«ECDC TECHNICAL REPORT Guide to public health measures to reduce the impact of influenza pandemics in Europe: ‘The ECDC Menu’ ...»
Guide to public health
measures to reduce the impact of
influenza pandemics in Europe:
‘The ECDC Menu’
ECDC TECHNICAL REPORT
Guide to public health measures
to reduce the impact of
influenza pandemics in Europe:
‘The ECDC Menu’
Stockholm, September 2009
© European Centre for Disease Prevention and Control, 2009 Reproduction is authorised, provided the source is acknowledged.
TECHNICAL REPORT Guide to public health measures to reduce the impact of influenza pandemics in Europe Table of contents Glossary
A. What will happen naturally
B. Scientific evidence and experience
C. Diversity in the characteristics and severity between pandemics
D. Diversity within the pandemic
E. The diversity of Europe
F. Isolated communities
G. Secondary, social and perverse effects
H. Timing of use, triggering and sustainability
I. Dealing with the first outbreaks in a European country
J. Complete or partial protection?
K. Multiple layered measures — ‘defence in depth’
L. The necessity of intersectoral planning and preparation
M. Legal issues, liability and ethics
N. General versus selective measures
O. Early recognition and diagnosis
P. Planning, preparation and practice
R. Special groups — special considerations
S. Protective sequestration — children and adolescents
T. European interoperability
U. Research implications
The menu of measures
Travel measures: Restrictions on international travel
1. Travel advice
2. Entry screening
3. Border closures
Personal protective measures
4. Regular hand-washing
5. Respiratory hygiene (proper use and disposal of tissues)
6. General mask-wearing outside the home
7. Mask-wearing in healthcare settings
8. Mask-wearing in other high-risk situations
9. Mask-wearing by people with respiratory symptoms
10. Voluntary isolation of cases not requiring hospitalisation
11. Voluntary quarantine of household contacts (including protective sequestration)
Social distancing measures
12. Internal travel restrictions
Educational setting and day care-based interventions
13. Reactive school and day care closures
14. Proactive school and day care closures
Measures in the workplace and public places
15. Reactive workplace closures
16. Home working, reducing meetings, safety in the workplace
17. Cancelling public gathering, international events, etc.
Early treatment with antivirals
18. Early treatment of all those with symptoms
19. Healthcare or other exposed key workers
Antiviral prophylaxis following a single case
20. Given to family (household) members of influenza cases
21. Family and social contacts
22. Family and geographical contacts
23. Healthcare workers or other key workers
Vaccines – Human avian influenza vaccines (pre-pandemic vaccines)
24. Whole population
25. Health or social care workers or key workers
26. Children vaccinated first
27. Pandemic vaccine
Annex: WHO global influenza preparedness plan
iv TECHNICAL REPORT Guide to public health measures to reduce the impact of influenza pandemics in Europe Executive summary Application of public health measures (see Summary Tables) will, to some extent, reduce the number of people who are infected, need medical care and die during an influenza pandemic. They will probably also reduce the numbers affected by severe epidemics of seasonal influenza. By lowering and perhaps delaying the peak of a pandemic curve (Figure 1) the measures could also mitigate the secondary consequences of pandemics that result when many people fall sick at once, i.e. the impact of mass absenteeism on key functions such as delivering healthcare and maintaining food supplies, fuel distribution and utilities, etc. Public health measures may even delay the peak of the epidemic curve of a pandemic until nearer the time a pandemic vaccine starts to become available, thereby possibly also reducing the total numbers affected. In addition, theoretically, they may delay the peak until influenza transmission declines naturally in the summer months.
Figure 1. Objectives of applying public health measures during a pandemic Delay and flatten epidemic peak Reduce peak burden on healthcare systems and threat to other essential systems through high levels of absenteeism Somewhat reduce total number of cases Buy a little time
A range of measures have been suggested (see Summary Tables), including personal actions, like hand-washing and mask-wearing, and pharmaceutical interventions such as antivirals, human avian influenza vaccines (also called pre-pandemic vaccines) and, late in the pandemic, specific vaccines, as well as community social distancing measures. It is thought by many that combinations of measures will be even more effective than single measures, so called ‘defence in depth’ or ‘layered interventions’. Both modelling work and common sense suggest that early interventions will be more effective than waiting until a pandemic is well advanced.
It is hard to imagine that measures like those within the category of social distancing would not have some positive impact by reducing transmission of a human respiratory infection spreading from human to human via droplets and indirect contact. However, the evidence base supporting each individual measure is often weak. It is also unclear how a number of them will interact. Specifically, will the effect of social distancing measures be cumulative? In some cases this lack of clarity is due to a lack of research (see Section ‘Research implications’).
More often it is because the measures are hard to evaluate with any experimental approach and when measures have been implemented in real situations they have been used in combination. Hence the absolute positive effect and relative strengths of different measures are extremely hard to judge. Also, the strength of effect could quite reasonably vary with the characteristics of the pandemic. For example, interventions targeting children might have been quite effective during the 1957 pandemic where transmission in younger age groups seems to have been especially important, but they would have been less effective during the 1918–19 and 1968 pandemics. Hence it will not be possible to have fixed plans that fit every pandemic. Furthermore, the effectiveness, feasibility and costs of social distancing measures will presumably vary among European countries or even within countries (for example, dense urban areas compared with rural areas).
The experience of previous pandemics and related events like SARS shows that to some extent public health measures are applied according to local customs and practice. In the United States during the 1918–19 pandemic these were organised and often proactive (Markel 2007), while in Europe during pandemics and during SARS they were more often reactive (see list of tables, from page 36).
1 Guide to public health measures to reduce the impact of influenza pandemics in Europe TECHNICAL REPORT Hence there are good arguments that there should be default plans (plans that have been tested during exercises to be implemented in the absence of other information). Indeed there is WHO guidance to that effect and many European countries have been developing plans. However, given the above considerations, these plans should have considerable flexibility and command and control structures that will allow changes to be made quickly in the light of new data and experience.
All public health measures have costs and many also have secondary effects. The secondary effects of most measures can be considerable and many will require careful consideration. The more drastic societal measures that have been suggested (e.g. proactive school closures and travel restrictions) have significant costs and consequences that will themselves vary by their setting. These are also difficult to sustain. Hence for ordinary seasonal influenza or a mild pandemic their application, and especially their early application, could be more damaging than just allowing the infection to run its course and treating those with more severe illness.
Some of the measures are relatively straightforward to implement and are already recommended for even mild seasonal influenza (e.g. regular hand-washing and early self-isolation when developing a febrile illness). These also have the advantage of empowering individuals and giving them useful advice at a difficult and worrying time.
Others are going to be difficult to implement or are too costly (e.g. timely mass use of antivirals by those becoming sick) and others are potentially highly disruptive to societal functions and difficult to sustain (e.g. border closures, internal transport restrictions). Therefore all the measures require Planning, Preparation and Practice.
The point about costly and disruptive measures is crucial. During a pandemic with lesser severe disease and of fewer falling sick, such as those seen in 1957 and 1968, some possible community measures (proactive school closures, home working, etc.), though probably reducing transmission, can be more costly and disruptive than the effects of the pandemic itself. Hence such measures may only have a net benefit if implemented during a severe pandemic, for example one that results in high hospitalisation rates or has a case fatality rate comparable to that of the 1918–19 ‘Spanish flu’.
For these reasons, early assessment of the clinical severity of a pandemic globally and in European settings will be crucial. Though early implementation of measures is logical, application of the more disruptive interventions too early will be costly and may make them hard to sustain.
A number of European countries are now considering their policy options for these measures. Because of Europe’s diversity, no single combination of measures will suit every European setting: one size will not fit all. However, common discussions on the measures will be helpful and make for a more efficient decision-making process.
Further, some countries have already undertaken considerable relevant scientific work, some of which this document draws upon, but which all European countries could benefit from along with thinking from other countries.
Purpose: In the light of the above considerations, and given that ECDC’s mandate is to give scientific advice rather than prescribe actions, the intention with this document is to present a menu of possible measures, giving public health and scientific information on what is known or can be said about their likely effectiveness, costs (direct and indirect), acceptability, public expectations and other more practical considerations. This is to help European Member States and EU institutions, individually or collectively, decide which measures they will apply.
That said, there are some measures which are either so self-evident or so ineffective that simply laying out the evidence should make for easy policy decisions.
Audience: The primary intended audience is those who develop policy and decision-makers, though secondary audiences are all those concerned with influenza, the public and the media. The understanding by the latter of the measures and their limitations will be crucial to their successful application in a pandemic.
Scope: When the pandemic is spreading in Europe in WHO phases 5 or 6 of a pandemic. This document also applies when there are epidemics of seasonal influenza. It does not address the different circumstances of phase 4, the unique needs of the first emergence of a putative pandemic strain (the WHO Rapid Containment Strategy), nor the complex planning and policy issues that arise over how to sustain key services during a pandemic (socalled business continuity planning for a pandemic). The latter is, in any case, outside the remit of ECDC.
The document should be read along with previous guidance that ECDC has published on personal protective measures. This is summarised in the text and tables. Relevant scientific guidance concerning human avian influenza (pre-pandemic) H5N1 vaccines has been published and is referenced within the text. The guidance should be read with the current (2005) WHO guidance, which is reproduced in the Annex, with permission from WHO’s 2005 pandemic plan and the new WHO Guidance on countermeasures (to be published later in 2009).
Please note that this is an interim guidance as there will be further research findings and it is possible that new countermeasures will emerge. Therefore the menu will continue to be updated at intervals.
1 Evidence of effectiveness: A, B and C represent strongly, reasonably and poorly evidence-based recommendations, respectively.
Grade A: Systematic reviews where there are diverse primary studies to draw from (not primarily modelling), well-designed epidemiologic studies or especially experimental studies (randomised controlled trials).
Grade B: Represents evidence based on well-designed epidemiologic studies, substantial observational studies or experimental studies with 5 to 50 subjects, or experimental studies with other limitations (not having influenza as an endpoint, for example).
The code Bm indicates modelling work, with emphasis on studies that have good quality primary data available. Hence quality can be both Bm & C.
Grade C: Represents evidence based on case reports, small poorly controlled observational studies, poorly substantiated larger studies, application of knowledge of mode of transmission, infectiousness period etc. Cm refers to modelling with few or poor quality primary data.