A Logistic Regression-Based Technique for Predicting Type II Diabetes

Childhood obesityChildhood is one of most important risk factors for incidence of cardiovascular disease and diabetes in adulthood. Childhood not only is common in developed countries but increasingly affects developing countries. Because of consequences of childhood -- in adulthood, cardiovascular disease, diabetes, arthritis -- has become one of leading causes of death in world (Rapport sur I'organisation de la recherche ..., 2010).Childhood has taken form of an epidemic since mid-1990s (Rapport sur I'organisation de la recherche..., 2010). One single indicator, Body Mass Index (BMI), usually determines adult obesity. It is calculated as ratio of weight in kilograms to square of height in meters. Individuals whose BMI is 25 to 29.9 are diagnosed as being overweight. The diagnosis of of first degree is made when BMI is 30 to 34.9; of second degree is pronounced when BMI is 35 to 40. Morbid obesity is diagnosed when BMI is over 40. In case of children it is not enough to know only BMI; it is necessary to correlate it with an interval of rate of growth for relevant age. For example, in France about 10% of children (from 3 to 16 years old) are obese, and about 19% are overweight (Sassi, Cecchini, & Devaux, 2012).A report of Institut national de la recherche agronomique (France) (2010) indicated that number of children who were overweight didn't change from 1997 to 2007. According to Ministry of Health in Russia, about 19.6% of boys and 5.5% of girls from 5 to 15 years old are overweight, while in Moscow number of children with a BMI above norm is 30% more than in other regions (Briz, 2008).These statistics suggest a joyless future. Taking into account that the chance that a child suffering from will continue this pathology in adulthood ranges from 20 to 50%, if is diagnosed before adolescence; if diagnosis of is made for a teenager, this probability varies from 50% to 70% (Branca, Nikogosian, & Lobstein, 2007, p. 23). We can suggest that level of diseases (diabetes, cardiovascular disease, arthritis) associated with in adulthood will also increase, and they will be manifested at an increasingly early age (Prevention, depistage ...,2012).Studies that identify factors associated with show that a lack of physical activity and an unbalanced diet are main direct factors. Genetic factors are responsible for 1% to 6% of childhood (Klump, Kaye, & Strober, 2001). The economic and socioprofessional situation of parents is also associated with children's being overweight or obese, as are low income, poor social environment, and low parental education (Institut national de la recherche agronomique, 2010).Videogames in childhood: Do they matter?Videogames are based on approaches to play in human development. The effect of play on general child development is a running theme for some of most prominent educational psychologists. Play creates possibilities for children to (a) have social experiences and simulate alternative emotional states and feelings, which can then bring about resolution outside play context (Erikson, 1977), (b) reproduce real-life dilemmas and complicated situations and then work out perfect resolutions and overcome unpleasant feelings (Piaget, 1962), (c) have a basis for fostering development of social cognition (Piaget, 1962; Vygotsky, 19301934/1978).Today, games, gamifications, and videogames in particular have a primary role as media of human development. Previous decades saw an expanding role for new media in area of psychosocial development in childhood. As a medium for entertainment and education, videogames constitute a multi-billion-dollar industry. They are, for example, present in two of three U. …

Our aim was to evaluate liver transplant outcomes involving donors with high macrosteatosis grafts in the obese modern liver transplant recipient population. A high-steatosis graft was defined as donor graft macrosteatosis ≥30% on biopsy. Recipient obesity was defined as body mass index (BMI) >35 adjusted for ascites. Raw and adjusted recipient liver transplant survival were evaluated and compared between 4 cohorts: (1) high-steatosis graft in high-BMI recipient; (2) low-steatosis graft in high-BMI recipient; (3) high-steatosis graft in normal-BMI recipient; and (4) low-steatosis graft in normal-BMI recipient. After adjustment for multiple factors, recipient high-BMI remained an independent predictor of posttransplant mortality at 30 days (P < 0.0001) and persisted at 1 year (P = 0.009). A high-steatosis graft was the strongest independent predictor of mortality at 30 days (hazard ratio 2.05, 1.66-2.53; P < 0.0001) and that effect was diminished but persistent at 1 year (1.27, 1.10-1.46; P = 0.001). Recipient high-BMI and a high-steatosis graft are both significant independent and equally powerful predictors of mortality after modern liver transplant. High-steatosis grafts transplanted into obese recipients have the highest mortality. The increase in mortality associated with a high-steatosis graft into a normal-BMI recipient is similar in magnitude to a low-steatosis graft placed into a high-BMI recipient.

We sought to identify factors that are associated with hypotension after the induction of spinal anesthesia (SpA) by using an anesthesia information management system. Hypotension was defined as a decrease of mean arterial blood pressure of more than 30% within a 10-min interval, and relevance was defined as a therapeutic intervention with fluids or pressors within 20 min. From January 1, 1997, to August 5, 2000, data sets from 3315 patients receiving SpA were recorded on-line by using the automatic anesthesia record keeping system NarkoData. Hypotension meeting the predefined criteria occurred in 166 (5.4%) patients. Twenty-nine patient-, surgery-, and anesthesia-related variables were studied by using univariate analysis for a possible association with the occurrence of hypotension after SpA. Logistic regression with a forward stepwise algorithm was performed to identify independent variables (P < 0.05). The discriminative power of the logistic regression model was checked with a receiver operating characteristic curve. Calibration was tested with the Hosmer-Lemeshow goodness-of-fit test. The univariate analysis identified the following variables to be associated with hypotension after SpA: age, weight, height, body mass index, amount of plain bupivacaine 0.5% used for SpA, amount of colloid infusion before puncture, chronic alcohol consumption, ASA physical status, history of hypertension, urgency of surgery, surgical department, sensory block height of anesthesia, and frequency of puncture. In the multivariate analysis, independent factors for relevant hypotension after SpA consisted of three patient-related variables ("chronic alcohol consumption," odds ratio [OR] = 3.05; "history of hypertension," OR = 2.21; and the metric variable "body mass index," OR = 1.08) and two anesthesia-related variables ("sensory block height," OR = 2.32; and "urgency of surgery," OR = 2.84). The area of 0.68 (95% confidence interval, 0.63-0.72) below the receiver operating characteristic curve was significantly greater than 0.5 (P < 0.01). The goodness-of-fit test showed a good calibration of the model (H = 4.3, df = 7, P = 0.7; C = 7.3, df = 8, P = 0.51). This study contributes to the identification of patients with a high risk for hypotension after SpA induction, with the risk increasing two- or threefold with each additional risk factor. By using automated data collection, 5 (chronic alcohol consumption, history of hypertension, body mass index, sensory block height, and urgency of surgery) of 29 variables could be detected as having an association with hypotension after spinal anesthesia induction. The knowledge of these risk factors should be useful in increasing vigilance in those patients most at risk for hypotension, in allowing a more timely therapeutic intervention, or even in suggesting the use of alternative methods of spinal anesthesia, such as titrated continuous or small-dose spinal anesthesia.

BackgroundBody mass index (BMI) is strongly associated with the development of type 2 diabetes. However, the association between long-term time in target range (TTR) for BMI and the incidence of new-onset diabetes remains unclear.Methods and resultsThis study utilized a non-diabetic population aged 45 years or older from the China Health and Retirement Longitudinal Study (CHARLS). BMI-TTR was assessed in Waves 1, 2, and 3 over a 5-year period, with the target range defined as 18.5 kg/m² ≤ BMI < 23 kg/m². New-onset diabetes in Waves 2, 3, and 4 over a 6-year follow-up served as the study endpoint. After applying exclusion criteria, 6662 participants (3143 men and 3519 women; mean age 58.93 ± 8.85 years) were enrolled. Participants were categorized into four groups (TTR1–TTR4) based on the number of times BMI was within the target range (0–3 times). The risk of new-onset diabetes decreased progressively with increasing BMI-TTR during follow-up. Compared with the TTR1 group, participants in the TTR4 group exhibited a significantly lower risk of diabetes (adjusted HR: 0.577, 95% CI: 0.463–0.720, P < 0.001), even after adjusting for baseline BMI (adjusted HR: 0.685, 95% CI: 0.537–0.872, P = 0.002). This effect was even more pronounced in female subgroup and in individuals aged under 60 years of age.ConclusionIn adults aged 45 years or older, regardless of baseline BMI, maintaining BMI within the target range over time was associated with a reduced risk of new-onset diabetes, particularly among women and individuals under 60 years of age. These findings highlight the importance of long-term weight management in diabetes prevention.

The paradox of high apolipoprotein A-I levels independently predicting incident type-2 diabetes among Turks

Association Between Body Size and Colorectal Adenoma Recurrence

The aim of this study was to develop a new risk score system to predict 5-year incident diabetes risk among middle-aged and older Chinese population. This prospective study included 17,690 individuals derived from the Dongfeng-Tongji cohort. Participants were recruited in 2008 and were followed until October 2013. Incident diabetes was defined as self-reported clinician diagnosed diabetes, fasting glucose ≥7.0mmol/l, or the use of insulin or oral hypoglycemic agent. A total of 1390 incident diabetic cases were diagnosed during the follow-up period. β-Coefficients were derived from Cox proportional hazard regression model and were used to calculate the risk score. The diabetes risk score includes BMI, fasting glucose, hypertension, hyperlipidemia, current smoking status, and family history of diabetes. The β-coefficients of these variables ranged from 0.139 to 1.914, and the optimal cutoff value was 1.5. The diabetes risk score was calculated by multiplying the β-coefficients of the significant variables by 10 and rounding to the nearest integer. The score ranges from 0 to 36. The area under the receiver operating curve of the score was 0.751. At the optimal cutoff value of 15, the sensitivity and specificity were 65.6 and 72.9%, respectively. Based upon these risk factors, this model had the highest discrimination compared with several commonly used diabetes prediction models. The newly established diabetes risk score with six parameters appears to be a reliable screening tool to predict 5-year risk of incident diabetes in a middle-aged and older Chinese population.

As reported by the Centers for Disease Control and Prevention (CDC), heart disease is the major cause of death for both sexes and members of most racial and ethnic groups. Cardiovascular disease claims one life in the United States every 34 seconds. The expense of medical services, medications, and lost productivity because of deaths from heart disease causes the United States to lose a significant amount of money each year. At least half of all Americans have one or more of the following three major risk factors for heart disease: high blood pressure, high cholesterol, and smoking. Other significant indicators are the presence of diabetes, obesity (high body mass index), a lack of physical activity, and excessive alcohol consumption. In recent times, the healthcare sector has discovered that machine learning is a practical and reliable decision-making tool. The latter tends to assist doctors in making informed decisions about the heart condition of a patient. It has become increasingly important to develop an effective early-stage cardiac disease prognosis using digital patient information to enhance clinical decision-making. This study intends to find the most accurate machine learning classifiers for these prognostic uses. To predict cardiac disease, several supervised machine-learning algorithms were used and compared in terms of accuracy and performance. The following methods were used in our research: Logistic Regression, Random Forest, Decision Tree, and K Nearest Neighbor. The heart disease dataset from Kaggle has been used to validate this research. The latter was created using data from the 2020 United States national health status survey of more than 410,000 people, performed yearly by the centers for disease control and prevention.

Health hazards of obesity have been recognized for centuries, appearing, for example, in writings attributed to Hippocrates. From the later decades of the 20th century through the present, there have been numerous epidemiological studies of the relationship between excess weight and the total, or all-cause, mortality rate,1 a critical cumulative measure of the public health impact of any health condition. Using body mass index (BMI), an indicator of relative weight for height (weight [kg]/height [m]2) and a frequently used surrogate for assessment of excess body fat, these studies have found linear, U-shaped, or J-shaped relationships between total mortality and BMI. That is, in some studies, both the thin and the obese were more likely to die than those in between. There is, however, always a point at which increasing BMI is associated with increasing mortality risk, but the BMI at which this occurs varies across studies and populations.2 Currently,3 overweight in adults is defined as a BMI of 25.0 to <30.0 kg/m2 and obesity as a BMI of ≥30.0 kg/m2 (Table 1). A number of studies have found no significant relationship between BMI in the overweight range and mortality rate4 and have shown the nadir of mortality risk to be in the overweight range. In particular, commentaries in both the lay press5–7 and scientific literature2,8,9 subsequent to recent reports from National Health and Nutrition Examination Surveys (NHANES)10,11 have highlighted the confusion and controversy regarding this issue. Some have interpreted the recent data to mean that overweight is not detrimental to health and is not in itself a public health concern and that drawing attention to the need for weight loss in this range will have negative effects on the health and well-being of the general population.8 Others have argued …

This study aimed to develop a diabetes prediction model. The model performance was compared with logistic regression, and the decision tree Chi-square automatic interaction detection (CHAID) was used to predict diabetes.In total, 3233 patients were included in the analysis. Of these, 589 patients with diabetes and 2644 patients without diabetes were included after analyzing the study sample from the Korean Genome and Epidemiology Study (KoGES)-8 data. In this study, Diabetes Mellitus (DM) diagnosis prediction was compared with logistic regression and prediction through machine learning (ML) using the CHAID decision classification tree. We performed statistical analysis using the CHAID method with International Business Machine (IBM) statistical program SPSS®.We performed logistic regression analysis to predict the classification of diabetes accurately, and the total classification accuracy of the analysis was 81.7%, and the CHAID decision tree classification accuracy was 81.8%. A diabetes diagnosis decision tree was created, which included seven terminal nodes and three depth levels. This analysis showed that a blood pressure problem and hospital visits were the most decisive variables at the time of classification, and two risk levels were created for diabetes diagnosis.The suggested method is a valuable tool for predicting diabetes. Patients who visit the hospital because of blood pressure problems are more likely to develop diabetes than under-treating hyperlipidemia. The diabetes prediction model can help doctors make decisions by detecting the possibility of diabetes early; however, it is impossible to diagnose diabetes using only the model without the doctor’s opinion.

e17570 Background: Ovarian cancer(OC) is the leading cause of death from gynecologic malignancy. Current challenges include lack of diagnostic tools, predictive biomarkers, and identifying appropriate surgical candidates. Machine learning(ML) is an emerging field that can make accurate projections by making inferences on data and may play a crucial role in OC.The objective of the current study was to review the literature on application of ML in OC and report the most commonly used algorithms and their performance in comparison to existing prediction tools and traditional regression models. Methods: This is a systematic review of published literature from January 1985 to March 2021 on the use of ML in OC. An extensive search of electronic library databases was conducted. Four independent reviewers screened the articles initially by title then full text. Quality was assessed using the MINORS criteria. P-values were generated using the Pearson’s Chi-squared(x2) test to compare performance of ML models with traditional statistics. No p-values were reported if only one study was available. Results: Among 4,295 articles screened, 88 studies on ML in OC were included. The mean age of OC patients was 54.7 years(11-90) and the most common stages at diagnosis were:Stage III (39.9%) and IV (34%). Applications of ML were in clinical datasets(33%, n = 29), preoperative diagnostics(30.7%, n = 27), serum biomarkers (21.6%, n = 19), genomics (12.5%, n = 11), and prediction of cytoreductive outcomes (2.3%, n = 2). The most commonly applied algorithms were Support Vector Machine [SVM](28%, n = 33)and Neural Networks[NN] (25.28%). Over the past decades, the number of publications on ML in OC increased three-fold from 20(1994-2010) to 67 (2011–2021). Only 9 (10%) studies compared ML techniques with existing prediction tools, or traditional regression models. Among 29 clinical dataset studies, 4 compared ML with traditional logistic regression(LR). Two studies reported better performance with ML compared to LR but not significant(accuracy: 0.88 vs 0.84, p = 0.15), one study performed comparably(accuracy: 0.1 vs 0.1) while one study performed worse(accuracy: 0.1 vs 0.97). Only one preoperative diagnostic study compared ML techniques with LR. SVM classifiers outperformed LR in classifying ovarian masses as benign or malignant(sensitivity: 0.88 vs. 0.70). One serum biomarker study compared LR with ML algorithms; LR performed better using two biomarkers for predicting OC(accuracy: 0.97 vs. 0.94). Among five studies reporting overall survival outcomes, only one study compared survival ML techniques using NN with LR and showed that NN classifiers outperformed LR in predicting overall survival(AUC: 0.72 vs. 0.62). Conclusions: This is the first systematic review exploring the literature on ML algorithms in OC. Most ML models outperformed traditional models. However, larger datasets would be required to validate findings.

IntroductionThe association between liver disease and diabetes has been described, but less information is available regarding mortality from liver disease among people with diabetes from a population perspective.AimTo compare mortality...

Socio-economic status has been associated with diabetes in cross-sectional studies. This study aimed to evaluate associations of household income and wealth with both prevalent and incident diabetes among older adults in the UK. It also evaluated the association between obesity and socio-economic status. A cohort of people aged ≥50 years was selected from the English Longitudinal Study of Ageing. The relation of prevalent and incident self-reported physician diagnosed diabetes to household income and wealth was evaluated in logistic regression models adjusting for education, social class, housing tenure, age, ethnicity, marital status, body mass index (BMI), smoking, alcohol use and physical activity stratified by sex. The relation of prevalent obesity to household income and wealth was also evaluated using logistic regression models. There were 9053 participants (4021 men and 5032 women) including 721 (8.0%) with diabetes at baseline. Among 8332 participants initially free from diabetes, 246 (3.0%) were diagnosed with diabetes during ∼4 years follow-up. The adjusted odds ratio for prevalent diabetes in the lowest quintile of wealth compared with the highest was 1.56 for men and 2.08 for women. Incident diabetes was associated with lower wealth (P for trend 0.05 for men and 0.004 for women) after adjusting for socio-economic and demographic factors, but attenuated after further adjustment for lifestyle and BMI. Prevalent obesity was significantly associated with lower wealth in women but not in men. Lower wealth, but not income, may be associated with prevalent and incident diabetes among older adults in UK.

ObjectivesThe objective of this study is to examine the mediating role of unhealthy behaviors and body mass index (BMI) in the relationship between high job strain and self-rated poor health in workers with a low education.MethodsA total of 8369 low educated workers, who participated in the Lifelines cohort study during the period 2012–2017, were included. Self-reported job strain, health behaviors (smoking, physical activity, and fruit and vegetable consumption), and BMI were assessed at baseline, and self-rated health after 2 years. To assess mediation by the health behaviors and BMI, structural equation modeling with logistic and multinomial regression analyses were performed.ResultsWorkers with high job strain had a higher odds of poor health (OR 1.34; 95% CI 1.13–1.60) compared to those with low job strain. Workers with high job strain were more likely to have a lack of physical activity (OR 1.14; 95% CI 1.01–1.28), but were not more likely to smoke, to be overweight or obese, or to have a low fruit or vegetable consumption. Workers who smoke, have a lack of physical activity or are overweight or obese are more likely to report poor health (OR 1.37; 95% CI 1.16–1.60, OR 1.25; 95% CI 1.08–1.43, OR 1.37; 95% CI 1.16–1.61, OR 2.25; 95% CI 1.86–2.72). Indirect (mediating) effects of unhealthy behaviors and BMI in the relationship between high job strain and poor health were small and not statistically significant.ConclusionsNo mediating effects of unhealthy behaviors or BMI were found in the relationship between high job strain and self-rated poor health among workers with a low educational level.

BackgroundAdvances in automated data processing and machine learning (ML) models, together with the unprecedented growth in the number of social media users who publicly share and discuss health-related information, have made public health surveillance (PHS) one of the long-lasting social media applications. However, the existing PHS systems feeding on social media data have not been widely deployed in national surveillance systems, which appears to stem from the lack of practitioners and the public’s trust in social media data. More robust and reliable data sets over which supervised ML models can be trained and tested reliably is a significant step toward overcoming this hurdle. The health implications of daily behaviors (physical activity, sedentary behavior, and sleep [PASS]), as an evergreen topic in PHS, are widely studied through traditional data sources such as surveillance surveys and administrative databases, which are often several months out-of-date by the time they are used, costly to collect, and thus limited in quantity and coverage.ObjectiveThe main objective of this study is to present a large-scale, multicountry, longitudinal, and fully labeled data set to enable and support digital PASS surveillance research in PHS. To support high-quality surveillance research using our data set, we have conducted further analysis on the data set to supplement it with additional PHS-related metadata.MethodsWe collected the data of this study from Twitter using the Twitter livestream application programming interface between November 28, 2018, and June 19, 2020. To obtain PASS-related tweets for manual annotation, we iteratively used regular expressions, unsupervised natural language processing, domain-specific ontologies, and linguistic analysis. We used Amazon Mechanical Turk to label the collected data to self-reported PASS categories and implemented a quality control pipeline to monitor and manage the validity of crowd-generated labels. Moreover, we used ML, latent semantic analysis, linguistic analysis, and label inference analysis to validate the different components of the data set.ResultsLPHEADA (Labelled Digital Public Health Dataset) contains 366,405 crowd-generated labels (3 labels per tweet) for 122,135 PASS-related tweets that originated in Australia, Canada, the United Kingdom, or the United States, labeled by 708 unique annotators on Amazon Mechanical Turk. In addition to crowd-generated labels, LPHEADA provides details about the three critical components of any PHS system: place, time, and demographics (ie, gender and age range) associated with each tweet.ConclusionsPublicly available data sets for digital PASS surveillance are usually isolated and only provide labels for small subsets of the data. We believe that the novelty and comprehensiveness of the data set provided in this study will help develop, evaluate, and deploy digital PASS surveillance systems. LPHEADA will be an invaluable resource for both public health researchers and practitioners.