Birnbaum–Saunders frailty regression models: Diagnostics and application to medical data

In survival models, some covariates affecting the lifetime could not be observed or measured. These covariates may correspond to environmental or genetic factors and be considered as a random effect related to a frailty of the individuals explaining their survival times. We propose a methodology based on a Birnbaum–Saunders frailty regression model, which can be applied to censored or uncensored data. Maximum‐likelihood methods are used to estimate the model parameters and to derive local influence techniques. Diagnostic tools are important in regression to detect anomalies, as departures from error assumptions and presence of outliers and influential cases. Normal curvatures for local influence under different perturbations are computed and two types of residuals are introduced. Two examples with uncensored and censored real‐world data illustrate the proposed methodology. Comparison with classical frailty models is carried out in these examples, which shows the superiority of the proposed model.     

Robust variance-components approach for assessing genetic linkage in pedigrees.

To assess evidence for genetic linkage from pedigrees, I developed a limited variance-components approach. In this method, variability among trait observations from individuals within pedigrees is expressed in terms of fixed effects from covariates and effects due to an unobservable trait-affecting major locus, random polygenic effects, and residual nongenetic variance. The effect attributable to a locus linked to a marker is a function of the additive and dominance components of variance of the locus, the recombination fraction, and the proportion of genes identical by descent at the marker locus for each pair of sibs. For unlinked loci, the polygenic variance component depends only on the relationship between the relative pair. Parameters can be estimated by either maximum-likelihood methods or quasi-likelihood methods. The forms of quasi-likelihood estimators are provided. Hypothesis tests derived from the maximum-likelihood approach are constructed by appeal to asymptotic theory. A simulation study showed that the size of likelihood-ratio tests was appropriate but that the monogenic component of variance was generally underestimated by the likelihood approach.     

Bayesian Semiparametric Frailty Selection in Multivariate Event Time Data

Biomedical studies often collect multivariate event time data from multiple clusters (either subjects or groups) within each of which event times for individuals are correlated and the correlation may vary in different classes. In such survival analyses, heterogeneity among clusters for shared and specific classes can be accommodated by incorporating parametric frailty terms into the model. In this article, we propose a Bayesian approach to relax the parametric distribution assumption for shared and specific‐class frailties by using a Dirichlet process prior while also allowing for the uncertainty of heterogeneity for different classes. Multiple cluster‐specific frailty selections rely on variable selection‐type mixture priors by applying mixtures of point masses at zero and inverse gamma distributions to the variance of log frailties. This selection allows frailties with zero variance to effectively drop out of the model. A reparameterization of log‐frailty terms is performed to reduce the potential bias of fixed effects due to variation of the random distribution and dependence among the parameters resulting in easy interpretation and faster Markov chain Monte Carlo convergence. Simulated data examples and an application to a lung cancer clinical trial are used for illustration.     

Network frailty and the geometry of herd immunity

The spread of infectious disease through communities depends fundamentally on the underlying patterns of contacts between individuals. Generally, the more contacts one individual has, the more vulnerable they are to infection during an epidemic. Thus, outbreaks disproportionately impact the most highly connected demographics. Epidemics can then lead, through immunization or removal of individuals, to sparser networks that are more resistant to future transmission of a given disease. Using several classes of contact networks-Poisson, scale-free and small-world-we characterize the structural evolution of a network due to an epidemic in terms of frailty (the degree to which highly connected individuals are more vulnerable to infection) and interference (the extent to which the epidemic cuts off connectivity among the susceptible population that remains following an epidemic). The evolution of the susceptible network over the course of an epidemic differs among the classes of networks; frailty, relative to interference, accounts for an increasing component of network evolution on networks with greater variance in contacts. The result is that immunization due to prior epidemics can provide greater community protection than random vaccination on networks with heterogeneous contact patterns, while the reverse is true for highly structured populations.     

The heritability of mortality due to heart diseases: a correlated frailty model applied to Danish twins.

Data of the Danish Twin Registry on monozygotic and dizygotic twins are used to analyse genetic and environmental influences on susceptibility to heart diseases for males and females, respectively. The sample includes 7955 like-sexed twin pairs born between 1870 and 1930. Follow-up was from 1 January 1943 to 31 December 1993 which results in truncation (twin pairs were included in the study if both individuals were still alive at the beginning of the follow-up) and censoring (nearly 40% of the study population was still alive at the end of the follow-up). We use the correlated gamma-frailty model for the genetic analysis of frailty to account for this censoring and truncation. During the follow-up 9370 deaths occurred, 3393 deaths were due to heart diseases in general, including 2476 deaths due to coronary heart disease (CHD). Proportions of variance of frailty attributable to genetic and environmental factors were analyzed using the structural equation model approach. Different standard biometric models are fitted to the data to evaluate the magnitude and nature of genetic and environmental factors on mortality. Using the best fitting model heritability of frailty (liability to death) was found to be 0.55 (0.07) and 0.53 (0.11) with respect to heart diseases and CHD, respectively, for males and 0.52 (0.10) and 0.58 (0.14) for females in a parametric analysis. A semi-parametric analysis shows very similar results. These analyses may indicate the existence of a strong genetic influence on individual frailty associated with mortality caused by heart diseases and CHD in both, males and females. The nature of genetic influences on frailty with respect to heart diseases and CHD is probably additive. No evidence for dominance and shared environment was found.     

Olfactory Dysfunction Predicts 5-Year Mortality in Older Adults

Prediction of mortality has focused on disease and frailty, although antecedent biomarkers may herald broad physiological decline. Olfaction, an ancestral chemical system, is a strong candidate biomarker because it is linked to diverse physiological processes. We sought to determine if olfactory dysfunction is a harbinger of 5-year mortality in the National Social Life, Health and Aging Project [NSHAP], a nationally representative sample of older U.S. adults. 3,005 community-dwelling adults aged 57–85 were studied in 2005–6 (Wave 1) and their mortality determined in 2010–11 (Wave 2). Olfactory dysfunction, determined objectively at Wave 1, was used to estimate the odds of 5-year, all cause mortality via logistic regression, controlling for demographics and health factors. Mortality for anosmic older adults was four times that of normosmic individuals while hyposmic individuals had intermediate mortality (p<0.001), a “dose-dependent” effect present across the age range. In a comprehensive model that included potential confounding factors, anosmic older adults had over three times the odds of death compared to normosmic individuals (OR, 3.37 [95%CI 2.04, 5.57]), higher than and independent of known leading causes of death, and did not result from the following mechanisms: nutrition, cognitive function, mental health, smoking and alcohol abuse or frailty. Olfactory function is thus one of the strongest predictors of 5-year mortality and may serve as a bellwether for slowed cellular regeneration or as a marker of cumulative toxic environmental exposures. This finding provides clues for pinpointing an underlying mechanism related to a fundamental component of the aging process.     

Random regression models for detection of gene by environment interaction

Two random regression models, where the effect of a putative QTL was regressed on an environmental gradient, are described. The first model estimates the correlation between intercept and slope of the random regression, while the other model restricts this correlation to 1 or -1, which is expected under a bi-allelic QTL model. The random regression models were compared to a model assuming no gene by environment interactions. The comparison was done with regards to the models ability to detect QTL, to position them accurately and to detect possible QTL by environment interactions. A simulation study based on a granddaughter design was conducted, and QTL were assumed, either by assigning an effect independent of the environment or as a linear function of a simulated environmental gradient. It was concluded that the random regression models were suitable for detection of QTL effects, in the presence and absence of interactions with environmental gradients. Fixing the correlation between intercept and slope of the random regression had a positive effect on power when the QTL effects re-ranked between environments.     

The effect of treatment delay on time-to-recovery in the presence of unobserved heterogeneity

We study the effect of delaying treatment in the presence of (unobserved) heterogeneity. In a homogeneous population and assuming a proportional treatment effect, a treatment delay period will result in notably lower cumulative recovery percentages. We show in theoretical scenarios using frailty models that if the population is heterogeneous, the effect of a delay period is much smaller. This can be explained by the selection process that is induced by the frailty. Patient groups that start treatment later have already undergone more selection. The marginal hazard ratio for the treatment will act differently in such a more homogeneous patient group. We further discuss modeling approaches for estimating the effect of treatment delay in the presence of heterogeneity, and compare their performance in a simulation study. The conventional Cox model that fails to account for heterogeneity overestimates the effect of treatment delay. Including interaction terms between treatment and starting time of treatment or between treatment and follow up time gave no improvement. Estimating a frailty term can improve the estimation, but is sensitive to misspecification of the frailty distribution. Therefore, multiple frailty distributions should be used and the results should be compared using the Akaike Information Criterion. Non-parametric estimation of the cumulative recovery percentages can be considered if the dataset contains sufficient long term follow up for each of the delay strategies. The methods are demonstrated on a motivating application evaluating the effect of delaying the start of treatment with assisted reproductive techniques on time-to-pregnancy in couples with unexplained subfertility.     

Dependence Functions and Frailties

This note considers association between nonnegative random variables in which the two observed survival times depend on an unobservable random variable via the proportional hazard model. When the random variables are subject to censoring, the conditional hazard functions provides a reasonable means of describing the association between the two variables. A numerical example demonstrating association in disease incidence in ordered pairs of individuals is analysed. Also, examples of distributions satisfying the notions of dependence considered are provided.     

Joint modeling of longitudinal and survival data via a common frailty

We develop a joint model for the analysis of longitudinal and survival data in the presence of data clustering. We use a mixed effects model for the repeated measures that incorporates both subject- and cluster-level random effects, with subjects nested within clusters. A Cox frailty model is used for the survival model in order to accommodate the clustering. We then link the two responses via the common cluster-level random effects, or frailties. This model allows us to simultaneously evaluate the effect of covariates on the two types of responses, while accounting for both the relationship between the responses and data clustering. The model was motivated by a study of end-stage renal disease patients undergoing hemodialysis, where we wished to evaluate the effect of iron treatment on both the patients' hemoglobin levels and survival times, with the patients clustered by enrollment site.     

How heritable is individual susceptibility to death? The results of an analysis of survival data on Danish, Swedish and Finnish twins.

Molecular epidemiological studies confirm a substantial contribution of individual genes to variability in susceptibility to disease and death for humans. To evaluate the contribution of all genes to susceptibility and to estimate individual survival characteristics, survival data on related individuals (eg twins or other relatives) are needed. Correlated gamma-frailty models of bivariate survival are used in a joint analysis of survival data on more than 31,000 pairs of Danish, Swedish and Finnish male and female twins using the maximum likelihood method. Additive decomposition of frailty into genetic and environmental components is used to estimate heritability in frailty. The estimate of the standard deviation of frailty from the pooled data is about 1.5. The hypothesis that variance in frailty and correlations of frailty for twins are similar in the data from all three countries is accepted. The estimate of narrow-sense heritability in frailty is about 0.5. The age trajectories of individual hazards are evaluated for all three populations of twins and both sexes. The results of our analysis confirm the presence of genetic influences on individual frailty and longevity. They also suggest that the mechanism of these genetic influences may be similar for the three Scandinavian countries. Furthermore, results indicate that the increase in individual hazard with age is more rapid than predicted by traditional demographic life tables.     

Modeling random effects for censored data by a multivariate normal regression model

A normal distribution regression model with a frailty-like factor to account for statistical dependence between the observed survival times is introduced. This model, as opposed to standard hazard-based frailty models, has survival times that, conditional on the shared random effect, have an accelerated failure time representation. The dependence properties of this model are discussed and maximum likelihood estimation of the model's parameters is considered. A number of examples are considered to illustrate the approach. The estimated degree of dependence is comparable to other models, but the present approach has the advantage that the interpretation of the random effect is simpler than in the frailty model.     

Testing independence between two sequential gap times in the presence of covariates

In the risk analysis of sequential events, the successive gap times are often correlated, e.g. as a result of an individual heterogeneity. Correlation is usually accounted for by using a shared gamma‐frailty model, where the variance φ of the random individual effect quantifies the correlation between gap times. This method is known to yield satisfactory estimates of covariate effects, but underestimates φ, which could result in a lack of power of the test of independence. We propose a new test of independence between two sequential gap times where the first is the time elapsed from the origin. The test is based on an approximation of the hazard of the second event given the first gap time in a frailty model, with a frailty distribution belonging to the power variance function family. Simulation results show an increased power of the new test compared with the test derived from the gamma‐frailty model. In the realistic case where hazards are event specific, and using event‐specific approaches, the proposed estimation of the variance of the frailty is less biased than the gamma‐frailty based estimation for a wide range of values ( with the set of parameters considered), and similar for higher values. As an illustration, the methods are applied to a previously analysed asthma prevention trial with results showing a significant positive association between the successive times to asthmatic events. We also analyse data from a cohort of HIV‐seropositive patients in order to assess the effect of risk factors on the occurrence of two successive markers of progression of the HIV disease. The results demonstrate the ability of the proposed model to account for negative correlations between gap times.     

Función y fragilidad: ¿qué tenemos que medir?

There is fairly widespread agreement on the distinction between frailty and disability. However, there is no validated and widely accepted concept of frailty, a condition that is frequently found in old age and which could constitute the main marker of pathological ageing. Given the absence of biological markers with good sensitivity and specificity that could be applied in practice, current approaches attempt to delimit the clinical manifestations that define vulnerable elderly individuals with a high risk of developing serious adverse events, including dependence, hospitalization, institutionalization, and death. Although the origin of frailty is multifactorial, its manifestations probably converge in an alteration in physical functioning. Early identification of the signs and symptoms related to this deterioration would contribute to the early detection of frail elderly individuals who could benefit from primary and secondary prevention, thus reducing or delaying the development of the above- mentioned adverse effects.     

EVOLUTION IN FLUCTUATING ENVIRONMENTS: DECOMPOSING SELECTION INTO ADDITIVE COMPONENTS OF THE ROBERTSON–PRICE EQUATION

We analyze the stochastic components of the Robertson–Price equation for the evolution of quantitative characters that enables decomposition of the selection differential into components due to demographic and environmental stochasticity. We show how these two types of stochasticity affect the evolution of multivariate quantitative characters by defining demographic and environmental variances as components of individual fitness. The exact covariance formula for selection is decomposed into three components, the deterministic mean value, as well as stochastic demographic and environmental components. We show that demographic and environmental stochasticity generate random genetic drift and fluctuating selection, respectively. This provides a common theoretical framework for linking ecological and evolutionary processes. Demographic stochasticity can cause random variation in selection differentials independent of fluctuating selection caused by environmental variation. We use this model of selection to illustrate that the effect on the expected selection differential of random variation in individual fitness is dependent on population size, and that the strength of fluctuating selection is affected by how environmental variation affects the covariance in Malthusian fitness between individuals with different phenotypes. Thus, our approach enables us to partition out the effects of fluctuating selection from the effects of selection due to random variation in individual fitness caused by demographic stochasticity.     

The response of amphibian larvae to environmental change is both consistent and variable

Many environments are undergoing rapid environmental change and there is a need to understand the mechanisms by which species can persist in altered environments. Model systems, such as amphibian metamorphosis, which can be generalized across many types of environmental change and across many species, are a powerful tool for understanding mechanisms that facilitate persistence in altered and disturbed environments. Amphibian larvae respond to environmental change by varying age at metamorphosis, or size at metamorphosis. Differential selection pressures on age or size at metamorphosis may result in a differential response among taxa to environmental change. Using a meta‐analysis, we investigated whether age at metamorphosis, size at metamorphosis, and larval growth rate vary within and among taxonomic families of amphibians in experiments that modified the environmental temperature, density of individuals, food, hydroperiod and the presence of predators. For all environmental factors except predators, the direction of the response was consistent across most of the studied taxa. However, there was considerable variation in effect size both within and among families. Results demonstrate that amphibian metamorphosis is a valuable model system for studying the effects of environmental change. Yet, we stress the need for caution in making generalizations about how individuals respond to environmental factors that have an indirect effect on physiology and require the perception of an environmental cue, such as the presence of predators. Synthesis As the current conditions of the environment are rapidly changing there is a need to understand how organisms respond to environmental change, and whether response of one species can be generalized to other species. Using a meta‐analyses, we tested whether the phenotypic response of amphibian larvae to five types of environmental change is consistent among and within taxonomic families. The phenotypic response to changes in environmental factors was consistent when the environmental factor has a direct effect on physiology, but varies among and within family if the environmental factor has an indirect effect on physiology or requires the perception of an environmental cue.     

A frailty model of segregation analysis: understanding the familial transmission of alcoholism

Coupled with environmental factors, genes contribute to numerous human diseases and traits. While there are many epidemiological methods to assess the familial clustering of traits, few are flexible enough to accommodate interactions between covariates and familial factors. In this paper, we propose and develop a frailty model that establishes an integrated framework to evaluate familial transmission of a disease by controlling for covariate effects and conveniently testing the interactions between covariates and familial factors. We also present a peeling algorithm that dramatically reduces the computational burden. This frailty model is employed to examine the familial transmission of major subtypes of alcoholism, namely, alcohol abuse and dependence. We conclude that alcohol dependence is strongly familial whereas alcohol abuse expresses a marginally significant pattern of familial transmission. Moreover, females manifest alcoholism at a lower threshold, and there is no sex-specific familial transmission of alcoholism after adjustment for the threshold effect.     

One operational definition by population: the need for local evaluations of frailty

Frailty is a health problem that increases the probability of developing adverse health outcomes in the elderly. A frequently used way to operationalize frailty is the construction of a frailty index, which is built from the addition of several health deficits that describe biological aging. However, there is no consensus about the number of health deficits for building a frailty index and about which deficits must be chosen. This lack of a standardized frailty index is assumed to be an obstacle for the advancement of research on frailty. The focus of the present article is to propose a theoretically plausible alternative way of operationalizing frailty by means of frailty indexes composed of deficits selected at a local level. These deficits would therefore be different for each given population. This "anthropological approach" is on the opposite side from current trends in frailty research, which is characterized by the search for a standardized operational definition of frailty. The anthropological approach would generate more reliable data by taking into account the specificity of the population to be studied for selecting frailty deficits. In this approach, emotions, motives, and beliefs are as important to determine individuals' health vulnerability as chronic diseases and physical function. Physiological anthropologists are well positioned to contribute to research on frailty by carrying out studies on the selection of the best deficits to operationalize frailty in different populations, with different socio-cultural determinants of health, and living in different environmental life spaces.     

Marginal hazard ratio estimates in joint frailty models for heart failure trials

This work is motivated by clinical trials in chronic heart failure disease, where treatment has effects both on morbidity (assessed as recurrent non‐fatal hospitalisations) and on mortality (assessed as cardiovascular death, CV death). Recently, a joint frailty proportional hazards model has been proposed for these kind of efficacy outcomes to account for a potential association between the risk rates for hospital admissions and CV death. However, more often clinical trial results are presented by treatment effect estimates that have been derived from marginal proportional hazards models, that is, a Cox model for mortality and an Andersen–Gill model for recurrent hospitalisations. We show how these marginal hazard ratios and their estimates depend on the association between the risk processes, when these are actually linked by shared or dependent frailty terms. First we derive the marginal hazard ratios as a function of time. Then, applying least false parameter theory, we show that the marginal hazard ratio estimate for the hospitalisation rate depends on study duration and on parameters of the underlying joint frailty model. In particular, we identify parameters, for example the treatment effect on mortality, that determine if the marginal hazard ratio estimate for hospitalisations is smaller, equal or larger than the conditional one. How this affects rejection probabilities is further investigated in simulation studies. Our findings can be used to interpret marginal hazard ratio estimates in heart failure trials and are illustrated by the results of the CHARM‐Preserved trial (where CHARM is the ‘Candesartan in Heart failure Assessment of Reduction in Mortality and morbidity’ programme).     

Testing homogeneity in Weibull-regression models

In survival studies with families or geographical units it may be of interest testing whether such groups are homogeneous for given explanatory variables. In this paper we consider score type tests for group homogeneity based on a mixing model in which the group effect is modelled as a random variable. As opposed to hazard-based frailty models, this model presents survival times that conditioned on the random effect, has an accelerated failure time representation. The test statistics requires only estimation of the conventional regression model without the random effect and does not require specifying the distribution of the random effect. The tests are derived for a Weibull regression model and in the uncensored situation, a closed form is obtained for the test statistic. A simulation study is used for comparing the power of the tests. The proposed tests are applied to real data sets with censored data.