A customised portable LogMAR chart with adjustable chart illumination for use as a mass screening device in the rural population
M Gouthaman, RP Raman, A Kadambi, R Padmajakumari, PG Paul, Tarun Sharma
Sankara Nethralaya,18, College Road, Chennai - 600006, Tamil Nadu, India
Sankara Nethralaya,18, College Road, Chennai - 600006, Tamil Nadu
Aim: To develop a customised, portable, cost-effective (logarithmic minimal angle resolution) LogMAR chart with adjustable illumination for use as a mass vision-screening device in the rural population.
Materials and Methods : Visual acuity of 100 individuals was evaluated with a customised chart and compared with the standard Early Treatment Diabetic Retinopathy Study (ETDRS) chart and Snellen«SQ»s Chart. Bland and Altman analytical techniques were used for analysis.
Results: Test-retest variability of the customised chart was just a one-line difference (95% CI for agreement), and so were the results with the standard ETDRS charts; a variability of 3-line was noted with Snellen«SQ»s chart. Two-line differences were observed when comparison was made with Standard ETDRS chart and 2 to 3-line differences with Snellen«SQ»s chart.
Conclusion: The customised portable LogMAR chart with adjustable illumination shows less test-retest variability and better agreement with standard ETDRS chart; therefore, it can be used as a mass vision-screening device in rural settings.
|How to cite this article:|
Gouthaman M, Raman R P, Kadambi A, Padmajakumari R, Paul P G, Sharma T. A customised portable LogMAR chart with adjustable chart illumination for use as a mass screening device in the rural population.J Postgrad Med 2005;51:112-115
|How to cite this URL:|
Gouthaman M, Raman R P, Kadambi A, Padmajakumari R, Paul P G, Sharma T. A customised portable LogMAR chart with adjustable chart illumination for use as a mass screening device in the rural population. J Postgrad Med [serial online] 2005 [cited 2020 Jun 6 ];51:112-115
Available from: http://www.jpgmonline.com/text.asp?2005/51/2/112/16373
The commonest tool for testing visual acuity in clinical practice is the Snellen chart. However, it has certain inherent design errors such as non-geometric progression of letter sizes, variable number of letters per line, and the lack of a standardized scoring system. ,, The relatively large gaps between acuity levels at the lower end of the acuity scale on a Snellen chart (6/60-6/24) can result in gross overestimation and underestimation of visual acuity; this makes the assessment of change in visual acuity difficult. In clinical research, measurement of change in visual acuity is made over time and this might be estimated by different examiners and perhaps in different locations. Therefore, standardization of measurement of visual acuity is particularly important. Gibson and Sanderson reported poor reproducibility of visual acuity with the Snellen chart: only one-third had the same visual acuity on two occasions, with 13% differing by two lines or more.
The new design of an Early Treatment Diabetic Retinopathy Study (ETDRS) chart was an alternative to the Snellen chart and frequently employed in prospective clinical research., The advantages of an ETDRS chart included: geometric progression of letter sizes, equal number of letters per line, and a scoring system., These rather expensive charts are currently used predominantly in clinical research, but not as screening tools. Variable intensities of ambient light and chart illumination in rooms with open windows, particularly in rural settings additionally limit their use as screening devices.
We developed a portable ETDRS chart with adjustable chart illumination (to provide standard illumination all the time), and this was used as a screening device in the rural population. The visual acuity measurements were compared with those obtained using standard ETDRS and Snellen charts, to observe test-retest variability (TRV) and agreement between the charts.
Materials and Methods
The actual sample size was 97; it was calculated based on Bland and Altman's model of repeatability study for two observations with Type I error at 0.05 and Type II error at 0.2. The subjects were healthy individuals who accompanied their relatives for a diabetic retinopathy screening camp in rural south India. Participants without ocular disease were selected in order to avoid fluctuations in visual function that might occur in patients with ocular disease. We enrolled 100 such subjects who agreed to participate in the study.
The customised portable chart [Figure 1] and [Figure 2] was designed keeping in mind the standard design principles of the ETDRS chart, and included logarithmic letter size progression, inter-letter spacing, interline spacing, the avoidance of words or acronyms, and the use of Landolt's broken ring. The chart was prepared using milky-white mat-shade paper (not on glossy paper which could cause undue reflection). The dimensions of the customised chart (portable one) were 60 × 29 × 5 cm as compared to 62.5 × 62.5 × 17.5 cm of a standard ETDRS chart with the light box (back lit); the weight was just 5 kg instead of 14 kg of standard ETDRS chart.
The customised chart was externally illuminated from below by two halogen bulbs, aligned in such a way that they uniformly illuminated the front part of the chart without glare (front-lit chart). As specified by the British Standards Institute, the illumination at the chart was always kept constant ( 1000-1200 Lux) and modulated by a rheostat and a sensor.
The customised chart was kept at a distance of four meters from the subject. One eye of each subject based on toss of a coin was assessed. Visual acuity of all subjects was measured in four randomly chosen clinical settings:
The Snellen Charts (Clement Clarke UK)ETDRS Charts (Lighthouse Low Vision Products)Customised logMAR chart (with windows closed)
Two versions of each chart with different letter sequencing were used. The subjects were allowed to choose the order of the chart and their visual acuity was measured either unaided or with habitual spectacle correction, if any. The charts were hung at a height such that the top of the third row of letters (0.8 logMAR) was 125 + 5 centimetres from the floor. All measurements were made by a single examiner. The end point for Snellen visual acuity as well as logMAR acuity was the lowest line wherein at least half the letters were read correctly. The charts were scored using line assignment method. The Snellen acuity data were converted to the logMAR format by taking the log to the base 10 of the reciprocal of the Snellen acuity fraction. All acuity scores were adjusted for testing distance, before analysis.
To determine the test-retest variability (TRV) of each of the 4 clinical settings and elucidate agreement among all the charts, the analytical methods of Bland and Altman were used.
Of the 100 enrolled subjects, 62 were male, and 38 female. The mean age was 49.1 years (range, 21-85 years). [Table 1] summarizes the results of the test-retest variability (TRV) of each chart (a difference of 0.1 for TRV at 95% CI for the agreement is equal to one-line difference between the two charts, and a difference of 0.2 is equal to a 2-line difference; likewise if the mean difference was negative, the second chart for TRV overestimated the visual acuity). [Table 1] shows that 95% confidence interval for agreement for TRV for Snellen chart was more variable (± 0.30 to 0.37 logMAR; that is, overestimation of 3-lines by the second chart) than ETDRS chart and customised charts (±0.02, ±0.05 to ±0.07, and ±0.01 respectively; that is, less than 1-line difference).
[Table 2] demonstrates the agreement between customised charts versus ETDRS and Snellen charts. When ETDRS chart was compared with customised charts (windows closed and windows open), 95% CI for agreement and mean difference showed that customised chart with windows closed overestimated visual acuity by 2 lines, and with windows open, either over or underestimated visual acuity by 2 lines. On the other hand, the difference in overestimated visual acuity was 2-3 lines on Snellen versus customised chart with windows closed, and 1-2 lines with windows open.
Despite being the most widely used test for assessing visual function, measurement of visual acuity does not follow a standardized procedure in most clinical settings. Though modifications of the ETDRS chart have been made to reduce the test time, their applicability in the rural settings with variable ambient illumination has not been reported. Yang et al reported lack of adequate illumination when visual acuity was measured in schools.
The results of the present study showed that the test-retest variability of our customised chart was comparable to the standard ETDRS chart; viz. less than 1-line variability. The results also highlighted that our chart was in close agreement with the ETDRS chart - a difference of 2 lines - and this difference remained irrespective of surrounding illumination (windows open or closed). In a rural environment with little control over ambient illumination, we did not observe a greater difference in estimation of visual acuity with our chart. This was possible because chart illumination was kept constant by using a rheostat. However, as expected, these differences in the estimation of visual acuity were more (nearly 3-line) when compared with the Snellen chart.
It has been standard clinical practice to consider a change of at least 2-lines on Snellen chart as a cut-off value for clinically meaningful change in visual acuity. Also, considering the maximal difference of 0.05 logMAR (Mean difference) tolerated by the international standard DIN EN ISO 8597, visual acuity testing by customized chart is comparable to standard ETDRS chart.
Other advantages included its portability (5 kg vs. 14 kg) and cost just 5% of the standard ETDRS chart (about 112 US dollars vs. 2700 US dollars). In a country like India where economics does have an important role in the health care decision-making process, use of these customised charts would make a great impact in collecting data in a standardized format for epidemiological studies.
In this study, the optotypes used were Landolt rings for two reasons: population in the rural area was illiterate (no alphabet barrier) and Landolt rings were adopted by the NAS-NRC (National Academy of Science - National Research Council Committee on Vision). It was more practical for the patients to choose from 4 alternatives of Landolt rings than make10 choices from Sloan letters; another advantage was equal difficulty in interpreting each ring as four positions of gap were produced by rotating the target around a fixed point, thus minimizing artefacts and secondary clue.
To conclude, our customised portable ETDRS chart with adjustable illumination provides a simple standard cost-effective procedure in measuring visual acuity, particularly in rural areas with variable ambient illumination, while retaining the accuracy of a standard ETDRS chart. However, the utilization of this customised chart in various ocular pathological states warrants further studies.
|1||Mc Graw P, Winn B, Whitaker D. Reliability of the Snellen chart. BMJ 1995;310: 1481-2.|
|2||Ferris FL 3rd, Kassoff A, Bresnick GH, Bailey I. New visual acuity charts for clinical research. Am J Ophthalmol 1982;94:91-6.|
|3||Bailey IL, Lovie JE. New design principles for visual acuity letter charts. Am J Optom Physiol Opt 1976;53:740-5.|
|4||Gibson SA, Sanderson HF. Observer variation in ophthalmology. Br J Ophthalmol 1980;64:457-60.|
|5||Mbulaiteye SM, Reeves BC, Karabalinde A, Ruberantwari A, Mulwanyi F, Whitworth JA, et al. Evaluation of E-optotypes as a screening test and the prevalence and causes of visual loss in a rural population in SW Uganda. Ophthalmic Epidemiol 2002;9:251-62. |
|6||Bourne RR, Rosser DA, Sukudom P, Dineen B, Laidlaw DA, Johnson GJ. Evaluating a new logMAR chart designed to improve visual acuity assessment in population- based surveys. Eye 2003;17:754-8. |
|7||Bland JM, Altman DG. Measurement error. BMJ 1996;312:1654.|
|8||British Standards Institute. BS 4274:1968. Specification for test charts for determining distance visual acuity. London: British Standards Institute, 1968. Available from: http://avrc.city.ac.uk/o1handouts/BS4274:1968.|
|9||Sheedy JE. Standards for visual acuity measurement. In: Eye care technology forum proceedings. Bethesda, Maryland: National Institutes of Health, 1993. |
|10||Vanden Bosch ME, Wall M. Visual acuity scored by the letter-by- letter or probit methods has lower retest variability than the line assignment method. Eye 1997;11:411-7.|
|11||Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10.|
|12||Yang YF, Cole MD. Visual acuity testing in schools: what needs to be done. BMJ 1996;313:1053.|
|13||Bailey IL, Bullimore MA, Raasch TW, Taylor HR. Clinical grading and the effect of scaling. Invest Ophthalmol Vis Sci 1991;32:422-32.|
|14||Wesemann W. Visual acuity measured via the Freiburg visual acuity test (FVT), Bailey Lovie chart and Landolt Ring chart. Klin Monatsbl Augenheilkd 2002;219:660-7.|
|15||Recommended stardard procedures for the clinical measurement and specification of visual acuity. Report of working group 39. Committee on vision. Assembly of Behavioral and Social Sciences, National Research Council, National Academy of Sciences, Washington, D.C. Adv Ophthalmol 1980;41:103-48.|