Table 3.
Meta-regression analysis results
| Number of studies | HR (95% CI) | p value | ||
|---|---|---|---|---|
| Risk of bias | .. | .. | 0·76 | |
| Low | 2 | 1·73 (0·85–3·50) | .. | |
| Moderate | 12 | 1·28 (1·18–1·39) | .. | |
| High | 1 | 1·17 (0·63–2·16) | .. | |
| Vision chart | .. | .. | 0·082 | |
| ETDRS chart | 3 | 1·46 (1·29–1·64) | .. | |
| Snellen Tumbling E chart | 2 | 1·35 (1·14–1·61) | .. | |
| logMAR* | 5 | 1·28 (1·17–1·40) | .. | |
| Other or not reported | 5 | 1·12 (0·95–1·31) | .. | |
| Eye | .. | .. | 0·24 | |
| Better eye | 9 | 1·34 (1·19–1·50) | .. | |
| Other | 6 | 1·22 (1·12–1·33) | .. | |
| Vision assessment method | .. | .. | 0·0055 | |
| Best-corrected visual acuity | 7 | 1·45 (1·31–1·60) | .. | |
| Presenting visual acuity | 8 | 1·22 (1·13–1·31) | .. | |
| Follow-up duration, years | .. | .. | 0·58 | |
| <10 | 7 | 1·35 (1·13–1·60) | .. | |
| ≥10 | 8 | 1·27 (1·17–1·39) | .. | |
| Lower age limit, years | .. | .. | 0·27 | |
| <50 | 7 | 1·35 (1·19–1·54) | .. | |
| ≥50 | 8 | 1·24 (1·15–1·34) | .. | |
| Study design | .. | .. | 0·38 | |
| Cohort | 13 | 1·27 (1·17–1·39) | .. | |
| Randomised controlled trial | 2 | 1·40 (1·18–1·66) | .. | |
Estimated HRs for mortality in people with a visual acuity of <6/12 versus those with a visual acuity of ≥6/12, subcategorised by seven variables, with p values from the meta-regression analysis. ETDRS=Early Treatment Diabetic Retinopathy Study. HR=hazard ratio. logMAR=logarithm of minimum angle of resolution.
*
Includs Bailey-Lovie charts.