Dear Kevin, Im not familiar with the package metafor. Im familiar with package meta, although, which gives nice graphical output. I did some analysis with observational studies with stata back in 2006 in my master degree. Similar to your case I did have data from case-control, cross-sectional, and cohort investigations. At the time, the odds ratio were very popular to summarize data for trials too. Therefore I decided to use the OR for all investigations in way compare results and summarize them all. OR may not give the exact same results as RR but may give results close enough. Another way around is not summarize considering the potential bias in the summary estimate and the few studies included in the review. :(
Considering the earlier answer by Wolfgan, you could stratifiy the result by design if possible or you could estimate the standrd error in each point estimate of each investigation included and use the metagen function in meta package. The input of this function is OR (or RR) and SE (or logSE memory failed :-) ). Perhaps it will help if I got the problem right! Good luck, Dr. Pedro Emmanuel A. A. do Brasil Instituto de Pesquisa Clínica Evandro Chagas Fundação Oswaldo Cruz Rio de Janeiro - Brasil 2010/1/27 <r-sig-epi-requ...@stat.math.ethz.ch> > Send R-sig-Epi mailing list submissions to > r-sig-epi@stat.math.ethz.ch > > To subscribe or unsubscribe via the World Wide Web, visit > https://stat.ethz.ch/mailman/listinfo/r-sig-epi > or, via email, send a message with subject or body 'help' to > r-sig-epi-requ...@stat.math.ethz.ch > > You can reach the person managing the list at > r-sig-epi-ow...@stat.math.ethz.ch > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of R-sig-Epi digest..." > > > Today's Topics: > > 1. meta-analysis (Lu, Kevin) > 2. Re: meta-analysis (Viechtbauer Wolfgang (STAT)) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 27 Jan 2010 09:25:05 +0100 > From: "Lu, Kevin" <kevin...@wur.nl> > To: <r-sig-epi@stat.math.ethz.ch> > Subject: [R-sig-Epi] meta-analysis > Message-ID: > <124584afb3a4d449834b67071accd55501a7d...@scomp0040.wurnet.nl> > Content-Type: text/plain > > To someone who could help me: > > I am a PhD student, working in the Wageningen University, the > Netherlands. I am now using metafor package to do meta-analysis. > > Since there are few studies conducted on my topic, I have to combine the > results from 5 case-control studies and 1 cohort study. But I do not > know how to do it. > > I tried this way, first, use the escalc (measure="OR", ai, bi, ci, gi) > to calculate the effect estimate of yi and variance of vi. > > I could extract the RR and 95% confidence interval from the cohort study > to get the point estimate of LnRR and SE. > > but I can not combine the results from case-control study and the > results from cohort study due to the difference of vi and SE. > > For example, if I use rma(yi, vi, data,), I can not get vi (variance) > for the cohort study. I can only get SE for cohort study. How could > solve it? > > Thanks for your time. > > Best wishes, > > Yingchang (Kevin) Lu > > PhD student, > > Division of Human Nutrition > > Wageningen University > > Tel: 0031-317-485300 > > Fax:0031-317-482782 > > > [[alternative HTML version deleted]] > > > > ------------------------------ > > Message: 2 > Date: Wed, 27 Jan 2010 10:43:05 +0100 > From: "Viechtbauer Wolfgang (STAT)" > <wolfgang.viechtba...@stat.unimaas.nl> > To: "r-sig-epi@stat.math.ethz.ch" <r-sig-epi@stat.math.ethz.ch> > Subject: Re: [R-sig-Epi] meta-analysis > Message-ID: > <077e31a57da26e46ab0d493c9966ac7305c6052...@um-mail4112.unimaas.nl> > Content-Type: text/plain; charset="us-ascii" > > Dear Kevin, > > The standard error of the estimate (SE) is the square root of the sampling > variance of the estimate (vi). So, just square 'SE' and use that for 'vi'. > > Best, > > -- > Wolfgang Viechtbauer http://www.wvbauer.com/ > Department of Methodology and Statistics Tel: +31 (43) 388-2277 > School for Public Health and Primary Care Office Location: > Maastricht University, P.O. Box 616 Room B2.01 (second floor) > 6200 MD Maastricht, The Netherlands Debyeplein 1 (Randwyck) > > > ----Original Message---- > From: r-sig-epi-boun...@stat.math.ethz.ch > [mailto:r-sig-epi-boun...@stat.math.ethz.ch] On Behalf Of Lu, Kevin > Sent: Wednesday, January 27, 2010 09:25 To: r-sig-epi@stat.math.ethz.ch > Subject: [R-sig-Epi] meta-analysis > > > To someone who could help me: > > > > I am a PhD student, working in the Wageningen University, the > > Netherlands. I am now using metafor package to do meta-analysis. > > > > Since there are few studies conducted on my topic, I have to combine > > the results from 5 case-control studies and 1 cohort study. But I do > > not know how to do it. > > > > I tried this way, first, use the escalc (measure="OR", ai, bi, ci, > > gi) to calculate the effect estimate of yi and variance of vi. > > > > I could extract the RR and 95% confidence interval from the cohort > > study to get the point estimate of LnRR and SE. > > > > but I can not combine the results from case-control study and the > > results from cohort study due to the difference of vi and SE. > > > > For example, if I use rma(yi, vi, data,), I can not get vi (variance) > > for the cohort study. I can only get SE for cohort study. How could > > solve it? > > > > Thanks for your time. > > > > Best wishes, > > > > Yingchang (Kevin) Lu > > > > PhD student, > > > > Division of Human Nutrition > > > > Wageningen University > > > > Tel: 0031-317-485300 > > > > Fax:0031-317-482782 > > > > > > [[alternative HTML version deleted]] > > > > _______________________________________________ > > R-sig-Epi@stat.math.ethz.ch mailing list > > https://stat.ethz.ch/mailman/listinfo/r-sig-epi > > > > ------------------------------ > > _______________________________________________ > R-sig-Epi mailing list > R-sig-Epi@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/r-sig-epi > > > End of R-sig-Epi Digest, Vol 38, Issue 3 > **************************************** > [[alternative HTML version deleted]]
_______________________________________________ R-sig-Epi@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-epi
