Non-parametric estimation of the age-at-onset distribution from a cross-sectional sample.

We propose and study a simple and innovative non-parametric approach to estimate the age-of-onset distribution for a disease from a cross-sectional sample of the population that includes individuals with prevalent disease. First, we estimate the joint distribution of two event times, the age of disease onset and the survival time after disease onset. We accommodate that individuals had to be alive at the time of the study by conditioning on their survival until the age at sampling. We propose a computationally efficient expectation-maximization (EM) algorithm and derive the asymptotic properties of the resulting estimates. From these joint probabilities we then obtain non-parametric estimates of the age-at-onset distribution by marginalizing over the survival time after disease onset to death. The method accommodates categorical covariates and can be used to obtain unbiased estimates of the covariate distribution in the source population. We show in simulations that our method performs well in finite samples even under large amounts of truncation for prevalent cases. We apply the proposed method to data from female participants in the Washington Ashkenazi Study to estimate the age-at-onset distribution of breast cancer associated with carrying BRCA1 or BRCA2 mutations.

On the Paper by Leonid A. Gavrilov and Natalia S. Gavrilova entitled "Trends in Human Species-Specific Lifespan and Actuarial Aging Rate" Published in Biochemistry (Moscow), Vol. 87, Nos. 12-13, pp. 1622-1633 (2022).

The methodology used for analyzing the survival process should keep in mind heterogeneity in empirical data. Cross-sectional data are more heterogeneous in comparison with birth-cohort data.

Impacts of grandparenting on older Chinese adults' mental health: a cross-sectional study.

BACKGROUND: The practice of grandparenting has been growing in popularity worldwide, particularly in Asian societies. Nevertheless, there is a lack of thorough studies investigating the mental health effects of grandchild care on grandparents, particularly within the family context. The present study aimed to explore the impact of grandparenting on depressive symptoms in older Chinese adults, taking into account the functional role of intergenerational support. METHODS: Using the China Longitudinal Aging Social Survey (CLASS, 2014 and 2018, N = 9,486), we employed the Pooled Ordinary Least Square method (POLS) to explore the association between depressive symptoms and grandparenting intensity as well as include the interaction terms to investigate the role of intergenerational support among grandparents aged from 60 to 80. RESULTS: After adjusting for control variables, both non-intensive (-0.17; 95% CI: -0.30, -0.03) and intensive (-0.69; 95% CI: -0.95, -0.43) childcare, as well as giving financial support to adult children (-0.06; 95% CI: -0.08, -0.04) and emotional closeness with them (-0.94; 95% CI: -1.15, -0.72), were found to have a positive impact on the mental health of grandparents. Giving financial support (non-intensive: -0.04, 95% CI: -0.07, -0.01; intensive: -0.06, 95% CI: -0.13, -0.01) and providing instrumental support to adult children (non-intensive: -0.12, 95% CI: -0.24, -0.01; intensive: -0.19, 95% CI: -0.37, -0.02) moderated the association between grandparenting and the mental health of older adults, regardless of care intensity. However, the moderating role of receiving financial support (-0.07; 95% CI: -0.12, -0.02) only existed when non-intensive childcare was provided. CONCLUSION: Grandchild care predicted better mental health in grandparents, mainly when they engaged in intensive grandparenting. Emotional closeness and providing financial support to adult children brought mental health benefits to grandparents involved in childcare. Giving financial support and providing instrumental support to adult children moderated the association between grandparenting and the mental health of older adults. However, the moderating role of receiving financial support from adult children only existed when non-intensive childcare was provided.

The epidemiology of airplane headache: A cross-sectional study on point prevalence and characteristics in 50,000 travelers.

BACKGROUND: The current knowledge on the epidemiology and clinical manifestation of airplane headache is mostly derived from case series and small cohort studies without evidence from large populations. METHODS: This cross-sectional study was conducted over a five-month period in the arrival area of two international airports in Germany. 50,000 disembarking passengers were addressed about headaches during their flight to determine headache prevalence, and those confirming and willing to participate underwent a structured interview. RESULTS: Headache during travel was reported by 374 passengers (0.75%), and 301 underwent a structured interview. One hundred and one (0.2%) met the diagnostic criteria of airplane headache. Six passengers suffered from migraines and 134 from tension-type headaches. The differences in the age and gender distribution between the airplane headache and non-airplane headache groups were not statistically significant. The onset (79.2%), duration (82.2%), and location (73.3%) of airplane headache mostly complied with current diagnostic criteria but pain intensity (42.6%) and quality (42.6%) did less so. CONCLUSION: Our data suggest a substantially lower prevalence of airplane headaches than previously reported. The pain intensity and quality seem less characteristic than assumed, suggesting a need to refine the current diagnostic criteria.

Dynamic modeling of the Italians' attitude towards Covid-19.

We analyze repeated cross-sectional survey data collected by the Institute of Global Health Innovation, to characterize the perception and behavior of the Italian population during the Covid-19 pandemic, focusing on the period that spans from April 2020 to July 2021. To accomplish this goal, we propose a Bayesian dynamic latent-class regression model, that accounts for the effect of sampling bias including survey weights into the likelihood function. According to the proposed approach, attitudes towards covid-19 are described via ideal behaviors that are fixed over time, corresponding to different degrees of compliance with spread-preventive measures. The overall tendency toward a specific profile dynamically changes across survey waves via a latent Gaussian process regression, that adjusts for subject-specific covariates. We illustrate the evolution of Italians' behaviors during the pandemic, providing insights on how the proportion of ideal behaviors has varied during the phases of the lockdown, while measuring the effect of age, sex, region and employment of the respondents on the attitude toward covid-19.

Alcohol consumption is associated with widespread changes in blood DNA methylation: Analysis of cross-sectional and longitudinal data.

DNA methylation may be one of the mechanisms by which alcohol consumption is associated with the risk of disease. We conducted a large-scale, cross-sectional, genome-wide DNA methylation association study of alcohol consumption and a longitudinal analysis of repeated measurements taken several years apart. Using the Illumina HumanMethylation450 BeadChip, DNA methylation was measured in blood samples from 5606 Melbourne Collaborative Cohort Study (MCCS) participants. For 1088 of them, these measures were repeated using blood samples collected a median of 11 years later. Associations between alcohol intake and blood DNA methylation were assessed using linear mixed-effects regression models. Independent data from the London Life Sciences Prospective Population (LOLIPOP) (N = 4042) and Cooperative Health Research in the Augsburg Region (KORA) (N = 1662) cohorts were used to replicate associations discovered in the MCCS. Cross-sectional analyses identified 1414 CpGs associated with alcohol intake at P < 10-7 , 1243 of which had not been reported previously. Of these novel associations, 1078 were replicated (P < .05) using LOLIPOP and KORA data. Using the MCCS data, we also replicated 403 of 518 previously reported associations. Interaction analyses suggested that associations were stronger for women, non-smokers, and participants genetically predisposed to consume less alcohol. Of the 1414 CpGs, 530 were differentially methylated (P < .05) in former compared with current drinkers. Longitudinal associations between the change in alcohol intake and the change in methylation were observed for 513 of the 1414 cross-sectional associations. Our study indicates that alcohol intake is associated with widespread changes in DNA methylation across the genome. Longitudinal analyses showed that the methylation status of alcohol-associated CpGs may change with alcohol consumption changes in adulthood.

Estimating group fixed effects in panel data with a binary dependent variable: How the LPM outperforms logistic regression in rare events data.

Estimating fixed effects models can be challenging with rare events data. Researchers often face difficult trade-offs when selecting between the Linear Probability Model (LPM), logistic regression with group intercepts and the conditional logit. In this paper, I survey these tradeoffs and argue that, in fact, the LPM with fixed effects produces more accurate estimates and predicted probabilities than maximum likelihood specifications when the dependent variable has less than 25 percent of ones. I use Monte Carlo simulations to show when the LPM with fixed effects should be preferred. I perform these simulations on common time-series cross-sectional (TSCS) data structures found in the literature as well as big data. This paper provides clarity around fixed effects models in TSCS data and a novel technique to identify which one to use as a function of the frequency of events in y.

Adult low-risk drinkers and abstainers are not the same.

BACKGROUND: Alcohol consumption, even at low-levels, can not be guaranteed as safe or risk free. Specifically, the 2009 Australian National Health and Medical Research Council drinking guidelines recommend that adults should not drink more than two standard drinks on any day on average, and no more than four drinks on a single occasion. Nearly 40% of Australians aged 12 years and older drink alcohol but don't exceed these recommended limits, yet adult low-risk drinkers have been largely overlooked in Australian alcohol survey research, where they are usually grouped with abstainers. This paper examines the socio-demographic profile of low-risk drinking adults (18+ years old), compared to those who abstain. METHODS: Data from the 2013 National Drug Strategy Household Survey were used. In the past 12 months, 4796 Australians had not consumed alcohol and 8734 had consumed alcohol at low-risk levels, accounting for both average volume and episodic drinking (hereafter low-risk). RESULTS: Multivariate logistic regression results indicated that low-risk drinkers were more likely to be older, married, Australian-born, and reside in a less disadvantaged neighbourhood compared with abstainers. There was no significant difference by sex between low-risk drinkers and abstainers. CONCLUSIONS: The socio-demographic profile of low-risk drinkers differed from that of abstainers. Combining low-risk drinkers and abstainers into a single group, which is often the practice in survey research, may mask important differences. The study may support improved targeting of health promotion initiatives that encourage low-risk drinkers not to increase consumption or, in view of increasing evidence that low-risk drinking is not risk free, to move towards abstinence.

Is exposure to secondhand smoke associated with current depression (PHQ-8) among never-smokers? Results from a survey among German adults.

BACKGROUND: Findings on the association between exposure to secondhand smoke (SHS) and depression are contradictory. Comparability of existing research is limited due to varied methods and measures. This study examines the potential association between exposure to SHS and depression and a potential moderation by sex using representative data from Germany. METHODS: For our study, we used data from the German Health Update (GEDA) 2014/2015 on n = 10,274 never-smokers. We calculated a logistic regression model with an interaction term for potential sex-exposure interactions. We used the self-reported duration of exposure to predict current depression of any type as defined by the Patient Health Questionnaire PHQ-8 (score ≥ 10), accounting for a large number of confounders. RESULTS: In our sample, prevalence of depression was 8.8% in women and 5.3% in men. 90.4% were never or almost never exposed to SHS, while 7.1% were exposed < 1 h per day and around 2.5% reported being exposed for ≥1 h per day. While SHS exposure for < 1 h per day was not associated with current depression (OR = 1.54; 95%-CI: 0.93-1.61), SHS exposure for at least 1 h per day was associated with increased odds for current depression (OR = 1.59; 95%-CI: 1.08-2.35). No sex-specific differences were found. CONCLUSIONS: Higher levels of SHS exposure are associated with current depression, although the nature and direction of the association are still unclear. We identified no differences in the association between men and women. More studies, particularly using longitudinal data, are needed to determine the nature of the association.

Estimating the decision curve and its precision from three study designs.

The decision curve plots the net benefit (NB) of a risk model for making decisions over a range of risk thresholds, corresponding to different ratios of misclassification costs. We discuss three methods to estimate the decision curve, together with corresponding methods of inference and methods to compare two risk models at a given risk threshold. One method uses risks (R) and a binary event indicator (Y) on the entire validation cohort. This method makes no assumptions on how well-calibrated the risk model is nor on the incidence of disease in the population and is comparatively robust to model miscalibration. If one assumes that the model is well-calibrated, one can compute a much more precise estimate of NB based on risks R alone. However, if the risk model is miscalibrated, serious bias can result. Case-control data can also be used to estimate NB if the incidence (or prevalence) of the event ( Y=1 ) is known. This strategy has comparable efficiency to using the full (R,Y) data, and its efficiency is only modestly less than that for the full (R,Y) data if the incidence is estimated from the mean of Y. We estimate variances using influence functions and propose a bootstrap procedure to obtain simultaneous confidence bands around the decision curve for a range of thresholds. The influence function approach to estimate variances can also be applied to cohorts derived from complex survey samples instead of simple random samples.

Multinomial additive hazard model to assess the disability burden using cross-sectional data.

Population aging is accompanied by the burden of chronic diseases and disability. Chronic diseases are among the main causes of disability, which is associated with poor quality of life and high health care costs in the elderly. The identification of which chronic diseases contribute most to the disability prevalence is important to reduce the burden. Although longitudinal studies can be considered the gold standard to assess the causes of disability, they are costly and often with restricted sample size. Thus, the use of cross-sectional data under certain assumptions has become a popular alternative. Among the existing methods based on cross-sectional data, the attribution method, which was originally developed for binary disability outcomes, is an attractive option, as it enables the partition of disability into the additive contribution of chronic diseases, taking into account multimorbidity and that disability can be present even in the absence of disease. In this paper, we propose an extension of the attribution method to multinomial responses, since disability is often measured as a multicategory variable in most surveys, representing different severity levels. The R function constrOptim is used to maximize the multinomial log-likelihood function subject to a linear inequality constraint. Our simulation study indicates overall good performance of the model, without convergence problems. However, the model must be used with care for populations with low marginal disability probabilities and with high sum of conditional probabilities, especially with small sample size. For illustration, we apply the model to the data of the Belgian Health Interview Surveys.

Influence of Parental Expressed Emotions on Children's Emotional Eating via Children's Negative Urgency.

We investigated whether parental expressed emotion (criticism and emotional overinvolvement) is related to children's emotional eating and whether this relationship is mediated by children's negative urgency. One hundred children, aged 8 to 13 years, either healthy or have binge-eating disorder and/or attention-deficit/hyperactivity disorder, completed the questionnaires, along with their parents. Parental criticism and, to a lesser extent, parental emotional overinvolvement were both positively related to children's emotional eating, and this relationship was mediated by children's negative urgency. Further exploratory analyses revealed that the mediating role of children's negative urgency in the relationship between parental criticism and children's emotional eating was pronounced in the clinical group of children with binge-eating disorder and attention-deficit/hyperactivity disorder but almost absent in the healthy control group. Copyright © 2016 John Wiley & Sons, Ltd and Eating Disorders Association.

Factors affecting the livestock herd size among smallholder households in Zambia.

Structural changes in the economic and demographic landscapes in the developing countries create a host of opportunities for smallholder livestock producers to tap into the economic growth from which they have largely been excluded. However, small livestock herd sizes limit the opportunities available to farmers to commercialize and benefit from the growing demand for livestock and livestock products. Therefore, understanding what it takes to build and maintain the herd size is a critical step in addressing the problem of low market participation. The study used five cross-sectional nationally representative data sets to arrive at the following conclusions. While purchases are a primary source for the initial stock, births are the principal way of increasing the herd size. Livestock deaths as reflected by high mortality rates are the major outflow channel through which livestock leaves the herd, and reducing the mortality rates has a positive effect on the livestock population as evidenced within the cattle sub-sector. Thus, provision of adequate animal health services is key to addressing the high mortality rates which subsequently can lead to increased herd size. As births are the primary means of increasing the herd size, it is necessary to implement interventions that address the reproduction rates. The interventions include adequate feeding of livestock on high-quality feed and strategic feeding to address the dry season challenges. These interventions can facilitate the growth of the young livestock into mature breeding stock and thus increase the herd size.

The effect of snacking and eating frequency on dietary quality in British adolescents.

PURPOSE: To describe the effects of number of eating occasions and snacks on dietary quality (DQ), defined as adherence to dietary recommendations. METHODS: A sample of 884 adolescents (11-18 years) in the UK National Diet and Nutrition Survey (NDNS) were included. The Diet Quality Index for Adolescents (DQI-A) was implemented. The total number of eating occasions and snacks was frequency of food or beverages consumed over 24 h and frequency of foods or beverages consumed outside of the three mealtimes, respectively. Results were generated with and without low-energy food under 210 kJ (50 kcal). Regression models were generated with DQ score as the outcome variable and number of eating occasions and snacks as predictors. RESULTS: The mean (95 % CI) DQ score was 31.1 % (30.2, 32.0). The mean number of eating occasions and snacks was 7.5 (7.3, 7.7) and 2.6 (2.6, 2.7) times/day, respectively. When low-energy events were excluded, the mean number of eating occasions and snacks reduced to 6.2 (6.1, 6.4) and 2.0 (2.0, 2.1) times/day, respectively. DQ score increased by 0.74 points (0.42, 1.05; p < 0.01) and 0.55 points (-0.08, 0.69; p = 0.17) for total eating occasions and snacks, respectively. When low-energy events were excluded, DQ score increased by 0.30 points (-0.84, 0.69; p = 0.13) for each eating occasion and decreased by 1.20 points (-2.1, -0.3; p < 0.01) for each snack. CONCLUSION: Eating more frequently improves dietary quality especially if some eating occasions are low in energy. A focus on replacing high-energy snacks with low-energy alternatives rather than reducing the number of eating occasions may result in improved dietary quality in adolescents.

An introduction to latent variable mixture modeling (part 1): overview and cross-sectional latent class and latent profile analyses.

OBJECTIVE: Pediatric psychologists are often interested in finding patterns in heterogeneous cross-sectional data. Latent variable mixture modeling is an emerging person-centered statistical approach that models heterogeneity by classifying individuals into unobserved groupings (latent classes) with similar (more homogenous) patterns. The purpose of this article is to offer a nontechnical introduction to cross-sectional mixture modeling. METHOD: An overview of latent variable mixture modeling is provided and 2 cross-sectional examples are reviewed and distinguished. RESULTS: Step-by-step pediatric psychology examples of latent class and latent profile analyses are provided using the Early Childhood Longitudinal Study-Kindergarten Class of 1998-1999 data file. CONCLUSIONS: Latent variable mixture modeling is a technique that is useful to pediatric psychologists who wish to find groupings of individuals who share similar data patterns to determine the extent to which these patterns may relate to variables of interest.

Investigating seasonal changes in factors associated with COVID-19 concerns: Results from a serial cross-sectional survey study in Germany between 2020 and 2023.

Objective: COVID-19 risk perceptions are discussed to be volatile and have been shown to be connected to the adoption of preventive public health behaviors. This study aimed to investigate changes in COVID-19 concerns and influencing factors as a function of season among the German public. Methods: Sixty-three waves of cross-sectional telephone surveys with German participants aged 14 years and older conducted at least monthly between June 2020 and April 2023 provided the data basis (N = 63,471). After pooling participants of different waves by season (spring, summer, fall, winter), data were analyzed with regard to changes in physical health, mental health, economic, and social COVID-19 concerns. Individual characteristics (e.g., age), COVID-19 behavior (e.g., hygiene practices), and related perceptions (e.g., controllability of risk) were considered as predictors of composite concerns in different seasons. Results: Results showed a higher between-seasons than within-seasons variability in concerns, with rises in physical and mental health and social concerns during fall. Multivariate regressions revealed being female, lower education, adopting protective measures, and higher perceived probability of infection in both public and private settings to be consistent predictors of higher COVID-19 concerns. Coefficients of these predictors remained comparatively stable over seasons and years. Conclusion: Results indicate re-occurring changes in concerns during a prolonged crisis, with distinct characteristics being consistently associated with higher reported concerns. To ensure the application of protective measures, communicators should consider that risk perceptions are subject to fluctuations, but that certain groups of individuals tend to develop them and therefore deserve particular focus.

Utilizing Artificial Neural Networks for Establishing Hearing-Loss Predicting Models Based on a Longitudinal Dataset and Their Implications for Managing the Hearing Conservation Program.

Background: Though the artificial neural network (ANN) technique has been used to predict noise-induced hearing loss (NIHL), the established prediction models have primarily relied on cross-sectional datasets, and hence, they may not comprehensively capture the chronic nature of NIHL as a disease linked to long-term noise exposure among workers. Methods: A comprehensive dataset was utilized, encompassing eight-year longitudinal personal hearing threshold levels (HTLs) as well as information on seven personal variables and two environmental variables to establish NIHL predicting models through the ANN technique. Three subdatasets were extracted from the afirementioned comprehensive dataset to assess the advantages of the present study in NIHL predictions. Results: The dataset was gathered from 170 workers employed in a steel-making industry, with a median cumulative noise exposure and HTL of 88.40 dBA-year and 19.58 dB, respectively. Utilizing the longitudinal dataset demonstrated superior prediction capabilities compared to cross-sectional datasets. Incorporating the more comprehensive dataset led to improved NIHL predictions, particularly when considering variables such as noise pattern and use of personal protective equipment. Despite fluctuations observed in the measured HTLs, the ANN predicting models consistently revealed a discernible trend. Conclusions: A consistent correlation was observed between the measured HTLs and the results obtained from the predicting models. However, it is essential to exercise caution when utilizing the model-predicted NIHLs for individual workers due to inherent personal fluctuations in HTLs. Nonetheless, these ANN models can serve as a valuable reference for the industry in effectively managing its hearing conservation program.

Variation in Trust in Cancer Information Sources by Perceptions of Social Media Health Mis- and Disinformation and by Race and Ethnicity Among Adults in the United States: Cross-Sectional Study.

BACKGROUND: Mis- and disinformation on social media have become widespread, which can lead to a lack of trust in health information sources and, in turn, lead to negative health outcomes. Moreover, the effect of mis- and disinformation on trust in information sources may vary by racial and ethnic minoritized populations. OBJECTIVE: We evaluated how trust in multiple sources of cancer information varied by perceptions of health mis- and disinformation on social media and by race and ethnicity. METHODS: Cross-sectional, nationally representative survey data from noninstitutionalized adults in the United States from the 2022 Health Information National Trends Survey 6 (HINTS 6) were analyzed (N=4137). The dependent variable measured the level of trust in cancer information sources. The independent variables were perceptions about health mis- and disinformation on social media and race and ethnicity. Multivariable logistic regression models were adjusted for survey weight and design, age, birth gender, race and ethnicity, marital status, urban/rural designation, education, employment status, feelings about household income, frequency of social media visits, and personal and family history of cancer. We also tested the interaction effect between perceptions of social media health mis- and disinformation and participants' self-reported race and ethnicity. RESULTS: Perception of "a lot of" health mis- and disinformation on social media, relative to perception of "less than a lot," was associated with a lower likelihood of high levels of trusting cancer information from government health agencies (odds ratio [OR] 0.60, 95% CI 0.47-0.77), family or friends (OR 0.56, 95% CI 0.44-0.71), charitable organizations (OR 0.78, 95% CI 0.63-0.96), and religious organizations and leaders (OR 0.64, 95% CI 0.52-0.79). Among White participants, those who perceived a lot of health mis- and disinformation on social media were less likely to have high trust in cancer information from government health agencies (margin=61%, 95% CI 57%-66%) and family or friends (margin=49%, 95% CI 43%-55%) compared to those who perceived less than a lot of health mis- and disinformation on social media. Among Black participants, those who perceived a lot of health mis- and disinformation on social media were less likely to have high trust in cancer information from religious organizations and leaders (margin=20%, 95% CI 10%-30%) compared to participants who perceived no or a little health mis- and disinformation on social media. CONCLUSIONS: Certain sources of cancer information may need enhanced support against the threat of mis- and disinformation, such as government health agencies, charitable organizations, religious organizations and leaders, and family or friends. Moreover, interventions should partner with racial and ethnically minoritized populations that are more likely to have low trust in certain cancer information sources associated with mis- and disinformation on social media.

Reconstructing multi-strain pathogen interactions from cross-sectional survey data via statistical network inference.

Infectious diseases often involve multiple pathogen species or multiple strains of the same pathogen. As such, knowledge of how different pathogens interact is key to understand and predict the outcome of interventions targeting only a subset of species or strains involved in disease. Population-level data may be useful to infer pathogen strain interactions, but most previously used inference methods only consider uniform interactions between all strains or focus on marginal pairwise interactions. As such, these methods are prone to bias induced by indirect interactions through other strains. Here, we evaluated statistical network inference for reconstructing heterogeneous interactions from cross-sectional surveys detecting joint presence/absence patterns of pathogen strains within hosts. We applied various network models to simulated survey data, representing endemic infection states of multiple pathogen strains with potential interactions in acquisition or clearance of infection. Satisfactory performance was demonstrated by the estimators converging to the true interactions. Accurate reconstruction of interaction networks was achieved by regularization or penalization for sample size. Although performance deteriorated in the presence of host heterogeneity, this was overcome by correcting for individual-level risk factors. Our work demonstrates how statistical network inference could prove useful for detecting multi-strain pathogen interactions and may have applications beyond epidemiology.

A taxonomy of multiple regression methods for immunologists.

Multiple regression is a powerful tool in the immunologist's toolbox. This paper defines multiple regression, discusses availability and accessibility, provides some additional helpful definitions, treats the topics of transformation and extreme value screening, and establishes the paper's scope and philosophy. Then eleven methods of multiple regression are detailed, giving strengths and limitations. Throughout an emphasis is placed on application to immunological assays. A flowchart to guide selection of multiple regression methods is provided.