In order to fully appreciate the benefits of third-party training and certification, it is important to understand the critical role that standardization of all aspects of Visual Acuity measurement methodology plays in ensuring the most reliable and repeatable trial data. It is helpful to look at historic developments in standardization to understand the current state of the ophthalmic field and the widespread issues with standardization that exist today.
Visual Acuity is the most often measured parameter of vision. The word “acuity” is derived from the Latin “acuitas”, which translates as sharpness. Visual Acuity measures the clarity or clearness of a subject’s vision. In simple terms, it is a measure of how well a subject can see. An alternative definition is that visual acuity measures the ability of the subject’s eyes to correctly distinguish shapes and details at specified distances.
When measuring Visual Acuity, one eye is tested at a time, as each eye will often differ in its performance. It is, of course, imperative to measure visual acuity in a standardized way, in order to correctly detect changes in vision over time.
Visual Acuity is normally tested using letter recognition but testing can also involve pictures, numbers or symbols. However, this is not straightforward as it involves more than merely the eye’s ability to resolve an image – complicating factors include, inter alia, cognition (to recognize each letter) and motor ability (to communicate answers to the examiner).
The test can help to determine if the patient suffers from visual conditions such as myopia, hyperopia and astigmatism to neural issues including age-related macular degeneration (AMD), a detached retina and amblyopia (lazy eye).
Visual Acuity - A Brief Timeline
In the long history of Visual Acuity and ocular improvement, developments in scientific theory, practice and treatment often occurred within the social and cultural frameworks of the time. We can find evidence of formal assessment of visual acuity over the last 4 centuries.
17th Century Spain
In 1623 Benito Daza de Valdés (1591–1634) wrote a book commonly known as ‘The Use of Eyeglasses’, which was an illustrated, technical manual for practitioners of his age that covered, inter alia, how to classify lenses and how to graduate eyesight. (The full title was ‘Use of spectacles for all genres of views in which it is taught to know the degrees that each one is missing from their sight, and the that they have any glasses and also what time they are to be used, and how they will be ordered in absentia, with other important notices, to the utility and preservation of sight’). In doing so, he is often said to have become the father of modern-day optometrists.
Daza de Valdés’ developed an early form of a Snellen chart. His method for assessing visual acuity involved using small, regularly sized objects (mustard seeds) and measuring the maximum distance at which subjects could no longer count them. By determining the optical correction required, Daza de Valdés was able to determine how concave or convex the corrective lenses needed to be and was able to prescribe lenses for his subjects accordingly. This was the first known attempt at visual acuity measurement.
1830s – 1840s Germany
In 1836, Heinrich Georg Kuchler (1811 – 1873), a German ophthalmologist, also attempted to introduce a standardized measure of visual acuity by developing an early form of a Cardiff Acuity test. His chart consisted of representations of common objects such as birds, frogs, farm tools, cannons and more. These decreased in size and were glued to sheets of paper to make a series of charts.
Later, in 1843 Kuchler developed a modified series of eye charts that used one word on each line, with twelve lines in total and each word decreasing in size down the page. He developed three instances of the chart in total to prevent memorization by his subjects. The chart set also came with detailed instructions for use by practitioners. Although innovative with a greater shift towards the standardization of testing visual acuity (not dissimilar to modern-day ETDRS charts), Kuchler’s letters were not of a uniform size. His work was widely discarded by his peers and all but forgotten soon after its introduction.
In 1864 Edward Jaeger von Jaxtthal (1818 – 1884), an ophthalmologist from Vienna, published a series of high-quality reading samples in several languages, which proved popular. However, the lack of a defined standard typeface meant that his visual testing was not standardized. Moreover, von Jaxtthal failed to define a standard viewing distance so he had no way to reliably compare performance at different distances.
In the 1860s, Franz Cornelis Donders (1818—1889), a Dutch professor of physiology, sought to improve the standardization of Visual Acuity measurement and testing. He introduced the forerunners of many concepts that exist to this today in the field of ophthalmology including refraction, astigmatism, accommodation, hypermetropia, ametropia, presbyopia, aphakia, convergence and squint.
In 1864, Professor Donders published “On the anomalies of accommodation and refraction of the eye”, a seminal book that focused on separated errors of refraction and accommodation.
Donders devised a formula to define the acuity or sharpness of vision. This involved defining a “standard eye” and then comparing the subject’s performance against the reference standard, noting the magnification the subject required to reach the same standard. Donders’ defined the measurement of Visual Acuity as the reciprocal of the magnification the subject needed. Thus, if the subject needed two times standard, his or her Visual Acuity was half standard, i.e. 0.5.
In 1862, Doctor Herman Snellen (1834 – 1908), an associate of Professor Donders, worked on a visual acuity standard to implement his “standard eye” formula. He developed charts, initially based on passages of text, then using abstract shapes, before finally settling on using individual letters as a better option.
Rather than using existing typefaces, Snellen designed stylized letters with proportional spacing on each row, which were only used for measuring visual acuity. He referred to them as “optotypes” and designed each based on a strict 5x5 grid. The ten letters used were chosen so as to be mistakable for others in the series but the progression of letter sizes was irregular and varied from chart to chart. These charts are still the most commonly used eye chart in clinical practice today.
In 1867-1868, Doctor John Green (1835 – 1913), Sr. of St Louis introduced a revised version of Snellen’s chart. Doctor Green had in fact previously completed a fellowship with Snellen. Whereas Snellen’s chart had rows of letters varying from 14 to 100 percent, with the letter sizing and spacing essentially dictated by the available space on the chart, Green regularised the progression in letter sizes. He redesigned them to be uniformly proportional to each other, with a geometric progression of 25 percent between lines.
Initially, Doctor Green’s chart had six steps from 20/20 and 20/200, matching Snellen’s non-logarithmic chart. This was later revised to 10 logarithmic steps in accordance with the decimal system.
Doctor Green also suggested a non-serif font (sans-serif i.e. without curlicues or flags), but this innovation was rejected by his contemporaries, ostensibly because “it looked unfinished”.
In 1888 Edmund Landolt (1846 – 1926), based in Paris, proposed the broken ring symbol. Landolt had previously worked with Snellen and they had realized that Snellen’s optotypes were not all equally recognizable. Landolt’s devised a broken ring symbols chart also known as the “Landholt C”. Each ring only had one gap therefore every symbol resembled the letter “C”. The only variation between them was in their orientation. The Landholt C continues to be widely used in laboratory studies today.
In 1959, Doctor Louise Sloan (1898 – 1982) of Baltimore introduced the Sloan Letters, a new set of 10 non-serif, perfectly square letters, designed for equal legibility with a mix of horizontals, verticals and diagonals. At that time, the most-popular Snellen letters and Landolt C’s were deemed sufficient for screening “within normal limits” but found to be insufficient at the low vision end (20/200 and 20/400). Doctor Sloan sought to address this and the Sloan Letters are commonly used in modern visual acuity tests including in logMAR charts.
In 1976, Ian Bailey and Jan E Lovie-Kitchin proposed a new chart, known as the Bailey-Lovie chart using the Sloan Letters. Whereas the Snellen chart had a variable number of letters per line forming a rectangular text format, this chart had five letters spaced on each line leading to an inverted triangular text format. This chart was also designed to be scored in logMar units enabling the examiner to know the exact size of letters on the chart and thereby make adjustments to visual acuity scores based on non-standardized viewing distances easier.
In 1982, the Bailey-Lovie chart was modified based on the recommendations of Doctor Rick Ferris of the National Eye Institute (NEI). The equally spaced Sloan letters were combined with the Green and Bailey-Lovie logarithmic progression layout to produce standardized charts for use in the Early Treatment of Diabetic Retinopathy Study (ETDRS). These charts thereby became known as ETDRS charts. The NEI subsequently mandated this chart format for all clinical trials, popularising them worldwide. To this day, this ETDRS format is the de-facto international gold standard for measuring Visual Acuity in clinical trials.
The Distance for Visual Acuity Measurement
As Visual Acuity is defined by the angle under which letters are viewed, it can technically be measured at any distance, as long as the scale is adjusted according to the distance used.
Snellen originally settled on 20 feet as the standard test distance and instructed all his patients to stand at that distance from the chart whilst they read aloud all the characters they could discern clearly.
Later, when the Metric system was introduced in France, he converted his charts to 5 and 6 meters. The ETDRS protocol further reduced the viewing distance to 4 meters in order to allow a more manageable (smaller) chart size.
Which Optotypes to Use for VA Testing
Snellen’s original letter chart used the English alphabet and letters are still the obvious first choice for literate adults. Many different letter sets have been used since but, since the introduction of the ETDRS protocol, Sloan Letters have become the norm.
European Wide optotypes are ETDRS charts that have been modified to contain only those capital letters that are used by all three of the European language scripts (Latin, Cyrillic and Greek). They are therefore recognizable by a much larger proportion of the population. After their introduction, the modified charts were validated against the original ETDRS charts in a number of countries including Belgium, Bulgaria and Greece and were found to produce comparable visual acuity.
After Sloan Letters, the second choice for adults is the use of Numbers. Even if illiterate, most adults can usually recognize numbers. In addition, deaf adults or those with speech impediments can always replicate the number seen using their fingers.
PV Numbers have a layout based on a Snellen 5×5 grid and are calibrated against Sloan Letters.
Tumbling Es are more suited for use with very young children. They are often used in studies conducted in developing countries.
Landolt Cs are rarely used in clinical practice and are more often encountered in laboratory studies. They are widely accepted as the standard against which to calibrate other optotypes.
Tumbling Es use 4 directional orientations whereas Landolt Cs use 8 directional orientations (including diagonals). Their successful use requires the subject to be able to indicate each of these orientations, which means they may prove problematic for children with developmental issues.
HOTV charts, like Tumbling Es, also offer only four choices. The four letters used are H, O, T and V, and these were chosen because they are R/L symmetrical. Occasionally, the letters A and X are also used.
Various picture cards have been proposed but one common problem is that, given their different backgrounds, not all children are equally familiar with all pictures.
Patti Pics are symbols with a layout based on the 5×5 Snellen grid. They are calibrated for equal recognition to Sloan Letters such that, when used correctly, visual acuity readings should not show any material change when child subjects progress from Patti Pics to regular letter charts.
Beyond Visual Acuity
Visual Acuity may be the most frequently measured aspect of vision, but overall visual function is certainly much more than just visual acuity. Despite the advances in visual acuity measurement, it could be argued that modern measurement techniques still lack functional relevance. In real life, other components also play a role, including decay in visual acuity during the inter-blink interval, the rate at which someone can read information and the impact of lighting levels and glare. More sophisticated testing may yet need to become standardized in order to take full account of such factors.
Moreover, Visual Acuity is just one parameter of visual function. Other parameters such as visual field and contrast sensitivity are just as important in determining a subject’s overall visual capabilities. For example, ETDRS eye charts have high contrast letters with almost 100% percent reflectance and almost contrast. Objects in the real world, however, tend to have much less reflectance and far lower contrast against their backgrounds. For this reason, contrast sensitivity may also be measured in order to better understand a patient’s overall visual function.
The “gold standard” for primary outcomes of ophthalmic clinical trials is Visual Acuity testing. Therefore, the standardization of Visual Acuity testing in ophthalmic clinical trials is of critical importance. Visual Acuity is measured using specialized eye charts which normally consist of uppercase letters which are arranged in rows. The biggest-sized letters are positioned at the top and progressively smaller letters are positioned further down.
In order to be valid and reliable, Visual Acuity testing should give precise, repeatable results free from the influence of external factors. The data obtained will then identify only changes related to disease progression or treatment efficacy. Unfortunately, in the real world, Visual Acuity testing can also be influenced by a number of other factors such as the experience of the examiner, lighting levels in the testing rooms, and the design of the test charts used. Here we will focus on the potential unwelcome variability introduced by the choice of testing charts used to measure Visual Acuity.
A number of chart types can be used for Visual Acuity measurement. In clinical practice, Snellen charts are most common and are widely available and relatively straightforward to use. However, in clinical trials, ETDRS charts are the current “gold standard” for Visual Acuity testing. Indeed, FDA registration trials specifically require ETDRS charts for measuring visual outcomes, which in large part explains their popularity in clinical trials worldwide.
Deficiencies in Snellen Charts
Snellen charts have a number of important disadvantages:
Firstly, scoring is done using the line assignment method, whereby patients are scored for lines read, not letters. Since there are a variable number of letters on each line (e.g. poor vision lines usually contain only 1 or 2 letters, whereas good vision lines may contain 8 letters), this introduces a lack of standardization in terms of the progression between lines. Thus, a single missed letter on the poor vision lines is much more significant than missing a single letter on the good acuity lines. In terms of scoring, a single letter may sometimes result in one line less being scored than in other instances, and this is more likely to be the case for poor vision lines. This design feature may have been chosen for the simple aesthetic reason of making each row of letters the same approximate length. But, in doing so, it introduced this significant lack of standardization between rows.
Second, the progression of letter sizes from line to line is arbitrary and irregular, rather than standardized. This can lead to visual acuity being overestimated at the lower end of vision in particular. Hence, the loss or gain in a line of vision will not have the same level of significance in all lines on the chart.
Third, the line assignment scoring method and the consequent lack of standardized progression from line to line serve to prevent the measurement of Visual Acuity on a fine scale. This in turn means that Visual Acuity measured using Snellen charts is not suited to statistical assessment and therefore unsuitable for use in clinical trials.
Fourth, the legibility (or ease of reading) of different letters on a Snellen chart is not always the same. For example, letters such as C, D, E, G, and O are considered easier to read and less prone to mistaking for other letters than say A, J, or L.
Fifth, the distance between each of the letters and rows in the chart is not standardized. It has been demonstrated that when letters are placed too close to each other, a “crowding effect” from the adjacent contours of too tightly packed letters can occur, leading to diminished visual acuity on that chart. This crowding effect will further vary across the chart (depending on the chart’s design) – as the poor vision lines will have larger spaces and therefore no risk of crowding whereas the good acuity lines are more prone to crowding. As such, patients may fail to read a line because of this crowding phenomenon and not because they cannot see the letters.
Finally, the term “Snellen chart” has itself never undergone standardization i.e. there are no clear standards or criteria as to what can be labeled a Snellen chart design. As long as the broad characteristics are used, different manufacturers will label their charts as Snellen charts despite differences in fonts, letters, spacing ratios, and even illumination or projection method between charts.
Essentially, the testing of any given line on a chart should present an equivalent task for the patient apart from the single variable – the size of the letters. But as we can see, Snellen charts can introduce a number of additional, unstandardized variables.
If the Visual Acuity of a single patient is tested repeatedly on the same chart, we should theoretically expect there to be no difference in scores. However, in practice, we will find a distribution of scores. This range of scores will reflect the underlying variability introduced by the chart design and its associated measurement protocol (independent of any clinical change). This variability between patient visits is known as test-retest variability (TRV). As the “noise” of TRV increases, our ability to detect true changes in vision decreases.
The numerous shortcomings of the Snellen chart mean that its TRV can be very large, varying from ±5 to ±16.5 letters in normal vision patients and up to 3.3 lines in poor vision patients according to one study. Another study showed that patients had a 2-line difference in vision following multiple tests on a Snellen chart. In short, Snellen charts’ lack of standardization and associated issues can lead to large differences in Visual Acuity measurement that are due to chance rather than true changes in vision.
Advantages of ETDRS Charts
ETDRS charts are the result of a number of improvements over time and their design was finalized by Dr. Rick Ferris for use in the landmark Early Treatment Diabetic Retinopathy Study (ETDRS). They have the following design features:
First, the letters have almost equal legibility. While not quite the perfect legibility as the “Landolt C” or “illiterate E” letters, the letters are non-serif and designed on a precise 5x5 grid. Thus, the letter size is the single factor affecting the difficulty of each line. Spacing between each letter is consistent and each row of letters is proportional to the letter size. There is 1 letter-width between letters and 1-letter height between rows of letters (based on the smaller row size). This serves to eliminate the crowding phenomenon prevalent with the Snellen chart design.
A standardized geometric size progression exists for each line in the chart – the letters double in size every 3 lines. Therefore, by precisely varying the size of the letters, testing distance can be changed as desired without affecting the standardization of Visual Acuity scores.
As a result of these design features, ETDRS charts can be used to measure Visual Acuity scores by letter rather than by line. By using single-letter, forced-choice testing, this chart was able to demonstrate more consistent TRV on different days, between different examiners, and even across different clinical sites. The TRV varied from ±3.5 to ±10 letters, depending on whether the patients had normal vision or low vision – a considerable improvement over Snellen charts.
The lower TRV figures mean that ETDRS charts are more reliable than Snellen charts at measuring longitudinal patient vision, whether they had high or low vision. Interpatient differences in vision were also more reliable, making them ideal for multi-site clinical trials.
In terms of statistical analysis, acuity scores from ETDRS charts can be converted into logMAR notation. Since the chart is standardized across any and all sites, consistent statistical analysis can be performed on Visual Acuity recorded using these charts.
The Continuing Popularity of Snellen Charts
Despite ETDRS charts being shown to be more accurate, many peer-reviewed articles published in major ophthalmology journals have used Snellen charts to measure visual acuity. A principal reason is that clinical testing with ETDRS charts is considered to take longer, require specialized lanes, and be more difficult to administer than testing with Snellen charts. As such, widespread adoption of ETDRS charts has not occurred. In contrast, FDA registration trials require ETDRS charts for visual outcomes. This makes it difficult to correlate results from “real world” vision and smaller clinical trials with findings from phase III clinical trials.
The use of Snellen charts in clinical practice is problematic. For example, it is normally the case that patients will have better visual acuity scores on ETDRS charts compared to Snellen charts. As a result, it is not uncommon when screening patients for their Visual Acuity to “improve” from their Snellen score when tested on the ETDRS chart – even to the point where the patient is no longer eligible for the trial. Since the average increase was often far greater for subjects with low vision (up to 5 lines in some cases!), it is in these cases that measurements from the two charts are the least comparable.
Interestingly, historic reporting of Snellen Visual Acuity results in ophthalmic literature has sometimes focused on numbers of lines or fractions of lines, sometimes the results were converted and discussed in logMAR format and, occasionally, they were even converted to and discussed in decimal format. This variation in approach makes it impossible to compare the Visual Acuity outcomes between these studies.
This lack of standardization is more than just an academic issue; it can have important implications in the real world. For example, if a new ophthalmic drug is trialed using Snellen charts and appears to show similar results as previously published for an existing successful drug that was trialed using ETDRS charts, one might assume that, given the apparent similarity in the drugs, and based on the improvements in the vision of trial subjects on the newer drug, that it worked just as well as the older one. However, the lack of standardization in Snellen Charts means that results cannot be reliably compared with trials using ETDRS protocol charts. Thus, even if the new drug is as effective as the old one, a high degree of confidence cannot be leveled until both drugs are tested using the ETDRS protocol for Visual Acuity.
Snellen vs ETDRS Charts
The measurement of vision is the primary outcome of most ophthalmic clinical trials and indeed all FDA-registered ophthalmic trials. Visual Acuity is used to measure both disease and change in the eyes of patients and trial subjects. Consistent and repeatable Visual Acuity measurement is essential in clinical trials to reliably measure treatment response. Ideally, the choice of chart should not influence the results of Visual Acuity measurement. However, in practice, our ability to measure Visual Acuity is, arguably, still relatively crude. Accurately ascertaining Visual Acuity can be influenced by any number of factors, including light intensity; number, size, contrast, and shape of the optotypes; and the design of the test chart.
Given the increasing focus on evidence-based medicine to guide treatments, it is important to be able to compare clinical trials to clinical practice. Nonetheless, the chart type and testing method used in clinical trials give sufficiently different Visual Acuity measurements to indicate that this is not possible. Despite its known deficiencies, including unreliability and poor repeatability, the Snellen chart is the most widely accepted protocol used to assess Visual Acuity in clinical practice despite its significant deficits. Clearly, it is a convenient and easy-to-use protocol and has become the incumbent standard. This is even true for many retrospective case series and medicolegal decisions. Unfortunately, the shortcomings of the Snellen chart do not allow appropriate quantification and comparison of vision in these important areas.
To reduce the variability and enhance the precision of Visual Acuity testing, the ETDRS charts were introduced, adding a standardized administration and scoring protocol. These are the current “gold standard” for Visual Acuity testing in clinical trials worldwide. Moreover, all FDA registration ophthalmic trials require the use of ETDRS charts for Visual Acuity measurement. They allow for far more precise quantification of Visual Acuity and are a far more reliable measure of vision change. It is hoped that the perceived issues with ETDRS charts (the large size of the chart, the unfamiliarity with the testing and scoring protocol, the time it takes to perform, and the number of letters on the chart) are overcome and that this standard becomes even more widely accepted in clinical practice and in medicolegal scenarios.
For more on this topic, refer to Prospective Evaluation of Visual Acuity Assessment: A Comparison of Snellen Versus ETDRS Charts in Clinical Practice by Peter K. Kaiser, MD
The Need For Training and Certification
Whilst the use of ETDRS charts can reduce the unwelcome Test-Retest Variability introduced by Snellen charts, only consistent training and certification by expert certifiers can reduce the additional variability that can be introduced by investigators in trials. In our experience, differences in examiner skill, training levels, and methodology is an independent source of variability that can, if not controlled for, lead to significant differences in Visual Acuity measurement that worsen the quality of trial data.
The Importance of Standardization
Visual Acuity measurements are incredibly sensitive; small differences in measurement methodology or other factors can lead to large variability in results. This means that when Visual Acuity is an end-point for ophthalmic clinical trials, it is vital that it be measured and recorded in a highly standardized way if the trial data is to be both reliable and repeatable. In the absence of such standardization, trial data may prove to be of much lower overall clinical integrity.
For multi-site studies, standardization is required between examiners, between sites, and indeed between countries. For multi-year studies, it is also critical to ensure standardization of all these factors over time.
Standardization has a number of dimensions:
Standardization of Equipment
The charts used for a study must be standardized. The ETDRS chart and protocol regulate the following factors:
• size, number, shape, and contrast of the optotypes;
• design of the test chart;
• light intensity.
This standardization of charts allows for more precise quantification and measurement of vision and therefore vision change. The ETDRS chart and protocol is therefore the current “gold standard” in ophthalmic clinical research.
Standardization of equipment over the duration of a study is also important. It is not uncommon for room conditions to deteriorate over time, to the point that rooms initially fail certification by the time their annual re-certification is due. The periodic recalibration and re-certification process is therefore critical in checking and maintaining room and equipment standards.
Standardization of BCVA Practices
In clinical trials, inconsistencies in BCVA measurement lead to inconsistent and unrepeatable results.
A 2008 study looking at the uniformity of Visual Acuity measures in published studies discovered significant inconsistencies with regards to methods of BCVA measurement in clinical trials with various methods employed at the site level, ultimately leading to inconsistent and unrepeatable results.
A 2011 study by the Wilmer Eye Institute at Johns Hopkins Hospital provided a shocking analysis in regards to how poorly BCVA processes are understood and interpreted at the site level. Just over 20% of examiners (ophthalmologists in this case) were able to interpret logMAR scores correctly with 0% able to interpret final ETDRS scores correctly!
Stringent training and certification of all participating sites is therefore an indispensable prerequisite if reliable and repeatable results are sought in any clinical trial.
Standardization of Examiners
All participating examiners must be regularly trained and tested to ensure their understanding/adherence to the BCVA protocol. After certifying thousands of examiners at thousands of sites worldwide for multiple studies for multiple sponsors, our internal analysis reveals that many examiners involved in trials actually fail the certification process.
Firstly, we have found the training/certification process, just like any internal recruitment or appraisal process, is paramount in being able to ensure the quality of examiners selected for trial participation. Many who are initially selected are academically qualified but fall short in practice and fail their initial certification. Initial examiner training and certification at the outset, therefore, serve as an important baseline at the start of a trial.
Secondly, relapse of examiner ability over time is a common issue. Examiners fail consistently, and often at the time of annual re-certification – when they have had an additional year’s worth of experience with the protocol – when one would assume they would pass.
It should be noted also that we regularly have to offer aftercare and refresher training for even 'certified' examiners. Errors are often reported in the Refraction and BCVA process well into a trial. These errors can often result in differences in BCVA scores of up to 15-20 letters.
Thus, the ongoing training and recertification of examiners are critical in ensuring ongoing standards and adherence to protocols.
Finally, we have language barriers to consider. These also play a huge part in having comparable and consistent results. Since the onset of COVID, we have very successfully switched over to remote online training and multilingual-supported certification. This has proved instrumental in ensuring examiners are performing well, whilst also making the service efficient and cost-effective for our sponsors.
Standardization With Training & Certification
Only by controlling all of these conditions and factors, across all trial sites and over the full duration of the study, can trial sponsors have confidence in the reliability of their trial data.
A comprehensive, periodic training and testing program for sites and examiners is therefore critical to achieving the required level of standardization in ophthalmic studies.
A 2020 study looking at quality control procedures (including for ETDRS BCVA) from a multi-site ocular safety study concluded that “the data illustrate the benefit of identical equipment, stringent on-site instruction, and training, quality control, certification, and validation methods. The latter are recommended for planning and conducting multicenter trials … to monitor safety and/or efficacy of treatment intervention.”
Furthermore, it stated that such “stringent quality control procedures and reliable reference values are indispensable prerequisites for informative clinical trials.”
To date, only a limited number of formal clinical trials have been conducted that specifically address the need for and benefits of training and certification for multi-site ophthalmic studies. But what research does exist only serves to confirm our long experience in the field, that the certification process is a vital factor to ensure repeatable and reliable data is collected.
In time this dearth of research will be addressed, but until then, the onus is upon trial sponsors to insist on certification when running their trials if they wish to ensure their study data is of the highest quality.