【正文】 , using( if _by==1, boxopt(mcolor(red))) using( if _by==2, boxopt(mcolor(green))) lcols(evrate) rcols(u_1_1_ V_1_1_) nooverall nostats nowt ( I s q u a r e d = 0 . 0 %, p = 0 . 9 6 4 )S u b t o t a lG E T C B 0 5 C B 8 6L I L L EE O R T C 0 8 8 6 1b e l g i u mA g g r e g a t e( I s q u a r e d = 0 . 0 %, p = 0 . 7 1 0 )S u b t o t a lK O R E AI T A L YG E T C B 0 4 C B 8 6S L O V E N I AMR C L U 1 1C A MSL C S G 7 7 3I P Dr e f e r e n c e n u mb e rt r i a l0 . 5 00 . 6 40 . 4 30 . 8 30 . 6 90 . 8 10 . 5 10 . 6 80 . 8 50 . 7 80 . 5 80 . 7 2r a t eE v e n t3 . 2 43 . 0 6 4 . 5 04 . 9 5 2 . 5 6 2 . 4 81 . 0 74 . 7 7o E ( o )2 2 9 . 62 2 . 41 3 . 23 1 . 61 5 . 65 9 . 44 4 . 94 1 . 0V ( o )1 4. 2 5Summary and conclusion ? IPD is increasingly used, and its advantages widely accepted ? Large numbers of MA scientists use twostage models for analysing IPD ? Currently only AD MA (. metan) and onestage IPD (. xtmixed) mands exist in Stata ? ipdmetan is a universal mand for twostage IPD MA ? forestplot is a flexible forest plot mand ? does not carry out analysis itself, thus not restricted by it ? may be useful outside the MA context (. presenting trial subgroups) Further information ? Other related programs (all call forestplot by default): ? admetan: calls ipdmetan to analyse AD (direct alternative to metan) ? ipdover: fit model within series of subgroups ? petometan: perform metaanalysis of timetoevent data using the Peto (logrank) method ? SSC and Stata Journal article in near future Thankyou! ? Questions, requests, bug reports: ? Thanks to: ? Jayne Tierney, Patrick Royston ? Ross Harris (author of metan) for advice amp。 save the dataset . qui ipdmetan, study(trialid) by(subgroup) nooverall nograph saving() : stcox arm if subgroup==1, strata(sex) (Aside: Contents of ) _use trialid _labels _ES _seES _lci _uci _wgt _NN 1 1 belgium 202 1 8 EORTC 08861 105 1 12 LILLE 163 1 15 GETCB 05CB86 539 Inclusion of aggregate data: Syntax . ipdmetan, study(trialid) eform nooverall ad(, byad) : stcox arm if subgroup==0, strata(sex) Do not pool IPD and aggregate together Aggregate data syntax estimation_mand “byad” = treat IPD amp。 extendable model structure Natural extension of AD MA Easily presentable in forest plots Applicable to any set of effect estimates and SEs (incl. interactions) Negligible difference to 1S in most mon scenarios Cons Requires more statistical expertise Challenging in certain situations, . randomeffects with timetoevent data Not a natural fit with forest plots Only a single estimate can be pooled, which limits plexity (. interactions) Theoretically inferior in (at least) some scenarios Example data ? IPD MA of randomised trials of postoperative radiotherapy (PORT) in nonsmall cell lung cancer ? Trial ID (k=11) ? Patient ID (n=2343) ? Treatment arm ? Oute is censored time to overall survival (death from any cause) ? Time to event (from randomisation) ? Event type (death or censorship) ? Certain covariate measurements also available, not necessarily for all trials or patients ? Disease stage (factor, but treat as continuous) ? (+ others) ipdmetan syntax ipdmetan, study(trialid) eform : stcox arm, strata(sex) ipdmetan options after ma, before colon estima