Statistics in Medicine最新文献

Panel count data is common in cancer screening. In the context of colorectal cancer screening, our work focuses on the prediction of the probability of advanced adenoma conditional on patient-level risk factors and/or event history. We implement the joint frailty model proposed by Huang et al. which involves a non-stationary Poisson process for recurrent adenoma events and informative screening time using semi-parametric Cox models correlated by a latent frailty variable, where coefficients and baseline intensity functions are estimated by estimating equations. Subject-specific frailty value is estimated by the borrow-strength method. In addition, marginal models for the adenoma and screening events are also applicable when average covariate effects on the population level are of interest. Predictions based on the marginal model and predictions based on the frailty models for patients with or without a screening history are compared. When a patient's screening history is available and sufficient adenoma events are observed, the predictions based on the frailty model with estimated subject-specific frailty are superior. However, in the cases of early censoring when adenoma events are not observed for most patients and screening history is not available, the prediction based on the marginal model has better performance. Furthermore, for patients' future screening, the individualized screening intervals derived from the dynamic predictions of advanced adenoma risks will detect adenoma events earlier with shorter lag time between the adenoma state transition and the screening compared to the current practice of regular screening intervals.

Risk differences allow decision makers to easily estimate the excess safety risk associated with a medical product relative to the potential benefits. However, in post-market observational surveillance studies that actively monitor (e.g., sequentially over time) for safety risk of new medical products, available methods target a relative measure (e.g., odds ratio and relative risk), which can be especially unstable in the rare event setting. These studies are typically conducted within distributed healthcare networks (e.g., Food and Drug Administration [FDA] Sentinel and Centers for Disease Control [CDC] Vaccine Safety Datalink) with patient-level data protected behind firewalls, but sharing of aggregate, deidentified data for centralized analyses. We propose an inverse probability of treatment weighting (IPTW) method that uses site-specific propensity scores to estimate site-specific risk differences that are combined to create an overall stratified risk difference estimate. This method is tailored to the rare event setting and requires minimal data sharing. The stratified IPTW approach is then extended to the active post-market surveillance setting by incorporating group sequential monitoring boundaries using a novel permutation approach. A simulation study is conducted to evaluate the performance of the new methods relative to two centralized analysis approaches, and the methods are applied to a safety surveillance study comparing the risk of febrile seizure between two vaccines using FDA Sentinel Data from three healthcare organizations.

The ability to retrieve pseudo-individual patient data (IPD) from published survival study results is important to facilitate meta-analysis, evidence synthesis or secondary data analyses for the purpose of decision modeling for cost effectiveness analysis. While established methods exist for retrieving pseudo-IPD from Kaplan-Meier plots, these algorithms are not easily extendable to other types of survival data, nor do they allow all available information to be incorporated. An optimization-based approach is proposed where the task of reconstructing the IPD is formulated as a quadratic program (QP) with linear constraints. The method easily allows auxiliary information such as marked censoring times. Moreover, the same approach can be used to reconstruct patient-level competing risks survival data from published cumulative incidence functions. In simulation, the QP-based method is shown to outperform existing algorithms particularly when data on numbers at risk and marked censoring times are available. The methods are illustrated through reconstruction of data from a published study on patients with advanced stage follicular lymphoma.

Cause-of-death data is fundamental for understanding population health trends and inequalities as well as designing and evaluating public health interventions. A significant proportion of global deaths, particularly in low- and middle-income countries (LMICs), do not have medically certified causes assigned. In such settings, verbal autopsy (VA) is a widely adopted approach to estimate disease burdens by interviewing caregivers of the deceased. Recently, latent class models have been developed to model the joint distribution of symptoms and perform probabilistic cause-of-death assignment. A large number of latent classes are usually needed in order to characterize the complex dependence among symptoms, making the estimated symptom profiles challenging to summarize and interpret. In this paper, we propose a flexible Bayesian tensor decomposition framework that balances the predictive accuracy of the cause-of-death assignment task and the interpretability of the latent structures. The key to our approach is to partition symptoms into groups and model the joint distributions of group-level symptom sub-profiles. The proposed methods achieve better predictive accuracy than existing VA methods and provide a more parsimonious representation of the symptom distributions. We show our methods provide new insights into the clustering patterns of both symptoms and causes using the PHMRC gold-standard VA dataset.

The envelope model provides a dimension-reduction framework for multivariate linear regression. However, existing envelope methods typically assume normally distributed random errors and do not accommodate repeated measures in longitudinal studies. To address these limitations, we propose the robust longitudinal envelope model (RoLEM). RoLEM employs a scale mixture of matrix-variate normal distributions to model random errors, allowing it to handle potential outliers, and incorporates flexible correlation structures for repeated measurements. In addition, we introduce new prior and proposal distributions on the Grassmann manifold to facilitate Bayesian inference for RoLEM. Simulation studies and real data analysis demonstrate the superior performance of the proposed method.

Monte Carlo simulations are an important tool in modern statistical research. The data-generating process is foundational to any simulation. In survival analysis, a competing risk is an event whose occurrence precludes the occurrence of the primary event of interest. Two data-generating processes have been described for simulating competing risk data: one based on all the cause-specific hazard functions for the different types of events, and one based on a subdistribution hazard model for the primary event of interest. There is a paucity of research on the impact of the choice of data-generating process. We used a series of Monte Carlo simulations to evaluate the impact of the choice of data-generating process on the performance of prediction models when assessing discrimination using the time-dependent AUC and accuracy using the time-dependent Brier score. We also assessed the impact of the choice of competing risk regression used for computing smoothed event probabilities for use when computing the calibration metrics ICI (integrated calibration index), E50, and E90. The impact of discordance between the fitted model and the data-generating process on both the time-dependent AUC and the time-dependent Brier score was minimal. When computing the ICI, E50, and E90, we recommend that researchers use a model for computing smoothed event probabilities that is concordant with the type of model whose calibration is being assessed.

Difference-in-differences (DID) is popular because it can allow for unmeasured confounding when the key assumption of parallel trends holds. However, there exists little guidance on how to decide a priori whether this assumption is reasonable. We attempt to develop such guidance by considering the relationship between a causal diagram and the parallel trends assumption. This is challenging because parallel trends is scale-dependent and causal diagrams are generally scale-independent. We develop conditions under which, given a nonparametric causal diagram, one can reject or fail to reject parallel trends. In particular, we adopt a linear faithfulness assumption, which states that all graphically connected variables are correlated, and which is often reasonable in practice. We show that parallel trends can be rejected if either (i) the treatment is affected by pre-treatment outcomes, or (ii) there exist unmeasured confounders for the effect of treatment on pre-treatment outcomes that are not confounders for the post-treatment outcome, or vice versa. We also argue that parallel trends should be strongly questioned if (iii) the pre-treatment outcomes causally affect the post-treatment outcomes, since there exist reasonable semiparametric models in which such an effect violates parallel trends. When (i-iii) are absent, a necessary and sufficient condition for parallel trends is that the association between unmeasured confounders and potential outcomes is constant on an additive scale, pre- and post-treatment. We discuss our approach in the context of the effect of Medicaid expansion under the US Affordable Care Act on health insurance coverage rates.

The absolute change in the angle measured immediately after surgery and after bone healing is a clinically relevant endpoint to judge the stability of an osteotomy. Assuming that the difference in angles is normally distributed, the absolute difference follows a folded normal distribution. The confidence interval (CI) for the absolute angle change of a novel fixation screw compared to a standard fixation screw may be used for evaluating non-inferiority. We suggest that the Welch statistic may serve as the basis for CI calculations of the difference between two folded normals. The coverage probabilities of derived CIs are investigated by simulations. We illustrate the approaches with data from a randomized controlled trial and an observational study on bunion surgery, in which magnesium-based and titanium-based fixation screws were compared. In the simulation studies, asymptotic and both non-parametric and parametric bootstrap CIs based on the Welch statistic were close to nominal coverage levels. In case of different sample sizes between groups, the t-statistic-based CIs were not always meeting the nominal coverage levels. Methods based on chi-square distributions were not deemed appropriate for comparing two folded normals. The re-analysis of the bunion trial permitted the conclusion of non-inferiority for the primary endpoint, the absolute difference between baseline and six months follow-up for the distal metatarsal articular angle, between magnesium-based and titanium-based fixation screws. We recommend the use of CIs based on the Welch statistic to evaluate non-inferiority in trials where the stability of angles after osteotomy is to be compared.

Factor analysis (FA) can be used to identify key biomarkers in biological processes by assuming that latent biological pathways (statistically, "latent factors") drive the activity of measurable biomarkers ("observed variables"). However, biological pathways often interact, meaning that the classical FA assumption of independence between factors is questionable. Motivated by sparsely and irregularly collected longitudinal measurements of metabolites in a COVID-19 study, we propose a dynamic factor analysis model that accounts for cross-correlations between pathways via a multi-output Gaussian processes (MOGP) prior on the factor trajectories. To mitigate against overfitting caused by sparsity of longitudinal measurements, we introduce a roughness penalty upon MOGP hyperparameters and allow for non-zero mean functions. We also propose a scalable stochastic expectation maximization (StEM) algorithm that, in simulations, is both 20 times faster and provides more accurate and stable MOGP hyperparameter estimates than a previously-proposed Monte Carlo Expectation Maximization algorithm. In the motivating COVID-19 study, our methodology identifies a kynurenine pathway that affects the clinical severity of patients with COVID-19 disease and uncovers the role of the biomarker taurine. Our R package DFA4SIL implements the proposed method.

Survival trees are popular alternatives to Cox or Aalen regression models that offer both modeling flexibility and graphical interpretability. This paper introduces a new algorithm for survival trees that relaxes the assumption of independent censoring. To this end, we use the copula-graphic estimator to estimate survival functions. This allows us to flexibly specify shape and strength of the dependence of survival and censoring times within survival trees. For splitting, we present a permutation test for the null hypothesis of equal survival. Our test statistic consists of the integrated absolute distance of the groups' copula-graphic estimators. A first simulation study shows a good type I error and power behavior of the new test. We thereby assess simulation settings of various group sizes, censoring percentages, and grades of dependence generated by Clayton and Frank copulas. Using this test as a splitting criterion, a second simulation study studies the performance of the resulting trees and compares it with that of the usual logrank-based tree. Lastly, the tree algorithm is applied to real-world clinical trial data.