[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
"Kieran McCaul" <Kieran.McCaul@uwa.edu.au> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
st: RE: RE: RE: Choosing control variables in survival analysis |

Date |
Tue, 16 Jun 2009 06:17:09 +0800 |

... -epiconf- was written for Stata Version 5, before -stcox- was developed. So, when it won't recognise -stcox-. So, you will probably have to model the data by hand, observing the change in the exposure HR estimate as you add potential confounders. Or you could try modifying -epiconf- so that it will accept -stcox-. Looking at the ado file, it doesn't look like this would be too difficult. The options for poisson, logit, and cox are simply handled by -if- statements in the ado file. If you located all the -- if "`model'"=="cox" -- blocks most of these could be changed to -- if "`model'"=="cox" | "`model'"=="stcox" -- The only exception would be -if- blocks that actually ran the model. For example, lines 354 to 356: else if "`model'" == "cox" { qui xi: `model' `list' ``i'', dead(`dead') } These would have to stay as they are, but immediately below them you would add: else if "`model'" == "stcox" { qui xi: `model' `list' ``i'' } Or, for example, lines 423 to 425 else if "`model'" =="cox" { qui xi: `model' `yvar' `xvar' `list', dead(`dead') } You would add the following below this: else if "`model'" =="stcox" { qui xi: `model' `xvar' `list' } And anywhere else where there is an -if- block that runs the old -cox- command. Also change line 60 to allow -stcox- as an option. ______________________________________________ Kieran McCaul MPH PhD WA Centre for Health & Ageing (M573) University of Western Australia Level 6, Ainslie House 48 Murray St Perth 6000 Phone: (08) 9224-2701 Fax: (08) 9224 8009 email: Kieran.McCaul@uwa.edu.au http://myprofile.cos.com/mccaul http://www.researcherid.com/rid/B-8751-2008 ______________________________________________ If you live to be one hundred, you've got it made. Very few people die past that age - George Burns -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Polis, Chelsea B. Sent: Tuesday, 16 June 2009 5:11 AM To: statalist@hsphsun2.harvard.edu Subject: st: RE: RE: Choosing control variables in survival analysis Many thanks for your response, Kieran! I am indeed interested in the effect of a specific variable on time to death. I tried to use epiconf, but think that it might not be possible for survival data with multiple records with gaps, since a singular "time" variable is required. Am I missing something? Cheers, Chelsea -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Kieran McCaul Sent: Wednesday, June 10, 2009 5:12 PM To: statalist@hsphsun2.harvard.edu Subject: st: RE: Choosing control variables in survival analysis Hi Chelsea, The strategy will differ depending on what the purpose of the survival analysis is. Broadly speaking, there are two different reasons for modelling data: (1) You are interested in the effect of a particular variable and you want to estimate this by accounting for the possible confounding effects of other variables, or (2) you are interested in finding a parsimonious set of predictor variables. In (1), statistical significance plays no role in selecting confounders. Confounding is a bias. A variable either confounds your exposure the estimate or it doesn't. In (2), the fit of the model is important, so the significance of the variables you choose will important. It sounds like you have an exposure that you're interested in, so I would suggest that you select variables based on their confounding effect. The strategy is called the "change-in-estimate" method. Have a look at: Greenland S (1989). Modeling and variable selection in epidemiologic analysis. Am J Public Health 79(3): 340-9. This outlines the strategy and Zhiqiang Wang wrote an ado file -epiconf-, which you may find useful. Also, a chi-square test of exposure against outcome will generally be of no use because it does not account for time to event. ______________________________________________ Kieran McCaul MPH PhD WA Centre for Health & Ageing (M573) University of Western Australia Level 6, Ainslie House 48 Murray St Perth 6000 Phone: (08) 9224-2701 Fax: (08) 9224 8009 email: Kieran.McCaul@uwa.edu.au http://myprofile.cos.com/mccaul http://www.researcherid.com/rid/B-8751-2008 ______________________________________________ Man is a credulous animal, and must believe something; in the absence of good grounds for belief, he will be satisfied with bad ones. Bertrand Russell -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Polis, Chelsea B. Sent: Thursday, 11 June 2009 2:04 AM To: statalist@hsphsun2.harvard.edu Subject: st: Choosing control variables in survival analysis Dear Statalisters, I am doing a survival analysis, which uses multiple records per individual to incorporate time-varying exposure information. I plan on building two multivariate Cox regression models: (Model A) includes potential confounders based strictly on statistical significance, and (Model B) includes those variables plus others I think should be included based on theoretical concerns or comparability to previous studies. I have been including a variable in Model A if a variable is associated with my dichotomous exposure (in chi2 tests) and time to event (in univariate Cox regression). However, I am not sure whether to include it in Model A if it is NOT significantly associated with time to event in Cox regression, but IS significantly associated with having HAD the event in a chi2 test? Your thoughts would be much appreciated. * * 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/ * * 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/ * * 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/ * * 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/

**References**:**st: Choosing control variables in survival analysis***From:*"Polis, Chelsea B." <cpolis@jhsph.edu>

**st: RE: Choosing control variables in survival analysis***From:*"Kieran McCaul" <Kieran.McCaul@uwa.edu.au>

**st: RE: RE: Choosing control variables in survival analysis***From:*"Polis, Chelsea B." <cpolis@jhsph.edu>

- Prev by Date:
**AW: st: Re: overlay option for predxcon** - Next by Date:
**Re: st: Re: Generating index numbers with panel data in long format** - Previous by thread:
**st: RE: RE: Choosing control variables in survival analysis** - Next by thread:
**st: dropping observation** - Index(es):

© Copyright 1996–2017 StataCorp LLC | Terms of use | Privacy | Contact us | What's new | Site index |