Time T 0 time of first test Lots of reasons for variability eg visit GP only when ill GP decides to measure Y why measurement repeated soon because it was abnormal outcome dependent ID: 558230
Download Presentation The PPT/PDF document "Informative measurement process in routi..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Informative measurement process in routinely collected health data
Time
T
0
(time of first test)
Lots of reasons for variability, e.g.
- visit GP
only
when ill!
- GP decides to measure Y (why?)
- measurement repeated soon because it was abnormal (outcome dependent)
Matthew.Sperrin@Manchester.ac.ukSlide2
Informative measurement process in routinely collected health data
Outcome process y(t) = f(x(t),u)
Measurement process (hazard) h(t)=
(y(t),x(t),u)
May be interested in both – other may be a ‘nuisance process’
‘ignorable’ measurement process missing at random (MAR)‘nonignorable’ not missing at random (NMAR) Matthew.Sperrin@Manchester.ac.uk