DeepNull models non-linear covariate effects to improve phenotypic prediction and association power

Genome-wide association studies (GWASs) examine the association between genotype and phenotype while adjusting for a set of covariates. Although the covariates may have non-linear or interactive effects, due to the challenge of specifying the model, GWAS often neglect such terms. Here we introduce DeepNull, a method that identifies and adjusts for non-linear and interactive covariate effects using a deep neural network. In analyses of simulated and real data, we demonstrate that DeepNull maintains tight control of the type I error while increasing statistical power by up to 20% in the presence of non-linear and interactive effects. Moreover, in the absence of such effects, DeepNull incurs no loss of power. When applied to 10 phenotypes from the UK Biobank (n = 370K), DeepNull discovered more hits (+6%) and loci (+7%), on average, than conventional association analyses, many of which are biologically plausible or have previously been reported. Finally, DeepNull improves upon linear modeling for phenotypic prediction (+23% on average).

Multivariate functional additive mixed models

Multivariate functional data can be intrinsically multivariate like movement trajectories in 2D or complementary such as precipitation, temperature and wind speeds over time at a given weather station. We propose a multivariate functional additive mixed model (multiFAMM) and show its application to both data situations using examples from sports science (movement trajectories of snooker players) and phonetic science (acoustic signals and articulation of consonants). The approach includes linear and nonlinear covariate effects and models the dependency structure between the dimensions of the responses using multivariate functional principal component analysis. Multivariate functional random intercepts capture both the auto-correlation within a given function and cross-correlations between the multivariate functional dimensions. They also allow us to model between-function correlations as induced by, for example, repeated measurements or crossed study designs. Modelling the dependency structure between the dimensions can generate additional insight into the properties of the multivariate functional process, improves the estimation of random effects, and yields corrected confidence bands for covariate effects. Extensive simulation studies indicate that a multivariate modelling approach is more parsimonious than fitting independent univariate models to the data while maintaining or improving model fit.

Evaluating the transitions in care for children presenting with acute asthma to emergency departments: a retrospective cohort study

Background Acute asthma is a common presentation to emergency departments (EDs) worldwide and, due to overcrowding, delays in treatment often occur. This study deconstructs the total ED length of stay into stages and estimates covariate effects on transition times for children presenting with asthma. Methods We extracted ED presentations in 2019 made by children in Alberta, Canada for acute asthma. We used multivariable Cox regressions in a multistate model to model transition times among the stages of start, physician initial assessment (PIA), disposition decision, and ED departure. Results Data from 6598 patients on 8270 ED presentations were extracted. The individual PIA time was longer (i.e., HR < 1) when time to the crowding metric (hourly PIA) was above 1 h (HR = 0.32; 95% CI:0.30,0.34), for tertiary (HR = 0.65; 95% CI:0.61,0.70) and urban EDs (HR = 0.77; 95% CI:0.70,0.84), for younger patients (HR = 0.99 per year; 95% CI:0.99,1.00), and for patients triaged less urgent/non-urgent (HR = 0.89; 95% CI:0.84,0.95). It was shorter for patients arriving by ambulance (HR = 1.22; 95% CI:1.04,1.42). Times from PIA to disposition decision were longer for tertiary (HR = 0.47; 95% CI:0.44,0.51) and urban (HR = 0.69; 95% CI:0.63,0.75) EDs, for patients triaged as resuscitation/emergent (HR = 0.51; 95% CI:0.48,0.54), and for patients arriving by ambulance (HR = 0.78; 95% CI:0.70,0.87). Times from disposition decision to ED departure were longer for patients who were admitted (HR = 0.16; 95% CI:0.13,0.20) or transferred (HR = 0.42; 95% CI:0.35,0.50), and for tertiary EDs (HR = 0.93; 95% CI:0.92,0.94). Conclusions All transition times were impacted by ED presentation characteristics. The sole key patient characteristic was age and it only impacted time to PIA. ED crowding demonstrated strong effects of time to PIA but not for the transition times involving disposition decision and ED departure stages.

Analysis of multivariate longitudinal substance use outcomes using multivariate mixed cumulative logit model

Background Longitudinal assessments of usage are often conducted for multiple substances (e.g., cigarettes, alcohol and marijuana) and research interests are often focused on the inter-substance association. We propose a multivariate longitudinal modeling approach for jointly analyzing the ordinal multivariate substance use data. Methods We describe how the binary random slope logistic regression model can be extended to the multi-category ordinal outcomes. We also describe how the proportional odds assumption can be relaxed by allowing differential covariate effects on different cumulative logits for multiple outcomes. Data are analyzed from a P01 study that evaluates the usage levels of cigarettes, alcohol and marijuana repeatedly across 8 measurement waves during 7 consecutive years. Results 1263 subjects participated in the study with informed consent, among whom 56.6% are females. Males and females show significant differences in terms of the time trend for substance use. Specifically, males showed steeper trends on cigarette and marijuana use over time compared to females, while less so for alcohol. For all three substances, age effects appear to be different for different cumulative logits, indicating the violation of proportional odds assumption. Conclusions The multivariate mixed cumulative logit model offers the most flexibility and allows one to examine the inter-substance association when proportional odds assumption is violated.

The comparison of censored quantile regression methods in prognosis factors of breast cancer survival

The Cox proportional hazards model is a widely used statistical method for the censored data that model the hazard rate rather than survival time. To overcome complexity of interpreting hazard ratio, quantile regression was introduced for censored data with more straightforward interpretation. Different methods for analyzing censored data using quantile regression model, have been introduced. The quantile regression approach models the quantile function of failure time and investigates the covariate effects in different quantiles. In this model, the covariate effects can be changed for patients with different risk and is a flexible model for controlling the heterogeneity of covariate effects. We illustrated and compared five methods in quantile regression for right censored data included Portnoy, Wang and Wang, Bottai and Zhang, Yang and De Backer methods. The comparison was made through the use of these methods in modeling the survival time of breast cancer. According to the results of quantile regression models, tumor grade and stage of the disease were identified as significant factors affecting 20th percentile of survival time. In Bottai and Zhang method, 20th percentile of survival time for a case with higher unit of stage decreased about 14 months and 20th percentile of survival time for a case with higher grade decreased about 13 months. The quantile regression models acted the same to determine prognostic factors of breast cancer survival in most of the time. The estimated coefficients of five methods were close to each other for quantiles lower than 0.1 and they were different from quantiles upper than 0.1.

Getting Time Right: Using Cox Models and Probabilities to Interpret Binary Panel Data

Logit and probit (L/P) models are a mainstay of binary time-series cross-sectional (BTSCS) analyses. Researchers include cubic splines or time polynomials to acknowledge the temporal element inherent in these data. However, L/P models cannot easily accommodate three other aspects of the data’s temporality: whether covariate effects are conditional on time, whether the process of interest is causally complex, and whether our functional form assumption regarding time’s effect is correct. Failing to account for any of these issues amounts to misspecification bias, threatening our inferences’ validity. We argue scholars should consider using Cox duration models when analyzing BTSCS data, as they create fewer opportunities for such misspecification bias, while also having the ability to assess the same hypotheses as L/P. We use Monte Carlo simulations to bring new evidence to light showing Cox models perform just as well—and sometimes better—than logit models in a basic BTSCS setting, and perform considerably better in more complex BTSCS situations. In addition, we highlight a new interpretation technique for Cox models—transition probabilities—to make Cox model results more readily interpretable. We use an application from interstate conflict to demonstrate our points.