Sex- and time-specific associations of obesity with glycaemic traits: A two-step multivariate Mendelian randomization study.

AIM: To assess the sex- and time-specific causal effects of obesity-related anthropometric traits on glycaemic traits. MATERIALS AND METHODS: We used univariate and multivariate Mendelian randomization to assess the causal associations of anthropometric traits (gestational variables, birth weight, childhood body mass index [BMI], BMI, waist-to-hip ratio [WHR], BMI-adjusted WHR [WHRadj BMI]) with fasting glucose and insulin in Europeans from the Early Growth Genetics Consortium (n ≤ 298 142), the UK Biobank, the Genetic Investigation of Anthropometric Traits Consortium (n ≤ 697 734; females: n ≤ 434 794; males: n ≤ 374 754) and the Meta-Analyses of Glucose and Insulin-related traits Consortium (n ≤ 151 188; females: n ≤ 73 089; males: n ≤ 67 506), adjusting for maternal genetic effects, smoking, alcohol consumption, and age at menarche. RESULTS: We observed a null association for gestational variables, a negative association for birth weight, and positive associations for childhood BMI and adult traits (BMI, WHR, and WHRadj BMI). In female participants, increased birth weight causally decreased fasting insulin (betaIVW , -0.07, 95% confidence interval [CI] -0.11 to -0.03; p = 1.92 × 10-3 ), but not glucose levels, which was annulled by adjusting for age at menarche. In male participants, increased birth weight causally decreased fasting glucose (betainverse-variance-weighted (IVW) , -0.07, 95% CI -0.11 to -0.03; p = 3.22 × 10-4 ), but not insulin levels. In time-specific analyses, independent effects of birth weight were absent in female participants, and were more pronounced in male participants. Independent effects of childhood BMI were attenuated in both sexes; independent effects of adult traits differed by sex. CONCLUSIONS: Our findings provide evidence for causal and independent effects of sex- and time-specific anthropometric traits on glycaemic variables, and highlight the importance of considering multiple obesity exposures at different time points in the life course.

A cross-sectional study on turnover intention of nurses in eastern China.

BACKGROUND: This study aimed to investigate the turnover intention among nurses in eastern China and explore the association between turnover intention and personal characteristics, family factors, and work-related factors. METHODS: A total of 2504 nurses participated in a cross-sectional survey administered in 26 hospitals in Eastern China from October to November 2017. In December 2021, a survey was conducted on nurses who resigned between December 2017 and November 2021. RESULTS: The turnover intention score of in-service nurses was 15 (12-17), and 43% of nurses had a high turnover intention, which was mainly due to the following reasons: age < 40 years, raising two or more children, monthly income of USD786.10-1572.20 or < USD786.10, occupation was assigned or selected according to parental wishes, ≤ 1 or ≥ 2-night shifts per week, contractual or third-party personnel agents, full-time nurses with part-time jobs, and high job stress. Among 102 retired nurses, 80.4% reported family reasons for leaving, 39.2% for work reasons, and 21.6% for other personal reasons. CONCLUSION: Nurses' intention to leave their occupation is high in Eastern China. Age < 40 years old, > 1 child, low income, involuntary career selection, frequent night shifts, informal employment, part-time, and high job stress are significant factors associated with nurses' willingness to leave. Government and hospital administrators should consider ways to address these factors to retain nurses in hospitals in eastern China and improve the quality of nursing services.

An improved score-type confidence interval for stratified risk differences involving rare events.

A stratified analysis of the differences in proportions has been widely employed in epidemiological research, social sciences, and drug development. It provides a useful framework for combining data across strata to produce a common effect. However, for rare events with incidence rates close to zero, popular confidence intervals for risk differences in a stratified analysis may not have appropriate coverage probabilities that approach the nominal confidence levels and the algorithms may fail to produce a valid confidence interval because of zero events in both the arms of a stratum. The main objective of this study is to evaluate the performance of certain methods commonly employed to construct confidence intervals for stratified risk differences when the response probabilities are close to a boundary value of zero or one. Additionally, we propose an improved stratified Miettinen-Nurminen confidence interval that exhibits a superior performance over standard methods while avoiding computational difficulties involving rare events. The proposed method can also be employed when the response probabilities are close to one.

Stratified modestly weighted log-rank tests in settings with an anticipated delayed separation of survival curves.

Delayed separation of survival curves is a common occurrence in confirmatory studies in immuno-oncology. Many novel statistical methods that aim to efficiently capture potential long-term survival improvements have been proposed in recent years. However, the vast majority do not consider stratification, which is a major limitation considering that most large confirmatory studies currently employ a stratified primary analysis. In this article, we combine recently proposed weighted log-rank tests that have been designed to work well under a delayed separation of survival curves, with stratification by a baseline variable. The aim is to increase the efficiency of the test when the stratifying variable is highly prognostic for survival. As there are many potential ways to combine the two techniques, we compare several possibilities in an extensive simulation study. We also apply the techniques retrospectively to two recent randomized clinical trials.

Analysis and asymptotic theory for nested case-control designs under highly stratified proportional hazards models.

Nested case-control sampled event time data under a highly stratified proportional hazards model, in which the number of strata increases proportional to sample size, is described and analyzed. The data can be characterized as stratified sampling from the event time risk sets and the analysis approach of Borgan et al. (Ann Stat 23:1749-1778, 1995) is adapted to accommodate both the stratification and case-control sampling from the stratified risk sets. Conditions for the consistency and asymptotic normality of the maximum partial likelihood estimator are provided and the results are used to compare the efficiency of the stratified analysis to an unstratified analysis when the baseline hazards can be semi-parametrically modeled in two special cases. Using the stratified sampling representation of the stratified analysis, methods for absolute risk estimation described by Borgan et al. (1995) for nested case-control data are used to develop methods for absolute risk estimation under the stratified model. The methods are illustrated by a year of birth stratified analysis of radon exposure and lung cancer mortality in a cohort of uranium miners from the Colorado Plateau.

Harmonizing Genetic Ancestry and Self-identified Race/Ethnicity in Genome-wide Association Studies.

Large-scale multi-ethnic cohorts offer unprecedented opportunities to elucidate the genetic factors influencing complex traits related to health and disease among minority populations. At the same time, the genetic diversity in these cohorts presents new challenges for analysis and interpretation. We consider the utility of race and/or ethnicity categories in genome-wide association studies (GWASs) of multi-ethnic cohorts. We demonstrate that race/ethnicity information enhances the ability to understand population-specific genetic architecture. To address the practical issue that self-identified racial/ethnic information may be incomplete, we propose a machine learning algorithm that produces a surrogate variable, termed HARE. We use height as a model trait to demonstrate the utility of HARE and ethnicity-specific GWASs.

Patterns of sexual behaviour associated with repeated chlamydia testing and infection in men and women: a latent class analysis.

BACKGROUND: Adolescents and young adults are at higher risk of acquiring Chlamydia trachomatis infection (chlamydia), so testing is promoted in these populations. Studies have shown that re-testing for chlamydia is common amongst them. We investigated how sexual risk behaviour profiles are associated with repeated testing for chlamydia. METHODS: We used baseline data from a cohort of 2814 individuals recruited at an urban STI -clinic. We applied latent class (LC) analysis using 9 manifest variables on sexual behaviour and substance use self-reported by the study participants. We fitted ordered logistic regression to investigate the association of LC membership with the outcomes repeated testing during the past 12 months and lifetime repeated testing for chlamydia. Models were fit separately for men and women. RESULTS: We identified four LCs for men and three LCs for women with increasing gradient of risky sexual behaviour. The two classes with the highest risk among men were associated with lifetime repeated testing for chlamydia: adjOR = 2.26 (95%CI: 1.50-3.40) and adjOR = 3.03 (95%CI: 1.93-4.74) as compared with the class with lowest risk. In women, the class with the highest risk was associated with increased odds of repeated lifetime testing (adjOR =1.85 (95%CI: 1.24-2.76)) and repeated testing during past 12 months (adjOR = 1.72 (95%CI: 1.16-2.54)). An association with chlamydia positive test at the time of the study and during the participant's lifetime was only found in the male highest risk classes. CONCLUSION: Prevention messages with regard to testing for chlamydia after unprotected sexual contact with new/casual partners seem to reach individuals in highest risk behaviour classes who are more likely to test repeatedly. Further prevention efforts should involve potentially more tailored sex-specific interventions taking into consideration risk behaviour patterns.

Multiple imputation analysis of Miettinen-Nurminen interval for difference in proportions.

The standard multiple imputation technique focuses on parameter estimation and assumes that the parameter estimator is asymptotically normally distributed with a Wald-type confidence interval for the parameter of interest. On the other hand, the Miettinen-Nurminen (MN) method for difference in proportions (Miettinen O, Nurminen M. Stat Med. 1985;4:213-226) constructs the confidence interval using an asymptotic score method and hence is not directly amenable to the standard multiple imputation technique. We propose a multiple imputation analysis that is applicable to the MN method for difference in proportions. We use simulation studies to compare the proposed method with that of Li, Mehrotra and Barnard (LMB), which is based on effective sample sizes (Li X, Mehrotra DV, Barnard J. Stat Med. 2006;25:2107-2124). We show that both methods produce confidence intervals with adequate coverage, while the proposed method produces slightly shorter confidence intervals than the LMB method. In addition, the proposed method is evaluable for all datasets, while the LMB method cannot be used when the imputed cell counts are zero across imputations for a treatment group.

The utilities and pitfalls of stratified analysis in challenging situations.

Stratified analysis is commonly used to control confounding factors in evaluating treatment effect at both design and analysis stages. Motivated by a real-life experience, this article illustrates the utilities and pitfalls of stratified analysis through continuous and binary endpoints by comparing some typical analysis methodologies. Stratified analyses in some challenging situations, including propensity score stratification are discussed.

Long-term declining in carbon monoxide (CO) at a rural site of Beijing during 2006-2018 implies the improved combustion efficiency and effective emission control.

Carbon monoxide (CO) is primarily the result of incomplete combustion, which has important impacts on the atmospheric chemical cycle and climate. Improved quantitative characterization of long-term CO trends is important for both atmospheric modeling and the design and implementation of policies to efficiently control multiple pollutants. Due to the limitations of high time-resolution and high-quality long-term observational data, studies on long-term trends in the CO concentration in China are quite limited. In this study, the observational data of the concentration of CO in a rural site of Beijing during 2006-2018 was used to analyze the long-term trend in CO concentration in Beijing. The Theil-Sen method and the generalized additive model (GAM)-based method, were used to conduct the trend estimation analysis. We found that the concentration of CO at the Shangdianzi site showed a significant downward trend during 2006-2018, with a decline rate of 22.8 ± 5.1 ppbV per year. The declining trend in CO also showed phasic characteristics, with a fast decreasing rate during the period of 2006-2008, stable variations during the period of 2009-2013 and a continuous downward trend after 2013. The declining trend in the CO concentration in the south to west (S-W) sectors where the polluted air masses come from is more rapid than that in the sectors where the clean air masses come from. The declining trend in the CO concentration implies the improved combustion efficiency and the successful air pollution control policies in Beijing and the surrounding area.

A compartment modeling approach to reconstruct and analyze gender and age-grouped CoViD-19 Italian data for decision-making strategies.

BACKGROUND: Available national public data are often too incomplete and noisy to be used directly to interpret the evolution of epidemics over time, which is essential for making timely and appropriate decisions. The use of compartment models can be a worthwhile and attractive approach to address this problem. The present study proposes a model compartmentalized by sex and age groups that allows for more complete information on the evolution of the CoViD-19 pandemic in Italy. MATERIAL AND METHODS: Italian public data on CoViD-19 were pre-treated with a 7-day moving average filter to reduce noise. A time-varying susceptible-infected-recovered-deceased (SIRD) model distributed by age and sex groups was then proposed. Recovered and infected individuals distributed by groups were reconstructed through the SIRD model, which was also used to simulate and identify optimal scenarios of pandemic containment by vaccination. The simulation started from realistic initial conditions based on the SIRD model parameters, estimated from filtered and reconstructed Italian data, at different pandemic times and phases. The following three objective functions, accounting for total infections, total deaths, and total quality-adjusted life years (QALYs) lost, were minimized by optimizing the percentages of vaccinated individuals in five different age groups. RESULTS: The developed SIRD model clearly highlighted those pandemic phases in which younger people, who had more contacts and lower mortality, infected older people, characterized by a significantly higher mortality, especially in males. Optimizing vaccination strategies yielded different results depending on the cost function used. As expected, to reduce total deaths, the suggested strategy was to vaccinate the older age groups, whatever the baseline scenario. In contrast, for QALYs lost and total infections, the optimal vaccine solutions strongly depended on the initial pandemic conditions: during phases of high virus diffusion, the model suggested to vaccinate mainly younger groups with a higher contact rate. CONCLUSION: Because of the poor quality and insufficient availability of stratified public pandemic data, ad hoc information filtering and reconstruction procedures proved essential. The time-varying SIRD model, stratified by age and sex groups, provided insights and additional information on the dynamics of CoViD-19 infection in Italy, also supporting decision making for containment strategies such as vaccination.

Using causal forests to assess heterogeneity in cost-effectiveness analysis.

We develop a method for data-driven estimation and analysis of heterogeneity in cost-effectiveness analyses (CEA) with experimental or observational individual-level data. Our implementation uses causal forests and cross-fitted augmented inverse probability weighted learning to estimate heterogeneity in incremental outcomes, costs and net monetary benefits, as well as other parameters relevant to CEA. We also show how the results can be visualized in relevant ways for the analysis of heterogeneity in CEA, such as using individual-level cost effectiveness planes. Using a simulated dataset and an R package implementing our methods, we show how the approach can be used to estimate the average cost-effectiveness in the entire sample or in subpopulations, explore and analyze the heterogeneity in incremental outcomes, costs and net monetary benefits (and their determinants), and learn policy rules from the data.

Analyses stratified by maternal age and recombination further characterize genes associated with maternal nondisjunction of chromosome 21.

OBJECTIVE: In our previous work, we performed the first genome-wide association study to find genetic risk factors for maternal nondisjunction of chromosome 21. The objective of the current work was to perform stratified analyses of the same dataset to further elucidate potential mechanisms of genetic risk factors. METHODS: We focused on loci that were statistically significantly associated with maternal nondisjunction based on this same dataset in our previous study and performed stratified association analyses in seven subgroups defined by age and meiotic recombination profile. In each analysis, we contrasted a different subgroup of mothers with the same set of fathers, the mothers serving as cases (phenotype: meiotic nondisjunction of chromosome 21) and the fathers as controls. RESULTS: Our stratified analyses identified several genes whose patterns of association are consistent with generalized effects across groups, as well as other genes that are consistent with specific effects in certain groups. CONCLUSIONS: While our results are epidemiological in nature and cannot conclusively prove mechanisms, we identified a number of patterns that are consistent with specific mechanisms. In many cases those mechanisms are strongly supported by available literature on the associated genes.

Association of lipid accumulation product with chronic kidney disease in Chinese community adults: a report from the REACTION study.

BACKGROUND: Limited studies regarding the correlation of lipid accumulation product (LAP) with a decreased estimated glomerular filtration rate (eGFR) have yielded conflicting findings, and no report has demonstrated the relationship of LAP with chronic kidney disease (CKD), defined as the presence of albuminuria and/or a decreased eGFR. The purpose of this study was to estimate the possible correlation of LAP with CKD prevalence in Chinese community adults. METHOD: In this cross-sectional study, LAP level of 7202 participants (age ≥ 40 years) was determined, and its possible association with CKD was evaluated by a multiple logistic regression model. RESULTS: Compared with subjects with non-CKD, non-albuminuria, and high eGFR, LAP levels significantly increased in female not male subjects with CKD, albuminuria, and low eGFR, respectively (all P < 0.001). The univariate logistic regression analysis revealed that LAP level of female not male subjects were significantly and positively associated with the prevalence of CKD (P < 0.001). The multivariate logistic regression analysis showed that the risk of CKD prevalence in female not male subjects progressively increased across LAP quartiles (P for trend < 0.01), and the risk of CKD prevalence of subjects in Q4 significantly increased compared to those in Q1 after adjustment for potential confounding factors in Models 4 (odds ratio [OR]: 1.382, 95% confidence intervals [CI] 1.002-1.906, P < 0.05). Stratified analysis revealed positive associations of LAP quartiles with risk of CKD prevalence in people with the following characteristics: women, older, overweight, with hypertension, normal glucose tolerance, appropriate low-density lipoprotein cholesterol, nonsmokers, nondrinkers, and no cardiovascular disease events. CONCLUSIONS: High LAP levels might be significantly associated with risk of CKD prevalence in community-dwelling Chinese female adults, which may inform both public health recommendations and clinical practice.

Deriving stratified effects from joint models investigating gene-environment interactions.

BACKGROUND: Models including an interaction term and performing a joint test of SNP and/or interaction effect are often used to discover Gene-Environment (GxE) interactions. When the environmental exposure is a binary variable, analyses from exposure-stratified models which consist of estimating genetic effect in unexposed and exposed individuals separately can be of interest. In large-scale consortia focusing on GxE interactions in which only the joint test has been performed, it may be challenging to get summary statistics from both exposure-stratified and marginal (i.e not accounting for interaction) models. RESULTS: In this work, we developed a simple framework to estimate summary statistics in each stratum of a binary exposure and in the marginal model using summary statistics from the "joint" model. We performed simulation studies to assess our estimators' accuracy and examined potential sources of bias, such as correlation between genotype and exposure and differing phenotypic variances within exposure strata. Results from these simulations highlight the high theoretical accuracy of our estimators and yield insights into the impact of potential sources of bias. We then applied our methods to real data and demonstrate our estimators' retained accuracy after filtering SNPs by sample size to mitigate potential bias. CONCLUSIONS: These analyses demonstrated the accuracy of our method in estimating both stratified and marginal summary statistics from a joint model of gene-environment interaction. In addition to facilitating the interpretation of GxE screenings, this work could be used to guide further functional analyses. We provide a user-friendly Python script to apply this strategy to real datasets. The Python script and documentation are available at https://gitlab.pasteur.fr/statistical-genetics/j2s.

Contribution of CYP24A1 variants in coronary heart disease among the Chinese population.

BACKGROUND: Cytochrome P450 (CYPs) participate in the mechanisms of cardiovascular disease. The purpose of this research was to evaluate the contributions of CYP24A1 variants to coronary heart disease (CHD) among the Chinese Han population. METHODS: This study included 505 CHD cases and 508 controls. Four variants of CYP24A1 (rs2762934, rs1570669, rs6068816 and rs2296241) were chosen and genotyped by the Agena MassARRAY system among the Chinese population. The linkage between CYP24A1 variants and CHD risk were assessed by logistic regression to compute the odds ratio (OR) and 95% confidence interval (CI). Then, multifactor dimensionality reduction (MDR) was applied to analyze the interactions of CYP24A1 variants. RESULTS: The results of this study showed that CYP24A1 rs6068816 significantly enhanced CHD risk in multiple genetic models (allele: P = 0.014; codominant: P = 0.015; dominant: P = 0.043; recessive: P = 0.040; additive: P = 0.013), whereas rs2296241 was likely to protect individuals from CHD (codominant: P = 0.019; recessive: P = 0.013; additive: P = 0.033). Stratification analysis revealed that CYP24A1 polymorphisms had strong relationships with CHD risk that were dependent on age, sex, Gensini grade and smoking status (P <  0.05). Moreover, a four-locus model (rs2762934, rs1570669, rs6068816 and rs2296241) had significant impact on CHD risk in MDR analysis. CONCLUSION: It revealed that CYP24A1 variants were significantly linked with CHD susceptibility in the Chinese population.

Sociodemographic, behavioral, and medical risk factors associated with visual impairment among older adults: a community-based pilot survey in Southern District of Hong Kong.

BACKGROUND: The last visual survey of older adults in Hong Kong was a district-level study in 2002, with no assessment of behavioral and medical risk factors for visual impairment (VI). Our objectives were to determine the latest VI prevalence among older adults, significance of any spatial and temporal differences on the prevalence, and any associations of sociodemographic, behavioral and medical risk factors with VI from a multi-perspective analysis. METHODS: Community-based pilot survey of residents from a suburb of Hong Kong, aged ≥50, using a standardized questionnaire, was conducted in 2016. RESULTS: Of the 222 subjects, crude rates of bilateral and unilateral VI were 9.46 and 32.88%, respectively, or corresponding age-and-gender-adjusted rates of 6.89 and 30.5%. Older age and lower educational were associated with higher risk for unilateral VI, while older age, temporary housing, obesity and hyperlipidemia were associated with higher risk for bilateral VI. Smoking and alcohol-drinking status were not associated with unilateral or bilateral VI. Relative changes in ORs of hypertension or educational level on unilateral or bilateral VI were >  10% after adjusting for age. Interaction term between hyperlipidemia and gender or obesity was significant for unilateral VI. Gender, hypertension and cataract were not associated with unilateral or bilateral VI in general population of pooled analysis but were identified as risk factors in specific subgroups of stratified analysis. Refractive error (myopia or hyperopia) was significantly associated with VI in the eye-level analysis after adjusting the inter-eye correlation. CONCLUSIONS: Sociodemographic and medical risk factors contributed to VI, but behavioral risk factors did not. Sociodemographic disparities of visual health existed. Age was the confounders of the VI-hypertension or VI-educational level relationships. Gender and obesity were more likely to have multiplicative effect on unilateral VI when combined with hyperlipidemia. Stratified analysis should be conducted to provide further insight into the risk factors for VI in specific populations. Uncorrected refractive error remains a significant cause of impaired vision. The spatial and temporal differences in bilateral VI prevalence from the previous local study indicates a territory-wide survey is needed to assess regional differences and overall prevalence of VI in Hong Kong.

Sex and Gender-Related Issues in Heart Failure.

Understanding the role of sex- and gender-related factors, when dealing with a global growing epidemic such as heart failure, is a much needed and unmet goal for health care providers and scientists in order to design targeted strategies, aimed at improving both clinical and patient reported outcomes measures in women and men with heart failure. The present review provides an overview of the current available evidence on sex- and gender-related differences in heart failure.

Structure estimation of binary graphical models on stratified data: Application to the description of injury tables for victims of road accidents.

Graphical models are used in many applications such as medical diagnostics and computer security. Increasingly often, the estimation of such models has to be performed on several predefined strata of the whole population. For instance, in epidemiology and clinical research, strata are often defined according to age, gender, treatment, or disease type. In this article, we propose new approaches dedicated to the estimation of binary graphical models on such strata. These approaches are implemented by combining well-known methods that have been developed in the context of a single binary graphical model, with penalties encouraging structured sparsity, which have recently been shown to be appropriate when dealing with stratified data. Empirical comparisons on synthetic data highlight that our approaches generally outperform its competitors. We present an application of the approach to study associations among the injuries suffered by victims of road accidents according to road user type.

Relationships between Alcohol Intake and Ischemic Stroke Severity in Sex Stratified Analysis for Japanese Acute Stroke Patients.

BACKGROUND: The aim of this study is to evaluate in the relationships between alcohol intake and ischemic stroke severity in Japanese stroke patients. METHODS: Of a total of 101,165 registered cases of stroke in Japan, patient data of total strokes (n = 60,836), cardiogenic strokes (n = 11,894), and noncardiogenic strokes (n = 29,129) were extracted. The National Institutes of Health Stroke Scale (NIHSS) score at admission and the modified Rankin Scale (mRS) score at discharge were used. Significant alcohol consumption in regard to stroke severity at admission and discharge was identified by multivariate adjusted logistic regression analysis. RESULTS: Among all subject and males, nondrinkers had significantly poorer NIHSS scores of cardiogenic strokes as did those consuming greater than or equal to 60 g/day, whereas there was a significant difference in past heavy drinkers with noncardiogenic strokes. Among females, nondrinkers had a significance of cardiogenic stroke and non-cardiogenic strokes. Among all subjects and males, 20-39 g/day and 40-59 g/day were significantly associated with protective effects on mRS severity of cardiogenic strokes, whereas 40-59 g/day also had significant protective effects for noncardiogenic strokes. Among females, greater than or equal to 60 g/day had a significant protective effect for cardiogenic strokes. CONCLUSIONS: For males, alcohol intake of less than 60 g/day played a protective role in functional prognosis at discharge and showed a J-shape relationship. For females, although there was a limitation that the number of female drinkers was small, negative effects were shown at admission with less consumption than males.