Letters

# Integration of hepatitis B vaccination into national immunisation programmes

BMJ 1998; 316 (Published 14 February 1998) Cite this as: BMJ 1998;316:552

# Authors should have taken open minded view of all relevant evidence

1. W J Edmunds, Wellcome Fund postdoctoral research fellowa
1. a Department of Biological Sciences, University of Warwick, Coventry CV4 7AL
2. b World Health Organisation, Geneva, Switzerland
3. c Centre for the Evaluation of Vaccination, Epidemiology and Community Medicine, University of Antwerp, Belgium

Editor—There is currently considerable interest in strategies of vaccination against hepatitis B in areas of low endemicity, such as the United Kingdom. Since there is little argument over whether this vaccination is safe, effective, and desirable given sufficient resources, the debate has largely focused on the relative cost effectiveness of selective versus mass immunisation options.

In their article Van Damme et al used economic (and other) arguments to advocate mass immunisation against hepatitis B.1 Unfortunately, their section on economic evaluation is misleading. Specifically, on page 1035 they state that “cost effectiveness studies performed in countries with low endemicity (Belgium, Canada, United Kingdom, United States) consistently find that universal vaccination is economically attractive.” They support this statement by referencing four studies. They do not, however, reference those studies that do not agree with this statement (such as that by Williams et al on the cost effectiveness of vaccination against hepatitis B in the United Kingdom2). The evidence is therefore not consistent.

More serious than this apparent error of omission is that the authors proceed to reinforce their argument by stating that “Cost effectiveness ratios varied from $1000 to$20 000 (£625-12 500) per discounted life year gained depending on the country's epidemiological and organisational characteristics,” which, they argue, compares favourably with such ratios in other, existing, prevention programmes. The clear implication of this statement is that the United Kingdom study that they quote (by Mangtani et al3) found that the cost per discounted life year gained for universal immunisation lies in this range. But in fact Mangtani et al estimated the cost per discounted life year gained to be £51 817 and £94 821 for mass immunisation of adolescents and of infants respectively—roughly four to eight times higher (four to eight times less cost effective) than the upper limit that Van Damme et al quote. Ironically, in their discussion of the economic evidence Van Damme et al state that “Health policy makers should look carefully at these studies.” I agree wholeheartedly, provided that policy makers and those who advise them take an open minded view of all the relevant evidence.

The case for mass vaccination in the United Kingdom remains to be made convincingly. In the end, the best argument for mass vaccination against hepatitis B may be that global eradication of the virus is unlikely to be achieved without it.

## References

1. 1.
2. 2.
3. 3.

1. M Kane, Medical officer, Global Programme for Vaccines and Immunisationb,
2. P Van Damme, Researcherc,
3. A Meheus, Professorc
1. a Department of Biological Sciences, University of Warwick, Coventry CV4 7AL
2. b World Health Organisation, Geneva, Switzerland
3. c Centre for the Evaluation of Vaccination, Epidemiology and Community Medicine, University of Antwerp, Belgium

Editor—In our article we used different arguments, including economic ones, to advocate routine immunisation against hepatitis B. Edmunds focuses on the economic arguments, and his letter is based on two issues: that “the debate has largely focused on the relative cost effectiveness of selective versus mass immunisation options” and that we may have misled readers by quoting only studies that supported our position.

With respect to the first issue, although Edmunds and some other workers in Britain and Scandinavia believe that selective immunisation (for high risk groups) is a viable strategy for controlling hepatitis B, we do not agree. No country has succeeded in vaccinating a significant proportion of the high risk groups. The failure of selective immunisation against hepatitis B is due not only to the potentially correctable fact that few resources have been devoted to the problem but, more importantly, to the fact that these high risk groups are not effective targets for immunisation programmes. If selective immunisation is not an effective public health strategy for controlling hepatitis B on a community basis, cost effectiveness analyses that recommend it as the primary option are not realistic and give health officials and politicians who are reluctant to devote resources to control of hepatitis B an excuse not to do so.

With respect to the second issue, we chose to cite four studies which we thought were reasonably well done. The United Kingdom study by Williams et al was published in December 1996, after we sent our article to the BMJ (date of acceptance 10 September 1996).1 Our article did not focus on economic evaluations or on the situation in the United Kingdom specifically. We agree that the range of all published economic evaluations in countries with low endemicity is wider than that mentioned in our article.2 Our article is an excerpt from a chapter published elsewhere3 and refers to the majority of studies performed in regions of low endemicity. Edmunds is correct in noting that the results of Mangtani et al's study4 were not in the range mentioned in our article, but we did not intend to deceive readers. Because of different methodologies and different assumptions on possible alternatives and input data, direct comparison of different studies is impossible and wide ranges of cost effectiveness outcomes are reported. However, all the cited studies reach a similar conclusion: that universal hepatitis B immunisation programmes showed cost effectiveness ratios that are comparable to those of other well accepted public health programmes that have already been implemented.

## References

1. 1.
2. 2.
3. 3.
4. 4.
View Abstract