A Bayesian joint model for zero‐inflated integers and left‐truncated event times with a time‐varying association: Applications to senior health care

Abstract

Population aging in most industrialized societies has led to a dramatic increase in emergency medical demand among the elderly. In the context of private health care, an optimal allocation of the medical resources for seniors is commonly done by forecasting their life spans. Accounting for each subject's particularities is therefore indispensable, so the available data must be processed at an individual level. We use a large and unique dataset of insured parties aged 65 and older to appropriately relate the emergency care usage with mortality risk. Longitudinal and time-to-event processes are jointly modeled, and their underlying relationship can therefore be assessed. Such an application, however, requires some special features to also be considered. First, longitudinal demand for emergency services exhibits a nonnegative integer response with an excess of zeros due to the very nature of the data. These subject-specific responses are handled by a zero-inflated version of the hierarchical negative binomial model. Second, event times must account for the left truncation derived from the fact that policyholders must reach the age of 65 before they may begin to be observed. Consequently, a delayed entry bias arises for those individuals entering the study after this age threshold. Third, and as the main challenge of our analysis, the association parameter between both processes is expected to be age-dependent, with an unspecified association structure. This is well-approximated through a flexible functional specification provided by penalized B-splines. The parameter estimation of the joint model is derived under a Bayesian scheme. K E Y W O R D S age-related mortality hazard, left truncation, morbidity, private health care, shared-parameter joint model, time-varying association, zero inflation 1 INTRODUCTION 1.1 Medical motivation Nearly all developed countries have experienced a gradual population aging process over the course of the past decades. This has led to a substantial increase in elderly individuals' demand for health care, along with the related financial

Healthcare coordination and medical expenditures for Sickle Cell Disease patients with different levels of health utilization risk

“…Because many SCD patients had no inpatient expenditures, a two-part expenditure analysis based on a statistical decomposition of the distribution of the outcome into a process that generates zeros and a process that generates non-zero positive values 35 was conducted using the GLMMadaptive (v0.7.15) R package. The analysis accommodates the semi-continuous expenditure data; that is, a continuous model allowing for data with excess zeros was fitted to the data 36,37 .…”

Modeling unmeasured baseline information in observational time-to-event data subject to delayed study entry

“…One advantage of Bayesian penalized splines is that the choice of the number of knots is not critical because overfitting may be corrected by the penalty term. 25 As explained earlier and investigated in more depth in the remainder of the paper, the novelty of the application of model ( 1) and model ( 2) is that the combination of both models is not used to estimate the effect of the time-dependent value m i (t), typically with right-censored data, but utilizes the entire available longitudinal covariate information to recover missing baseline information m i (0) in left-truncated data, such that the baseline effect 𝜇 can properly be estimated. In other words, our aim is a standard Cox analysis with baseline covariates, which, however, is unfeasible as a consequence of left truncation.…”

A Bayesian multilevel time‐varying framework for joint modeling of hospitalization and survival in patients on dialysis

“…Furthermore, none of the aforementioned joint modeling approaches for three‐level hierarchical data consider modeling time‐varying effects. More specifically, only few works in the joint modeling literature allow for time‐varying effects, either time‐varying response‐predictor relationships or time‐varying relationships between the outcomes modeled 19‐21 . However, these methods only allow for a two‐level hierarchy, that is, repeated measurements nested within subjects.…”

“…More specifically, only few works in the joint modeling literature allow for time-varying effects, either time-varying response-predictor relationships or time-varying relationships between the outcomes modeled. [19][20][21] However, these methods only allow for a two-level hierarchy, that is, repeated measurements nested within subjects. Our proposed Bayesian multilevel time-varying joint model (BMT-JM) accounts for time-varying effects of multilevel risk factors on both hospitalization and mortality as well as the time-varying relationship between the two outcomes, while also accommodating the three-level hierarchical data structure of the USRDS data through multilevel REs.…”