Sir,

We would like to thank Dr Anderson and colleagues for the time and effort they made to respond to our article.1 We are in hearty agreement with them that systemic side effects are an important consideration in β blocker therapy.2, 3 This issue was considered extensively in our model and is noted in the 4th paragraph under ‘Markov model: medical aspects’ on the 2nd page of the article. To account for persistency rates, we used numerous published studies that included discontinuations for reasons of systemic side effects. The additional costs of these side effects were accounted for by the inclusion of the extra visit to change medications. We did not include costs for the primary care physician because of lack of evidence in the literature that such visits occur, even occasionally, because of the acute systemic side effects from β blockers. We believe this happens probably because ophthalmologists are generally adept at avoiding patients who might have acute, serious, systemic problems with topical β blockers.

Dr Anderson and colleagues’ letter, however, does raise an important point in that there are few data regarding the general costs of chronic β blocker therapy. This is why we developed our Markov model, to begin to address this issue. In addition, there are few data specifically over the costs of systemic side effects for which they wrote their letter. This is the reason that we had to use persistency figures. Consequently, many questions remain about β blockers and systemic side effects:

  1. 1)

    With the greater use of systemic β blockers in primary therapy in heart failure and prevention of heart attack, how does this change how ophthalmologists view this class of medicine? Otherwise, in the long term, do topical β blockers protect or hurt patients systemically from a cardiopulmonary standpoint?

  2. 2)

    From a CNS standpoint, how do β blockers affect the patient's mentation and mobility? Otherwise, is there a cost to society from topical β blockers in lessening the patient's independence and their activities of daily living?

  3. 3)

    Can we develop better prescribing guidelines to assist physicians in knowing in which patients to avoid this class of medicine to better prevent adverse CNS and cardiopulmonary effects?

  4. 4)

    Is there yet a β blocker that could be developed that would better avoid systemic side effects?

Again, we would like to thank Dr Anderson and his associates for their interest in our article. We congratulate them on their commitment, as we all should have, to understanding the effect of this important class of medicine in our patients’ lives.