Article Text
PDF
Statistics from Altmetric.com
Editor,—The letter by Dayanet al 1 raises some interesting questions regarding the examination and referral of patients by optometrists for further investigation in relation to open angle glaucoma.
While the authors base their comments upon largely anecdotal evidence from a series of only 11 subjects from one referral source, there are nevertheless some important points to be answered from these comments.
Firstly, the College of Optometrists guidelines offer guidance based on clinical evidence to optometrists conducting eye examinations. They specifically encourage optometrists to conduct the appropriate tests on any individual patient as a matter of best practice. In the case of glaucoma, recommendations are made that visual field tests should be conducted on subjects over the age of 40, those with a family history of glaucoma, and those with suspicious optic discs or other risk factors.2
The majority of visual field tests used in optometric practice are based upon static perimetry (College of Optometrists annual clinical survey, 1998) and decisions then need to be made on the results obtained. Therefore, if one or two points are missed on a central field test does that mean an abnormality is present? The optometrist needs to make a judgment on this issue in the light of the complete clinical findings rather than simply exert a pass/fail criterion from a screening test.
Not surprisingly most glaucoma patients in hospital clinics are referrals from optometrists but evidently this is at the cost of a relatively high false positive rate.3 Not surprisingly also, visual field test designers are continually trying to develop programs with high specificity and high sensitivity,4which help the user to make an appropriate decision regarding normal or abnormal findings and thus referral.
Whether the optometrist refers a patient to the general practitioner for further investigation is influenced by a number of factors. Optometrists are legally required to refer patients if an abnormality is suspected or found.
Therefore, realistically, individual practitioners are more concerned about missing pathology than referring false positives. This feeling is further strengthened by legal cases reaching the General Optical Council where optometrists have been sued (successfully) for not detecting and referring ocular pathology.
This defensive attitude could be changed. Clearly, ophthalmologists do not wish to spend clinical time seeing large numbers of “normal” eyes although there must be a grey area where a patient will be referred with a reasonable measure of doubt as to the presence of an abnormality. What is an acceptable false positive rate with respect to referrals to a hospital eye clinic?
One of the main ways in which the referral process could be improved is by a formal notification system from ophthalmologists to optometrists to let individual optometrists assess the accuracy of their clinical findings/decisions/results of the cases that they refer. The health service referral forms are designed to allow this to happen since the patient can be asked to sign giving consent to the general practitioner or ophthalmologist to pass information back to the referring optometrist. Ideally the ophthalmologist will send a copy of the reply to the general practitioner and to the referring optometrist. Currently and historically this process does not happen in many areas of the country (personal communication with members).
Reply
Editor,—We thank Dr Stevenson for his interest in our letter. Though he rightly comments that our evidence is anecdotal there has been no systematic study to justify the use of routine visual field screening in patients over 40 attending their optometrist.
The best way the improve the positive predictive value of visual field screening would be to apply it to a group with a higher pretest probability of having pathology—for example, patients over 60 years of age. We doubt that forthcoming technological developments will improve the trade off between sensitivity and specificity to a point where routine screening of patients over 40 can be justified. Furthermore, a test of adequate positive predictive value is a prerequisite for a screening programme not a post hoc modification.
With regard to the legal position, this has in part been created by the guidelines themselves since a practitioner who does not comply with them can be held to have been negligent.
The fault does not lie with individual optometrists who cannot be blamed for making the referrals; having identified a field defect they have little option but to refer to the hospital eye service. Informing the optometrist of the outcome of all referrals is undoubtedly good practice but we do not think it would have any significant impact on the false positive referral rate. It is the guidelines themselves that are the problem.
We note that there is some disquiet among optometrists as well as ophthalmologists about the guidelines. This is a difficult area in practice and there are no easy answers; however, it is time to review the cut off age for routine visual field screening.
Author’s reply
Editor,—I thank Dr Griffiths for his reply to my original comments. However, I would like to answer some of the points that he raises.
The use of visual field screening in patients above the age of 40 in optometric practice is based upon the fact that the incidence of glaucoma increases above that age. Therefore, measuring intraocular pressure and visual fields and combining this with assessment of the optic disc is highly relevant in that group of the population. The highest positive predictive value (PPV) is demonstrated when information on all three factors is included.
Logically it follows that the PPV of visual fields as a “stand alone” test increases if it is only used in a group with a higher cut off age—that is, 60 years of age as suggested by Griffiths, given the age related nature of many ocular conditions including glaucoma. This approach is likely to miss a significant number of possible early chronic simple glaucoma cases. Conditions other than glaucoma may also be detected by routine visual fields testing, a point made in Griffiths’s original letter.
The legal position of the college guidelines may be misunderstood by Griffiths in that, by not complying with them, he suggests that a practitioner may be held negligent. This is not the case since the guidelines are not set by the General Optical Council as law but are simply College of Optometrists professional guidelines given to members to represent current best practice.
Most optometrists feel that if they receive feedback on referrals to indicate a high false positive referral rate, referral criteria would be modified to correct this problem. Recent ophthalmological opinion does suggest that feedback on referrals would help case finding.
Finally, I do agree that it would be welcome and timely to have dialogue between ophthalmologists and optometrists regarding the age above which individuals should have their visual fields tested at routine eye test appointments. Much of the misunderstanding regarding referrals in this area could then be clarified.
This is also relevant in light of the current review of the GOC referral guidelines to optometrists relating to the present obligation to refer to a general practitioner when an abnormal finding is detected in the routine eye test.