity, including mental disorders and hypertension, play a dominant role in determining a disease load in the developing world. This article investigates the associations between the diagnoses of primary hypertension and generalized anxiety disorder (GAD) and the severity of GAD in individuals with primary hypertension. The association of GAD and age, sex, marriage, education level, income, chronic medical conditions, family history of mental disorders, adverse life events, and hypertension was assessed in 470 patients with the diagnosis of primary hypertension. Data analysis was performed with IBM SPSS Statistics software version 16. A significant relationship was found between the prevalence of GAD and the following variables: history of mental disorders ( p < 0.0001), chronic medical conditions ( p < 0.0001), and adverse life events ( p < 0.0001). The mean anxiety score was higher among patients with uncontrolled blood pressure, and a significant relationship was observed between the prevalence of GAD and blood pressure ( p < 0.0001). Because of the significant association between GAD and primary hypertension, it is recommended that anxiety disorders be considered in patients in whom primary hypertension is not controlled easily. This may lead to more proper control of hypertension while taking fewer antihypertensive medications.

The impact of antihypertensive treatment of mild to moderate hypertension during pregnancy on maternal and neonatal outcomes: An updated meta-analysis of randomized controlled trials.

Currently, there is controversy regarding the treatment of pregnant patients with mild hypertension (blood pressure 140-159/90-109 mm Hg). While guidelines do not recommend this treatment, results from recent clinical trials are supportive of the treatment. This meta-analysis aimed to clarify if active treatment of mild hypertension during pregnancy results in better maternal and fetal outcomes. All of the potentially eligible randomized controlled trials were retrieved through a systematic database search investigating the impact of pharmacological treatment in mild hypertensive patients on maternal, fetal, and neonatal outcomes. Relative risk (RR) and 95% confidence interval (CI) were calculated using a random-effects model. Data from 12 trials comprising 4461 pregnant women diagnosed with mild to moderate hypertension (2395 in the intervention group and 2066 in the control group) were extracted for quantitative synthesis. Antihypertensive treatment was associated with better outcomes in seven out of the 19 analyzed outcomes: Severe hypertension (RR = 0.53; 95% CI = [0.38;0.75]), preeclampsia (RR = 0.71; 95% CI = [0.54; 0.93]), placental abruption (RR = 0.48; 95% CI = [0.26; 0.87]), changes in electrocardiogram (RR = 0.43; 95% CI = [0.25; 0.72]), renal impairment (RR = 0.42; 95% CI = [0.34; 0.51]), pulmonary edema (RR = 0.46; 95% CI = [0.25; 0.84]), and neonatal mortality (RR = 0.72; 95% CI = [0.57; 0.92]). The primary safety outcome of small for gestational age was not different between the treatment group and the control group (RR = 1.12; 95% CI = [0.80; 1.57]). The results of this meta-analysis are in favor of the beneficial impact of pharmacological treatment of mild hypertension on both maternal and neonatal outcomes and without significant adverse events for the fetus.

Critical appraisal skills training to undergraduate medical students: A Randomized Control Study.

Introduction: Critical appraisal skills are presumed to perform a principal part in evidence-based health education and practice methods. This study aimed to design an educational intervention around critical appraisal in this context, evaluate the efficacy of the teaching methods, and investigate critical appraisal skills training to undergraduate medical students. Methods: In this prospective Randomized Control Trial (RCT), 256 sixth-year undergraduate medical students from Shiraz University of Medical Science, Shiraz, Iran, participated. 124 of the students were not taught critical thinking methods as a control group, and 132 participants were taught critical thinking methods by multi-media and attended an online Critical Appraisal Skills (CAS) training workshop as an intervention group. We used the Critical Appraisal Skills Programme (CASP) questionnaire to evaluate students' knowledge, attitude, confidence, and appraisal skills. This questionnaire was translated into Persian for the first time in Iran by Shokooh Varai, et al. They investigated its validity and reliability, the reliability of the questionnaire being obtained by the Richardson Index (0.75), and the validity being confirmed by some faculty members of Tehran Nursing and Midwifery School. To compare both control and intervention groups, we used an independent t-test and a Chi-Square test at a significance level of 5%, and to analyze the demographic information, we applied some descriptive statistics: frequency, frequency percentage, mean and standard deviation. All of the statistical approaches were analyzed, using SPSS 22. Results: In all dimensions of the critical appraisal skills, medical students who completed the critical appraisal multi-media training and workshop performed better than those who did not (control group), and this difference was statistically significant (p<0.05). Moreover, the Chi-Square test results showed no statistical relationship between the groups regarding the demographic variables (p>0.05). Also, the Cohen's D effect size values in the knowledge and confidence dimensions were greater than 0.5; this meant a large effect. Regarding the attitude and appraises Skill dimensions, the effect size was between 0.2 and 0.5, which meant a medium effect. Conclusions: Teaching critical appraisal skills through multi-media and CAS online workshops to medical students effectively improves the students' knowledge and confidence in appraising articles. This teaching also indicated a medium effect on students' attitudes and behavior. Our findings can justify implementing critical appraisal skills teaching modules in the undergraduate medical education curriculum.

Teaching clinical reasoning to undergraduate medical students by illness script method: a randomized controlled trial.

BACKGROUND: The illness script method employs a theoretical outline (e.g., epidemiology, pathophysiology, signs and symptoms, diagnostic tests, interventions) to clarify how clinicians organized medical knowledge for clinical reasoning in the diagnosis domain. We hypothesized that an educational intervention based on the illness script method would improve medical students' clinical reasoning skills in the diagnosis domain. METHODS: This study is a randomized controlled trial involving 100 fourth-year medical students in Shiraz Medical School, Iran. Fifty students were randomized to the intervention group, who were taught clinical reasoning skills based on the illness script method for three diseases during one clinical scenario. Another 50 students were randomized to the control group, who were taught the clinical presentation based on signs and symptoms of the same three diseases as the intervention group. The outcomes of interest were learner satisfaction with the intervention and posttest scores on both an internally developed knowledge test and a Script Concordance Test (SCT). RESULTS: Of the hundred participating fourth-year medical students, 47 (47%) were male, and 53 (53%) were female. On the knowledge test, there was no difference in pretest scores between the intervention and control group, which suggested a similar baseline knowledge in both groups; however, posttest scores in the intervention group were (15.74 ± 2.47 out of 20) statistically significantly higher than the control group (14.38 ± 2.59 out of 20, P = 0.009). On the SCT, the mean score for the intervention group (6.12 ± 1.95 out of 10) was significantly higher than the control group (4.54 ± 1.56 out of 10; P = 0.0001). Learner satisfaction data indicated that the intervention was well-received by students. CONCLUSION: Teaching with the illness script method was an effective way to improve students' clinical reasoning skills in the diagnosis domain suggested by posttest and SCT scores for specific clinical scenarios. Whether this approach translates to improved generalized clinical reasoning skills in real clinical settings merits further study.