Key Points
Question
How do authors currently report visual acuity (VA), and do they provide Snellen equivalents when using non-Snellen formats?
Findings
In this study, about half of all articles published in 4 ophthalmology clinical journals from November 2015 through October 2016 reporting VA used a non-Snellen format. Among these articles, 145 also provided a Snellen equivalent while 266 provided only a non-Snellen format.
Meaning
Since many ophthalmologists do not understand Snellen equivalents of logMAR or ETDRS letter scores, these data suggest that editors and publishing staff should encourage authors to provide Snellen equivalents when VA data are reported in a non-Snellen format.
Abstract
Importance
Visual acuity results in publications typically are reported in Snellen or non-Snellen formats or both. A study in 2011 suggested that many ophthalmologists do not understand non-Snellen formats, such as logarithm of the Minimum Angle of Resolution (logMAR) or Early Treatment Diabetic Retinopathy Study (ETDRS) letter scores. As a result, some journals, since at least 2013, have instructed authors to provide approximate Snellen equivalents next to non-Snellen visual acuity values.
Objective
To evaluate how authors currently report visual acuity and whether they provide Snellen equivalents when their reports include non-Snellen formats.
Design
From November 21, 2016, through December 14, 2016, one reviewer evaluated visual acuity reporting among all articles published in 4 ophthalmology clinical journals from November 2015 through October 2016, including 3 of 4 journals that instructed authors to provide Snellen equivalents for visual acuity reported in non-Snellen formats.
Main Outcomes and Measures
Frequency of formats of visual acuity reporting and frequency of providing Snellen equivalents when non-Snellen formats are given.
Results
The 4 journals reviewed had the second, fourth, fifth, and ninth highest impact factors for ophthalmology journals in 2015. Of 1881 articles reviewed, 807 (42.9%) provided a visual acuity measurement. Of these, 396 (49.1%) used only a Snellen format; 411 (50.9%) used a non-Snellen format. Among those using a non-Snellen format, 145 (35.3%) provided a Snellen equivalent while 266 (64.7%) provided only a non-Snellen format.
Conclusions and Relevance
More than half of all articles in 4 ophthalmology clinical journals fail to provide a Snellen equivalent when visual acuity is not in a Snellen format. Since many US ophthalmologists may not comprehend non-Snellen formats easily, these data suggest that editors and publishing staff should encourage authors to provide Snellen equivalents whenever visual acuity data are reported in a non-Snellen format to improve ease of understanding visual acuity measurements.
This study evaluates visual acuity reporting according to use of Snellen and non-Snellen formats in ophthalmology clinical journals.
Introduction
Visual acuity (VA) in publications can be reported in different ways. For example, a Snellen VA of 20/40 could be reported in non-Snellen formats as logarithm of the Minimum Angle of Resolution [logMAR] (0.3); Early Treatment Diabetic Retinopathy Study [ETDRS] letter score (70); decimal notation (0.50); or metric fractional notation (6/12). The inconsistency in VA reporting makes comparisons across clinical research reports potentially difficult. Standardized methods of VA measuring and reporting have been recommended. In addition, a study in 2011 suggested that many ophthalmologists do not understand Snellen equivalents of logMAR or ETDRS letter scores. In that study, among the 14 ophthalmologists who completed a survey asking them to provide the Snellen equivalent of VA measurements given in the non-Snellen formats discussed above, none correctly provided Snellen equivalents within 1 line of the VA measurement for all non-Snellen formats tested. In addition, most did not provide a correct Snellen equivalent within 2 lines of VA measurements given in logMAR or ETDRS letter scores. As a result, some journals since at least 2013 have instructed authors to provide approximate Snellen equivalents (such as 20/20 or 20/40) next to all VAs not reported in Snellen format.
Methods
From November 21, 2016, through December 14, 2016, one reviewer (B.C.T.) evaluated VA reporting among all articles published in 4 ophthalmology clinical journals (American Journal of Ophthalmology, JAMA Ophthalmology, Ophthalmology, and Retina) from November 2015 through October 2016. These 4 journals reviewed had the fifth, fourth, second, and ninth highest impact factors, respectively, for ophthalmology journals in 2015. The American Journal of Ophthalmology, JAMA Ophthalmology, and Retina instructed authors to provide Snellen equivalents for VA reported in non-Snellen formats; Ophthalmology did not give authors specific instructions regarding VA reporting. The occurrence of various VA reporting formats and how frequently VA reports in non-Snellen format provided Snellen equivalents were determined.
Results
A total of 1881 articles were reviewed; 807 (42.9%) of these provided a VA measurement, among which 396 (49.1%) provided a VA measurement only using Snellen format, while 411 (50.9%) used a non-Snellen format (Table 1). Among the 411 articles reporting VA in a non-Snellen format, 266 (64.7%) used logMAR, 88 (21.4%) used ETDRS letter scores, 32 (7.8%) used decimal notation, and 25 (6.1%) used metric fractional notation (Table 2). Among the 411 using a non-Snellen format, 145 (35.3%) provided a Snellen equivalent while 266 (64.7%) used only a non-Snellen format, including 150 (56.4%) providing VA using only logMAR, 65 (24.4%) providing VA using only ETDRS letter score, 27 (10.2%) providing VA using only decimal notation, and 24 (9.0%) providing VA using only metric fractional notation (Table 3). In total, 266 (33%) of 807 articles reporting VA provided the VA only in a non-Snellen format.
Table 1. Summary of Snellen and Non-Snellen Formats in Articles That Report Visual Acuity.
| Journal (No.) | Format, No. (%) | |
|---|---|---|
| Snellen | Non-Snellen | |
| American Journal of Ophthalmology (190) | 63 (33.2) | 127 (66.8) |
| JAMA Ophthalmology (162) | 123 (75.9) | 39 (24.1) |
| Ophthalmology (203) | 95 (46.8) | 108 (53.2) |
| Retina (252) | 115 (45.6) | 137 (54.4) |
| Total (807) | 396 (49.1) | 411 (50.9) |
Table 2. Summary of Non-Snellen Formats in Articles That Report Visual Acuity in Non-Snellen Format.
| Journal (No.) | No. (%) | |||
|---|---|---|---|---|
| logMAR | ETDRS Letter Score | Notation | ||
| Decimal | Metric Fractional | |||
| American Journal of Ophthalmology (127) | 92 (72.4) | 17 (13.4) | 12 (9.4) | 6 (4.7) |
| JAMA Ophthalmology (39) | 17 (43.6) | 5 (12.8) | 7 (17.9) | 10 (25.6) |
| Ophthalmology (108) | 58 (53.7) | 37 (34.3) | 9 (8.3) | 4 (3.7) |
| Retina (137) | 99 (72.3) | 29 (21.2) | 4 (2.9) | 5 (3.6) |
| Total (411) | 266 (64.7) | 88 (21.4) | 32 (7.8) | 25 (6.1) |
Abbreviation: ETDRS, Early Treatment Diabetic Retinopathy Study.
Table 3. Summary of Snellen Conversions in Articles That Report Visual Acuity in Non-Snellen Format.
| Journal (No.) | Equivalent, No. (%) | |
|---|---|---|
| Snellen | Non-Snellen | |
| American Journal of Ophthalmology (127) | 29 (22.8) | 98 (77.2) |
| JAMA Ophthalmology (39) | 16 (41.0) | 23 (59.0) |
| Ophthalmology (108) | 31 (28.7) | 77 (71.3) |
| Retina (137) | 69 (50.4) | 68 (49.6) |
| Total (411) | 145 (35.3) | 266 (64.7) |
Discussion
Among the articles from November 2015 through October 2016 reviewed in 4 journals with the second, fourth, fifth, and ninth highest impact factors for ophthalmology journals in 2015, VA reporting format was divided roughly in half between Snellen and non-Snellen formats. The most common non-Snellen format was logMAR, followed by ETDRS letter score, decimal notation, and metric fractional notation. However, the majority of the articles with VA reported in non-Snellen formats did not provide a Snellen equivalent despite previous recommendations by some journals to provide such equivalents. The majority of articles with VA reported in non-Snellen formats without a Snellen equivalent used logMAR and ETDRS letter score.
Limitations
Limitations to these findings include the possibility that the results from the time period chosen are not representative of articles prior to or since the evaluation, the journals chosen may not be representative of other ophthalmology journals, and the lack of evidence as to whether the failure to understand the VAs reported also lead to a failure of understanding the results and relevance of the papers that only used non-Snellen formats for VA reporting. Despite these limitations, the results are likely to be representative because a large time span (1 year) and recent time period (2015 to 2016) were chosen for the evaluation of this study; 3 commonly referenced, clinically oriented journals and a commonly referenced subspecialty journal were chosen for this analysis; and previous reports suggest that ophthalmologists across various stages of their career do not readily understand non-Snellen formats of VA, which likely would interfere with understanding a paper’s results and its relevance.
Conclusions
Although prior studies suggested that many ophthalmologists do not understand Snellen equivalents of logMAR or an ETDRS letter score and some journals recently have instructed authors to provide a Snellen equivalent when VA is not in Snellen format, more than half of all articles reporting VA in 4 frequently cited ophthalmology clinical journals still fail to provide Snellen equivalents when non-Snellen VA is used. To improve ease of understanding VA measurements, the current data suggest that editors and publishing staff should encourage authors to provide Snellen equivalents whenever VA data are reported in a non-Snellen format.
References
- 1.Ferris FL III, Bailey I. Standardizing the measurement of visual acuity for clinical research studies: guidelines from the Eye Care Technology Forum. Ophthalmology. 1996;103(1):181-182. [DOI] [PubMed] [Google Scholar]
- 2.Moutray TN, Williams MA, Jackson AJ. Change of visual acuity recording methods in clinical studies across the years. Ophthalmologica. 2008;222(3):173-177. [DOI] [PubMed] [Google Scholar]
- 3.Lopes MS, Zayit-Soudry S, Moshiri A, Bressler SB, Bressler NM. Understanding and reporting visual acuity measurements in publications of clinical research. Arch Ophthalmol. 2011;129(9):1228-1229. [DOI] [PubMed] [Google Scholar]
- 4.Impact factor list 2015. http://www.citefactor.org/journal-impact-factor-list-2015.html. Accessed March 11, 2017.