Bayesian joint modeling of multivariate longitudinal and survival outcomes using Gaussian copulas.

There is an increasing interest in the use of joint models for the analysis of longitudinal and survival data. While random effects models have been extensively studied, these models can be hard to implement and the fixed effect regression parameters must be interpreted conditional on the random effects. Copulas provide a useful alternative framework for joint modeling. One advantage of using copulas is that practitioners can directly specify marginal models for the outcomes of interest. We develop a joint model using a Gaussian copula to characterize the association between multivariate longitudinal and survival outcomes. Rather than using an unstructured correlation matrix in the copula model to characterize dependence structure as is common, we propose a novel decomposition that allows practitioners to impose structure (e.g., auto-regressive) which provides efficiency gains in small to moderate sample sizes and reduces computational complexity. We develop a Markov chain Monte Carlo model fitting procedure for estimation. We illustrate the method's value using a simulation study and present a real data analysis of longitudinal quality of life and disease-free survival data from an International Breast Cancer Study Group trial.

Novel non-linear models for clinical trial analysis with longitudinal data: A tutorial using SAS for both frequentist and Bayesian methods.

Longitudinal data from clinical trials are commonly analyzed using mixed models for repeated measures (MMRM) when the time variable is categorical or linear mixed-effects models (ie, random effects model) when the time variable is continuous. In these models, statistical inference is typically based on the absolute difference in the adjusted mean change (for categorical time) or the rate of change (for continuous time). Previously, we proposed a novel approach: modeling the percentage reduction in disease progression associated with the treatment relative to the placebo decline using proportional models. This concept of proportionality provides an innovative and flexible method for simultaneously modeling different cohorts, multivariate endpoints, and jointly modeling continuous and survival endpoints. Through simulated data, we demonstrate the implementation of these models using SAS procedures in both frequentist and Bayesian approaches. Additionally, we introduce a novel method for implementing MMRM models (ie, analysis of response profile) using the nlmixed procedure.

Factors associated with different types of freight crashes: A macro-level analysis.

INTRODUCTION: Despite evidence showing higher fatality rates in freight-related crashes, there has been limited exploration of their spatial distribution and factors associated with such distribution. This gap in the literature primarily stems from the focus of existing studies on micro-level factors predicting the frequency or severity of injuries in freight crashes. The present study delves into the factors contributing to freight crashes at the neighborhood level, particularly focusing on different types of freight crashes: collisions involving a freight vehicle and a passenger vehicle, crashes between freight vehicles, and freight vehicle-non-motorized crashes. METHOD: This study analyzes traffic crash data from the urbanized region of Seoul, collected between 2016 and 2019. To effectively deal with spatial autocorrelation and model different types of crashes in a unified framework, a Bayesian multivariate conditional autoregressive model was employed. RESULTS: Findings show substantial differences in the factors associated with various types of freight crashes. The predictors for crashes between freight vehicles diverge significantly from those for freight vehicle-non-motorized crashes. Crashes between freight vehicles are relatively more influenced by road network structure, while freight crashes involving non-motorized users are relatively more affected by the built environment and freight facilities than the other crash types examined. Freight vehicle-passenger vehicle crashes fall into an intermediate category, sharing most predictors with either of the other two types of freight crashes. CONCLUSIONS AND PRACTICAL APPLICATIONS: The findings of this study offer valuable lessons for transportation practitioners and policymakers. They can guide the formulation of effective land use policies and infrastructure planning, specifically designed to address the unique characteristics of different types of freight crashes.

Correlation between metabolic syndrome and hyperuricemia in perimenopausal women / 中华内分泌代谢杂志

Objective:To assess the prevalence of metabolic syndrome(MS) and its relationship with hyperuricemia(HUA) in perimenopausal women in Anning city, Yunnan province.Methods:This is a cross-sectional survey. In May 2021, a multi-stage stratified sampling method was used to collect demographics and clinical data [ethnicity, living community, height, weight, waist circumference, blood pressure, fasting plasma glucose, triglycerides(TG), serum uric acid, high density lipoprotein-cholesterol(HDL-C), alanine transaminase(ALT), etc] in a total of 6 721 perimenopausal women aged 45-60 years.Results:A total of 6 721 perimenopausal women were included in this study. The prevalences of MS and HUA were 14.05%(95% CI 13.22%-14.88%) and 6.46%(95% CI 5.88%-7.07%), respectively. The average age, HDL-C, urea, direct bilirubin, and albumin levels in the perimenstrual HUA population were lower than those in the non-HUA population while the levels of TG, ALT, heart rate, body mass index(BMI), and creatinine were higher(all P<0.05). The prevalence of HUA in perimenopausal women with ethnic minorities and family history of chronic diseases was higher than that in Han nationality and without family history of chronic diseases. The prevalence of MS in perimenopausal women was increased with the increase of serum uric acid( Z=-15.313 8, P<0.001). Multivariate logistic regression model showed that HUA was positively correlated with MS( OR=1.526, 95% CI 1.192-1.954) after adjusting for covariates such as BMI and ethnicity, and the incidence of MS in perimenopausal women in HUA group was 1.526 folds higher than that in non-hyperuricemia group. Conclusion:HUA is highly positively correlated with MS in perimenopausal women. The management of uric acid level in perimenopausal women should be strengthened.

Evaluation of bedside tests and proposal of a model for predicting difficult laryngoscopy: an observational prospective study / Avaliação de testes à beira leito e proposta de modelo para prever laringoscopia difícil: estudo prospectivo observacional

Abstract Background and objectives: The prediction of difficult laryngoscopy is based on tests that assess anatomic characteristics of face and neck. We aimed to identify the most accurate tests and propose a multivariate predictive model. Methods: This prospective observational study included 1134 patients. Thyromental Distance (TMD), Sternomental Distance (STMD), Ratio of Height-to-Thyromental Distance (R-H/TMD), Neck Circumference (NC), Ratio of Neck Circumference-to-Thyromental Distance (R-NC/TMD), Hyomental Distance with head in Neutral Position (HMD-NP) and at Maximal Extension (HMD-HE), Ratio of Hyomental Distance at Maximal head extension-to-hyomental distance in neutral position (R-HMD), Mallampati Class (MLC), Upper Lip Bite Test (ULBT), Mouth Opening (MO) and Head Extension (HE) were assessed preoperatively. A Cormack-Lehane Grade ≥ 3 was defined as Difficult Laryngoscopy. Sensitivity, specificity, positive and negative predictive values were assessed for all tests. Multivariate analysis with logistic regression was used to create the predictive models. Results: A model incorporating MLC, ULBT, HE, HMD-HE and R-NC/TMD showed high prognostic accuracy; x2(5) = 109.12, p < 0.001, AUC = 0.86, p < 0.001). Its sensitivity, specificity and negative predictive value were 82.3%, 74.8% and 97.4%, respectively. A second model including two measurements not requiring patient's cooperation (R-NC/TMD and HMD-HE) exhibited good prognostic performance; x2(2) = 63.5, p < 0.001, AUC = 0.77, p < 0.001. Among single tests, HE had the highest sensitivity (78.5%) and negative predictive value (96%). Conclusions: A five-variable model incorporating MLC, ULBT, HE, HMD-HE and R-NC/TMD showed satisfyingly high predictive value for difficult laryngoscopy. A model including R-NC/TMD and HMD-HE could be useful in incapable patients. The most accurate single predictor was HE.

Resumo Justificativa e objetivos: A previsão de laringoscopia difícil se baseia em testes que avaliam as características anatômicas da face e pescoço. Nosso objetivo foi identificar os testes mais precisos e propor modelo preditivo multivariado. Método: Estudo observacional prospectivo incluiu 1134 pacientes e avaliou no pré-operatório: Distância Tireomentoniana (DTM), Distância Esternomentoniana (DEM), razão Altura-Distância Tireomentoniana (A/DTM), Circunferência Cervical (CC), razão Circunferência Cervical-Distância Tireomentoniana (CC/DTM), Distância Hiomentoniana com a cabeça na Posição Neutra (DHM-PN) e em Extensão Máxima (DHM-EM), razão Distância Hiomentoniana com Cabeça em Extensão Máxima/Distância Hiomentoniana na posição Neutra (DHME/DHMN), Classe Mallampati (CML), Teste da Mordida do Lábio Superior (TMLS), Abertura da Boca (AB) e Extensão da Cabeça (EC). Grau Cormack-Lehane ≥ 3 foi definido como Laringoscopia Difícil. A sensibilidade, especificidade, valores preditivos positivo e negativo foram avaliados para todos os testes. A análise multivariada com regressão logística foi usada para criar modelos preditivos. Resultados: Um modelo incorporando CML, TMLS, EC, DHM-EM e CC/DTM demonstrou alta precisão prognóstica (x2(5) = 109,12, p < 0,001, AUC = 0,86, p < 0,001). A sensibilidade, especificidade e valor preditivo negativo foram 82,3%, 74,8% e 97,4%, respectivamente. Um segundo modelo incluindo duas medidas que não necessitavam da cooperação do paciente (CC/DTM e DHM-EM) demonstrou bom desempenho prognóstico (x2 (2) = 63,5; p < 0,001; AUC = 0,77, p < 0,001). Entre os testes individuais, EC teve a maior sensibilidade (78,5%) e valor preditivo negativo (96%). Conclusões: O modelo de cinco variáveis incorporando CML, TMLS, EC, DHM-EM e CC/DTM mostrou valor preditivo satisfatoriamente alto para laringoscopia difícil. Um modelo que incluísse CC/DTM e DHM-EM poderia ser útil em pacientes com incapacidade. O preditor individual mais preciso foi EC.