Characteristics of included studies [ordered by study ID]

Guymer 2013
Methods	Study design: randomized controlled trial Number randomized: 114 total; 57 simvastatin; 57 placebo Exclusions after randomization: none Number analyzed: at 36 months: 114 total; 57 simvastatin; 57 placebo Unit of analysis: individuals Losses to follow up: 34 participants total; 20 simvastatin; 14 placebo How was missing data handled?: last-observation-carried-forward method used for 34 participants; 11 participants with baseline data only and 23 participants who missed the 3-year follow-up visit Power calculation: 58 participants in each arm for power of 80% at alpha 0.05 to detect a 50% reduction in progression of disease
Participants	Country: Australia Mean age: 74.6 years overall; 74.8 years for simvastatin group; 74.4 years for placebo group Gender: 77/114 (68%) women 37/114 (32%) men total 39/57 (68%) women 18/57 (32%) men in the simvastatin group 38/57 (67%) women 19/57 (33%) men in the placebo group Inclusion criteria: 1) males and females aged 50 years and older; 2) able to assess the macula in at least one eye; 3) visual acuity ≥ 20/60 in study eye; 4) high risk drusen in both eyes: one or more large soft drusen, > 10 intermediate drusen, or late AMD in one eye and any drusen or pigment change in study eye; 5) normal cholesterol levels; and 6) not currently on cholesterol-lowering medications Exclusion criteria: 1) bilateral end-stage AMD; 2) medical or ophthalmic conditions which could potentially affect visual function, such as cataract, diabetes, glaucoma; 3) use of medications that may affect visual function, such as plaquenil, chloroquine, major tranquillizers; 4) currently on cholesterol-lowering medication; 5) use of statins is contraindicated; 6) alanine aminotransferase (ALT) two times the upper limit of normal; and 7) previous severe adverse or allergic reactions to statins Equivalence of baseline characteristics: no; more participants in simvastatin group had unilateral advanced AMD as compared with placebo; less smokers in placebo group than simvastatin group
Interventions	Intervention 1: two tablets of simvastatin (40 mg daily) for three years Intervention 2: placebo with an identical appearance for three years Length of follow-up: Planned: three years Actual: three years
Outcomes	Primary outcome, as defined in study reports: “Primary outcome was progression of non-advanced AMD to either advanced AMD or higher severity scores of non-advanced AMD”, evaluated every 6 months. “Advanced AMD was defined as presence of either CNV or geographic atrophy (GA). CNV was confirmed on angiography and GA was defined as an area of hypopigmentation 175 mm with a choroidal vessel in its base on colour photography.” Secondary outcomes, as defined in study reports: (1) change in visual function over time; (2) genotype as an effectmodifier of the association between statins and progression of AMD Adverse events reported: yes Intervals at which outcomes assessed: 1, 6, 12, 18, 24, 30, and 36 months
Notes	Funding sources: Ian Potter Foundation, John Reid Charitable Trust and Royal Victorian Eye and Ear Hospital; National Health and Medical Research Council (NHMRC) supported the study through a Centre for Clinical Research Excellence award to CERA (#529923), a Practitioner Fellowship (#529905) and a Senior Research Fellowship (#1028444); Wagstaff Fellowship; Victorian Government Disclosures of interest: co-author Paul Baird is a PLOS ONE Editorial Board member Study period: 3 years; 2003 to 2006 Reported subgroup analyses: yes Trial investigators provided information on loss to follow-up by intervention at three year follow-up (email communication) Trial reported at ARVO (abstract); trial registration number: ACTRN12605000320651 (registered at WHO International Clinical Trials Registry Platform)
Risk of bias
Bias	Authors’ judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	“Randomization was performed by a bio- statistician using permuted blocks of randomly varying size.”
Allocation concealment (selection bias)	Low risk	“The hospital pharmacist packed the medication into identical containers according to the randomization code. The sequentially numbered containers were allocated to the participants by the study coordinator in order of enrolment.” “The allocation list was stored at a remote site.”
Masking (performance bias and detection bias)	Low risk	“The study staff, the participants, and data analysts were masked to treatment allocation until the analysis was finalised.”
Incomplete outcome data (attrition bias) All outcomes	High risk	Data missing for 34/114 (30%) participants at 3 years follow-up: 20/57 (35%) in the simvastatin group and 14/57 (25%) in the placebo group. Reasons for missing the 3-year visit were: personal, poor health, unable to contact, adverse reaction to study medication, reached late AMD, sick at 3-year follow-up, deceased, or developed macular hole. The study investigators imputed missing data using the last-observation-carried-forward method
Selective reporting (reporting bias)	Low risk	Primary and secondary outcomes reported in the 2013 results paper matched the protocol published in 2008
Other bias	Unclear risk	“Analysis was done ‘by person’ and used the data from the eye showing greatest progression. If one eye of a person worsened and the other eye showed improvement, the person was classified as having progressed”, but AMD progression by eye also was reported; at baseline, “the number of participants with unilateral advanced AMD was twice as large in the simvastatin group compared to the placebo group (x2 = 9.2, P = 0.002). Smoking also was less prevalent in the placebo group; the difference was marginally significant (x2 = 3.5, P = 0.06).”
Martini 1991
Methods	Study design: randomized controlled trial Number randomized: 30 participants total; 15 in each group Exclusions after randomization: none reported Number analyzed: 30 participants total; 15 in each group Unit of analysis: individuals Losses to follow up: none reported How was missing data handled?: not applicable Power calculation: none reported
Participants	Country: Italy Age: greater than 60 years Gender: not reported Inclusion criteria: participants with drusen (no CNV), good visual acuity (mean 0.52 LogMAR) and serum cholesterol level > 260 mg/dL Exclusion criteria: participants’ age less than 60 years Equivalence of baseline characteristics: not reported
Interventions	Intervention 1: simvastatin (20 mg daily) for 3 months Intervention 2: placebo for 3 months Length of follow-up: Planned: 4.5 months Actual: 4.5 months
Outcomes	Primary outcome, as defined in study reports: 1) serum cholesterol levels Secondary outcomes, as defined in study reports: 2) visual acuity; 3) microscopic eye examination; 4) fluorescein angiography; 5) electroretinography, and 6) visual evoked potentials Adverse events reported: no Intervals at which outcomes assessed: baseline, 3 months and 4.5 months
Notes	Funding sources: not reported Disclosures of interest: not reported Study period: not reported Reported subgroup analyses: no Attempted to contact trial investigators, but no reliable contact information found Article in Italian
Risk of bias
Bias	Authors’ judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No description was found in the article
Allocation concealment (selection bias)	Unclear risk	No description was found in the article
Masking (performance bias and detection bias)	Unclear risk	No description was found in the article
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No description was found in the article
Selective reporting (reporting bias)	Unclear risk	No description was found in the article
Other bias	Unclear risk	Inadequate information reported

AMD: age-related macular degeneration

CNV: choroidal neovascularization

mg/dL: milligrams per deciliter