Winter Respiratory Viruses and Health Care Use: A Population-Based Study in the Northwest United States

Transcription

1 MAJOR ARTICLE Winter Respiratory Viruses and Health Care Use: A Population-Based Study in the Northwest United States Kathleen M. Neuzil, 1,3 Charles Maynard, 2,4 Marie R. Griffin, 5 and Patrick Heagerty 2,4 1 Department of Medicine, University of Washington School of Medicine, 2 University of Washington School of Public Health and Community Medicine, 3 Department of Medicine and 4 Epidemiologic Research and Information Center, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; and 5 Departments of Medicine and Preventive Medicine, Vanderbilt University School of Medicine and Geriatric Research Education and Clinical Center, Veterans Affairs, Tennessee Valley Healthcare System, Nashville, Tennesseee To quantify health care use among adults during influenza and respiratory syncytial virus (RSV) seasons, we identified a cohort of veterans aged 18 years who used Department of Veterans Affairs (VA) facilities in Oregon and Washington states as their source of health care. During , veterans accrued 237,159 person-years of follow-up. Using VA data sources, we measured acute cardiopulmonary hospitalizations and primary care and urgent care visits. Differences between rates of study events when influenza and/or RSV were circulating and event rates when neither virus was circulating were used to calculate winter virus attributable morbidity. Inpatient and outpatient event rates were consistently higher during winter virus season, compared with non winter virus season. Annual rates of cardiopulmonary hospitalizations attributable to influenza or RSV infection ranged from 0.8 (95% confidence interval [CI], ) per 1000 low-risk individuals aged years, to 10.6 (95% CI, ) per 1000 high-risk individuals aged 65 years. Each year, circulation of influenza and RSV coincide with predictable increases in medical care use. For centuries, increased mortality rates during the winter season and a correlation between winter respiratory illnesses and increased mortality rates have been recognized. Since the isolation of influenza virus in 1933, and since the advent of widespread influenza virus surveillance, the relationship between influenza virus circulation, influenza-like illnesses, and pneumonia and influenza-related hospitalizations and deaths has been more clearly elucidated [1 7]. Although influenza virus circulates yearly, its impact may vary considerably from year to year [2, 4, 6 8]. These variations have Received 21 January 2003; accepted 13 March 2003; electronically published 7 July Financial support: Unrestricted educational grants from GlaxoWellcome Worldwide Epidemiology and Aventis-Pasteur. Reprints or correspondence: Dr. Kathleen M. Neuzil, Medical Service 111-ID, 1660 S. Columbian Way, Seattle, WA (kneuzil@u.washington.edu). Clinical Infectious Diseases 2003; 37:201 7 This article is in the public domain, and no copyright is claimed /2003/ been attributed primarily to the circulating strain of influenza virus, whereas relatively little attention has been paid to the circulation of other respiratory viruses [2, 4, 6 13]. In the Northern Hemisphere, winter outbreaks of respiratory syncytial virus (RSV) infection occur annually between November and April and thus may coincide with outbreaks of influenza [14, 15]. Peak RSV infection season is easily recognized in infants because of the increase in the specific clinical entity of bronchiolitis [14, 16]. In adults, RSV infection may be difficult to distinguish from influenza on the basis of clinical criteria and confounds ecologic estimates of influenza-associated morbidity and mortality [17 20]. RSV caused a substantial proportion of influenza-like illnesses among community-dwelling adults during 3 successive winters in the United Kingdom [20]. Among hospitalized and institutionalized adults, illnesses associated with RSV may be as common and severe as illnesses caused by influenza virus [17 19]. Likewise, Respiratory Viruses and Health Care Use CID 2003:37 (15 July) 201

2 RSV infection may be a significant contributor to wintertime hospitalizations and mortality [21 24]. Recently, the Centers for Disease Control and Prevention (Atlanta, GA) adjusted its methodology for calculating influenza-related mortality to include RSV circulation [24]. For the first time, this model allowed estimates of national RSV infection associated mortality for all age groups. In this study, we examined health care use in a population of veterans in the northwest United States in relation to influenza and RSV circulation. We determined rates of hospitalizations and outpatients visits among veterans aged 18 years who received health care at the Department of Veterans Affairs (VA) Medical Centers in Oregon and Washington during the periods when influenza virus and/or RSV circulated, as well as during times when they did not circulate. METHODS We conducted a 3-year, retrospective, cohort study of RSV infection and influenza-associated morbidity among all veterans observed in the VA Health Care System in Oregon and Washington during the fiscal years 1997, 1998, and Eligible patients were observed during the next 3 consecutive fiscal years, from 1998 to 2000 (1 October 1997 through 30 September 2000) to determine the incidence of hospitalizations and outpatient visits for acute cardiopulmonary conditions throughout the year. The incidences of these study outcomes were compared between winter virus season (defined as periods of influenza and/or RSV infection circulation) and non winter virus season (defined as periods of time before and after winter virus season). The study was approved by the Human Subjects Division of the University of Washington and the Research and Development Committee of VA Puget Sound Health Care System (Seattle, Washington). We used encrypted Social Security numbers to link the VA patient treatment file (inpatient file), which contains abstracts on all patients discharged from all VA hospitals, and the VA outpatient clinic file, which contains records for every outpatient visit to a VA facility. The Social Security numbers were used to link these files with the VA beneficiary record locator system death file (BIRLS), which lists all deaths regardless of place of death. Death ascertainment with VA death records has a sensitivity of 91% for veterans who are hospitalized, and it has a sensitivity of 74% for those who have used only outpatient services, compared with state death certificates [25]. Information in the computerized VA databases included demographic characteristics, hospital and outpatient dates of service, and associated diagnoses and procedures. Study population. Eligibility for this study was determined by use of medical care services. In the VA database, such services are organized by fiscal year, which begins on 1 October of the preceding year and ends on 30 September of the current year (e.g., the 1998 fiscal year began on 1 October 1997 and ended on 30 September 1998). For each of the 3 study years (1998, 1999, and 2000), the respective prior year (1997, 1998, and 1999) was used to identify eligible patients and to determine demographic characteristics and the presence of comorbid conditions. All veterans with 2 visits to a VA facility in Oregon or Washington in the year before the study year (indicating regular receipt of care in the baseline year) and no indication of death in that year or the study year (i.e., hospital discharge status of dead or date of death as recorded in BIRLS) were included in the study cohort on 1 October of the study year. The patient remained in the study cohort until 30 September of the following calendar year. With these criteria, a person could be observed for up to 3 years. Thus, all veterans accrued 1 3 person-years of follow-up. Definition of RSV infection and influenza seasons. Viral infection seasons were determined by respiratory virus surveillance conducted at the University of Washington Virology Laboratory at Children s Hospital (Seattle, WA) [26]. Although the surveillance included all specimens submitted from the community, most specimens were obtained from children. We used viral surveillance in children to define our seasons on the basis of studies that demonstrate that the timing and isolation of respiratory viruses in children correlates with the timing and isolation of respiratory viruses in adults in the community and in elderly care centers [20, 27]. Influenza season was defined each year as the period from the first day after the first 7-day period in which there were 4 influenza virus isolates through the week after the last 7-day period with 4 influenza virus isolates. RSV infection season was similarly defined using a threshold of 8 RSV isolates. The 7-day periods were based on prespecified weeks used by the CDC for influenza surveillance that begin with the first Sunday in January. Because of the overlap between influenza and RSV seasons (figure 1), and because the timing of the maximal number of isolates for each virus was similar in 2 of the 3 study years, we combined these 2 seasons for analysis (winter virus season). All time periods that did not meet the definition of either influenza season or RSV season were termed non winter virus season. Non winter virus season consisted of a period that began before the start of winter virus season and lasted until the criteria were met for either influenza or RSV infection season, as well as a period at the end of winter virus season that began when RSV infection season ended. Exposure definition and measures. We categorized all patients by age, race, sex, health care location, VA eligibility, and comorbid medical conditions, as determined by inpatient or 202 CID 2003:37 (15 July) Neuzil et al.

3 Figure 1. Circulation of influenza and respiratory syncytial viruses from October 1997 through September 2000, Seattle, Washington. Graphs show the no. of isolates of influenza virus (solid black line) and respiratory syncytial virus (hatched line) per week during (a), (b), and (c). outpatient diagnoses. The patients age as of 1 October of the study year was recorded. Patients were determined to be high risk if they had 1 of the following conditions: chronic lung disease (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM], codes 277.0, , , , and 519.9), diabetes (ICD-9-CM codes 250 and 648.0), chronic heart disease (ICD-9-CM codes 093, , , , 416, 424, 425, 428, 429, 440, 745, and 746), malignancy (ICD-9-CM codes , except 173), and chronic renal disease (ICD-9-CM codes , 585, and 587 or procedure codes indicating dialysis). Patients who did not have a specified high-risk condition were considered to be at low risk. Outcome measures. Outcome measures were all hospitalizations for cardiopulmonary conditions (ICD-9-CM discharge diagnosis of pneumonia [codes ], influenza [code 487], other acute respiratory conditions [codes ], other respiratory conditions [codes ], and heart failure or myocarditis [codes 422, 427, and 428]), and all primary care or urgent care outpatient visits. Analysis. The primary outcome was hospitalizations for cardiopulmonary conditions. The incidences of study outcomes during both seasons (winter virus season and non winter virus season) were calculated for all veterans combined, for 3 age strata (18 49, 50 64, and 65 years), and for 2 risk strata (low and high) by dividing the number of hospitalizations per selected period by the person-time during that period, with adjustment for fiscal year. Because we used person-time methods, the estimated rates account for the different season lengths during the 3 study years. Also, with 3 years of data and 2 seasons per year, each subject contributes hospitalization or outpatient data in up to 6 time periods. To account for the repeated measurements, we used generalized estimating equation regression methods [28] in the Genmod procedure (SAS software; SAS Institute) to obtain the corrected SE. The winter rate (i.e., the number of excess events during influenza and RSV infection seasons) was calculated by subtracting the adjusted rates during the non winter virus season from adjusted rates during winter virus season. This rate difference was then multiplied by the length of winter virus season Respiratory Viruses and Health Care Use CID 2003:37 (15 July) 203

4 to generate annual excess event rates. Winter excess outpatient events were calculated using a similar method. RESULTS Study population. Eligible veterans accrued 237,159 personyears of follow-up during the 3 study years. Demographic characteristics were similar to national figures for veterans seen at VA health care facilities (table 1) [29]. The population was predominantly male, with 34% of the subjects aged years, 31% aged years, and 35% aged 65 years. The prevalence of high-risk medical conditions increased with age. Pulmonary disease was the most common identified high risk medical condition in veterans aged!50 years, whereas cardiac disease was the most common condition in those aged 50 years. Viral surveillance. On the basis of influenza and RSV surveillance, the duration of winter virus season was weeks, and this period accounted for 35.3% of the total period of the study. During 2 of the 3 years, winter virus season began in early January, and the maximum number of viruses were isolated in late February or early March (figure 1a, 1b). In the third season ( ), winter virus season began in late November. The number of influenza isolates was greatest in January, and the number of RSV isolates was greatest in February (figure 1c). Study outcomes. Cardiopulmonary hospitalization rates during both seasons increased with increasing age. Within all age and risk categories, hospitalization rates were higher during winter virus season, when influenza virus and/or RSV were circulating in the community, compared with non winter virus season. (table 2) During non winter virus season, the rate of hospitalizations for cardiopulmonary conditions was 141 hospitalizations per 1000 person-years among high-risk veterans. This increased by 18%, to 167 hospitalizations per 1000 personyears during winter virus season. Low-risk veterans had substantially lower rates of hospitalization than did high-risk veterans during both seasons. However, the effect of winter virus season was relatively greater in low-risk individuals, in whom rates during winter virus season increased by 30%, from 20 hospitalizations per 1000 person-years during winter virus season to 27 hospitalizations per 1000 person-years during non winter virus season. We assumed that the rate of events during non winter virus season represented baseline values. Thus, events attributable to influenza virus or RSV circulation were calculated by subtracting these rates from the rates for winter virus season. These rate differences were annualized to estimate the contribution of respiratory viruses to event rates in the population (table 2). Among veterans with high-risk conditions, there were an estimated 6.8 (95% CI, ), 4.0 (95% CI, ), and 10.6 (95% CI, ) hospitalizations annually per 1000 Table 1. Demographic characteristics of and health care use among veterans seeking medical care at Veterans Affairs facilities in Oregon and Washington, fiscal years Variable Age, years Total person-years 80,158 74,488 82,513 Sex Male Female Care location Oregon Washington High-risk conditions Cardiac disease Pulmonary disease Diabetes Renal disease Cancer At least 1 of the above NOTE. Data are percentage of patients, unless otherwise indicated. high-risk veterans aged 15 49, 50 64, and 65 years, respectively, attributed to winter respiratory virus infections. Overall rates were much lower for those without high-risk conditions but remained higher in winter virus season (compared with non winter virus season) for all age groups (table 2). Similar to the trends seen for hospitalizations, outpatient visit rates were higher in older age groups (compared with younger age groups), in patients with high-risk conditions (compared with those without high-risk conditions), and during winter virus season (compared with non winter virus season; table 3). Overall, winter virus season was associated with an estimated annual excess of outpatients visits with high-risk conditions. Among those without high-risk conditions, an estimated annual excess of outpatient visits occurred. DISCUSSION In this study, which included a large cohort of veterans in the northwest United States and encompassed 3 winter respiratory virus seasons, hospitalizations and outpatient rates were highest when influenza virus and/or RSV circulated in the community. The association between excess health care use and winter respiratory virus circulation was seen in all age categories and among patients with and without high-risk medical conditions. Veterans aged 65 years and younger veterans with highrisk medical conditions had the highest excess rates of hospitalization for cardiopulmonary conditions during winter virus season. Previous population-based studies estimated annual av- 204 CID 2003:37 (15 July) Neuzil et al.

5 Table 2. Cardiopulmonary hospitalizations and annual winter-virus-attributable cardiopulmonary hospitalizations, by age and risk group. Age, years Persons with high-risk conditions Hospitalizations per 1000 person-years hospitalizations Persons without high-risk conditions Hospitalizations per 1000 person-years Nonwinter Winter Nonwinter Winter hospitalizations ( ) ( ) 6.8 ( ) 13.7 ( ) 15.9 ( ) 0.8 ( ) ( ) ( ) 4.0 ( ) 23.5 ( ) 28.0 ( ) 1.6 ( ) ( ) ( ) 10.6 ( ) 33.6 ( ) 47.2 ( ) 4.8 ( ) NOTE. The winter excess events were calculated by multiplying the event rate difference between non winter virus season and winter virus season by the proportion of overall time in winter virus season. erages of 1 6 hospitalizations for influenza with high-risk conditions, hospitalizations per 1000 persons aged!65 years without high-risk conditions, and 1 2 hospitalizations aged 65 and older without high-risk conditions [4, 5, 7, 13]. Our estimates of hospitalizations per 1000 high-risk individuals and hospitalizations per 1000 low-risk individuals are higher in part because they included RSV infection season in addition to influenza season. In a population-based study that was restricted to persons enrolled in Tennessee Medicaid, Griffin et al. [21] estimated that there were excess hospitalizations for influenza and RSV infection per 1000 adults with chronic lung disease. Possible reasons for the higher rates in that study include differences in overall health and socioeconomic status and in sex and geographic variations in health care use [30]. High influenza vaccine coverage rates among veterans would also be expected to affect our estimates, because vaccination of elderly and high-risk persons is associated with reductions of 30% 57% in the number of hospitalizations for pneumonia and influenza [31, 32]. Although the ability of both influenza virus and RSV to cause acute respiratory disease in otherwise healthy adults is well documented, to our knowledge, this is the first populationbased study to determine rates of disease in inpatients and outpatients associated with circulation of these viruses in adults without high-risk conditions. Our data suggest that persons per 1000 individuals without identified high-risk conditions will have an outpatient visit associated with influenza or RSV infection. Among high-risk individuals, the impact is even greater, with an estimated persons per 1000 individuals having an additional outpatient visit during the winter virus season. These results add to the growing body of evidence that implicates both influenza and RSV infection as major contributors to the excess respiratory disease associated morbidity seen during winter [20, 21, 24]. Other respiratory pathogens, including rhinoviruses, coronaviruses, and adenoviruses, are less seasonal and are isolated throughout the year [33 35]. The isolation of parainfluenza viruses, which may cause serious disease in elderly adults or persons with chronic lung disease, tends to peak in autumn or spring, as it did during the 3 years of this study [16, 36, 37]. A newly described human metapneumovirus has been isolated from children and adults with acute respiratory disease in The Netherlands, the United States, and Canada during winter months. Whether this organism will prove to be an important winter pathogen has yet to be determined [38, 39]. We did not measure environmental factors, such as humidity or temperature, which could affect morbidity and lead to overestimates of the contributions of influenza and RSV infection. Of interest, the association between winter season Table 3. group Outpatient (primary and urgent care) visits by season and annual winter-virus-attributable outpatient visits, by age and risk Age years Persons with high-risk conditions Outpatient visits per 1000 person-years outpatient visits Persons without high-risk conditions Outpatient visits per 1000 person-years Nonwinter Winter Nonwinter Winter outpatient visits ( ) ( ) ( ) ( ) ( ) 46.5 ( ) ( ) ( ) ( ) ( ) ( ) 67.0 ( ) ( ) ( ) ( ) ) ( ) 42.6 ( ) NOTE. The winter excess events were calculated by multiplying the event rate difference between non winter virus season and winter virus season by the proportion of overall time in winter virus season. Respiratory Viruses and Health Care Use CID 2003:37 (15 July) 205

6 and respiratory morbidity persists in residents of nursing homes, where the environment is controlled but where respiratory viruses still circulate [40]. The strengths of our study include the large number of subjects, which provided the power to analyze data by age and risk group, and the inclusion of 3 winter seasons. Our study also has several important limitations. We obtained all data for our analysis from VA databases. Although veterans may have a financial incentive to access VA hospitals exclusively for care, veterans in this system may access non veteran providers. Hospitalizations or outpatient visits at non-va facilities would not be included. Medical comorbidity was based solely on ICD-9 codes; this may have led to some misclassification of the risk status of veterans in this study. Finally, our cohort was predominantly male, which may limit the generalizability of our results. Our study of excess hospitalizations and outpatient visits for cardiopulmonary conditions provides further evidence of the impact of influenza and RSV infection on adult morbidity and the health care system. Seasonal stresses on the health care system due to these viruses should be anticipated and may be eased with advanced planning, such as decreasing the number of elective admissions or routine visits during winter virus season, or by anticipating increased staffing needs [41]. Such planning will be key to responding to the next influenza pandemic or other unanticipated events that may affect health care workers and a substantial segment of the community. Cocirculation of influenza virus and RSV during the 3 years of this study precluded us from estimating the individual impact of these viruses in this population. RSV circulation may partly account for the wide variability in influenza-associated morbidity and mortality from year to year and for the difficulty in determining the effect of influenza vaccination coverage on respiratory illness within a community. Laboratory-based studies that differentiate RSV- and influenza-associated morbidity will be important for determining the impact of influenza vaccines, antiviral medications, and future RSV vaccines on improving wintertime health in adults. Acknowledgment We are indebted to Anne Cent (Children s Hospital and Medical Center; Seattle, WA) for her dedication to organizing and maintaining the viral surveillance data. References 1. Glezen WP, Payne AA, Snyder DN, Downs TD. Mortality and influenza. J Infect Dis 1982; 146: Simonsen L, Clarke MJ, Williamson DG, Sroup DF, Arden NH, Schonberger LB. The impact of influenza epidemics on mortality: introducing a severity index. Am J Public Health 1997; 87: Lui KJ, Kendal AP. Impact of influenza epidemics on mortality in the United States from October 1972 to May Am J Public Health 1987; 77: Centers for Disease Control and Prevention. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2002; 51(RR-03): Barker WH, Mullooly JP. Impact of epidemic type A influenza in a defined adult population. Am J Epidemiol 1980; 112: Simonsen L, Fukuda, K, Schonberger LB, Cox NJ. Impact of influenza epidemics on hospitalizations. J Infect Dis 2000; 181: Neuzil KM, Reed GW, Mitchel EF Jr, Griffin MR. Influenza-associated morbidity and mortality in young and middle-aged women. JAMA 1999; 281: Griffin MR, Neuzil KM. The global implications of influenza in Hong Kong. N Engl J Med 2002; 347: Neuzil KM, Mellen BG, Wright PF, Mitchel EF, Griffin MR. The impact of influenza on hospitalizations, outpatient visits, and antibiotic prescriptions in children. N Engl J Med 2000; 342: Izurieta HS, Thompson WW, Piotr K, et al. Influenza and the rates of hospitalization for respiratory disease among infants and young children. N Engl J Med 2000; 342: Mullooly JP, Barker WH. Impact of type A influenza on children: a retrospective study. Am J Public Health 1982; 72: Neuzil KM, Wright PF, Mitchel EF, et al. The burden of influenza illness in children with asthma and other chronic medical conditions. J Pediatr 2000; 137: Perrotta DM, Decker M, Glezen WP. Acute respiratory disease hospitalizations as a measure of impact of epidemic influenza. Am J Epidemiol 1985; 122: Boyce TG, Mellen BG, Mitchel EF Jr, Wright PF, Griffin MR. Rates of hospitalization for respiratory syncytial virus infection among children in Medicaid. J Pediatr 2000; 137: Gilchrist S, Tdrdk, TJ, Gary, HE Jr, Alexander JP, Anderson LJ. National Surveillance for Respiratory Syncytial Virus, United States, J Infect Dis 1994; 170: Hall CB. Respiratory syncytial virus and parainfluenza virus. N Engl J Med 2001; 344: Falsey AR, Cunningham CK, Barker WH, et al. Respiratory syncytial virus and influenza A infections in the hospitalized elderly. J Infect Dis 1995; 172: Dowell SF, Anderson LJ, Gary HE, et al. Respiratory syncytial virus is an important cause of community-acquired lower respiratory infection among hospitalized adults. J Infect Dis 1996; 174: Falsey AR, Treanor JJ, Betts RF, Walsh EE. Viral respiratory infections in the institutionalized elderly: clinical and epidemiologic findings. J Am Geriatr Soc 1992; 40: Zambon MC, Stockton JD, Clewley JP, Fleming DM. Contribution of influenza and respiratory syncytial virus to community cases of influenza-like illness: an observational study. Lancet 2001; 358: Griffin MR, Coffey CS, Neuzil KM, et al. Winter viruses: influenza and respiratory syncytial virus-related morbidity. Arch Intern Med 2002; 162: Nicholson KG. Impact of influenza and respiratory syncytial virus on mortality in England and Wales from January 1975 to December Epidemiol Infect 1996; 116: Fleming DM, Cross KW. Respiratory syncytial virus or influenza. Lancet 1993; 342: Thompson WW, Shay DK, Weintraub E, et al. Mortality associated with influenza and respiratory syncytial virus in the United States. JAMA 2003; 289: Dominitz JA, Maynard C, Boyko EJ. Assessment of vital status in Department of Veterans Affairs national databases: comparison with state death certificates. Ann Epidemiol 2001; 11: University of Washington Clinical Virology Laboratory. Respiratory virus detections. 3 January Available at: washington.edu/rspvirus/viral_detections_1999.htm. Accessed 29 August CID 2003:37 (15 July) Neuzil et al.

7 27. Falsey AR, McCann RM, Hall WJ, et al. Acute respiratory tract infection in daycare centers for older persons. J Am Geriatr Soc 1995; 43: Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika 1986; 73: Ashton CM, Petersen NJ, Wray NP, Yu HJ. The Veterans Affairs medical care system: hospital and clinic utilization statistics for Med Care 1998; 36: Ashton CM, Petersen NJ, Souchek J, et al. Geographic variation in utilization rates in veterans affairs hospitals and clinics. N Engl J Med 1999; 340: Nichol KL, Margolis KL, Wuorenma J, von Sternberg T. The efficacy and cost-effectiveness of vaccination against influenza among elderly persons living in the community. N Engl J Med 1994; 331: Nichol KL, Baken L, Nelson A. Relation between influenza vaccination and outpatient visits, hospitalization, and mortality in elderly persons with chronic lung disease. Ann Intern Med 1999; 130: Greenberg SB, Allen M, Wilson J, Atmar RL. Respiratory viral infections in adults with and without chronic obstructive pulmonarydisease. Am J Respir Crit Care Med 2000; 162: Monto AS, Cavallaro JJ. The Tecumeseh study of respiratory illness. II. Patterns of occurrence of infection with respiratory pathogen Am J Epidemiol 1971; 94: Monto AS, Sullivan KM. Acute respiratory illness in the community: frequency of illness and the agents involved. Epidemiol Infect 1993; 110: Reed G, Jewett PH, Thompson J, Tollefson S, Wright PW. Epidemiology and clinical impact of parainfluenza virus infections in otherwise healthy infants and young children!5 years old. J Infect Dis 1997; 175: Glezen WP, Greenberg SB, Atmar RL, Piedra PA, Couch RB. Impact of respiratory virus infections on persons with chronic underlying conditions. JAMA 2000; 283: Van Den Hoogen BG, De Jong JC, Groen J, et al. A newly discovered human pneumovirus isolated from young children with respiratory tract disease. Nat Med 2001; 7: Peret TCT, Boivin G, Li Y, Couillard M, et al. Characterization of human metapneumoviruses isolated from patients in North America. J Infect Dis 2002; 185: Ellis SE, Coffey CS, Mitchel EF, Dittus RS, Griffin MR. Influenza and respiratory syncytial virus-associated morbidity and mortality in the nursing home population. J Am Geriatr Soc 2003; 51: Glaser CA, Gilliam S, Thompson WW, et al. Medical care capacity for influenza outbreaks, Los Angeles. Emerg Infect Dis 2002; 8: Respiratory Viruses and Health Care Use CID 2003:37 (15 July) 207