Sir,

Dr Verma restates Wilson and Jungner's1 principles of screening. As these principles have been debated at length elsewhere, and reference frequently made to how few of them are actually satisfied by the majority of our existing societal screening programmes,2, 3 I propose to address only the fifth; namely, the requirement for a valid and reproducible screening test. I feel entitled to address this issue because it was actually the point of our editorial,4 and Dr Verma adds usefully to the debate on this subject drawing our attention to a recent paper from the Preferential Hyperacuity Perimetry Research Group.5 He also highlights points 7 and 9, which are indeed crucial to the wider debate, but do not affect our argument, namely that more prediagnostic steps merely threaten to reduce the yield of screening.

Economics aside, reducing yield is a serious concern for health professionals who wish to help as many patients as possible with this disabling and highly prevalent condition. However, of course, one cannot put economics aside. The opportunity cost of screening for and treating ARMD must be considered responsibly within the broader obligations of the NHS to all patients and all diseases. As Dr Verma correctly states, addressing the ninth point in his letter, advances in treatment such as Bevacizumab (although probably not PDT, Pegaptanib sodium, or Ranibizumab) promise highly acceptable management costs. This is entirely independent of the cost of case-finding, however, which is the other half of point 9, and which we dealt with briefly in our editorial.

If we are to reliably detect and promptly refer people with early neovascular ARM, what we need is a good screening test. A test that is safe, preferably inexpensive, and certainly valid (has acceptable sensitivity and specificity). Early indications, which still need confirmation by other workers, indicate that PHP may fulfil this role. This is indeed exciting news. The test will still be opportunistic, unless the government embraces the concept of a formal screening programme of course, so by no means everyone will benefit. However, that is unduly disingenuous. We have to start somewhere and where better than with improving early detection in second eyes and self-referring elderly patients in the high street. If the test lives up to its early promise and is operator independent, the real goal surely would be to roll this out to all optometric practices. Why have two tiers of optometrist when, as we have argued, it is inherent to such a hierarchy that more cases will be missed? In other words, our original question: What will the new optometrist with special interest achieve? is answered, even in the presence of better tests that we now possess. It will still reduce yield.

However, it may also introduce another problem. The PHP research group excluded one-third of enrolled cases. There will always be a group of patients ineligible for PHP. The best way to help this population would surely be direct referral to a hospital retinal specialist for flourescein angiography. Again, these patients would gain little from a further prehospital consultation. Indeed, this would delay potentially eligible patients accessing treatment at the earliest stages of the disease. Some lesions may even become too advanced for treatment by the time of hospital attendance.

The cost of case-finding, mentioned by Wilson and Jungner, is more than merely the cost of the test itself, however. It includes all the costs incurred along the patient journey. Making this journey longer will of necessity make it more expensive per case detected.

Dr Verma's letter does raise one final interesting point, however, namely the issue of compromising interests in journal publication. He suggests, no doubt amicably of course, that we have such a compromising interest, namely a ‘resistance to change’ and even an imputed motive; a less than professionally courteous relationship with optometry. We are all familiar, of course, with the disavowal of compromising interests when they threaten the intellectual neutrality of original work. Are the same rigours inapplicable to letters and editorials? Of course not. Indeed, such is the human tendency to accept argumentum ad verecundiam (appeal to authority) that they should be at least equally transparent. I accept that there are problems with disclosure,6 but of course the risks of nondisclosure are significantly greater. For example, I am sure that the fact that Dr Verma is the founding director of an independent profitable concern that markets detailed screening tests for macular degeneration will not compromise his neutrality in this debate, but it would seem that readers should be aware of the facts and allowed to draw their own conclusions.