Search

Racial and ethnic minority subpopulations experience a disproportionate burden of asthma and adverse childhood experiences (ACEs). These disparities result from systematic differences in risk exposure, opportunity access, and return on resources, but we know little about how accumulated differentials in ACEs may be associated with adult asthma by racial/ethnic groups. We used Behavioral Risk Factor Surveillance System data (N = 114,015) from 2009 through 2012 and logistic regression to examine the relationship between ACEs and adult asthma using an intersectional lens, investigating potential differences for women and men aged 18 and older across seven racial/ethnic groups. ACEs were significantly related to asthma, adjusting for race/ethnicity and other covariates. Compared to the reference group (Asians), asthma risk was significantly greater for Black/African American, American Indian and Alaska Native (AIAN), White, and multiracial respondents. In sex-stratified interactional models, ACEs were significantly related to asthma among women. The relationship between ACEs and asthma was significantly weaker for Black/African American and AIAN women compared to the reference group (Asian women). The findings merit attention for the prevention and early detection of ACEs to mitigate long-term health disparities, supporting standardized screening and referrals in clinical settings, evidence-based prevention in communities, and the exploration of strategies to buffer the influence of adversities in health.

Household food insecurity (HFI) is a significant problem in the developing world. Relationships between HFI and nutrition, physical growth, and development have been elucidated; less is known about the non-nutritional impacts among individuals living in rural areas in low-income countries. The aim of this study was to determine if HFI is a risk factor for suboptimal mental health and overall health in rural Honduras. In a population of 24,696 adults with 176 isolated villages in western Honduras, we collected data on household food insecurity and physical and mental health outcome measures. Using logistic regression with and without adjusting for village and household level unobservables invariant across individual respondents, we show that females (OR: 1.11, p, Highlights • Article identifies the potential causes and consequences of household food insecurity in a high migration area in rural Honduras. • Study addresses food insecurity in the context of gender issues and social connectedness in a marginalized rural community in Honduras. • More individuals reported stress and anxiety about being food insecure than to actually running out of food. • Household food insecurity is a significant risk factor for both mental and overall health in a vulnerable population in the 'Dry Corrider' area in Honduras.

Ageing people with hemophilia (PWH) have a higher prevalence of hypertension than the general population. This study aimed to determine whether macroscopic hematuria was associated with hypertension in PWH in a post hoc analysis using data from a cross-sectional study conducted by the ADVANCE Working Group (the H3 study), which included PWH ≥ 40 years of age. Data from 16 contributing centers, located in 13 European countries and Israel, were analyzed using logistic regression models. Of 532 recruited PWH in the H3 study, 117 had hypertension and a positive family history of hypertension (hypertension FH+), 75 had hypertension and a negative family history of hypertension (hypertension FH−), 290 had no diagnosis of hypertension, and the remaining 50 had missing hypertension data. Logistic regressions showed that macroscopic hematuria was associated with hypertension FH+, both in the univariate (OR = 1.84 [1.17–2.90], P = .01) and in the multivariate model (OR = 1.80 [1.03–3.16], P = .04). Macroscopic hematuria was not associated with hypertension FH−. Moreover, in a multivariate logistic regression the odds of hypertension FH+ were increased with the number of macroscopic hematuria episodes. The association between macroscopic hematuria and hypertension was significant for PWH with a family history of hypertension.

In proof-of-payment transit systems, fare evasion has recently captured increasing attention because of the relevant implications it produces. Research has investigated how sociodemographic, travel behaviour, and situational determinants affect the intention to evade fares for segments of passengers clustered according to ‘Gender’ and ‘Age’. Conversely, no study has isolated these determinants in segments clustered according to ‘Employment’. This paper fills this gap by analyzing students, workers, and unemployed passengers. Key determinants are isolated by logistic regression models. The findings show that gender, age, and having been fined are the common determinants that make all these segments more likely to evade fares. In addition, some specific determinants are identified for each segment. Hence, the overall findings may support transit operators by anticipating preventive and corrective strategies tailored to specific segments, which can positively impact other segments.

The purpose of this study was to examine physiological and physical determinants of ice-hockey performance in order to assess their impact on the result during a selection for ice hockey. A total of 42 ice hockey players took part in the selection camp. At the end of the camp 20 best players were selected by team of expert coaches to the ice hockey team and created group G1, while the second group (G2) consisted of not selected players (non-successful group Evaluation of goodness of fit of the model to the data was based on the Hosmer Lemeshow test Ice hockey players selected to the team were taller 181.95±4.02 cm, had lower % body fat 13.17±3.17%, a shorter time to peak power 2.47±0.35 s , higher relative peak power 21.34±2.41 W • kg-1 and higher relative total work 305.18±28.41 J • kg-1. The results of the aerobic capacity test showed significant differences only in case of two variables. Ice hockey players in the G1 had higher VO2max 4.07±0.31 l • min-1 values than players in the G2 as well as ice hockey players in G1 showed a higher level of relative VO2max 51.75±2.99 ml • min-1 • kg-1 than athletes in G2. Ice hockey players selected to the team (G1) performed better in the 30 m Forwards Sprint 4.28±0.31 s; 6x9 Turns 12.19±0.75 s; 6x9 stops 12.79±0.49 s and Endurance test (6x30 m stops) 32.01±0.80 s than players in G2. The logistic regression model showed that the best predictors of success in the recruitment process of top level ice hockey players were time to peak power, relative peak power, VO2max and 30 m sprint forwards on ice. On the basis of the constructed predictive logistic regression model it will be possible to determine the probability of success of the athletes during following the selection processes to the team.

Objective To externally validate the diagnostic and prognostic value of three previously identified microRNAs in emergency department patients with sepsis. Methods Patients meeting consensus criteria for sepsis and septic shock were compared to controls. Three microRNAs (miR-150, miR-146a, and miR-223) were measured using real-time quantitative PCR, and levels of miRNAs were compared among the three cohorts. The association between miRNAs and both inflammatory markers and Sequential Organ Failure Assessment (SOFA) score were compared. To assess the prognostic value of each miRNA, unadjusted and adjusted logistic regression models were constructed using in-hospital mortality as the dependent variable. Results Ninety-three patients were enrolled; 24 controls, 29 with sepsis, and 40 with septic shock. We found no difference in serum plasma miR-146a or miR-223 between cohorts, and found no association among these microRNAs and either inflammatory markers or SOFA score. miR-150 demonstrated a significant correlation with SOFA score ( P = 0.31, P =0.01) and IL-10 ( P =0.37, P =0.001), but no IL-6 or TNF-α ( P =0.046, P =0.59). Logistic regression demonstrated miR-150 to be independently associated with mortality, even after adjusting for SOFA score ( P =0.003) or initial lactate ( P =0.01). Conclusions miR-146a and miR-223 demonstrated no significantly diagnostic or prognostic ability in this cohort. miR-150 was associated with inflammation, severity of illness, and mortality. Given the independent predictive value of miR-150, additional research regarding its role in sepsis is warranted.

TEZ11114 Tez (Yüksek Lisans) -- Çukurova Üniversitesi, Adana, 2018. Kaynakça (s. 125-129) var. xvii, 130 s. : tablo ; 29 cm. Sağlık alanında sınıflama modelleri hastalıkların seyrinin tahmin edilmesinde ve tanı koymada sık kullanılan modellerdir. Bu modeller arasında en sık kullanılan ve en iyi bilineni Lojistik regresyon(LR) dur ancak son yıllarda kullanımı artan sınıflama modellerinin performansları konusunda henüz yeterli bilgiye sahip değiliz. Bu tezde kullanılan Karar Ağacı(KA), Random Forest(RF), Destek Vektör Makineleri(DVM) ve Naive Bayes(NB) sınıflama modelleri yanıt değişkenin iki değerli ve açıklayıcı değişkenlerin kategorik ve/veya sürekli olabildiği modellerden seçilmiştir. Bu tezde, LR yöntemi ile diğer yöntemlerin farklı örnek büyüklüğü, prevelans, açıklayıcı değişken tipi ve tanımlayıcılık katsayısı durumunda, etkileşim terimlerini bulunduran ve etkileşim terimlerini bulundurmayan 2 farklı model ile veri setleri üretilerek sınıflama modellerinin performanslarının karşılaştırılması amaçlanmıştır. Ayrıca literatürden alınan 12 gerçek veri seti için performansları karşılaştırılmıştır. Genel olarak etkileşim terimlerini bulundurmayan modelde NB yönteminin performansı diğer yöntemlerden daha yüksek ve LR yöntemi ile benzer sonuçlar verdiği gözlenmiştir. Etkileşim terimlerini bulunduran modelde, düşük örnek büyüklüğünde NB yönteminin diğer yöntemlerden daha iyi performans göstermiştir. Orta ve büyük veri setlerinde DVM ve RF yöntemlerinin daha iyi performans göstermektedir. Bunula birlikte KA ve DVM yöntemleri düşük prevalans, düşük tanımlayıcılık katsayısı ve küçük örnek büyüklüğünde sınıflama yapamadığı durumların çok fazla olduğu(%50) gözlenmiştir. Gerçek veri setleri analizlerde DVM ve RF yöntemleri daha iyi performans gösterdiği gözlenmiştir. Classification models are frequently used in health for prediction purposes and diagnosis. Logistic regression (LR) is the most common used and best-known among these models. In the past decades many new classification models were proposed. The performance of these relative new methods has not been studied well enough. The Decision Tree (DT), Random Forest (RF), Support Vector Machines (SVM), and Naive Bayes (NB) classification models used in this thesis are selected from the models in which the bivariate and explanatory variables of the response variable are categorical and / or continuous. In this thesis we aimed to compare the performance of classification models by simulating data sets using 2 different models (with and without interaction terms) where sample size, prevalence, and coefficient of determination combinations changed. In addition, their performances were compared for 12 real data sets from the literature. In simulations without interaction terms, the performance of the NB method was higher than the other methods and comparable with the LR method. In simulations with interaction terms, the NB method performed better than the other methods at low sample size but SVM and RF methods performed better in medium and large data sets. We observed that DT and SVM methods were not able to make classifications (50%) in simulation settings with low prevalence and low coefficient of determination and small sample size. Real data set analysis showed that SVM and RF methods perform better than LR, DT and NB in some real data sets. Bu çalışma Ç.Ü. Bilimsel Araştırma Projeleri Birimi tarafından desteklenmiştir. Proje No: BAP TYL2018-10004.

Lojistik regresyon modelinde en yaygın kullanılan parametre tahmin yöntemi en çok olabilirliktir. Çoklu iç ilişkinin var olması durumunda en çok olabilirlik tahmin edicinin boyu ve asimptotik varyansı aşırı büyüktür. Bunun sonucu olarak, istatistiksel çıkarımlar hatalı olur. Çoklu iç ilişkinin etkilerini azaltmak için literatürde en çok olabilirlik tahmin ediciye alternatif bazı parametre tahmin yöntemleri önerilmiştir. Bu çalışmada ise, lojistik regresyonda çoklu iç ilişki olması durumunda birinci-sıra jackknife ridge lojistik tahmin edici, birinci-sıra r-k ve r-d sınıf lojistik tahmin ediciler önerilmiş ve bu tahmin edicilerin özellikleri ve performansları incelenmiştir. Teorik olarak elde edilen sonuçlar, hem sayısal örneklerle hem de simülasyon çalışmaları üzerinde gösterilmiştir.Anahtar Kelimeler: Lojistik regresyon, Çoklu iç ilişki, Birinci-sıra jackknife ridge lojistik tahmin edici, Birinci-sıra r-k sınıf lojistik tahmin edici, Birinci-sıra r-d sınıf lojistik tahmin edici In logistic regression model, the most popular parameter estimation method is maximum likelihood. In the existence of multicollinearity, the length and the asymptotic covariance of the maximum likelihood estimator are excessively large. As a result, statistical inferences become imprecise. In order to reduce the effects of multicollinearity, some alternative parameter estimation methods to maximum likelihood were proposed in the literature. In this study, first-order jackknifed ridge logistic estimator, first-order r-k and r-d class logistic estimators are introduced. The properties and the performances of these estimators are examined. The theoretical results are illustrated via numerical examples and simulation studies.Key Words: Logistic regression, Multicollinearity, First-order jackknifed ridge logistic estimator, First-order r-k class logistic estimator, First-order r-d class logistic estimator 112

TEZ8944 Tez (Yüksek Lisans) -- Çukurova Üniversitesi, Adana, 2013. Kaynakça (s. 67-70) var. x, 72 s. : res. ; 29 cm. Today the effect of the fact of Supply Chain Management on the performance and the life cycles of enterprises is great. For an enterprise keeping on its existence in the sector of manufacturing industry, one of the most important rings of supply chain is the data of order and the manufacturing performance of the enterprise. This work aims to prove that the order parameters are usable for estimating the probability of order realization at deadline. In this manner, it will be provided for the manufacturing industry enterprises and the enterprise decision makers, to be more accurate on making decisions about orders and reaching the results of those by this model. In addition to this, applicability of this approach for many rings of supply chain will be discussed. Günümüzde Tedarik Zinciri Yönetimi olgusunun işletmelerin yaşam süreçleri ve performansları üzerindeki ektisi büyüktür. İmalat sanayi sektöründe varlığını sürdüren bir işletmede ise, tedarik zinciri yönetiminin en önemli halkalarından birisi sipariş bilgileri ve işletmenin üretim performansıdır. Bu çalışma, sipariş parametrelerinin, siparişin zamanında gerçekleşme ihtimalini tahmin etmede kullanılabilir olduğunu kanıtlamayı amaçlamaktadır. Böylelikle imalat sanayi işletmelerinin ve işletme karar alıcılarının bu modelleme ile siparişler üzerine karar almada ve kararlarının sonucuna ulaşabilmede, daha isabetli olmaları sağlanabilinecektir. Bununla birlikte, tedarik zinciri yönetiminin birçok halkasında da bu yaklaşımın uygulanabilirliği tartışılacaktır.

Clinicians often face difficult decisions despite the lack of evidence from randomized trials. Thus, clinical evidence is often shaped by non-randomized studies exploiting multivariable approaches to limit the extent of confounding. Since their introduction, propensity scores have been used more and more frequently to estimate relevant clinical effects adjusting for established confounders, especially in small datasets. However, debate persists on their real usefulness in comparison to standard multivariable approaches such as logistic regression and Cox proportional hazard analysis. This holds even truer in light of key quantitative developments such as bootstrap and Bayesian methods. This qualitative review aims to provide a concise and practical guide to choose between propensity scores and standard multivariable analysis, emphasizing strengths and weaknesses of both approaches.

Complex diseases may be caused by interactions or combined effects between multiple genetic and environmental factors. One of the main limitations of testing for interaction between genetic loci in large whole genome studies is the high computational cost of performing such analyses. In this study a new methodology for interaction testing (commonly referred to in genetics as the epistatic effect) between two single nucleotide polymorphisms (SNPs) and a categorical phenotype is presented. It is shown that it provides reasonable approximations with a significantly shorter run time. The proposed measure based on the Pearson's chi-square additive property is compared to fitting a logistic regression model on a randomly selected subset of 218 SNP loci from a study that included 550,000 SNPs). For each possible pair of SNPs a chi-square test for the epistatic effect on case-control status is estimated by fitting a logistic regression model, and compared to the results of the proposed method. Results indicate strong agreement (Pearson's correlation r>0.95) while the proposed method is found to be 20 times faster. This provides a significant pragmatic advantage for the proposed method since the number of tests for epistasis can now be increased by a factor of 20 while the computational cost remains the same. © 2010 IEEE. 6194 6197 Conference code: 83008 Cited By :2

Logistic regression models were developed to identify significant factors that influence the location of existing wood-using bioenergy/biofuels plants and traditional wood-using facilities. Logistic models provided quantitative insight for variables influencing the location of woody biomass-using facilities. Availability of “thinnings to a basal area of 31.7m2/ha,” “availability of unused mill residues,” and “high density of railroad availability” had positive significant influences on the location of all wood-using faciities. “Median family income,” “population,” “low density of railroad availability,” and “harvesting costs for logging residues” had negative significant influences on the location of all wood-using faciities. For larger woody biomass-using mills (e.g., biopower) availability of “thinnings to a basal area of 79.2m2/ha,” “number of primary and secondary wood-using mills within an 128.8km haul distance,” and “amount of total mill residues,” had positive significant influences on the location of larger wood-using faciities. “Population” and “harvesting costs for logging residues” have negative significant influences on the location of larger wood-using faciities. Based on the logistic models, 25 locations were predicted for bioenergy or biofuels plants for a 13-state study region in the Southern United States.

Background To understand the influence of the number of years spent at an Australian rural clinical school (RCS) on graduate current, preferred current and intended location for rural workforce practice. Methods Retrospective online survey of medical graduates who spent 1–3 years of their undergraduate training in the University of New South Wales (UNSW) Rural Clinical School. Associations with factors (gender, rural versus non-rural entry, conscription versus non-conscript and number of years of RCS attendance) influencing current, preferred current and intended locations were assessed using χ2 test. Factors that were considered significant at P Results 214 graduates responded to the online survey. Graduates with three years of previous RCS training were more likely to indicate rural areas as their preferred current work location, than their colleagues who spent one year at an RCS campus (OR = 3.0, 95% CI = 1.2-7.4, P = 0.015). Also RCS graduates that spent three years at an RCS were more likely to intend to take up rural medical practice after completion of training compared to the graduates with one year of rural placement (OR = 5.1, 95% CI = 1.8-14.2, P = 0.002). Non-rural medicine entry graduates who spent three years at rural campuses were more likely to take up rural practice compared to those who spent just one year at a rural campus (OR = 8.4, 95% CI = 2.1-33.5, P = 0.002). Conclusions Increasing the length of time beyond a year at an Australian RCS campus for undergraduate medical students is associated with current work location, preferred current work location and intended work location in a rural area. Spending three years in a RCS significantly increases the likelihood of rural career intentions of non-rural students.