Characteristic	Kahn 2006	Ko 2006	Lebovitz 2001	Ovalle 2004	Phillips 2001	Raskin 2004
Intervention 1 (I1) / intervention 2 (I2) / control 1 (C1)	I1: rosiglitazone I2: metformin C1: glyburide	I1: rosiglitazone + (sulfonylurea +/‐ metformin) C1: bedtime isophane insulin + (sulfonylurea +/‐ metformin)	I1: rosiglitazone 2 mg I2: rosiglitazone 4 mg C1: placebo	I1: rosiglitazone + glimepiride + metformin + C1: glimepiride + metformin + 70/30 mixed human insulin	I1: rosiglitazone 4 mg od I2: rosiglitazone 2 mg bid I3: rosiglitazone 8 mg od I4: rosiglitazone 4 mg bid C1: placebo	I1: rosiglitazone I2: repaglinide C1: rosiglitazone + repaglinide
Randomised controlled clinical trial (RCT)	Y	Y	Y	Y	Y	Y
Non‐inferiority / equivalence trial	N	N	N	N	Primary hypothesis: superiority of rosiglitazone vs placebo; secondary hypothesis: equivalence of once daily vs twice daily administration of rosiglitazones	?
Controlled clinical trial	N	N	N	N	N	N
Design: parallel study	Y	Y	Y	Y	Y	Y
Design: crossover study	N	N	N	N	N	N
Design: factorial study	N	N	N	N	N	N
Crossover study: wash‐out phase	NA	NA	NA	NA	NA	NA
Crossover study: carryover effect tested	NA	NA	NA	NA	NA	NA
Crossover study: period effect tested	NA	NA	NA	NA	NA	NA
Method of randomisation	stratified according to sex in blocks of six	?	?	?	?	?
Unit of randomisation (individuals, cluster ‐ specify)	individuals	individuals	individuals	individuals	individuals	individuals
Randomisation stratified for centres	N	NA	?	NA	?	?
Randomisation ratio	1:1:1	1:1	1:1	1:1	1:1:1:1:1	1:1:2
Concealment of allocation	Y	?	?	?	?	?
Stated blinding (open; single, double, triple blind)	double‐bind	open	double‐blind	?	double‐blind	open
Actual blinding: participant	?	NA	Y	?	?	N
Actual blinding: caregiver / treatment administrator	?	NA	?	?	?	N
Actual blinding: outcome assessor	?	?	?	?	?	?
Actual blinding: others	Y	?	N	N	N	?
Blinding checked: participant	N	NA	N	N	N	NA
Blinding checked: caregiver / treatment administrator	N	NA	N	N	N	NA
Primary endpoint defined (power calculation)	Y	Y	N	N	N	Y
[n] of primary endpoint(s)	1	1	1	1	1	1
[n] of secondary endpoints	?	?	10	6?	8	7
Total [n] of endpoints	?	9	11	7	9	8
Prior publication of study design	Y	N	N	N	N	N
Outcomes of prior/current publication identical	?	NA	NA	NA	NA	NA
Power calculation	Y	Y	N	N	N	Y
[n] participants per group calculated	3600 (initially); 4182 (March 2002); further extension of trial was decided in February 2004 to compensate withdrawals	50	NA	NA	NA	total: 190
Non‐inferiority trial: interval for equivalence specified	NA	NA	NA	NA	Y	NA
Intention‐to‐treat analysis (ITT)	Y	Y	Y	?	Y	Y
Per‐protocol‐analysis	NA	N	?	?	N	?
ITT defined	N	N	Y	NA	Y	Y
Missing data: last observation carried forward (LOCF)	N	?	Y	?	Y	N
Missing data: Other methods	N	?	N	N	N	Y
LOCF defined	NA	?	Y	NA	N	NA
Analysis stratified for centres	N	NA	Y	NA	N	N
[n] of screened patients (I1 / I2/ C1/ total)	total: 6676	?	total: 623	total: ?	total: 1503	total: ?
[n] of randomised participants (I1/ I2 / C1 / total) ‐ primary endpoint	I1: 1456 I2: 1454 C1: 1441 total: 4351	I1: 56 C1: 56 total: 112	I1: ? I2: ? C1: ? total: 533	I1: 9 C1: 8 total: 17	I1: ? I2: ? I3: ? I4: ? C1: ? total: 959	I1: 62 I2: 63 I3: 127 total: 252
[n] of participants finishing the study (I1/ I2 / C1 / total)	I1: 917 I2: 903 C1: 807 total: 2627	I1: 50 C1: 52 total: 102	I1: ? I2: ? C1: ? total: 365	I1: ? C1: ? total: ?	I1: ? I2: ? I3: ? I4: ? C1: ? total: ?	I1: 37 I2: 38 I3: 106 total: 181
[n] of participants analysed (I1/ I2 / C1 / total) ‐ primary endpoint	I1: 1393 I2: 1397 C1: 1337 total: 4127	I1: ? C1: ? total: ?	I1: ? I2: ? C1: ? total: 472	I1: ? C1: ? total: ?	I1: 181 I2: 186 I3: 181 I4: 187 C1: 173 total: 908	I1: 55 I2: 59 I3: 126 total: 240
Description of discontinuing participants	N	Y	N	N	N	N
Drop‐outs (reasons explained)	N	Y partly	N	N	N	Y
Withdrawals (reasons explained)	Y	N	N	N	Y	N
Losses‐to‐follow‐up (reasons explained)	N	N	N	N	N	N
[n] of participants who discontinued (I1/ I2 / C1 / total)	I1: 539 I2: 551 C1: 634 total: 1724	I1: 6 C1: 2 total: 8	I1: 46 I2: 45 C1: 77 total: 168	I1: ? C1: ? total: ?	I1: ? I2: ? I3: ? I4: ? C1: ? total: 51	I1: 25 I2: 25 I3: 21 total: 71
[%] discontinuation rate (I1/ I2 / C1 / total)	I1: 37 I2: 38 C1:44 total: 44	I1: 10.7 C1: 3.6 total: 7.1	I1: 26 I2: 25 C1: 44 total: 32	I1: ? C1: ? total: ?	I1: ? I2: ? I3: ? I4: ? C1: ? total: 5%	I1: 40.3 I2: 39.7 I3: 16.5 total: 28.2
Discontinuation rate similar between groups	?	N	N	?	N "patients who withdrew from treatment were more poorly controlled at baseline"	N discontinuation rate lower for repaglinide/rosiglitazone combination therapy due to lack of efficiency in the monotherapy groups
[%] crossover between groups	?	?	?	?	?	?
Differences [n] calculated to analysed patients	additional patients were recruited during the study	N	NA	NA	NA	N
Adjustment for multiple outcomes / repeated measurements	N	?	Y	N	Y	N
Baseline characteristics: Clinically relevant differences	Y	Y gender, HbA1c, metformin dosage, antihypertensive and lipid‐lowering agents	N	? only few characteristics reported, numerical differences in age	N	previous sulfonylurea / metformin treatment
Treatment identical (apart from intervention)	Y	Y	Y	there was no titration period in the rosiglitazone group	Y	Y
Compliance measured	N	N	N	N	N	N
Other important covariates measured (specify)	N	N	N	N	N	N
Co‐morbidities measured	N	N	N	N	N	N
Co‐medications measured	N	Y	N	N	N	N
Specific doubts about study quality	N	N	N	Y	N	N
Funding: commercial	Y	?	?	Y	?	Y
Funding: non‐commercial	?	?	?	?	?	?
Publication status: peer review journal	Y	Y	Y	Y	Y	Y
Publication status: journal supplement	N	N	N	N	N	N
Publication status: abstract	N	N	N	N	N	N
Publication status: other	N	N	N	N	N	N
Notes	24 weeks treatment duration as inclusion criterion	./.	authors from a pharmaceutical company	./.	two authors hold stocks in pharmaceutical companies	./.
Footnotes Y = yes; N = no; ? = unclear I = intervention; C = control; (baseline) = if numbers for certain features could ne be derived from the text, numbers from baseline characteristics were used