A Joint Model for Longitudinal and Time-to-event Data in Social and Life Course Research: Employment Status and Time to Retirement

Longitudinal studies including a time-to-event outcome in social research often use a form of event history analysis to analyse the influence of time-varying endogenous covariates on the time-to-event outcome. Many standard event history models however assume the covariates of interest to be exogenous and inclusion of an endogenous covariate may lead to bias. Although such bias can be dealt with by using joint models for longitudinal and time-to-event outcomes, these types of models are underused in social research. In order to fill this gap in the social science modelling toolkit, we introduce a novel Bayesian joint model in which a multinomial longitudinal outcome is modelled simultaneously with a time-to-event outcome. The methodological novelty of this model is that it concerns a correlated random effects association structure that includes a multinomial longitudinal outcome. We show the use of the joint model on Danish labour market data and compare the joint model to a standard event history model. The joint model has three advantages over a standard survival model. It decreases bias, allows us to explore the relation between exogenous covariates and the longitudinal outcome and can be flexibly extended with multiple time-to-event and longitudinal outcomes.

Shared Reading and Risk of Social-Emotional Problems

BACKGROUND AND OBJECTIVES: The American Academy of Pediatrics recommends literacy promotion as well as routine developmental surveillance during well-child visits to improve academic, relational, and health outcomes. In this study, we examined the possible association between shared reading and social-emotional problems among young children. METHODS: We conducted a retrospective review of longitudinal records for children aged 30 to 66 months presenting for visits to an academic pediatric primary care center between July 1, 2013, and February 1, 2019. The outcome was evidence of social-emotional problems, defined by an Ages and Stages: Social Emotional Questionnaire (ASQ:SE) score above the established cutoff. The predictor was caregiver-reported frequency of shared reading (most = 5–7 days per week, some = 2–4 days per week, rarely = 0–1 days per week) at a previous visit. Generalized linear models with generalized estimating equations were used to assess the association between the longitudinal outcome and predictor, adjusting for child demographics and needs reported on routine social history questionnaires. RESULTS: Analyses included 5693 children who completed at least 1 ASQ:SE (total of 7302 assessments) and had shared reading frequency documented before each ASQ:SE assessment. Children were predominantly Black (75%) and publicly insured (80%). Sixteen percent of ASQ:SE scores were suggestive of social-emotional concerns; 6% of caregivers reported sharing reading rarely. Children with rare shared reading had a higher risk of an ASQ:SE above cutoff compared with those with shared reading on most days (adjusted risk ratio, 1.62; 95% confidence interval, 1.35–1.92). CONCLUSIONS: Less-frequent caregiver-reported shared reading was associated with higher risk of social-emotional problems in young children presenting for primary care. This highlights potential relational and social-emotional benefits of shared reading.

Joint Modelling Approaches to Survival Analysis via Likelihood-Based Boosting Techniques

Joint models are a powerful class of statistical models which apply to any data where event times are recorded alongside a longitudinal outcome by connecting longitudinal and time-to-event data within a joint likelihood allowing for quantification of the association between the two outcomes without possible bias. In order to make joint models feasible for regularization and variable selection, a statistical boosting algorithm has been proposed, which fits joint models using component-wise gradient boosting techniques. However, these methods have well-known limitations, i.e., they provide no balanced updating procedure for random effects in longitudinal analysis and tend to return biased effect estimation for time-dependent covariates in survival analysis. In this manuscript, we adapt likelihood-based boosting techniques to the framework of joint models and propose a novel algorithm in order to improve inference where gradient boosting has said limitations. The algorithm represents a novel boosting approach allowing for time-dependent covariates in survival analysis and in addition offers variable selection for joint models, which is evaluated via simulations and real world application modelling CD4 cell counts of patients infected with human immunodeficiency virus (HIV). Overall, the method stands out with respect to variable selection properties and represents an accessible way to boosting for time-dependent covariates in survival analysis, which lays a foundation for all kinds of possible extensions.

Predictors of long-term prognosis in rheumatoid arthritis-related interstitial lung disease

The aim of the study was to identify specific clinical and serum protein biomarkers that are associated with longitudinal outcome of RA-associated interstitial lung disease（RA-ILD). 60 RA patients with clinical and serological profiles were assessed by HRCT and pulmonary function tests (PFTs) at baseline (Year 0) and 5 years post enrollment (Year 5). Progression versus non-progression was defined based on changes in Quantitative Modified HRCT scores and PFTs over time. Specific serum protein biomarkers were assessed in serum samples at baseline and Year 5 by Multiplex enzyme-linked immunosorbent assays (ELISAs). At Year 5, 32% of patients demonstrated progressive RA-ILD, 35% were stable, and 33% improved. Baseline age and rheumatoid factor (RF) were significantly different between RA-ILD outcomes of progression vs. no-progression (p< 0.05). Changes in levels of CXCL11/I-TAC and MMP13 over 5 years also distinguished pulmonary outcomes (p< 0.05). A final binary logistic regression model revealed that baseline age and changes in serum MMP13 were associated with RA-ILD progression at Year 5 (p< 0.05), with an AUC of 0.7569. Collectively, these analyses demonstrated that baseline clinical variables (age, RF) and shifts in levels of selected serum proteins (CXCL11/I-TAC, MMP13) were strongly linked to RA-ILD outcome over time.

Predictors of Long-term Prognosis in Rheumatoid Arthritis-related Interstitial Lung Disease

Objective. To identify specific clinical and serum protein biomarkers that are associated with longitudinal outcome of RA-associated interstitial lung disease（RA-ILD).Methods. 60 RA patients with clinical and serological profiles were assessed by HRCT and pulmonary function tests (PFTs) at baseline (Year 0) and 5 years post enrollment (Year 5). Progression versus non-progression was defined based on the changes in Quantitative Modified HRCT scores and PFTs over time. Specific serum protein biomarkers were assessed in serum samples at baseline and Year 5 by Multiplex enzyme-linked immunosorbent assays (ELISAs).Results. Based on changes of Quantitative Modified HRCT scores, 32% of patients demonstrated progressive RA-ILD, 35% were stable, and 33% improved. Baseline age and rheumatoid factor (RF) were significantly different between RA-ILD outcomes of progression versus no-progression (p=0.002 and 0.027, respectively). In multivariate analyses, changes in the serum levels of CXCL11/I-TAC and MMP13 from Year 0 to Year 5 (log Year 5-log Year 0) also distinguished RA-ILD outcomes (p< 0.05). A final binary logistic regression model revealed that baseline age and changes of MMP13 were associated with RA-ILD progression (Yes vs. No) at Year 5 (p< 0.05). Receiver Operating Characteristics analysis indicated that these variables predicted progression with an AUC of 0.7569.Conclusion. While baseline age and RF predicted RA-ILD outcomes of progression versus no-progression, changed levels of CXCL11/I-TAC and MMP-13 over 5 years were also correlated with long-term prognosis of RA-ILD. In multivariate analyses, baseline age and changed levels of MMP-13 were associated with the risk of progression of RA-ILD.Trial registrationNo