Abstract
SummaryMany psoriatic arthritis patients do not progress to permanent joint damage in any of the 28 hand joints, even under prolonged follow‐up. This has led several researchers to fit models that estimate the proportion of stayers (those who do not have the propensity to experience the event of interest) and to characterize the rate of developing damaged joints in the movers (those who have the propensity to experience the event of interest). However, when fitted to the same data, the paper demonstrates that the choice of model for the movers can lead to widely varying conclusions on a stayer population, thus implying that, if interest lies in a stayer population, a single analysis should not generally be adopted. The aim of the paper is to provide greater understanding regarding estimation of a stayer population by comparing the inferences, performance and features of multiple fitted models to real and simulated data sets. The models for the movers are based on Poisson processes with patient level random effects and/or dynamic covariates, which are used to induce within‐patient correlation, and observation level random effects are used to account for time varying unobserved heterogeneity. The gamma, inverse Gaussian and compound Poisson distributions are considered for the random effects.
Highlights
Several researchers (Aguirre-Hernandez and Farewell (2004), Solis-Trapala and Farewell (2005) and recently O’Keeffe et al (2012)) have considered the existence of a stayer population with regard to clinical joint damage, after having discussed the clinical plausibility of such a population with Professor Dafna Gladman, who has established the largest and most comprehensively studied cohort of psoriatic arthritis patients in the world
After accounting for correlation through the multiplicative patient level random effect, it is natural to postulate that the negative association is indicative of there being fewer joints having the propensity to become damaged
As the patient level random effect is designed to reflect partly this type of correlation, the effect of this dynamic covariate will probably be confounded with the random effects
Summary
Several researchers (Aguirre-Hernandez and Farewell (2004), Solis-Trapala and Farewell (2005) and recently O’Keeffe et al (2012)) have considered the existence of a stayer population (those who do not have the propensity to experience the event of interest) with regard to clinical joint damage, after having discussed the clinical plausibility of such a population with Professor Dafna Gladman, who has established the largest and most comprehensively studied cohort of psoriatic arthritis patients in the world. This paper demonstrates that, when fitted to the same data (real psoriatic arthritis data and simulated data), these models can provide widely varying conclusions on the existence and proportion of a stayer population, demonstrating that, as is commonly done in the literature, a single analysis of a stayer population should not generally be adopted uncritically In light of these results and given the prevalence of longitudinal count data with excess 0s, this paper aims to provide greater understanding of mover–stayer counting process models, relating to estimation of the stayer proportion, by comparing the inferences, performance and model features of many models fitted to real and simulated data sets. We note that these results are analogous to those obtained on cure rate models (Boag, 1949; Farewell, 1977, 1982, 1986; Gordan, 1990; Ghitany et al, 1994) where it has been shown that the estimate of the cured fraction can be quite sensitive to the choice of the survival distribution (Yu et al, 2004)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Journal of the Royal Statistical Society Series C: Applied Statistics
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
