Longitudinal Missing Data

Longitudinal cohorts have been widely used in epidemiological studies. These cohorts have been vital for the discovery of important risk factors associated with various diseases and public health issues in the last few decades (refs). Long term follow up of these cohorts will continue to provide valuable data that will lead to the discovery of novel risk factors associated with disease.
One of the most important issue with longitudinal studies is the missing data at each follow up time (refs). This may occur in several forms such as, death of the patient, participants recruited at the start of the study drop out from the follow up studies due to various reasons, participants may intermittently fail to attend some follow ups. It is important to consider these scenarios of missing data separately. (Give two examples to show the importance) …show more content…
Missing data can introduce potential bias in parameter estimation and weaken the generalizability of the results ( refs ). Ignoring cases with missing data is not a n efficient option as it leads to the loss of information which in turn decreases statistical power and increases standard errors ( refs ). Significant research has been done in handling missing data (refs). However, several studies reported in top medical journals do not use appropriate methods to handle missing data (refs). A survey of journal editors has shown that improper handling of missing data in manuscripts is a significant concern for evaluating studies. Missing data is generally not reported properly, or curated before analysis to exclude individuals with missing data ignoring missingness. In the case where missing data were considered the assumptions of missingness were not tested