Influenza in elderly people in care homes

BMJ 2006; 333 doi: https://doi.org/10.1136/bmj.39050.408044.80 (Published 14 December 2006) Cite this as: BMJ 2006;333:1229
  1. Rachel E Jordan, senior scientist (r.e.jordan@bham.ac.uk)1,
  2. Jeremy I Hawker, head of public health development2
  1. 1Health Protection Research and Development Unit, Health Protection Agency, Department of Public Health and Epidemiology, University of Birmingham, Edgbaston, Birmingham B15 2TT
  2. 2Health Protection Agency, London WC1V 7PP

    New evidence strengthens policy to vaccinate healthcare workers

    Influenza causes substantial mortality and morbidity in elderly people, particularly those with chronic diseases. Excess deaths during influenza epidemics are not limited to obvious causes such as influenza and pneumonia but also include circulatory and other respiratory causes.w1 People in elderly care homes and hospital wards are at particular risk, because high risk individuals are concentrated in an environment susceptible to the spread of respiratory pathogens. In this week's BMJ, Hayward and colleagues report the impact of vaccinating healthcare workers in elderly peoples homes on mortality in residents.1

    Most developed countries offer elderly people vaccination against predicted influenza strains for the next season.2 However, the age related decline of immune function reduces the ability of elderly patients to respond to the influenza vaccine,3 and the vaccine is less effective in patients with chronic diseases.4 Also, as most of the evidence in elderly people comes from database cohort studies, effects may have been overestimated because healthier people are more likely to be vaccinated and the reported estimates may not have been fully adjusted for confounding factors.5 So even if all elderly people in residential care were vaccinated, the effect on reducing the risk of complications of influenza may be modest. It therefore makes sense to examine alternative strategies, such as vaccination of healthcare workers in elderly care establishments, which offer indirect protection by reducing the exposure of at-risk people.

    Until now, the best evidence in support of vaccinating healthcare workers came from two related trials conducted in long term geriatric care wards in Scotland in the 1990s.6 7 Both found that vaccination significantly reduced mortality in residents (in the larger trial 13.6% v 22.4% in the control arm7; in the smaller pilot study 10% v 17%6) in years when influenza activity was two to three times higher than recent years8 and when the vaccine match to the circulating strain was good7 or reasonable.6 Assuming the estimates are robust, such a policy is likely to be cost saving or at the worst highly cost effective.9

    However, the existing evidence has methodological limitations. The small number of clusters led to an imbalance in important confounding factors (patient vaccination rates and levels of disability) between the trial arms and uncertainty about the extent of benefit.9 w2 This uncertainty was aired in a recent article published in the BMJ,10 which has sparked debate about the value of influenza vaccination programmes, including vaccination of healthcare workers, by suggesting that the evidence does not justify the policy.

    The current paper by Hayward and colleagues1 provides robust evidence that vaccinating healthcare workers against influenza benefits elderly patients. The cluster randomised controlled trial was conducted over two seasons in 44 private care homes around the United Kingdom. Staff of 22 homes were offered influenza vaccination and those of 22 matched control homes were not (usual policy). During the 2003-4 periods of influenza activity, five fewer deaths occurred per 100 residents in intervention homes compared with control homes (95% confidence interval 2 to 7, P=0.002). Episodes of influenza-like illness, consultations with general practitioners for influenza-like illness, and hospital admissions for such illness also decreased significantly. In the following season, influenza activity was much lower and no significant differences in patient morbidity or mortality were seen.

    Despite an incomplete vaccine match,4 in a year of modest influenza activity and vaccine uptake of around 50% for employees and more than 70% for patients, vaccinating healthcare workers significantly reduced patient mortality. Notably, in the following year of very low activity no effect was seen on outcomes.

    Applying the uptake rates and mortality data from the new trial to our economic model9 confirms the original conclusions that, even in the most pessimistic scenario, vaccination of healthcare workers costs as little as £274 (€407; $542) per life year gained. The trial therefore strengthens the case that vaccination of healthcare workers is the correct policy. Similarly, well designed studies in other settings and in years with different levels of circulating influenza, vaccine match, and vaccine uptake would help define best practice.

    So should all healthcare workers in elderly care establishments be vaccinated? Evidence shows that healthcare workers themselves would benefit by reducing their risk of influenza with minimal adverse effects,9 employers may benefit by reduced absenteeism,9 and elderly people in care homes would benefit from reduced morbidity and mortality (although the quality of lives saved needs to be analysed).

    Most countries in Europe and North America2 have recommended for some years that healthcare workers should receive the influenza vaccine, but uptake remains poor—less than 25% in Europe.11 Surveys suggest that the main reasons for refusing the offer are fear of side effects, fear that vaccinations will cause influenza, dislike of injections, being unaware that the vaccination is useful or available, and lack of time or forgetfulness.9 The challenge to the health services is to overturn the misconceptions and provide an easy access service within which there are no reasons to refuse vaccination. Small studies suggest that mobile vaccination services can be beneficial,9 but this—and other novel methods of delivery—needs to be tested in a well designed randomised controlled trial. It will also be interesting to view the progress of the new policy in the United States, which recommends that all healthcare personnel should be offered annual influenza vaccination, and those who decline for non-medical reasons should provide a signed declination.12



    • Embedded Image
      Extra references w1-w2 are on bmj.com

    • Competing interests: None declared.


    View Abstract

    Log in

    Log in through your institution


    * For online subscription