Grant Hannaford discusses why practitioners get more from a lens with greater attention to quality measurements, and how this translates into better patient outcomes.
Most, if not all, practitioners who dispense optical appliances engage in the assessment of the quality of measurements on a daily basis. So often so, that it’s become an unconscious exercise.
Since the introduction of freeform lens generation techniques, lens manufacturers have implemented ever more complex surface geometry in their designs. Many of the benefits to be found are only realised once practitioners increase the precision of their measurements.
This is a common source of complaint where practitioners do not see any meaningful difference between older and newer lens designs, simply due to measurement techniques.
Precision, accuracy, uncertainty
These terms are often used interchangeably. While they all deal with aspects of the quality of a measurement, they speak to different parts of the process.
Precision generally refers to the smallest increment or scale of our measurement for example, 1mm or 0.01mm. It may seem like 0.01mm increments are better but if our equipment is poorly calibrated or maintained, then we may find that the larger increment is the better one to use as it is more reliable.
Uncertainty for measurements indicates the range in which our true value might be expected to fall and may be systemic (predictable and inherent to the system or device) or random. If we are using a millimetre ruler, then a reasonable uncertainty would be half the smallest increment i.e. 0.5mm.
That is to say, if we measured a pupillary distance (PD) as 32mm, then there is a chance the measurement might be 31.5mm or 32.5mm as well, simply because the smallest increment on the ruler doesn’t reliably allow for better measurements than this. Experienced practitioners may indicate they can give results to 0.3mm or even 0.1mm on a millimetre ruler, data that suggests that while they can record a result at this resolution, the results are not reliably replicated [1-4].
These elements, which contribute to the uncertainty of the data, accrue, so that a person using a PD rule to take measurements may feel they are working to millimetre precision. In reality the uncertainty for a full set of facial measurements can be as high as 3mm once all the variables are accounted for (using the RSS method)[5].
Finally, we can look at how accurate our measurements may be through their repeatability. If our measurements are always giving the same answer, then we have good quality data. However, if the measurements are constantly changing then we have an issue with the reliability of the data and our accuracy will be suspect.
The ‘error box’
Taking measurements to greater levels of accuracy ensures the patient’s visual system is aligned correctly with the corrective lens. In an ideal fitting, the eye will be aligned perfectly with the appropriate fitting point of the lens so the design can function correctly.
This can only occur once we have fitted the frames prior to taking measurements. In reality patients will have slightly different placement of their spectacles every time they put them on but using the concept of an error box we can minimise the effect of these changes in placement.
Figure 1 shows two placements of an eye in a hypothetical error box with a yellow fitting cross at its centre representing the area in which a lens will perform best. In the left image the eye is offset in the error box so that if the patient were to put their spectacles on crooked there is a high chance that their eye may be outside the area of best vision, making the lenses perform poorly.
The image on the right shows a more accurate placement with maximum tolerances in place to allow for changes in fitting or wearing placement.
Comparing measurement devices
In the second part of this article we will consider the different types of measuring techniques and systems used in practice, ranging from rulers through to tower-based systems.
The ideas above have been used to determine the systemic uncertainty and hence relative reliability of some of these systems for nine monocular measurements.
These uncertainties are solely systemic. Once we discuss the influence of the operator in part two of this article, we will see how potential errors for some techniques can exceed 4mm, even with a skilled operator.
REFERENCES:
1. McMahon, T.T., E.L. Irving, and C. Lee, Accuracy and repeatability of self-measurement of interpupillary distance. Optom Vis Sci, 2012. 89(6): p. 901-7.
2. Brooks, C.W. and I.M. Borish, System for ophthalmic dispensing. 3rd ed. 2007, St. Louis, Mo.: Butterworth Heinemann. xx, 665 p.
3. Brooks, C.W.R., Hubert D, Efect of Prescribed Prism on Monocular Interpupillary Distances and Fitting Heights for Progressive Add Lenses. Optometry and Vision Science, 1994. 71(6): p. 401-7.
4. Gerstman, D.R., Ophthalmic lens decentration as a function of reading distance. Br J Physiol Opt, 1973. 28(1): p. 34-7.
5. Gbur, G., Mathematical methods for optical physics and engineering. 2011, Cambridge ; New York: Cambridge University Press. xvii, 800 p.
ABOUT THEOR AUTHOR: Grant Hannaford is the co-founder and director of the Academy of Advanced Ophthalmic Optics. He has been practising in optics for more than two decades and works with optometry and dispensing students, as well as industry professionals.
