Summary of findings for the main comparison. Multivitamin versus placebo.
| Antioxidant multivitamin and mineral supplement versus placebo or no treatment | ||||||
| Patient or population: people with AMD Setting: community Intervention: antioxidant multivitamin and mineral supplement* Comparison: placebo or no treatment | ||||||
| Outcomes | Anticipated absolute effects** (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with placebo | Risk with Multivitamin antioxidant vitamin or mineral supplement | |||||
| Progression to late AMD (neovascular AMD, geographic atrophy or both) | Low | OR 0.72 (0.58 to 0.90) | 2445 (3 RCTs) | ⊕⊕⊕⊝ MODERATE1 | Average follow‐up in study contributing most of the events was 6 years | |
| 15 per 1000 | 11 per 1000 (9 to 14) | |||||
| High | ||||||
| 430 per 1000 | 352 per 1000 (304 to 404) | |||||
| Progression to neovascular AMD | Low | OR 0.62 (0.47 to 0.82) | 1206 (1 RCT) | ⊕⊕⊕⊝ MODERATE1 | Average follow‐up 6 years. Estimate of effect from study population including AMD category 3 & 4 only | |
| 10 per 1000 | 6 per 1000 (5 to 8) | |||||
| High | ||||||
| 300 per 1000 | 210 per 1000 (168 to 260) | |||||
| Progression to geographic atrophy | Low | OR 0.75 (0.51 to 1.10) | 1206 (1 RCT) | ⊕⊕⊕⊝ MODERATE1 | Average follow‐up 6 years. Estimate of effect from study population including AMD category 3 & 4 only | |
| 10 per 1000 | 8 per 1000 (5 to 11) | |||||
| High | ||||||
| 300 per 1000 | 243 per 1000 (179 to 320) | |||||
| Progression to visual loss (loss of 3 or more lines on logMAR chart) | Low | OR 0.77 (0.62 to 0.96) | 1791 (1 RCT) | ⊕⊕⊕⊝ MODERATE1 | Average follow‐up 6 years | |
| 15 per 1000 | 12 per 1000 (9 to 14) | |||||
| High | ||||||
| 430 per 1000 | 367 per 1000 (319 to 420) | |||||
| Quality of life assessed with: change in National Eye Institute Visual Function Questionnaire (NEI‐VFQ) score (higher scores better) | The mean change in NEI‐VFQ score in the control group was ‐8.7 | The mean NEI‐VFQ quality of life score in the intervention group was 12.3 higher (4.24 higher to 20.36 higher) | ‐ | 110 (1 RCT) | ⊕⊕⊝⊝ LOW 2,3 | Follow‐up 24 months |
| Adverse effects | Data from AREDS suggested no serious adverse effects associated with multivitamin use (hazard ratio for mortality 0.87, 95% CI 0.60 to 1.25) but participants in the antioxidant arms more frequently reported yellow skin (8.3% versus 6.0%, P = 0.008).. | ⊕⊝⊝⊝ VERY LOW 4 | ‐ | |||
| Resource use and costs | ‐ | ‐ | ‐ | ‐ | ‐ | Not reported |
| * Most of the evidence in this table is drawn from the AREDS study which studied antioxidants (vitamin C 500 mg, vitamin E 400 IU, beta‐carotene 15 mg daily) plus zinc 80 mg as zinc oxide, copper 2 mg as cupric oxide (daily) **The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). The assumed risk in the comparison group is estimated using data from AREDS: low risk = AREDS category 2; high risk = AREDS category 4. CI: Confidence interval; RR: Risk ratio; OR: Odds ratio; | ||||||
| GRADE Working Group grades of evidence High‐certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate‐certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low‐certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low‐certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect | ||||||
1 Downgraded one level for imprecision because upper confidence interval crosses line of minimum important difference (0.8 to 1.25)
2 Downgraded one level for risk of bias because study was not placebo‐controlled and at high risk of performance and detection bias
3 Downgraded one level for imprecision because confidence intervals included clinically insignificant effect
4 Downgraded for one level for imprecision (as included studies were underpowered to look at adverse effects), one level for risk of bias (adverse effects were inconsistently reported) and one level for inconsistency (inconsistent results reported).