Scholler 2014.

Methods	Number randomized (total and per group): 55 total: 29 in ranibizumab group and 26 in bevacizumab group Exclusions after randomization: “nine eyes were excluded” but meaning is unclear Number analyzed (total and per group): 55: 29 vs 26 at baseline. Number examined at follow‐up times not reported, but 26 in ranibizumab group and 20 in bevacizumab group finished the study. These 46 eyes could be analyzed in total Unit of analysis: unclear; appears to be eyes Losses to follow‐up: “nine eyes were excluded” appears to refer to such losses Compliance: 3 of 9 “excluded” were given ranibizumab instead of bevacizumab as assigned Intention‐to‐treat analysis: not reported; unclear as follow‐up denominators missing Reported power calculation: none Study design comment: none
Participants	Country: Austria Age: 79.5 in ranibizumab group and 80.8 in bevacizumab group (80.15 years total) Gender (per cent): ranibizumab: 24% men, 76% women; bevacizumab: 35% men, 65% women Inclusion criteria: over 50 years of age with nAMD, which was verified in fluorescence angiography (FLA). Only patients with VA between 20/40 and 20/320 were included Exclusion criteria: previous treatment for AMD, previous systemic administration of bevacizumab, vision‐impairing cataract or other ophthalmologic disease like glaucoma, active inflammation, diabetic retinopathy, and others Equivalence of baseline characteristics: yes; similar age, BCVA, CRT, FA lesion size, gender (all P values > 0.37) Diagnoses in participants: neovascular AMD
Interventions	Intervention 1: 3 ranibizumab injections (0.5 mg) at 30‐day intervals followed by PRN ranibizumab for another 10 months Intervention 2: 3 bevacizumab injections (1.25 mg) at 30‐day intervals followed by PRN bevacizumab for another 10 months Length of follow‐up: Planned: scheduled visits were performed monthly (30 ± 7 days) for 12 months Actual: 12 months (monthly)
Outcomes	Primary outcome, as defined: a difference in injection frequencies of ranibizumab and bevacizumab Secondary outcomes, as defined: effectiveness of ranibizumab and bevacizumab with respect to BCVA and CRT Adverse events (Y/N): yes; 3 reported, all in ranibizumab arm Intervals at which outcome assessed: monthly
Notes	Full study name: Differences of Frequency in Administration of Ranibizumab and Bevacizumab in Patients With Neovascular AMD Type of study: published Funding sources: “Birgit Weingessel and Pia Veronika Vécsei‐Marlovits received unrestricted grants from Novartis Austria during the last 5 years” Declarations of interest: study authors declare no conflicts of interest Study period: accrual: 1‐1‐2011 to 12‐31‐2011; follow‐up for 1 year yields January 2011 to December 2012 Reported subgroup analyses (Y/N): if yes, specify: no Registration: EK‐07‐192‐1007/EudraCT Nr.2007‐005157‐33 (Ethikkommission der Stadt Wien)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	“For allocation of the participants, a computer‐generated list of random numbers was used on the basis of simple randomization (1:1)”
Allocation concealment (selection bias)	Unclear risk	No information provided on how participants could or could not possibly foresee assignments
Masking of participants (performance bias)	Unclear risk	No information provided on how masking of participants was done for primary and secondary outcomes
Masking of study personnel (performance bias)	Unclear risk	No information provided on how masking of physicians was done for primary and secondary outcomes, but outcomes are not likely to be influenced by lack of masking or blinding
Masking of outcome assessment (detection bias)	High risk	No information provided on how masking of outcome assessors was done for primary and secondary outcomes, and this could influence participants to respond differently
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Cannot tell; number of participants who contributed outcome data at each follow‐up time not reported, but results show that 26 participants in ranibizumab group and 20 in bevacizumab group completed the study after exclusion of 9 eyes
Selective reporting (reporting bias)	Unclear risk	Neither FA lesion size @ 12 months nor change in FA lesion size from baseline to 12 months reported
Other bias	Unclear risk	Two study authors had received unrestricted grants from Novartis Austria within 5 years (Novartis is the distributor of ranibizumab in Europe)