st: maximum likelihood inference for left censored exposure variable

["Floud, Sarah" <s.floud07@imperial.ac.uk>]

Using maximum likelihood inference for left-censored exposure variables:

For an epidemiological study, I have exposure data (noise exposure) which is left-censored, with about 10% of cases set to a limit of detection (LOD). I am currently using logistic regression to estimate the effect of the noise exposure on the health outcomes. But I would like to be able to infer values below the LOD for the exposure variable.
 
A method proposed by:
Cole et al 2009 ("Estimating the odds ratio when exposure has a limit of detection" International Journal of Epidemiology 2009 1-7 doi: 10.1093/ije/dyp269) and by:
Lynn, H.S. 2001 ("Maximum likelihood inference for left-censored HIV RNA data" Statistics in Medicine 2001; 20:33-45) 
is to use maximum likelihood inference for the values set at the LOD instead of using other methods such as setting all at the LOD or LOD/2 etc.

I would like to know if there is a way to do this in STATA without having to program it myself, as I think that would be beyond me. I currently have STATA IC 10.

Thank you for your consideration.

Sarah Floud




Sarah Floud
Faculty of Medicine
Imperial College London
St Mary's Campus, Norfolk Place
LONDON W2 1PG





*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/