* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Influenza in the acute hospital setting
Sociality and disease transmission wikipedia , lookup
Neonatal infection wikipedia , lookup
Globalization and disease wikipedia , lookup
Marburg virus disease wikipedia , lookup
Henipavirus wikipedia , lookup
Common cold wikipedia , lookup
Vaccination wikipedia , lookup
Childhood immunizations in the United States wikipedia , lookup
Transmission (medicine) wikipedia , lookup
Hospital-acquired infection wikipedia , lookup
Review Nosocomial influenza Influenza in the acute hospital setting Cassandra D Salgado, Barry M Farr, Keri K Hall, and Frederick G Hayden Influenza poses special hazards inside healthcare facilities and can cause explosive outbreaks of illness. Healthcare workers are at risk of acquiring influenza and thus serve as an important reservoir for patients under their care. Annual influenza immunisation of high-risk persons and their contacts, including healthcare workers, is the primary means of preventing nosocomial influenza. Despite influenza vaccine effectiveness, it is substantially underused by healthcare providers. Influenza can be diagnosed by culturing the virus from respiratory secretions and by rapid antigen detection kits; recognition of a nosocomial outbreak is important in order to employ infection-control efforts. Optimal control of influenza in the acute-care setting should focus upon reducing potential influenza reservoirs in the hospital, including: isolating patients with suspected or documented influenza, sending home healthcare providers or staff who exhibit typical symptoms of influenza, and discouraging persons with febrile respiratory illness from visiting the hospital during a known influenza outbreak in the community. (Note: influenza and other respiratory viruses can cause non- Figure 1. Poster displayed at University of Virginia showing proportion of UVA healthcare febrile illness but are still transmissible.) workers receiving influenza vaccine. The antiviral M2 protein inhibitors Healthcare workers have increased risk of acquiring (amantadine, rimantadine) and neuraminidase inhibitors (zanamivir, oseltamivir) have proven influenza during known outbreaks since they are exposed to efficacy in treating and preventing influenza illness; infected individuals in the community as well as hospitalised however, their role in the prevention and control of patients with influenza. Whether infected at work or in the influenza in the acute hospital setting remains to be community, healthcare workers may become an important more fully studied. source of influenza for their patients.4,6,7 For example, a Lancet Infectious Diseases 2002; 2: 145–55 An important cause of morbidity and often mortality in most communities every winter, influenza poses special hazards inside healthcare facilities. Because of its short incubation period and efficient respiratory spread from person to person, it can cause explosive outbreaks of febrile respiratory illness. The hospital patient population has serious underlying illnesses, making influenza more lethal in this setting.1–5 THE LANCET Infectious Diseases Vol 2 March 2002 serological study in the UK, where healthcare worker immunisation had not been the standard of care, found that up to 23% of healthcare workers contracted clinical or subclinical influenza during community outbreaks.8 Up to CDS and KKH are infectious diseases fellows, BMF is hospital epidemiologist and FGH is professor of medicine and pathology at the University of Virginia Health System, Charlottesville, VA, USA. Correspondence: Professor Frederick G Hayden, Box 800473, University of Virginia Health System, Charlottesville, VA 22908, USA. Tel +1 434 924 5059; fax +1 434 924 9065; email [email protected] http://infection.thelancet.com 145 For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza impact, its modes of transmission and reservoirs of infectivity, and methods of nosocomial influenza detection, control, and prevention. Active surveillance within medical facilities is necessary to identify nosocomial outbreaks and prevent spread within the facility. All patients with suspected influenza should be tested, isolated under droplet precautions, and reported to infection control. Epidemiology of nosocomial influenza in the acute care setting Any patient who develops a flu-like illness 72 h after admission should be considered a nosocomial case. Hospital admissions The annual number of influenza cases and subsequent influenza-associated hospital admissions, complications, and deaths in the community are dependent upon the pattern of disease and circulating viral strains. Annual Influenza testing should be When a nosocomial outbreak influenza epidemics typically affect instituted when influenza is is suspected, infected patients 5–10% of the adults in the community detected during surveillance, should be isolated. In extreme and result in hospitalisation rates of especially when a documented situations, cohorting of patients 2–4/10 000 in otherwise young healthy nosocomial case occurs, or a on a ward staffed by designated adults.10,11 However, hospitalisation rates cluster (3) of cases of vaccinated personnel may suspected nosocomial influenza are much higher among infants and be necessary. occur on the same ward during young children, the elderly and certain a short (48–72 h) period. high-risk populations.12–14 In influenza epidemics from 1968–1997 in the USA, the number of disease-related hospital admissions ranged from Once a nosocomial case is 16 000–220 000 per epidemic with over confirmed, daily monitoring for flu-like illnesses should be half of these hospitalisations occurring done, especially on in those 65 years of age.12–15 Many of wards with high-risk patients. those hospitalised are considered to have exacerbations of underlying disease and are not always recognised as having All healthcare personnel with intercurrent influenza infection, so they flu-like illness should be tested may serve as potential sources of spread. and excluded from patient care Influenza-associated deaths are often pending results. the result of pneumonia or exacerbations of cardiopulmonary conditions, and the elderly population currently accounts for In addition, influenza antivirals approximately 90% of all influenzashould be considered to treat related deaths.16 Excess influenza-related infected patients and for deaths occurred in 19 of 23 influenza prophylaxis of exposed patients, epidemics with rates ranging from unvaccinated personnel, and 30–150/100 000 persons in those older those vaccinated <2 weeks before exposure. than 65 years in the USA.15 Of note, epidemics of influenza A H3N2 subtype viruses have been associated with greater Figure 2. Infection control measures for documented or suspected outbreaks. Adapted from hospitalisation and mortality rates. In www.cdc.gov/ncidod/hip/INFECT/flu_acute.htm addition to causing substantial morbidity and mortality, epidemics three-quarters of healthcare workers with influenza-like often have a deleterious, sometimes overwhelming, effect on illness have continued to work while ill9 and thus may healthcare delivery and cause marked increases in healthcare perpetuate transmission and foster outbreaks. Illness in costs. This results not only from a sudden surge in the healthcare workers may also disrupt patient care and result number of hospital admissions over a relatively short period in higher hospital expenditures. of time, but also disruption of services due to an increased Optimal prevention of nosocomial influenza should number of ill caregivers.15,17 Medical cost-containment involve multiple strategies that focus upon reducing efforts, resulting in decreased numbers of available beds and potential influenza reservoirs in the hospital. For example, staff, could exacerbate this problem and lead to shortages of campaigns encouraging healthcare workers to get vaccinated acute care hospital beds during epidemics. An even more dramatic and devastating effect on the is one way forward (figure 1).We will review relevant aspects of influenza epidemiology in the acute-care setting, its healthcare system has been described with pandemic 146 THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza influenza. The pandemic of 1918–1919 was responsible for nearly 40 million deaths and over 500 million infections worldwide.18 It disrupted essential community services including healthcare delivery in many communities.19 The world-wide cumulative number of deaths in the 1957 Asian and 1968 Hong Kong pandemics was estimated at nearly 6 million.18 Furthermore, pandemics are associated with excess hospitalisations and mortality in adults below the age of 65 years.16 One of the first well-documented descriptions of nosocomial influenza involved a severe outbreak on a general medical service during the 1957 pandemic.20 Estimates indicate that the next pandemic will cause between 89 000 and 207 000 deaths and between 314 000 and 734 000 hospitalisations in the USA alone.21 Nosocomial impact During periods of increased influenza activity in the community and subsequent higher numbers of patients with influenza in hospitals, the risk for nosocomial transmission from infected patients increases. In addition, nosocomial influenza may be acquired from healthcare workers, other hospital employees, and visitors. Influenza can cause sporadic or epidemic disease in hospitalised patients, and nosocomial outbreaks have been described in regular hospital wards, neonatal and adult intensive care units, transplant centres, and in nursing homes and other chronic care facilities. Nosocomial influenza outbreaks are not uncommon but are likely substantially under-reported. Outbreaks usually occur during periods of community influenza activity and are associated with notable morbidity, increased cost of care, and excess mortality.1–5,17 In a review of 12 nosocomial influenza outbreaks2-4,6,7,22-29(table), the outbreak duration ranged from 2–69 days (median 7 days) and reported patient attack rates ranged from 3–50% on epidemic wards and 0·7–20% throughout the hospital. Healthcare worker attack rates were infrequently reported but ranged from 11–59% among those caring for patients with influenza. Much lower healthcare worker infection rates have been documented at our hospital (less than 2% in any season during the past decade) where workers have been encouraged to be vaccinated, to report febrile respiratory illness, to be screened for influenza, and sent home when ill to prevent spread of infection to patients.30 Most nosocomial outbreaks are caused by influenza A with nosocomial influenza B infections occurring primarily in hospitalised children; however, influenza B adult outbreaks have been reported and one outbreak affected 20% of hospitalised adults, whose average age was over 60 years.4 Although valuable information is gained from review of these outbreak investigations, most were not prospective studies and may not be representative of all hospital experience with influenza. Mortality rates during nosocomial outbreaks vary according to the affected patient population and circulating strain. In neonatal ICU outbreaks, mortality may be low, except for deaths occurring in babies with severe underlying cardiopulmonary disease.24 In acute-care facilities, and geriatric hospitals, median mortality has been 16%,2,4,6 whereas in specialised populations such as transplant THE LANCET Infectious Diseases Vol 2 March 2002 or ICU patients, the mortality can be 33–60%.3,29 The immunocompromised population including transplant recipients, those receiving chemotherapy for malignancies, and those with advanced HIV disease deserve special consideration as illness secondary to influenza is often severe. Up to 70% of influenza infections in immunocompromised transplant and acute leukaemia patients have been nosocomial in origin in several studies.31–33 Pneumonia and death are common complications of influenza in immunosuppressed children, bone-marrow-transplant recipients, and those receiving chemotherapy for acute leukaemia.33–37 Patients at greatest risk of these complications seem to be those with the highest degree of immunosuppression.36 Information regarding the economic impact of nosocomial influenza is limited, but the financial burden can be substantial, in part because of prolongation of hospital stay and additional diagnostic and therapeutic interventions for affected patients. Excess hospital costs are also due in part to absenteeism of healthcare workers who become ill during an influenza outbreak, since hospitals must cover the salaries of the ill workers in addition to their replacements.17 One study estimated a mean excess hospital cost of over $7500 per episode of nosocomially acquired influenza.17 Transmission General principles The influenza virus spreads from person to person by respiratory droplets. When a person coughs or sneezes, particles of varying sizes are created ranging from large particle droplets that are filtered by the nose and pharynx to small particle droplets that reach the alveolus. The observation that influenza can be induced by intranasal drops38 suggests a role for direct contact spread, but several lines of evidence indicate more efficient transmission via smaller particle aerosols. In experimentally infected volunteers, the human infectious dose is at least 10–100 fold lower with inhalation of small particle aerosols delivered to the lower respiratory tract than with direct instillation of nasal drops.39 Further, intranasally administered influenza virus uncommonly causes cough or lower respiratory tract manifestations in experimentally infected volunteers,40 whereas early onset and protracted cough indicative of tracheobronchitis is characteristic of natural influenza.41 Studies of differing routes of administering antiviral drugs also suggest that intranasal inoculation is not the usual route of natural transmission of influenza. Intranasal zanamivir and intranasal interferon are protective against intranasally introduced influenza under experimental conditions, but both fail to prevent natural influenza in ambulatory persons.42–44 By contrast, inhaled zanamivir is protective against natural influenza in household and nursing home contacts.45 Because of the range in natural respiratory droplet particle size and a resulting gradient in where particles land in the respiratory tract between the nasopharynx and the alveolus, it is possible, however, that droplets of intermediate size (ie, those capable of passing the nose but not able to reach those more than about 5 feet http://infection.thelancet.com 147 For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza Table 1. Description of reported nosocomial outbreaks in the acute-care setting Author Outbreak population Muchmore6 Neurology ward Duration of outbreak Virus type Number of patients affected (attack rate) 1 month Influenza A2 “Asian Flu” 77 (50% on epidemic ward, 9% entire hospital) Bauer7 Neonatal intensive care unit Kapila2 Acute-care hospital ward 1 week Influenza A2 9 (30% on epidemic ward) Hall23 Acute-care infant ward *1 month Influenza A and B 12 (not stated, total number exposed not known) One week Influenza A/Victoria/3/75 (H3N2) 8 infants affected, 5 with documented infection (total number exposed not known) 1 month Influenza B/Singapore/222/ 79-like 28 (33% on epidemic ward; hospital attack rate for first 2 weeks 20%, 12% for last 2 weeks) Meibalane28 Neonatal ICU Not stated, but all cases occurred Influenza A/ within hours of each other Hong Kong/1/68 (H3N2) 3 (not stated, total number exposed not known) Van-Voris4 Acute-care hospital Malavaud25 Solid organ transplant unit 4 days Influenza A (H3N2) 4 (33·3%) CDC3 Medical-surgical ICU Not stated Influenza A 3 affected (total number exposed not stated) 1 week Influenza A (H3N2) Sydney 7 (25·9%) 4 (26·7%) Weinstock29 Adult bone marrow transplant unit Munoz24 Neonatal ICU 5 days Influenza A (H3N2) Adal26 Acute-care hospital 69 days Influenza A (H3N2) 10 (total number exposed not known) Gowda22 Geriatric unit (rehabilitation and 1 month admission wards) Influenza A (resembling A/ Victoria/3/75) 19 (30%) *Description of nosocomial influenza infections during a 1 month period of increased influenza activity in the community Table continues on page 149 from the coughing individual) could have an important role in influenza transmission. Of note, influenza viruses retain infectiousness after aerosolisation for at least 24 h at room temperature under conditions of low relative humidity.46 Some epidemiological observations also point to transmission through the air. One point source outbreak due to an infected passenger on a delayed aeroplane with a non-functioning ventilation system found a 72% clinical attack rate for people throughout the passenger cabin over the following 4 days; the risk of illness was dependent upon the amount of time spent on board during the repairs of the system.47 This outbreak is often used as evidence of smaller particle droplet transmission, although it remains possible that large particle droplet spread also had a role in transmission during this outbreak as passengers moved freely about the passenger compartment.47,48 Such observations suggest that most naturally occurring influenza infections occur by inhalation of droplets or droplet nuclei rather than direct intranasal inoculation (eg, from contaminated fingers). Nosocomial aspects In general, hospitalised patients with influenza are maintained in private rooms but are not placed into negative pressure isolation. Despite this, only rare instances of spread outside the isolation room have been recognised. For example, we have not recognised timespace clusters of cases of nosocomial influenza during the past 15 influenza seasons at the University of Virginia despite putting most cases of influenza into private, positive pressure ventilation rooms. However, when susceptible patients or groups are cohorted with an 148 infected index case, spread through the air seems to be responsible for some transmission. One study assessed influenza virus cross-infection during an influenza season in paediatric wards and found that patients admitted to wards with a greater proportion of beds in an open area were more likely to acquire nosocomial influenza than those who were admitted to wards with a greater proportion of private beds.49 While this suggested spread through the air, the data cannot clearly distinguish between large and small particle droplet spread because of lack of information about the spacing of the beds and any time-space clustering of the illnesses. Furthermore, one of the four wards in the group with more nosocomial infections had 100% private rooms, which accounted for 29% of all of the infections in this group.49 The rapid onset and explosive nature of nosocomial influenza outbreaks with secondary cases often occurring within hours of each other have been used to suggest that smaller particle aerosol transmission must be involved.7,24 In one of these outbreaks, two culture-proven and one suspected nosocomial case occurred within hours of one another, but each patient had been cared for by a healthcare worker with influenza.7 In the other outbreak four infants developed illness over a 4-day period; two of the babies’ cribs were adjacent to one another suggesting possible transmission. However, four neonatal ICU nurses were sent home just before onset of the cases for flu-like illness, although they had reportedly not worked with the affected infants.24 While the sources and transmission modes of influenza are unclear in such reports, they implicate healthcare workers.7,24 Influenza viruses can retain infectiousness on non-porous surfaces for 24–48 h and for at least 5 min on hands,50 and one THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza Table, continued Author Presenting symptoms Muchmore6 Headache, chills, muscle aches, and fever Healthcare worker attack rate Patient mortality rate 3 healthcare workers affected (total number with patient contact not known) 3 deaths occurred (all with Inter-ward transfers discontinued and underlying pulmonary disease) afebrile patients discharged to epidemic ward 16% hospital 4% decrease crowding Control efforts employed Bauer7 Fever, nasal drainage, 2 ill healthcare workers tachypnoea, raised white blood (total number with patient cell count contact not known); both were symptomatic and cared for infants before their diagnosis 0% Not stated Kapila2 Fever, bronchopulmonary No ill healthcare symptoms, and chest radioworkers reported graph findings of basilar infiltrates 67%. All patients had underlying illness. Not stated Hall23 Fever, dry cough, lethargy, irritability, white blood cell count, anorexia, and pulmonary infiltrates One physician caring for infants 0% affected (documented); however, 4 of 9 (44%) other medical personnel possibly affected but not tested Careful handwashing and gown use when handling affected infants Meibalane28 Lethargy, poor feeding, mottling, apnoea, fever, cough, pulmonaryinfiltrates. Duration of illness ranged from 4–11 days. Not stated but immediately before 0% the outbreak several patients and nursing staff had fever with upper respiratory symptoms Parents were not allowed to visit and nursing staff was sent home Van-Voris4 Malaise, chills, fever (0–8 days), cough, Excess hospital duration of 3·2 days. Excess hospital cost of $458 per patient infected 59% on epidemic ward 0% Patients clinically ill with influenza placed into cohorts, elective admissions limited, visitors limited, and ill workers sent home Malavaud25 Fever, headache, myalgias, cough, pulmonary infiltrates on chest x-ray 11·1% 0% Respiratory isolation of infected patients. CDC3 Not stated 1 healthcare worker affected; total 33% number exposed not known Not stated Weinstock29 Radiographic evidence of pneumonia 5 healthcare worker affected but 60% total number exposed not known 15 step infection control programme Munoz24 Rhinorrhea, cough, tachypnoea, Not stated fever, mild WBC elevation, and hypoxaemia 25%. All patients had underlying cardiac or pulmonary disease Adal26 Specific symptoms not stated 10%. Patient had underlying pulmonary illness Gowda22 Not stated Fever, tachycardia, hypotension, Not stated mental status changes, and patchy consolidation on chest 15·8% Respiratory isolation for ill patients and use of a mask, gown, and gloves for care. Potentially exposed infants cohorted Suspected cases placed into respiratory isolation, symptomatic visitors discouraged from entering hospital Not stated radiograph. ICU= intensive care unit; WBC=white blood cells; HCW=healthcare workers *Description of nosocomial influenza infections during a 1 month period of increased influenza activity in the community nursing home outbreak implicated transmission by the hands of healthcare providers.51 It seems reasonable to assume that multiple routes of nosocomial transmission are possible. Duration of contagiousness An important issue in acute-care settings is the type and duration of isolation for patients with documented or suspected influenza. The principal reservoir of human influenza is the respiratory tract, particularly nasopharyngeal secretions and sputum of infected people. The period of communicability is estimated by the duration of recovery of infectious virus from respiratory tract samples but is likely to be heavily influenced by the presence of THE LANCET Infectious Diseases Vol 2 March 2002 symptoms that generate infectious aerosols (eg, coughing). Viral shedding may begin at least 1 day before clinical illness and the duration of shedding depends on the age and immune status of the affected person. Outpatient children commonly shed for 7 days or longer, whereas nonhospitalised immunocompetent adults may shed for 4–5 days after symptom onset.52 Institutionalised elderly adults have shown more prolonged virus recovery (for up to 7 days),53 and nosocomially infected infants shed virus for up to 21 days.23 Immunosuppressed patients often also shed for prolonged periods of time. For example, one 7-year-old girl with advanced HIV disease had nasopharyngeal shedding of influenza for at least 9 weeks54 and among influenza-infected http://infection.thelancet.com 149 For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza adult bone-marrow-transplant or leukaemia patients, the median duration of shedding was 7 days but ranged up to 44 days.55,56 Of note, among rimantadine recipients, 83% of those shedding virus for 3 days or more had resistant variants. The duration of virus recovery is not well characterised in hospitalised adults but many have detectable virus in upper-respiratory-tract samples at 5 days after admission.57 The duration of viral replication in the lower respiratory tract has not been studied rigorously, nor has the effect of antiviral therapy on viral recovery been critically assessed in hospitalised patients. Prolonged shedding has important implications for nosocomial spread, in that such patients may serve as reservoirs of influenza virus, including drug-resistant variants. Sequential virological monitoring of patients at high risk for protracted viral excretion may be necessary. Detection and diagnosis Recognition of an outbreak is essential in order to employ control efforts such as isolation of patients suspected of having influenza, discouraging visits by those with suspected influenza, and sending home ill healthcare workers. The development of unexplained fever and/or respiratory illness in a hospitalised patient should raise the possibility of nosocomial influenza during influenza season. Hospitalised elderly adults may exhibit atypical features including gastrointestinal manifestations and altered cognition. In infants, a sepsis syndrome may occur with apnoea, lethargy, and poor feeding.28 Influenza may occur sometimes outside the usual seasonal patterns, so outbreaks of unexplained febrile illness during summer months warrant appropriate diagnostic studies. Influenza can be diagnosed by culturing the virus from respiratory secretions (nasal swabs, throat swabs, or nasal washes), or by more rapid influenza diagnostic tests for antigen (immunoassay or immunofluoresence), neuraminidase enzymatic activity, or viral RNA by reverse transcription PCR. The performance of one enzyme immunoassay, Directigen FLU-A (Becton Dickinson Microbiology Systems, Cockeysville, MD) and a direct immunofluorescence test, IFA, were compared with culture isolation of influenza A in geriatric patients.58 Directigen FLU-A had a sensitivity of 86·8% and a specificity of 99·1%, and IFA had a sensitivity of 92·5% and a specificity of 97·2%.58 By providing same-day diagnosis, rapid testing allows for early detection of nosocomial influenza outbreaks and enables infection control departments to institute appropriate containment measures throughout the hospital to prevent further spread. However, sending specimens for viral culture is important because only culture isolates can provide information regarding influenza subtypes and strains and potential drug resistance. Furthermore, the available rapid diagnostic tests vary in their performance characteristics according to assay type, sample type, time of sample collection, and age of the patient. When compared with influenza isolation, immunofluorescence assays are 50–90% sensitive for viral detection58–61 and enzyme immunoassays are 50–80% sensitive.62–65 In general, the sensitivity of rapid diagnostic tests is higher in children than adults and higher for nasal samples than throat specimens.59 150 False-positive results occur with some commercial antigen assays and have resulted in incorrect diagnosis of influenza as a cause of several outbreaks of respiratory illness. The positive predictive value of these tests outside known influenza epidemics may be low.60–63 Annual prospective surveillance of influenza activity in the community helps to define the period of risk for hospitalised patients. Passive surveillance by employee health services may not be helpful as the only method of community influenza surveillance because it often underestimates the amount of active influenza illness in the community. Once influenza is known to be circulating in the community, we recommend routine influenza testing in patients presenting to hospitals for admission with an acute febrile respiratory illness. Hospitalised patients who develop symptoms compatible with influenza during a known outbreak in the community should also be tested. Pseudooutbreaks due to laboratory contamination should be considered when unusual isolates or recovery from persons with atypical illness occur.66 Prevention and control Interrupting transmission of virus Effective control of outbreaks in acute-care facilities is challenging since several reservoirs for transmission exist, including patients, providers, staff, and visitors. Once a community outbreak has been recognised, patients presenting for hospital admission with febrile respiratory illness should be tested for influenza and placed into droplet precautions (private room, mask for healthcare workers and for patients when they leave the room) until their influenza status is known (figure 2). Similarly, droplet precautions should be employed for hospitalised patients with documented or suspected influenza.15 As influenza rapid antigen tests are relatively insensitive, patients with suspected influenza and a negative rapid antigen test should remain in droplet precautions for at least 5 days pending results of influenza cultures before removal from isolation. When appropriate, it is also important to consider and screen for alternative diagnoses, such as respiratory synctial virus, adenovirus, Mycoplasma pneumoniae, and others. Rapid tests may be useful in situations where patient cohorting becomes necessary. Patients with documented influenza of the same type may be placed in the same room or cohort ward. To prevent further spread, visitors with fever and respiratory symptoms should be discouraged from entering the hospital. This can be accomplished by public service announcements in local newspapers, on television news, over radio broadcasts, and on memos posted around hospital entrances. Another nosocomial reservoir is healthcare providers and other hospital employees who have patient contact. Sending home of these workers when they exhibit symptoms typical of influenza becomes an important control measure. Illness in healthcare workers is likely to be under-reported because these people are highly motivated and typically do not want to burden co-workers, consider their roles important, and may not feel “ill enough” to stay home. In addition, many paid healthcare workers may avoid using sick leave days that are combined THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza with vacation time. Educational efforts explaining the seriousness of illness caused by influenza, its mode of transmission and nosocomial spread, as well as ways they can participate in control and prevention activities may be necessary to gain compliance with furlough of infected workers. Immunisation Annual influenza immunisation of high-risk persons and their contacts, including healthcare providers, is the primary means of preventing nosocomial influenza; however, vaccination has been substantially underused, particularly by healthcare workers.15 acceptance by healthcare workers continues to be below recommended levels. One survey during the1986–87 season at a large acute-care hospital found that among healthcare workers, immunisation rates were less than 10%.9 Also, 35% developed influenza-like illness during that influenza season, and 77% took care of patients while ill.9 Even in the setting of community-wide epidemics, several studies have reported that one-third or fewer hospital workers accepted influenza vaccine,73,74 and some studies have found lower rates of healthcare worker acceptance than among non-medical employees in such institutions.75 More recent surveys in the USA have found that approximately 30% of healthcare workers receive influenza immunisations.76 Inactivated influenza vaccine in patients Increasing rates of pre-season immunisation among people in the community and healthcare providers would lessen the impact of influenza on the healthcare environment. Prevention of influenza has been hampered by inadequate vaccine use and by the fact that many high-risk patients do not develop a protective antibody response after immunisation. The Advisory Committee on Immunisation Practices (ACIP) of the Centers for Disease Control and Prevention (CDC) recommends annual immunisation vaccination of high-risk patients and their contacts.15 Highrisk is currently defined as age 50 years or more, residence in a nursing home or other long-term care facility, chronic pulmonary or cardiovascular disorders, underlying illness that requires regular medical follow-up or hospitalisation, long-term aspirin therapy, and pregnancy entering its second or third trimester during an influenza season. Generally, in healthy young adults, influenza vaccine efficacy has ranged from 70–90% depending upon seasonal virus circulation.67 Among the US population over the age of 65, influenza vaccination rates have increased from 33% in 198968 to 63% in 1998.69 The greatest proportion of those vaccinated are non-Hispanic whites, and rates are lower among blacks and Hispanics.69 According to CDC, only 23% of adults younger than 50 years with high-risk conditions were vaccinated last year in the USA, far short of the 60% goal.15 The effectiveness of vaccine in preventing influenza among the institutionalised elderly and those with chronic illness may be as low as 30–40%, but can be effective for reducing hospitalisations, complications, and deaths.70,71 One recent study suggested that an increased influenza vaccination rate among elderly patients in the community (to 65%) could have had an important effect (along with a significant increase in healthcare worker vaccination compliance rate) on risk for nosocomial influenza among patients in the local hospital.30 Inactivated influenza vaccine in healthcare workers ACIP has also recommended annual influenza immunisation for healthcare workers since 1981 and has targeted providers in both hospitals and outpatient care settings as a high priority group since 1986.72 The goals of this strategy are to reduce the risk of patient influenza exposure and to ensure provision of essential community services, so that patient care is not disrupted. Vaccine THE LANCET Infectious Diseases Vol 2 March 2002 Vaccine acceptance among healthcare workers Surveys of healthcare workers have identified various reasons for a lack of vaccine acceptance including inconvenience, necessity of obtaining written informed consent, concerns about vaccine effectiveness, concerns about vaccine side-effects including misconceptions that vaccine may cause influenza, and lack of understanding of the risks of acquiring infection and transmitting it to high-risk patients. One study found that the most frequently cited reasons for non-acceptance were fear of side-effects (35%), avoidance of medications (33%), reaction to vaccine in the past (24%), impression of low risk of acquiring influenza (18%), and dislike of shots (18%).75 Specific occupational groups have different concerns, which remain to be fully elucidated. Better understanding of epidemiological risk factors and barriers to influenza immunisation should translate into better acceptance of vaccine. For example, a study at the University of Iowa74 found that physicians were significantly more likely to accept the vaccine than other occupational groups. Advancing age, prior absenteeism (as surrogate marker for underlying illness), higher socioeconomic status (salary level), and marriage were associated with increased vaccine uptake in various target groups.74 Another study found that the predictors of acceptance were prior receipt of influenza vaccine, age of at least 50 years, and knowledge that vaccine does not cause influenza.75 Vaccine efforts need to be directed at all healthcare workers and other staff who deal directly with sick people. In this regard, the Iowa study concluded that professional support staff with patient contact, younger workers, those recently employed, and lower paid workers should be considered for targeted efforts at increasing immunisation.74 To maximise convenience and minimise disruption of usual clinical activities, the ACIP recommends using a mobile cart to deliver vaccine onsite to healthcare workers in the workplace, increased availability after hours, and follow-up immunisation programmes early in the course of recognised community outbreaks.15 At the University of Virginia, use of a chart, showing updated healthcare worker compliance rates with influenza vaccine, posted in frequented areas of the hospital, was partly responsible for increasing vaccination acceptance rates to nearly 70% (figure 1).30 http://infection.thelancet.com 151 For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza number of reported febrile days exceeded the number of work absence days suggested that the affected healthcare providers reported for work during febrile illness82 (figure 3). Inactivated influenza vaccine is highly effective in preventing influenza infection in younger healthcare workers, reduces days of illness, and probably days of absenteeism.82 Vaccine efficacy Convincing evidence that influenza immunisation of healthcare workers would prevent nosocomial crossinfection would be a compelling argument in favour of healthcare worker immunisation. Two studies conducted in Scottish geriatric hospitals, one retrospective and the other prospective, provided evidence that healthcare worker immunisation was associated with reduced mortality during influenza season in the resident patients.8,77 However, it was unclear whether the reduction in mortality was due to prevention of influenza, since no data were presented regarding specific rates of influenza-related illness or pneumonia in patients or workdays lost in healthcare providers.8 Another study from the University of Virginia documented an association between a significant improvement in vaccine compliance among hospital employees and a significant decrease in nosocomial influenza among hospital patients over the same period.30 Questions remain regarding the cost benefit of vaccinating healthy working adults because cost-benefit studies have not been consistent.78–81 A recent assessment of the societal economic impact of annual influenza vaccination of healthy working adults was conducted by Nichol.78 A mathematical model estimated that, vaccination prevented an average of 12 workday absenteeisms and three physician visits per 100 persons vaccinated per year. Vaccination also resulted in a net savings of almost $14 for each person vaccinated and was cost effective 95% of the time.78 Few studies have focused specifically on healthcare workers. One double-blind, placebo-controlled trial found that vaccine efficacy based on serological evidence of infection was high in healthcare workers, mostly resident physicians and medical students.82 Compared with the cumulative influenza attack rate of 14% in the control group, protection against laboratory-documented influenza infection averaged approximately 90% for both influenza A and B virus infections. Reductions in days of reported febrile illness or days of absence were observed in vaccine recipients, but these were not statistically significant, perhaps due to relatively small samples sizes, intercurrent non-influenza illness, and a high degree of motivation to return to work. The finding that the mean Live-attenuated, intranasal influenza vaccine A live-attenuated, cold-adapted influenza vaccine (CAIV) administered intranasally is under active consideration by the United States Food and Drug Administration (USFDA) for use in adults younger than 65 years and children. One question regarding CAIV is whether it is as efficacious as the injected inactivated vaccine in adults. A 5-year controlled, blinded trial involving 5210 healthy adult, over 80% of whom were aged 18–65 years, compared the efficacy of inactivated vaccine to attenuated intranasal CAIV formulations containing two influenza A viruses. For H1N1 infection, the efficacy of intranasal vaccine was 85% in preventing culture-positive illness compared with 76% for injected vaccine. For H3N2 infection the efficacies were 58% and 76%, respectively. Inactivated vaccine significantly increased serum haemagglutination-inhibiting (HAI) antibody titres and appeared to be more protective against seroconversion and reported illness.83 However, this trial was limited by use of a ten-fold lower dose of CAIV virus than current formulations, administration by drops instead of spray (perhaps more effective), and, unlike the situation for injected vaccine components, by lack of frequent intranasal CAIV virus updates to assure a good antigenic match with the circulating strains of influenza. A recent study in experimental intranasally induced human influenza A and B found that the efficacy of a trivalent CAIV in preventing influenza following wild-type challenge was higher with CAIV-T (85%) than with trivalent inactivated vaccine (71%).84 A related question is whether vaccine recipients, symptomatic or not, are at risk of transmitting vaccine virus infection to hospitalised patients. In immunised adults the titres of recoverable virus in the nasal secretions have usually been much lower than those needed to induce infection in seronegative children (103 50% tissue culture infectious doses [TCID50]/mL) or adults (105 TCID50/mL). Very rare instances of transmission of CAIV have been recognised from immunised children. % serological influenza Days of absence per 100 persons Antiviral chemoprophylaxis Amantadine and rimantadine Days of febrile respiratory illness per 100 persons Trivalent influenza vaccine Control 0 10 20 Figure 3. Influenza vaccine in healthcare workers: cumulative outcomes.82 152 30 40 50 Two classes of antiviral agents, the M2 ion-channel inhibitors (amantadine and rimantadine) and the neuraminidase inhibitors (oseltamivir and zanamivir) have proven efficacy in preventing influenza illness and infection. The first reported trial demonstrating the THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza efficacy of antiviral chemoprophylaxis for hospital-acquired influenza was Clinical conducted in 197285 (figure 4). The influenza A trial was a non-blinded randomised comparison of oral amantadine with no Subclinical treatment among patients admitted to influenza A either a general medicine or neurology ward during a recognised community epidemic of influenza A. Amantadine Total infection was given in a dose of 100 mg twice daily for the duration of hospitalisation. Four patients were admitted with active 0 5 10 15 20 25 influenza. The efficacy of amantadine in % of patients preventing influenza illness was 100% compared with an attack rate of 11% in No treatment (n=61) Amantadine (n=50) the group receiving no treatment. Almost all of the infections occurred in Figure 4. Prevention of nosocomial influenza with amantadine. those with a low admission HI titre (less than or equal to 1/20) and only a few patients had Oseltamivir and zanamivir The neuraminidase inhibitors (zanamivir and oseltamivir) received vaccine.85 Emergence and transmission of drug-resistant influenza have demonstrated safety and efficacy in prevention of may occur rapidly with use of amantadine or rimantadine, both influenza A and B in community and family-based specifically in institutions and households. Up to one-third studies.45,91 Oseltamivir may be given orally; however, of patients may shed resistant strains by 2–5 days after because of poor bioavailablity, zanamivir must be given by starting treatment for influenza56,86,87 Of note, among inhalation. One controlled trial of inhaled zanamivir for immunocompromised rimantadine recipients, 83% of those prevention of influenza in families with an index case of shedding virus for 3 days or more had resistant variants.56 illness demonstrated that zanamivir provided protection Failures of drug prophylaxis and at least one case of fatal against influenza A and B with a 79% reduction in the influenza due to probable nosocomial transmission of proportion of families with at least one affected contact.45 Also, importantly, no zanamivir-resistant variants were resistant virus have occurred in nursing home residents.88 The optimal duration of antiviral chemoprophylaxis for a recovered among index cases who received zanamivir. The nosocomial outbreak is not certain. Randomised trials in protective efficacy of long-term use of oseltamivir during nursing home populations, which employed short-term or an influenza season was 74% for influenza-like illness and long-term amantadine prophylaxis found that administration 87% for culture-proven disease.91 When used for postfor 14 days, and 7 days beyond the last confirmed illness on exposure prophylaxis in families, oseltamivir (75 mg once the unit, appeared to be adequate.88 The duration of antiviral daily for 7 days) was 90% effective.92 Clinical isolates with prophylaxis should be guided by ongoing surveillance on the reduced susceptibility to zanamivir and oseltamivir after specific ward or facility. Once drug use has been initiated, treatment have been reported uncommonly;{Roche virologic testing of any new illnesses is helpful in guiding Laboratories 2000 70 /id} but data are limited regarding duration and to look for other nosocomial pathogens (eg, possible transmission of these strains and their ultimate respiratory syncitial virus [RSV]). Persistent isolations of virus clinical impact. with new cases suggest circulation of drug-resistant variants or The use of zanamivir and oseltamivir during acute-care new introductions. For example, in nursing homes, outbreaks outbreaks of influenza has not been reported, but evidence of parainfluenza virus, RSV, and influenza B virus have been exists regarding effective influenza prevention in other documented during periods of community influenza A institutional settings. One randomised, unblinded study activity or there have been out-of-season outbreaks not temporally associated with community activity.89 Search strategy and selection criteria Amantadine and rimantadine can cause central nervous Medline searches of English language publications from system (CNS) and gastrointestinal side-effects. The incidence 1966–2001 on the major topic headings: influenza, cross of CNS side-effects seems to be greater for those taking infection, nosocomial influenza, morbidity, mortality, amantadine. One nursing home study evaluated CNS toxicity death, patient outcome, influenza vaccine, infection control, among patients receiving amantadine or rimantadine for cost, cost benefit analysis, prevention, and treatment (and influenza A prophylaxis during the 1997–1998 influenza combinations thereof) as well as personal Reference season.90 Amantadine was associated with a significantly Manager files and bibliography lists of textbook chapters higher rate of CNS toxicity than rimantadine (18·6% vs 1·9%, were used as the database for this review. References were p<0·01) and more patients discontinued amantadine (17·3% selected for use if they addressed the impact, transmission, vs 1·9%, p<0·001). Confusion was the predominant symptom prevention, or control of influenza in the acute-care associated with CNS toxicity and after multivariate analysis, hospital, or the efficacy and recommendations for influenza male sex, reduced creatinine clearance, and amantadine use vaccine and chemoprophylaxis. were all significant risk factors for CNS toxicity.90 85 THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com 153 For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza compared zanamivir chemoprophylaxis with rimantadine during an influenza A nursing home outbreak and to no prophylaxis during a subsequent influenza B outbreak.93 During the prophylaxis period for the influenza A outbreak, only one laboratory-confirmed case of influenza A occurred in a volunteer taking rimantadine and during the prophylaxis period for the influenza B outbreak, only one laboratory- confirmed case of influenza B occurred in a volunteer not receiving prophylaxis.93 Even though this study was small with a low overall influenza attack rate, it demonstrated that zanamivir was comparable to the standard of care at the time. A follow-up, blinded trial study compared 2-week prophylaxis with oral rimantadine (100 mg daily) with inhaled zanamivir (10 mg daily in nursing homes with documented influenza A outbreaks) and found that zanamivir was associated with 61% additional protective efficacy compared with rimantadine, References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Weingarten S, Friedlander M, Rascon, Ault M, Morgan M, Meyer R. Influenza surveillance in an acute-care hospital. Arch Intern Med 1988; 148: 113–16. Kapila R, Lintz D, Tecson F, Ziskin L, Louria D. A nosocomial outbreak of influenza A. Chest 1977; 71: 576–9. Anonymous, Centers for Disease Control and Prevention. Suspected nosocomial influenza cases in an intensive care unit. Morb Mortal Wkly Rep 1988; 37: 3–4. Van-Voris L, Belshe R, Shaffer J. Nosocomial influenza B virus infection in the elderly. Ann Intern Med 1982; 96: 153–8. Bean B, Rhame F, Hughes R, Weiler M, Peterson L, Gerding D. Influenza B: hospital activity during a community epidemic. Diag Microbiol Infect Dis 1983; 1: 177–83. Muchmore H, Felton F, Scott L. A confirmed hospital epidemic of Asian influenza. J Oklahoma State Med Assoc 1960; 53: 142–5. Bauer C, Elie K, Spence L, Stern L. Hong Kong influenza in a neonatal unit. JAMA 1973; 223: 1233–35. Elder A, O’Donnell B, McCruden E, Symington I, Carman W. Incidence and recall of influenza in a cohort of Glasgow healthcare workers during the 1993-4 epidemic: results of serum testing and questionnaire. BMJ 1996; 313: 1241–42. Weingarten S, Riedinger M, Bolton L, Miles P, Ault M. Barriers to influenza vaccine acceptance: a survey of physicians and nurses. Am J Infect Control 1989; 17: 201–207. Barker W, Mullooly J. Impact of epidemic type A influenza in a defined adult population. Am J Epidemiol 1980; 112: 798–811. Neuzil K, Reed G, Mitchel E, et al. Influenzaassociated morbidity and mortality in young and middle aged women. JAMA 1999; 281: 901–907. Simonsen L, Fukuda K, Schonberger L, Cox N. Impact of influenza epidemics on hospitalisations. J Pediatr 2000; 181: 831–37. Noble G. Epidemiological and clinical aspects of influenza. In: Beare A, editor. Basic and applied influenza research. 1 ed. Boca Raton, FL: CRC Press, 1982: 27–38. Lui K, Kendal A. Impact on influenza epidemics on mortality in the United States from October 1972 to May 1985. Am J Public Health 1987; 77: 712–16. Anonymous. Prevention and control of influenza; recommendations of the Advisory Committee on Immunisation Practices (ACIP). Morb Mortal Wkly Rep 2001; 50: 1–44. Simonsen L, Clarke M, Schonberger L, Arden N, Cox N, Fukuda K. Pandemic versus epidemic influenza mortality: a pattern of changing age distribution. J Infect Dis 1998; 178: 53–60. Serwint J, Miller R. Why diagnose influenza infections in hospitalised pediatric patients? Pediatr Infect Dis J 1993; 12: 200–4. Oxford J. Influenza A pandemics of the 20th century with special reference to 1918: virology, pathology, and epidemiology. Rev Med Virol 2000; 10: 119–33. Schoch-Spana M. Implications of pandemic influenza 154 in part because of failures of rimantadine prophylaxis due to drug-resistant strains.48 In a highly immunised nursing home population, seasonal prophylaxis with oral oseltamivir 75 mg once daily was well-tolerated and 92% preventive against influenza compared with placebo.94 These newer antiviral agents are highly effective for prevention but more expensive than M2 inhibitors, and other than documented influenza B outbreaks, their role in prevention of nosocomial outbreaks in acute-care hospitals remains to be determined. Given the significant healthcare effects associated with nosocomial influenza outbreaks, CDC recommends that institutions consider prophylaxis for exposed patients, unvaccinated personnel, and those vaccinated less than 2 weeks before exposure. When choosing which antiviral to use for prophylaxis, sideeffects, cost, and influenza strain and susceptibility pattern should be considered. for bioterrorism response. Clin Infect Dis 2000; 31: 1409–13. 20 Blumenfeld H, Kilbourne E, Louria D, Rogers D. Studies on influenza in the pandemic of 1957-58: I. An epidemiologic, clinical, and serologic investigation of an intrahospital epidemic, with a note on vaccine efficacy. J Clin Invest 1959; 38: 199–212. 21 Meltzer M, Cox N, Fukuda K. The economic impact of pandemic influenza in the United States: priorities for intervention. Emerg Infect Dis 1999; 5: 659–71. 22 Gowda H. Influenza in a geriatric unit. Postgrad Med J 1979; 55: 188–91. 23 Hall C, Douglas R. Nosocomial influenza infection as a cause of intercurrent fevers in infants. Paediatrics 1975; 55: 673–7. 24 Munoz F, Campbell J, Atmar R, Garcia-Prats J, Baxter D, Johnson L, et al. Influenza A virus outbreak in a neonatal intensive care unit. Paediatr Infect Dis J 1999; 18: 811–15. 25 Malavaud S, Malavaud B, Sandres, Durand D, Marty N, Icart J, et al. Nosocomial outbreak of influenza virus A (H3N2) infection in a solid organ transplant department. Transplantation 2001; 72: 535–37. 26 Adal K, Flowers R, Anglim A, Hayden F, Titus M, Coyner B, et al. Prevention of nosocomial influenza. Infect Control Hosp Epidemiol 1996; 17: 641–48. 27 Glezen W, Falcao O, Cate T, Mintz A. Nosocomial influenza in a general hospital for indigent patients. Can J Infect Control 1991; 6: 65–67. 28 Meibalane R, Sedmak G, Sasidharan P, Garg P, Grausz J. Outbreak of influenza in a neonatal intensive care unit. J Pediatr 1997; 91: 974–76. 29 Weinstock D, Eagan J, Malak S, Rogers M, Wallace H, Kiehn T et al. Control of influenza A on a bone marrow transplant unit. Infect Control Hosp Epidemiol 2000; 21:730–32. 30 Salgado C, Giannetta E, Hayden F, Farr B. Preventing nosocomial influenza by improving clinicians’ vaccine acceptance. Abstracts of the 11th Annual Scientific Meeting of the Society for Healthcare Epidemiology of America; Toronto, Canada; 2001; 271. 97. 31 Aschan J, Ringdien O, Ljungman, Andersson J, Lewensohn-Fuchs I, Forsgren M. Influenza B in transplant patients. Scand J Infect Dis 1989; 21: 349–50. 32 Mauch T, Bratton S, Myers, Krane E, Gentry S, Kashtan C. Influenza B virus infection in pediatric solid organ transplant recipients. Paediatrics 1994; 94: 225–29. 33 Whimbey E, Elting L, Couch R, et al. Influenza A virus infections among hospitalised adult bone marrow recipients. Bone Marrow Transplant 1994; 13: 440. 34 Kempe A, Hall C, MacDonald N, et al. Influenza in children with cancer. J Paediatr 1989; 115: 33–9. 35 Hirschhorn L, McIntosh K, Anderson K, Dermody T. Influenzal pneumonia as a complication of autologous bone marrow transplantation. Clin Infect Dis 1992; 14: 786–87. 36 Yousuf H, Englund J, Couch R, et al. Influenza among hospitalised adults with leukemia. Clin Infect Dis 1997; 24: 1095–99. 37 Lin J, Nichol K. Excess mortality due to pneumonia or influenza during influenza seasons among persons with acquired immunodeficiency syndrome. Arch Intern Med 2001; 161: 441–46. 38 Couch R, Cate T, Douglas R, Gerone P, Knight V. Effect of route of inoculation on experimental respiratory viral disease in volunteers and evidence of airborne transmission. Bacteriol Rev 1966; 30: 517–29. 39 Alford R, Kasel J, Gerone P, Knight V. Human influenza resulting from aerosol inhalation. Proc Soc Exp Biol Med 1966; 122: 800–804. 40 Treanor J, Hayden F. Volunteer challenge studies. In: Nicholson K, Webster R, Hay A, editors. Textbook of Influenza. 1 ed. Malden, MA: Blackwell Science Ltd, 1998: 517–37. 41 Nicholson K. Human influenza. In: Nicholson K, Webster R, Hay A, editors. Textbook of Influenza. 1 ed. Malden, MA: Blackwell Science Ltd.; 1998. p. 219–64. 42 Calfee D, Peng A, Hussey E, Lobo M, Hayden F. Safety and efficacy of once daily intranasal zanamivir in preventing experimental human influenza A infection. Antiviral Therapy 1999; 4: 143–49. 43 Tannock G, Gillett S, Gillett R, Barry R, Hensley M, Herd R, et al. A study of intranasally administered interferon A (rIFN-a2A) for the seasonal prophylaxis of natural viral infections of the upper respiratory tract in healthy volunteers. Epidemiol Infect 1988; 101: 611–21. 44 Kaiser L, Henry D, Flack N, Keene O, Hayden F. Short-term treatment with zanamivir to prevent influenza: results of a placebo-controlled study. Clin Infect Dis 2000; 30: 587–89. 45 Hayden F, Gubareva L, Monto A, Klein T, Elliot M, Hammond J, et al. Inhaled zanamivir for the prevention of influenza in families. Zanamivir Family Study Group. N Engl J Med 2000; 343: 1282–89. 46 Hemmes J, Winkler K, Kool S. Virus survival as a seasonal factor in influenza and poliomyelitis. Nature 1960; 4748: 430–31. 47 Moser M, Bender T, Margolis H, Ritter D, Noble G, Kendal A, et al. An outbreak of influenza aboard a commercial airliner. Am J Epidemiol 1979; 110: 1–6. 48 Gravenstein S, Drinka P, Osterweil D, Schilling M, McElhaney J, Elliot M, et al. A multicenter prospective double-blind randomized controlled trial comparing the relative safety and efficacy of zanamivir to rimantidine for nursing home influenza outbreak control. Abstracts of the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy; Toronto, Canada; 2000. 1155: 270. 49 Gardner P, Court S, Brocklebank J, Downham M, Weightman D. Virus cross-infection in pediatric wards. BMJ 1973; 9: 571–75. 50 Bean B, Moore B, Sterner B, Peterson L, Gerding D, Balfour H. Survival of influenza on environmental surfaces. J Infect Dis 1982; 146: 47–51. 51 Morens D, Rash V. Lessons from a nursing home outbreak of influenza A. Infect Control Hosp Epidemiol 1995; 16: 275–80. 52 Frank A, Taber L, Wells C, Wells J, Glezen W, Paredes A. Patterns of shedding of myxoviruses and paramyxoviruses in children. J Infect Dis 1981; 144: 433–41. 53 Betts R, Treanor J, Graman P, Bentley D, Dolin R. Antiviral agents to prevent or treat influenza in the elderly. J Resp Dis 1987; Supplement: S56-S59. THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. Review Nosocomial influenza 54 Evans K, Kline M. Prolonged influenza A infection responsive to rimantadine therapy in a human immunodeficiency virus-infected child. Pediatr Infect Dis J 1995; 14: 332–4. 55 Englund J, Champlin R, Wyde P, Kantarjian H, Atmar R, Tarrand J, et al. Common emergence of amantadine- and rimantadine-resistant influenza A viruses in symptomatic immunocomprimised adults. Clin Infect Dis 1998; 26: 1418–24. 56 Klimov A, Rocha E, Hayden F, Shult P, Roumillat L, Cox N. Prolonged shedding of amantadine-resistant influenza A viruses by immunodeficient patients: detection by polymerase chain reaction-restriction analysis. J Infect Dis 1995; 172: 1352–5. 57 Ison MG, Gnann J, Treanor J, et al. Phase II study of the safety and efficacy of nebulized zanamivir in patients with serious influenza virus infections. Abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, IL, 2001: 289 58 Leonardi G, Leib H, Birkhead G, Smith C, Costello P, Conron W. Comparison of rapid detection methods for influenza A virus and their value in health-care management of institutionalized geriatric patients. J Clin Microbiol 1994; 32: 70–4. 59 Waner JL, Todd SJ, Shalaby H, Murphy P, Wall LV. Comparison of Directigen FLU-A with viral isolation and direct immunofluorescence for the rapid detection and identification of influenza A virus. J Clin Microbiol 1991; 29: 479–82. 60 Dominguez EA, Taber LH, Couch RB. Comparison of rapid diagnostic techniques for respiratory syncytial and influenza A virus respiratory infections in young children. J Clin Microbiol 1993; 31: 2286–90. 61 Claas EC, van Milaan AJ, Sprenger MJ, et al. Prospective application of reverse transcriptase polymerase chain reaction for diagnosing influenza infections in respiratory samples from a children’s hospital. J Clin Microbiol 1993; 31: 2218–21. 62 Duverlie G, Houbart L, Visse B, et al. A nylon membrane enzyme immunoassay for rapid diagnosis of influenza A infection. J Virolog Methods 1992; 40: 77–84. 63 Kok T, Mickan LD, Burrell CJ. Routine diagnosis of seven respiratory viruses and Mycoplasma pneumoniae by enzyme immunoassay. J Virolog Methods 1994; 50: 87–100. 64 Chomel JJ, Remilleux MF, Marchand P, Aymard M. Rapid diagnosis of influenza A. Comparison with ELISA immunocapture and culture. J Virol Methods 1992; 37: 337–43. 65 Atmar RL, Baxter BD. Typing and subtyping clinical isolates of influenza virus using reverse transcriptionpolymerase chain reaction. Clin Diagn Virology 1996; 7: 77–84. 66 Budnick L, Moll M, Hull H, Mann J, Kendal A. A pseudo-outbreak of influenza A associated with the use of laboratory stock strain. Am J Public Health 1984; 74: 607–9. 67 Palache A. Influenza vaccines: a reappraisal of their use. Drugs 1997; 54: 841–56. 68 Centers for Disease Control and Prevention. Influenza and pneumococcal vaccination coverage levels among persons aged 65 years, United States, 1973-93. Morb Mortal Wkly Rep 1995; 44: 506–7. 69 Singelton J, Lu P. Influenza vaccination levels in the United States, 1998. Abstracts of the 35th National Immunisation Conference; Atlanta, GA; 2001 (in press). 70 Dorrell L, Hassan L, Marshall S, Chakraverty P, Ong E. Clinical and serologic response to an inactivated influenza vaccine in adults with HIV infection, diabetes, obstructive airways disease, elderly adults, and healthy volunteers. Int J STD AIDS 1997; 8: 776–79. 71 Gross P, Hermogenes A, Sacks H, Lau J, Levandowski R. The efficacy of influenza vaccine in elderly persons. A meta-analysis and review of the literature. Ann Intern Med 1995; 123: 518–27. 72 Anonymous. Recommendations for prevention and control of influenza. Recommendations of the Immunisation Practices Advisory Committee. Ann Intern Med 1986; 105: 399–404. 73 Watanakunakorn C, Ellis G, Gemmel D. Attitude of healthcare personnel regarding influenza immunisation. Infect Control Hosp Epidemiol 1993; 14: 17–20. 74 Doebbeling B, Edmond M, Davis C, Woodin J, Zeitler R. Influenza vaccination of health care workers: evaluation of factors that are important in acceptance. Preventive Med 1997; 26: 68–77. 75 Heimberger T, Chang H, Shaikh M, Crotty L, Morse D, Birkhead G. Knowledge and attitudes of healthcare workers about influenza: why are they not getting vaccinated? Infect Control Hosp Epidemiol 1995; 16: 412–15. 76 Walker F, Singleton J, Strikas. National Immunisation Programme CAG: are U.S. health care workers receiving annual influenza vaccinations? Abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy; San Fransisco, CA; 1999. 752: 606. 77 Potter J, Stott D, Roberts M, Elder A, O’Donnell B, Knight P, et al. Influenza vaccination of health care workers in long-term care hospitals reduces the mortality of elderly patients. J Infect Dis 1997; 175: 1–6. 78 Nichol K. Cost-benefit analysis of a strategy to vaccinate healthy working adults against influenza. Arch Intern Med 2001; 161: 749–59. 79 Bridges C, Thompson W, Meltzer M, Reeve G, Talamonti W, Cox N, et al. Effectiveness and costbenefit of influenza vaccination of healthy working adults: A randomized controlled trial. JAMA 2000; 284: 1655–63. 80 Campbell D, Rumley M. Cost-effectiveness of the influenza vaccine in a healthy, working-age population. J Occup Environ Med 1997; 39: 408–14. 81 Kumpulainen V, Makela M. Influenza vaccination among healthy employees: a cost benefit analysis. THE LANCET Infectious Diseases Vol 2 March 2002 http://infection.thelancet.com Scand J Infect Dis 1997; 29: 181–85. 82 Wilde J, McMillan J, Serwint J, Butta J, O’Riordan M, Steinhoff M. Effectiveness of influenza vaccine in health care professionals: a randomized trial. JAMA 1999; 281: 908–13. 83 Edwards K, Dupont W, Westrich M, Plummer W, Palmer P, Wright P. A randomized controlled trial of cold-adapted and inactivated vaccines for the prevention of influenza A disease. J Infect Dis 1994; 169: 68–76. 84 Treanor J, Kotloff K, Betts R, Belshe R, Newman F, Iacuzio D, et al. Evaluation of trivalent, live, coldadapted (CAIV-T) and inactivated (TIV) influenza vaccines in prevention of virus infection and illness following challenge of adults with wild-type influenza A (H1N1), A (H3N2), and B viruses. Vaccine 1999; 18: 899–906. 85 O’Donoghue J, Ray C, Terry D, Beaty H. Prevention of nosocomial influenza infection with amantadine. Am J Epidemiol 1973; 97: 1616–20. 86 Hayden F, Belshe R, Clover R, Hay A, Oakes M, Soo W. Emergence and apparent transmission of rimantadine-resistant influenza A virus in families. New Engl J Med 1989; 321: 1696–702. 87 Mast E, Harmon M, Gravenstein, et al. Emergence and possible transmission of amantadine-resistant viruses during nursing home outbreaks of influenza A (H3N2). Am J Epidemiol 1991; 134: 988–97. 88 Drinka P, Gravenstein S, Schilling M, Krause P, Miller B, Shult P. Duration of antiviral prophylaxis during nursing home outbreaks of influenza A: a comparison of 2 protocols. Arch Intern Med 1998; 158: 2155–59. 89 Drinka P, Gravenstein S, Krause P, Langer E, Barthels L, Dissing M, et al. Non-influenza respiratory viruses may overlap and obscure influenza activity. J Am Geriatr Soc 1999; 47: 1087–93. 90 Keyser L, Karl M, Nafziger A, Bertino J. Comparison of central nervous system adverse effects of amantadine and rimantadine used as sequential prophylaxis of influenza A in elderly nursing home patients. Arch Intern Med 2000; 160: 1485–88. 91 Hayden F, Atmar R, Schilling M, Johnson C, Poretz D, Parr D, et al. Use of the selective oral neuraminidase inhibitor oseltamivir to prevent influenza. N Engl J Med 1999; 341: 1336–43. 92 Welliver R, Monto A, Carewicz O, Schatteman E, Hassman M, Hedrick J, et al. Effectiveness of oseltamivir in preventing influenza in household contacts. JAMA 2001; 285: 748–54. 93 Schilling M, Povinellit L, Krause O, Gravenstein M, Ambrozaitis A, Jones H, et al. Efficacy of zanamivir for chemoprophylaxis of nursing home influenza outbreaks. Vaccine 1998; 16: 1771–74. 94 Peters P, Gravenstein S, Norwood P, DeBock V, Van Couter A, Gibbens M et al. Long-term use of oseltamivir for the prophylaxis of influenza in a vaccinated frail older population. J Am Geriatr Soc 2001; 49: 1–7. 155 For personal use. Only reproduce with permission from The Lancet Publishing Group.