st: CI of an index of quality performance based on logistic regression models

["Kaaresen Per Ivar" <Per.Ivar.Kaaresen@unn.no>]

Ikke sensitiv - Please ignore this message - due to in-house  security    

Dear all

Hosmer and Lemeshow developed a method of calculating CI estimates of an index of quality performance based on logistic regression models when you only have access to published coefficients of the fitted model (Statistics in Medicine 1995;14:2161-2172).

You can take the coefficients and input them in a _matrix_  as described in (R) logistic and then use the _lfit_ with the beta(b) option to obtain the expected events and then calculate the ratio Observed/Expected. 

My question is: How can you calculate the CI's as described by Hosmer and Lemeshow in Stata (I need what they called the S1 parametric method). Has anyone a do/ado file to perform this task or can anybody explain how I'm doing this?

Thank you in advance

Regards

Per Ivar Kaaresen MD
Tromsoe
Norway



*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/