R help - Dec 2007 - SAS PROC NLMIXED into R

Dear R friends

A while a go I sent an email to the epi-list and later to the help-list and
no answer could fully illuminate my question. So Im trying again with a more
specific matter.

Im trying to work on a script (function) to analyse data from a diagnostic
test meta-analysis with random effects. This was first described by an
author using SAS witn PROC NLMIXED.

Im not an expert in R and much worst in SAS.

So my obstacle right now how to translate this syntax, specially to choose
the correct function in R that better fit this SAS syntax with PROC NLMIXED.

nlme or lme4? or they both would work well?

Original SAS syntax (if it makes any difference?).....

proc mixed data=bi_meta method=reml cl ; /* option cl will give confidence
intervals */
   class study_id modality;
model logit  dis*modality non_dis*modality / noint cl df=1000, 1000, 1000,
1000, 1000, 1000;
random dis non_dis / subject=study_id type=un ;
repeated / group=rec;
parms / parmsdata=cov hold=4 to 91;

Best regards to all,

Pedro Emmanuel Brasil
Chagas disease clinical research laboratory
IPEC-FIOCRUZ
Rio de Janeiro - Brasil

	[[alternative HTML version deleted]]