Re: st: Once more - combining confidence intervals?

["b. water" <barleywater@hotmail.com>]

Allan,

I hope I have not misunderstood the query in writing this response. One 
question beforehand though, if it is so heterogeneous, is it sensible to 
combine to produce a pooled estimate of the prevalence?

One approach is may be to pooled the point estimate like you would pooled in 
a meta-analysis and weighted it according to the number of males and females 
in each stratum. You then also do likewise with the upper and lower 
confidence interval limits. There is no specific user-written Stata ado file 
for this that I am aware of.

Another approach is Bayesian modelling (so I have been told when I 
considered this problem and similar problems in the past) using WinBUGS to 
obtain the point estimate of the prevalence and limit of the confidence 
intervals. However, one trip to its website enough to put me off, then and 
now.

Hope this help in some way and that you have not already considered it and 
or discounted it.

Best wishes,
BW





> From: Alan Kelly <akelly@gemini.tcd.ie>
> Reply-To: statalist@hsphsun2.harvard.edu
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Once more - combining confidence intervals?
> Date: Thu, 23 Jan 2003 10:28:50 +0000
>
> Dear all,
> A little surprised that I had no takers on the following question (posted 
> on the 17th).
> I'm trying again - just in case!
> best wishes,
> Alan Kelly
>
>
> Dear all,
> Following stratified modelling (log-linear) to derive a population 
> prevalence, I have a number of point prevalence estimates and their 
> associated confidence intervals for various age/sex groups (15 - 24 years, 
> 25 - 34 years and 35 - 54 years for males and females separately).  I need 
> to have an estimate of the prevalence for the entire 15 - 54 population.  
> Fitting a global model produces a very poor fit (due to heterogeneity).  
> Summing the point estimates to obtain an overall prevalence seems 
> reasonable. However, although I have seen this done, summing the individual 
> lower and upper confidence intervals to give a CI for the total prevalence 
> does not seem intuitive to me...for a start,  the  individual age/sex 
> standard errors are based on different numbers of individuals.
> Any suggestions as to how I could tackle the above would be appreciated.
> best wishes,
> Alan Kelly
> +-------------------------------------------------------+
>  Dr. Alan Kelly          E-mail akelly@mail.tcd.ie
>  Biostatistician           Phone:  +353-1-608 1385 or 608 1087
> 		  Fax:      +353-1-403 1211 or 403 1212
>
>
>
> Director of the Small Area Health Research Unit
>
> Department of Community Health & General Practice
>
> Trinity Center for Health Sciences
> AMNCH
> Tallaght Hospital
> Dublin 24
> Ireland
>
>
>  http://www.tcd.ie/Community_Health/SAHRU/index.html
> +--------------------------------------------------------+
> *
> *   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/

_________________________________________________________________
The new MSN 8: smart spam protection and 2 months FREE*  
http://join.msn.com/?page=features/junkmail

*
*   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/