Nine steps long-term care facilities can make to control an influenza outbreak

February 2006

Long-term care facility residents are a particularly vulnerable population during an influenza epidemic. Even when an appropriate vaccine is available, many of the elderly do not respond with sufficient antibody to be fully protected, although complications are reduced.

The specter of a pandemic of H5N1 avian influenza has been raised. Although the next pandemic may not be due to H5N1, it is reasonable for each long-term care facility to have a plan for dealing with a pandemic. The plans will vary from facility to facility and depend on financial, social, geographical, structural and other considerations. This editorial is meant only as a template for a facility to consider, pick and choose what is feasible for its circumstances. It is recognized that some of the measures suggested below cannot be applied to demented residents.

Some of the facts that have been true about influenza in the past and are assumed here in terms of a new influenza strain causing a pandemic are the following:

• The incubation period of a case will probably be one to four days;
• An individual incubating influenza will be capable of transmitting the disease 24 hours before getting sick, although the infectivity will be relatively low. With onset of symptoms, the infectivity will become very high and then diminish over four days;
• A surgical mask and coughing/sneezing into tissues will help decrease spread;
• Effective antivirals will probably be much more effective in prophylaxis (ie, preventing symptomatic disease) than in treating symptomatic disease, but they will not prevent all infections, symptomatic or asymptomatic;
• Treatment will have an effect only if started within 48 hours of onset of symptoms and will decrease, but not eliminate, viral shedding; and
• There will be shortages of staff because of illness.

It must be stressed that if and when H5N1 mutates to a form capable of causing a pandemic, it will almost certainly behave differently from what we are observing now in Eurasia (and may behave differently from classical human influenza, such as H3N2) in terms of antigenic properties, susceptibilities to antivirals, incubation period, time over which virus is excreted, whether or not virus is abundant in stool (true of H5N1 but not of previous influenza), clinical syndrome and mortality rate.

The potential components of a comprehensive plan will be:

• Immunization;
• Early warning and surveillance;
• Universal precautions;
• Social distancing;
• Respiratory precautions;
• Isolation;
• Chemoprophylaxis;
• Therapy; and
• Response to lack of bed capacity in local hospitals.

Preparing for a pandemic

Immunization. If available, immunization is by far the single most important part of any plan. If a vaccine is available against the influenza strain sustained major human-to-human outbreak somewhere in the world, then the vaccine should be used. I would not favor use of a relatively untested vaccine in the absence of human-to-human spread for two reasons. First, it may be the wrong vaccine in terms of not being protective. Second, in 1976, there were severe and lethal consequences (ie, Guillain Barré syndrome) from use in millions of people of an influenza vaccine given to prevent a pandemic that did not occur.

It is critical to require immunization of paid and volunteer workers of all types (and if possible, visitors) who will enter the long-term care facility as well as all residents. It has been demonstrated that for protection of residents in a long-term care facility, it is very important to immunize personnel. With a H5N1 pandemic and lack of available s1. fic vaccine, it is conceivable that vaccination with the currently available vaccine may give some protection because of its N1 component.

Early warning and surveillance. Virtually all pandemics have started in the Far East, and there are many early warning systems that track human epidemics (eg, ProMED-mail, WHO, the CDC and Eurosurveillance). There will probably be at least weeks, if not months, of advance notice when a true pandemic becomes established in the United States. Those responsible for infection control in a long-term care facility should track the world as well as U.S. and local situations. Subscription to appropriate local and state health department list servers is important.

Once a pandemic strain of influenza enters the United States, it is essential for long-term care facilities to heighten their usual surveillance for respiratory illnesses and to use rapid testing for the new strain, if available, when outbreaks seem to be occurring.

Universal precautions. Hand hygiene and glove changes are standards of long-term care facilities and are especially important during outbreaks. Personnel should take these precautions before contact with residents. Gowns should be used if soiling with respiratory secretions is likely; the gown should be discarded after each contact.

Social distancing. The extent of social distancing will depend on what is feasible for a specific facility. Once H5N1 or another pandemic strain is in the neighboring community (or ideally anywhere in the United States), access to the facility should be restricted to essential personnel. New admissions should be cancelled. Personnel and anyone else entering the facility should wear surgical masks when in contact with others. Residents should only leave the facility for essential appointments and should wear masks during the trip. Assuming a four-day incubation period, on return, they should, if possible, be isolated in a single room for five days (if they remain well) and wear a surgical mask if they must enter common areas. No personnel should come to work if ill.

To the degree that it is possible, residents should avoid congregating in large numbers. For example, each unit should eat separately and, as much as possible, stay within the unit. Personnel should remain assigned to the same unit for the duration of the outbreak.

If visitors are allowed, long-term care facility staff should make efforts to restrict visitors to adults who have not been ill for at least one week. Personnel, residents and visitors should wear masks, and contact should be restricted to as few residents as possible. Personnel and residents should practice frequent hand hygiene, especially after coughing or sneezing.

Social distancing is a strategy to try to delay entrance of influenza into the facility and subsequent spread. The longer the delay, the better the chance to have vaccines and adequate amounts of antivirals. Although total avoidance of infection in the facility is a worthy goal, it is unlikely to be achieved.

Respiratory precautions. Everyone who is coughing must cough into a tissue or handkerchief and then use hand hygiene. If possible, residents who develop any upper respiratory symptoms at the time of an outbreak in the community, but not yet in the facility, should be isolated in a single (or cohorted) negative-pressure room or one with a portable HEPA filtration unit for at least five days. During this time, they should wear a surgical mask if it is necessary to be in a common area. Training must begin prior to a U.S. outbreak. Long-term care facility residents generally do not adjust well to wearing masks. Remember, influenza is transmissible for up to 24 hours before onset of illness.

Managing the pandemic

Isolation. Once influenza is inside the facility, the staff should attempt to isolate residents suspected of having influenza; this may not be possible. If not, cohorting may be possible and is worth trying. Cohorted, or isolated, residents are fed in their rooms, kept there with doors closed and treated with respiratory and contact isolation. The rooms should have HEPA filters, if possible, unless a negative pressure room is available. The residents should wear surgical masks when others enter the room or if they have to leave the room. Personnel should wear N95 respirators when entering the room. Training and fit testing are necessary for N95 masks; they are difficult to tolerate. Isolation is probably only necessary for five days after onset of symptoms.

Chemoprophylaxis. Administer rimantadine if the influenza strain is susceptible (as are some of the H5N1 Turkish strains); if not, then oseltamivir (Tamiflu, Roche) should be given to all residents and personnel at the first sign of an outbreak in the facility. Ideally, prophylaxis is continued until long-term care facility and community activity disappear. This could take eight weeks or longer.

Even if an effective vaccine has been given to personnel and all residents are receiving prophylaxis, prophylaxis of personnel is advisable. Personnel should wear surgical masks when in contact with other individuals and exclude themselves from work if ill.

Therapy. For suspected cases, therapy with an appropriate antiviral should be given (eg, rimantadine or oseltamivir) for at least five days. Some studies suggest that higher-than-recommended doses of oseltamivir are necessary for H5N1. Therapy must start on the first day (preferred) or, at the latest, the second day of illness. If prophylaxis has been started facility-wide and is effective, there will probably be relatively few residents who become ill.

Response to lack of bed capacity in local hospitals. A major problem may be inability to have patients admitted to the hospital because of limited bed capacity. Therefore, long-term care facilities must give consideration to alternatives, such as gearing up to manage very sick residents in the facility. One possibility might be to set up one unit as a hospital unit with higher nursing staffing and with IV and respiratory equipment availability. Under these circumstances, appropriate medical, nursing and respiratory technician coverage will be necessary.

Thanks to Keith Kaye, MD, MPH, and Jerry Zuckerman, MD, for their suggestions.

For more information:

• To view the updated management of influenza in long-term care settings, visit www.cdc.gov/flu/professionals/infectioncontrol/longtermcare.htm.

• Dr. Donald Kaye, is professor emeritus of Drexel University College of Medicine in Philadelphia, and a member of the Infectious Disease News Editorial Board.