Elsevier

Vaccine

Volume 22, Issues 17–18, 2 June 2004, Pages 2192-2198
Vaccine

The number needed to vaccinate (NNV) and population extensions of the NNV: comparison of influenza and pneumococcal vaccine programmes for people aged 65 years and over

https://doi.org/10.1016/j.vaccine.2003.11.052Get rights and content

Abstract

The number needed to treat (NNT) and population extensions of the NNT describe the population outcomes of disease treatment, usually for chronic diseases. We have developed similar measures for vaccine preventable diseases based on the number needed to vaccinate. These measures quantify the number of people, or the number of vaccine doses, needed to prevent one event due to disease per year and allow the calculation of the vaccine cost to prevent one event due to disease per year. We used this method to compare influenza and pneumococcal vaccine programmes for people aged 65 years or more. We calculated the vaccine cost ($AUD) to prevent a case of disease, a hospitalisation or a death due to influenza or invasive pneumococcal disease each year in this age group. For influenza, the vaccine costs were $598 to prevent one case per year, $10,787 per hospitalisation prevented and $74,801 per death prevented each year. We assumed all cases of invasive pneumococcal disease were hospitalised so the vaccine cost of $11,494 per case prevented was the same as the vaccine cost per hospitalisation prevented, while the vaccine cost per death prevented each year was $49,972. These results suggest the vaccine costs for the prevention of one hospitalisation or one death each year due to invasive pneumococcal disease among the elderly in Australia would be similar to the costs for the prevention of one hospitalisation or death due to influenza.

Introduction

Many countries require cost-effectiveness analyses prior to the introduction of new drugs and vaccines, especially when these agents may be publicly subsidised or funded [1]. These analyses typically involve the estimation of disability adjusted life-years (DALYs) and quality adjusted life-years (QALYs). In addition to disease incidence in defined populations, calculating these parameters involves a number of assumptions related to service delivery, intervention outcomes, uptake of the therapeutic agent in the community and costs [2].

For population based vaccine programmes, we propose an alternative approach that allows an estimate of the impact a vaccine programme is likely to have on its target population. This approach can be used to compare programmes that target the same population and captures an intuitive understanding of the likely effectiveness of a vaccine programme. We have defined four measures, in terms of the number of people, the number of vaccine doses or the associated cost to prevent one event per year that describe the impact of a vaccine programme. These measures are: the number of people needed to vaccinate (NNV); the number of people to target for vaccination; the number of vaccine doses needed and the vaccine cost. The measures are based on the expression of benefit in terms of annual absolute risk reduction. The number needed to vaccinate is analogous to the number needed to treat (NNT) [3] and the other measures are related to population extensions of the NNT [4], [5]. The aim of this study is to define and derive these measures for vaccine preventable diseases and use them to compare influenza and pneumococcal vaccine programmes for people aged 65 years and over.

Section snippets

The number needed to vaccinate

The number needed to vaccinate is defined as the number of people needed to vaccinate in order to prevent one event due to disease each year. Because vaccine preventable diseases generally have a short incubation period, a short duration and may recur, the NNV is considered on an annual basis. In this respect, the NNV differs from the NNT which is defined as the number of people needed to treat over a defined period [3]. In general the NNV can be calculated as NNV=1/(annual incidence of event

The number needed to target for vaccination

While vaccine programmes are generally funded to provide 100% coverage for a target population, vaccine coverage is almost always incomplete. This is due to wastage (vaccine provided but never used), leakage (vaccine provided to people not in the target population) and the refusal of some people to be vaccinated. Because of wastage, leakage and refusal in any vaccine programme, more people need to be targeted for vaccination than will ultimately be vaccinated. The number of people to target for

Vaccine doses needed

If one dose of vaccine confers life-long immunity, only one dose of vaccine is needed to provide life-long protection. However, many vaccines do not provide this immunity. Influenza vaccine, for example, is recommended annually. The number of vaccine doses needed to prevent one event due to disease per year accounts for differences in the expected duration of vaccine induced immunity. The number of vaccine doses needed is equal to the product of the number of people to target for vaccination in

Vaccine cost

The previous calculations describe the annual impact of a vaccine programme on the target population. From these measures we can calculate the vaccine cost (in units of currency) to prevent one event per year as the product of the number of vaccine doses needed and the vaccine cost per dose. The vaccine cost, like the other impact measures, will be incurred annually, since each impact measure is based on the prevention of one case per year. Table 1 provides a summary of the definition and

A constructed example for the prevention of one case of disease

If the attack rate for a given vaccine preventable disease was 1 per 100 persons per year, vaccinating 100 people would, on average, prevent one case of disease each year if the vaccine were 100% effective. If the vaccine were 50% effective, it would be necessary to vaccinate 200 people to prevent one case of disease per year. This is the NNV to prevent one event when the event is a case of disease. If the vaccine coverage achieved in the programme were only 50%, the number of people to target

Influenza and pneumococcal polysaccharide vaccine programmes for people aged 65 years and over

Influenza vaccine is already provided as a publicly funded programme to all Australians aged at least 65 years. We calculated the vaccine cost to prevent one case of disease, one hospitalisation and one death per year due to influenza in this age group and the same measures for the polysaccharide pneumococcal vaccine used to prevent invasive pneumococcal disease. We used Australian data wherever they were available.

Discussion

We have developed measures, based on the NNV, which describe the potential impact of population based vaccine programmes. These measures are useful when comparing vaccine programmes targeted at the same population, when the consequences of events due to disease and vaccine delivery methods should be directly comparable. Because this cannot be assumed when target populations are different, the comparison of impact measures for vaccine programmes is only justifiable for the same target

Conclusion

The measures for vaccine preventable diseases developed here are similar to impact measures developed for clinical epidemiology. The NNV is analogous to the NNT [3], although the NNV for the prevention of one event needs to be considered on an annual basis. Impact measures related to the number of people or vaccine doses, or the associated vaccine cost, to prevent one event attributed to disease each year give an indication of the relative impacts of different vaccines in a specific target

Acknowledgements

Ross Andrews was a Ph.D. scholar at the Australian National University and received a scholarship from the National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases during the course of this study. Dr. Paul Roche provided data on influenza-like illness and laboratory notifications of influenza for Australia.

References (27)

  • T. Vu et al.

    A meta-analysis of influenza vaccine in persons aged 65 years and over living in the community

    Vaccine

    (2002)
  • National Health and Medical Research Council. How to compare the costs and benefits: evaluation of the economic...
  • Gold M, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York: Oxford University...
  • A. Laupacis et al.

    An assessment of clinically useful measures of the consequences of treatment

    N. Engl. J. Med.

    (1988)
  • R.F. Heller et al.

    Disease impact number and population impact number: population perspective to measures of risk and benefit

    BMJ

    (2000)
  • J. Attia et al.

    Impact numbers in health policy decisions

    J. Epidemiol. Community Health

    (2002)
  • K.G. Nicholson et al.

    Acute viral infections of upper respiratory tract in elderly people living in the community: comparative, prospective, population based study of disease burden in the community

    BMJ

    (1997)
  • T.M. Govaert et al.

    The efficacy of influenza vaccination in elderly individuals. A randomized double-blind placebo-controlled trial

    JAMA

    (1994)
  • K.M. Sullivan

    Health impacts of influenza in the United States

    Pharmaco. Econ.

    (1996)
  • P.B. McIntyre et al.

    Epidemiology of invasive pneumococcal disease in urban New South Wales 1997–1999

    Med. J. Aust.

    (2000)
  • D.S. Fedson et al.

    The burden of pneumococcal disease among adults in developed and developing countries: what is and is not known

    Vaccine

    (1999)
  • P. Roche et al.

    Annual report of the national influenza surveillance scheme, 2001

    Commun. Dis. Intell.

    (2002)
  • E.D. Shapiro et al.

    The protective efficacy of polyvalent pneumococcal polysaccharide vaccine

    N. Engl. J. Med.

    (1991)
  • Cited by (58)

    • Number needed to immunize to prevent RSV with extended half-life monoclonal antibody

      2020, Vaccine
      Citation Excerpt :

      The NNV estimates impact by expressing the benefit in terms of number of people needed to vaccinate to prevent one disease-related event each year. NNV is the reciprocal of the annual vaccine attributable incidence rate reduction [20]. We describe a corollary measure for the NNV, the number needed to immunize (NNI), which is more relevant to the passive EHL-mAb immunization strategy.

    View all citing articles on Scopus
    View full text