A nice connection from last week’s journal club paper to this week’s: the errors in health information system data. Last week was about correcting the bias from missing individuals. This week is about correcting the bias from missing facilities.
From the key figure, it looks like missing individuals bias things more:
This week’s paper for journal club is short and has a nice figure:
Geng et al, Sampling-Based Approach to Determining Outcomes of Patients Lost to Follow-Up in Antiretroviral Therapy Scale-Up Programs in Africa
It looks like this corrected estimate is quite different than the uncorrected version!
I think the mathematics involved have an extended treatment in this work referenced by Geng et al: Addressing an idiosyncrasy in estimating survival curves using double-sampling in the presence of self-selected right censoring