Influenza vaccine for all?

As systematic reviews and meta-analyses have become established as methods for evidence-based decision-making, reviews on similar questions have been published, sometimes with discordant results. Recommended approaches to reconciling these differences include determining if the results truly differ or if the variation arises from the interpretation of the results.¹ Ross Upshur notes that different conclusions on vaccination of the general public were reached by the Canadian Task Force on Preventive Health Care (CTFPHC)^2,3 and the Cochrane Collaboration.⁴ In this case, the reviews covered different populations (healthy adults and children in the CTFPHC review, healthy adults only in the Cochrane review) and considered different interventions (vaccines and neuraminidase inhibitors in the CTFPHC review, vaccines only in the Cochrane review). There were also differences in methods: CTFPHC reviews are systematic qualitative reviews,^5,6 whereas the Cochrane reviews are generally quantitative reviews.^4,7 As noted in the Methods section and Fig. 1 of our review,² we reviewed the Cochrane database to find primary trials that might not have been identified in our literature search. The trials that were judged acceptable were not identical in the 2 reviews.

Perhaps the most important difference between the 2 reviews is in the interpretation of cumulative evidence for influenza vaccination in healthy people. The Cochrane reviewers concluded that the efficacy of inactivated influenza vaccines (the type of vaccine that is available in Canada) was 70% (95% confidence interval 56% to 80%) in healthy adults, but thought that this was insufficient evidence to support general vaccination.⁴ The CTFPHC concluded that vaccination was a moderately effective intervention to reduce influenza in adults and children, without evidence of harm, and recommended it.³ The clinical significance of a 70% reduction in influenza virus infection will likely be of variable importance to patients and their families, clinicians and other health care providers, and payers. The ultimate decision to offer influenza vaccination rests with those who must balance the broader issues of universal programs, such as the practicability of vaccinating large populations in a short period of time, public acceptance, vaccine procurement and the value of this intervention relative to other health prevention or treatment interventions.

Joanne M. Langley Associate Professor Departments of Pediatrics and of Community Health and Epidemiology Dalhousie University Halifax, NS John Feightner Chair Canadian Task Force on Preventive Health Care London, Ont.