Search

The case-cohort design allows analysis of multiple endpoints and only requires covariates to be measured for cases and non-cases in a random subcohort from the cohort. Stratification of subcohort sampling and weight calibration increase efficiency of estimates of log-relative hazards and covariate-specific pure risk, but they may require specifically adapted variance estimators. Some recent articles in epidemiology and medical journals used an inappropriate "robust" variance estimator. In addition, stratification, weight calibration and analysis of pure risk seem underutilized in case-cohort studies, possibly because practitioners are put off by the varied technical methodologic literature and lack of convenient software. We recently proposed a unified approach to variance estimation for Cox model log-relative hazards and pure risks, and we implemented it in an R package, CaseCohortCoxSurvival, available on CRAN, that allows appropriate and convenient analysis of case-cohort data, with and without stratification, weight calibration, or missing at random phase-two data. Here we illustrate how easy it is to use CaseCohortCoxSurvival to analyze data from the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial to estimate pure covariate-specific risk of prostate cancer incidence with various case-cohort design and analysis options. These analyses also indicate situations where the simple "robust" variance is too large.

The case-cohort design obtains complete covariate data only on cases and on a random sample (the subcohort) of the entire cohort. Subsequent publications described the use of stratification and weight calibration to increase efficiency of estimates of Cox model log relative hazards, and there has been some work estimating pure risk. Yet there are few examples of these options in the medical literature, and we could not find programs currently online to analyze these various options. We therefore present a unified approach and R software to facilitate such analyses. We used influence functions adapted to the various design and analysis options together with variance calculations that take the two-phase sampling into account. This work clarifies when the widely used "robust" variance estimate of Barlow is appropriate. The corresponding R software, CaseCohortCoxSurvival, facilitates analysis with and without stratification and/or weight calibration, for subcohort sampling with or without replacement. We also allow for phase-two data to be missing at random for stratified designs. We provide inference not only for log relative hazards in the Cox model, but also for cumulative baseline hazards and covariate-specific pure risks. We hope these calculations and software will promote wider use of more efficient and principled design and analysis options for case-cohort studies.

Studies of vaccine efficacy often record both the incidence of vaccine-targeted virus strains (primary outcome) and the incidence of non-targeted strains (secondary outcome). However, standard estimates of vaccine efficacy on targeted strains ignore the data on non-targeted strains. Assuming non-targeted strains are unaffected by vaccination, we regard the secondary outcome as a negative control outcome and show how using such data can (i) increase the precision of the estimated vaccine efficacy against targeted strains in randomized trials, and (ii) reduce confounding bias of that same estimate in observational studies. For objective (i), we augment the primary outcome estimating equation with a function of the secondary outcome that is unbiased for zero. For objective (ii), we jointly estimate the treatment effects on the primary and secondary outcomes. If the bias induced by the unmeasured confounders is similar for both types of outcomes, as is plausible for factors that influence the general risk of infection, then we can use the estimated efficacy against the secondary outcomes to remove the bias from estimated efficacy against the primary outcome. We demonstrate the utility of these approaches in studies of HPV vaccines that only target a few highly carcinogenic strains. In this example, using non-targeted strains increased precision in randomized trials modestly but reduced bias in observational studies substantially., Comment: Etievant, L., Sampson, J.N. and Gail, M.H. (2022) Increasing efficiency and reducing bias when assessing HPV vaccination efficacy by using nontargeted HPV strains. Biometrics. 1-12

The adaptive lasso refers to a class of methods that use weighted versions of the $L_1$-norm penalty, with weights derived from an initial estimate of the parameter vector to be estimated. Irrespective of the method chosen to compute this initial estimate, the performance of the adaptive lasso critically depends on the value of a hyperparameter, which controls the magnitude of the weighted $L_1$-norm in the penalized criterion. As for other machine learning methods, cross-validation is very popular for the calibration of the adaptive lasso, that this for the selection of a data-driven optimal value of this hyperparameter. However, the most simple cross-validation scheme is not valid in this context, and a more elaborate one has to be employed to guarantee an optimal calibration. The discrepancy of the simple cross-validation scheme has been well documented in other contexts, but less so when it comes to the calibration of the adaptive lasso, and, therefore, many statistical analysts still overlook it. In this work, we recall appropriate cross-validation schemes for the calibration of the adaptive lasso, and illustrate the discrepancy of the simple scheme, using both synthetic and real-world examples. Our results clearly establish the suboptimality of the simple scheme, in terms of support recovery and prediction error, for several versions of the adaptive lasso, including the popular one-step lasso., Comment: 23 pages, 4 figures

Partial Least Squares (PLS) refer to a class of dimension-reduction techniques aiming at the identification of two sets of components with maximal covariance, to model the relationship between two sets of observed variables $x\in\mathbb{R}^p$ and $y\in\mathbb{R}^q$, with $p\geq 1, q\geq 1$. Probabilistic formulations have recently been proposed for several versions of the PLS. Focusing first on the probabilistic formulation of the PLS-SVD proposed by el Bouhaddani et al., we establish that the constraints on their model parameters are too restrictive and define particular distributions for $(x,y)$, under which components with maximal covariance (solutions of PLS-SVD) are also necessarily of respective maximal variances (solutions of principal components analyses of $x$ and $y$, respectively). We propose an alternative probabilistic formulation of PLS-SVD, no longer restricted to these particular distributions. We then present numerical illustrations of the limitation of the original model of el Bouhaddani et al. We also briefly discuss similar limitations in another latent variable model for dimension-reduction.

For exposures $X$ like obesity, no precise and unambiguous definition exists for the hypothetical intervention $do(X = x_0)$. This has raised concerns about the relevance of causal effects estimated from observational studies for such exposures. Under the framework of structural causal models, we study how the effect of $do(X = x_0)$ relates to the effect of interventions on causes of $X$. We show that for interventions focusing on causes of $X$ that affect the outcome through $X$ only, the effect of $do(X = x_0)$ equals the effect of the considered intervention. On the other hand, for interventions on causes $W$ of $X$ that affect the outcome not only through $X$, we show that the effect of $do(X = x_0)$ only partly captures the effect of the intervention. In particular, under simple causal models (e.g., linear models with no interaction), the effect of $do(X = x_0)$ can be seen as an indirect effect of the intervention on $W$.

Many causal models of interest in epidemiology involve longitudinal exposures, confounders and mediators. However, repeated measurements are not always available or used in practice, leading analysts to overlook the time-varying nature of exposures and work under over-simplified causal models. Our objective is to assess whether - and how - causal effects identified under such misspecified causal models relates to true causal effects of interest. We derive sufficient conditions ensuring that the quantities estimated in practice under over-simplified causal models can be expressed as weighted averages of longitudinal causal effects of interest. Unsurprisingly, these sufficient conditions are very restrictive, and our results state that the quantities estimated in practice should be interpreted with caution in general, as they usually do not relate to any longitudinal causal effect of interest. Our simulations further illustrate that the bias between the quantities estimated in practice and the weighted averages of longitudinal causal effects of interest can be substantial. Overall, our results confirm the need for repeated measurements to conduct proper analyses and/or the development of sensitivity analyses when they are not available.