Table 2.
Results of studies not included in the meta-analysis
| Reason for exclusion from the meta-analysis | Vision impairment definition (vision assessment method, eye) | Comparison group | Minimally adjusted effect estimates (95% CI)* | Maximally adjusted effect estimates (95% CI) | |
|---|---|---|---|---|---|
| Anstey et al (2001)31 | Categorical analysis | 6/9 (BCVA, eye NR); 6/12 (BCVA, eye NR); and 6/18 to 6/60 (BCVA, eye NR) | 6/6 | 6/9, RR 0·95 (0·67–1·33); 6/12, 1·16 (0·84–1·59); and 6/18 to 6/60, 1·10 (0·80–1·53) | 6/9, RR 0·89 (0·63–1·25); 6/12, 1·10 (0·80–1·52); and 6/18 to 6/60, 1·01 (0·72–1·39) |
| Freeman et al (2005)37 | Effect estimate per unit change in visual acuity | Mild vision loss, 2–3 lines (PVA, eye NR); moderate vision loss, ≥3 lines (PVA, eye NR); and vision gain, ≥2 lines (PVA, eye NR) | No change in visual acuity | Mild vision loss, HR 0·92 (0·61–1·37); moderate vision loss, 2·23 (1·43–3·46); and vision gain, HR 0·47 (0·23–0·96) | Mild vision loss, HR 0·91 (0·61–1·36); moderate vision loss, 2·26 (1·45–3·52); and vision gain, 0·47 (0·23–0·95) |
| Jacobs et al (2005)33 | Did not report HRs | ≤6/12 (BCVA, better eye) | NA | OR 2·84 (1·48–5·46) | |
| Kim et al (2019)6 | Non-standard vision impairment thresholds | Mild vision loss, 6/30 to 6/100 (BCVA, better eye) or ≤6/300 (BCVA, worse eye); and severe vision loss, ≤6/300 (BCVA, better eye) | >6/30 | Mild vision loss, HR 1·17 (0·81–1·69); and severe vision loss, 1·90 (1·08–3·35) | Mild vision loss, HR 1·16 (0·81–1·67); and severe vision loss, 1·87 (1·06–3·29) |
| Kulmala et al (2008)21; 75-year-old cohort | Categorical analysis | ≤6/12 to ≥6/18 (PVA, better eye); and <6/18 (PVA, better eye) | >6/12 | ≤6/12 to ≥6/18, HR 1·98 (1·25–3·13); and <6/18, 1·90 (1·12–3·20) | ≤6/12 to ≥6/18, HR 2·11 (1·27–3·48); and <6/18, 1·34 (0·75–2·39) |
| Kulmala et al (2008)21; 80-year-old cohort | Categorical analysis | ≤6/12 to ≥6/18 (PVA, better eye); and <6/18 (PVA, better eye) | >6/12 | ≤6/12 to ≥6/18, HR 1·13 (0·74–1·72); and <6/18, 0·92 (0·47–1·78) | ≤6/12 to ≥6/18, HR 0·77 (0·48–1·26); and <6/18, 0·75 (0·33–1·67) |
| Kuper et al (2019)5 | Categorical analysis and did not report HRs | <6/12 to ≥6/18 (PVA, better eye); <6/18 to ≥6/60 (PVA, better eye); and <6/60 (PVA, better eye) | ≥6/12 | <6/12 to ≥6/18, RR 0·92 (0·57–1·50); <6/18 to ≥6/60, 1·75 (1·28–2·40); and <6/60, 1·98 (1·04–3·80) | <6/12 to ≥6/18, RR 0·82 (0·48–1·41); <6/18 to ≥6/60, 1·56 (1·14–2·15); and <6/60, 1·46 (0·80–2·68) |
| Li et al (2011)26 | Categorical analysis and did not report HRs | <6/18 to ≥3/60 (BCVA, better eye); <3/60 (BCVA, better eye) | ≥6/18 | NA | <6/18 to ≥3/60, OR 3·1 (1·5–6·4); and <3/60, 3·9 (2·1–7·2) |
| Taylor et al (2000)35 | Did not report HRs | ≤6/12 (BCVA, better eye) | >6/12 | NA | OR 2·42 (1·07–5·43) |
| Thiagarajan et al (2005)19 | Categorical analysis and did not report HRs | <6/6 to ≥6/9 (PVA, binocular); <6/9 to ≥6/18 (PVA, binocular); and <6/18 (PVA, binocular) | ≥6/6 | <6/6 to ≥6/9, RR 1·10 (1·01–1·19); <6/9 to ≥6/18, 1·32 (1·22–1·42); and <6/18, 1·60 (1·47–1·74) | <6/6 to ≥6/9, RR 1·06 (0·97–1·16); <6/9 to ≥6/18, 1·24 (1·14–1·35); and <6/18, 1·52 (1·39–1·66) |
| Thompson et al (1989)24 | Categorical analysis and did not report HRs | ≤6/7·5 to ≥6/9 (BCVA, better eye); ≤6/12 to ≥6/18 (BCVA, better eye); ≤6/24 to ≥6/60 (BCVA, better eye); and <6/60 (BCVA, better eye) | ≥6/6 | NA | ≤6/7·5 to ≥6/9, RR 1·62 (0·87–3·01); ≤6/12 to ≥6/18, 1·83 (0·93–3·63); ≤6/24 to ≥6/60, 1·72 (0·77–3·84); and <6/60, 0·35 (0·08–1·57) |
| Wang et al (2014)32 | Effect estimate per unit difference in visual acuity and did not report HRs | (BCVA, worse eye) | NA | OR 1·76 (1·35–2·29) | NA |
The table shows effect estimates of studies that were excluded from meta-analysis, with reasons for exclusion and definitions of vision impairment. BCVA=best-corrected visual acuity. HR=hazard ratio. NA=not applicable. NR=not reported. OR=odds ratio. PVA=presenting visual acuity. RR=risk ratio.
*
All estimates are, at minimum, adjusted for age.