Table 2 Trial quality assessment.

Trial	Allocation concealment adequate	Blinding		Measurement				Withdrawals (%)	ITT analysis or last observation carried forward
		Patient	Investigator	Equipment	Examiners*	Average of 3 readings in study eye?	Washout†
Akman et al8	NS	N	Y	GAT	Same	Y	NS	11	NS
Camras et al24	Y	N	Y	GAT	Different	Y	Y	19	Y
De Figueiredo and De Figueiredo23	N (alphabetical)	NS	NS	NS	Same	NS	Y	NS	NS
DuBiner et al9	Y	Y	Y	GAT	Different	NS	Y	6	N
Garcia‐Sanchez et al25	Y	N	Y	GAT	Different	Y	Y	12	Y
Inan et al10	NS	N	Y	GAT, colour Doppler ultrasound	Same	NS	Y	7	NS
Kampik et al16	NS	N	Y	GAT	Different	Y	Y	13	Y
Simmons and Samuelson11	NS	N‡	Y	Applanation tonometry	Different	NS	NS	0§	Y
Simmons et al12	Y	Y	Y	NS	Different	NS	NS	715	N
Sodhi et al13	Y	N14	Y	GAT	NS	NS	NS	6	N
Waldock et al26	NS	N	Y	GAT	Same	Y	NS	?¶	N
Liu et al18,19	Y (random number table)	N	Y	GAT, Ocular blood flow tonometer	Same	Y	Y	22	NS
Orzalesi et al17	Y (list of random numbers)	N14	Y	GAT and hand‐held tonometer	Different (2)	NS	Y	0	Y
Stewart et al20	NS	Y	Y	GAT	Different	N	N**	3††	N
Stewart et al21	NS	Y	Y	GAT	Different	N	N17	9	N

?, value unknown; GAT, Goldmann applanation tonometer; ITT, intention‐to‐treat; N, no; NS, not stated or uncertain; Y, yes.

*Unless otherwise stated, multicentre trials were assumed to have different examiners. †For trials in which all patients were treatment naive, this is reported as Yes (Y).

‡Patients masked to bottles, not to frequency (ie morning placebo not specified for latanoprost arm). §At 1 month. Unable to ascertain percentage (authors did not specify from which arms withdrawals originated). **Washout occurred before beginning the trial, but not between crossover arms.

††Insufficient information to determine whether all withdrawals were reported.