Abstract
Missing data is common in longitudinal studies. We present a package for Farewell's Linear Increments Model for Missing Data (the FLIM package), which can be used to fit linear models for observed increments of longitudinal processes and impute missing data. The method is valid for data with regular observation patterns. The end result is a list of fitted models and a hypothetical complete dataset corresponding to the data we might have observed had individuals not been missing. The FLIM package may also be applied to longitudinal studies for causal analysis, by considering counterfactual data as missing data - for instance to compare the effect of different treatments when only data from observational studies are available. The aim of this article is to give an introduction to the FLIM package and to demonstrate how the package can be applied.
Highlights
Longitudinal data consist of repeated measurements recorded for a group of individuals over a given time period
The FLIM package may be applied to longitudinal studies for causal analysis, by considering the counterfactuals as missing data
Applications of the linear increments model in the FLIM package are done with the core function flim: flim(formula, data, id, obstime, t.values = NULL, method = "locf", lambda = NULL, art.cens=NULL)
Summary
Longitudinal data consist of repeated measurements recorded for a group of individuals over a given time period. Farewell (2006) and Diggle et al (2007) introduced the linear increments model as a tool for dealing with missing data due to drop-out in a dynamic manner, that is, explicitly considering the time order of the responses and where drop-out occurred. In the linear increments model, the discrete time-independent censoring (DTIC) assumption is adopted. The FLIM package may be applied to longitudinal studies for causal analysis, by considering the counterfactuals as missing data. Work on the linear increments model focused on handling monotone missingness (i.e. when individuals are missing and do not return to the study; typically referred to as drop-out). The approach of Aalen and Gunnes (2010), which we adopt, includes both monotone and nonmonotone missingness (i.e. when individuals are missing for a certain amount of time, but later return to the study), and the required assumptions for both types of missingness are discussed
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
