| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Department of Community Health Sciences and Manitoba Centre for Health Policy, The University of Manitoba, Winnipeg, Canada
b Department of Medicine, The University of Manitoba, Winnipeg, Canada
Verena H. Menec, Dept. of Community Health Sciences, University of Manitoba, Winnipeg MB R3E 0W3, Canada E-mail: vmenec{at}cpe.umanitoba.ca.
| Abstract |
|---|
|
|
|---|
Methods. Using administrative data, all hospital admissions and deaths due to respiratory illnesses (pneumonia and influenza, chronic lung disease, and acute respiratory diseases) were identified for all individuals aged 65 and older living in Winnipeg, Canada (approximately 88,000 individuals) during four influenza seasons (19951996 to 19981999).
Results. Hospitalization and death rates for respiratory illnesses increased significantly during influenza seasons, compared to fall periods (e.g., 42.7 vs 25.2 hospitalizations per 1000 population aged 80 and older). Moreover, hospitalization rates for pneumonia and influenza, chronic lung disease, and acute respiratory diseases were higher among individuals living in senior residences (42.5 per 1000 for all respiratory illnesses combined) than their counterparts living in the community (22.8 per 1000). Furthermore, deaths due to pneumonia and influenza and chronic lung disease were higher among senior housing residents (4.2 per 1000) than community residents (2.6 per 1000) and were particularly high among nursing home residents (52.1 per 1000).
Conclusions. Individuals living in seniors residences are at increased risk of being hospitalized for and dying of respiratory illnesses during influenza seasons. Given that influenza vaccination is currently the best method to reduce influenza-associated illnesses among seniors, this suggests that influenza vaccination strategies should be targeted at this population.
INFLUENZA is an important cause of morbidity and mortality. In Canada, 70,000 to 75,000 hospitalizations and approximately 6700 deaths each year are attributed to influenza and pneumonia (1). Influenza and pneumonia together are the sixth leading cause of death in Canada (2). The complications of influenza are primarily respiratory in nature and include pneumonia, acute respiratory diseases, and exacerbation of underlying medical conditions, such as chronic lung disease (3)(4)(5)(6)(7)(8)(9). Risk factors for complications of influenza are age and preexisting disease (3)(4)(5)(6)(7)(8)(9)(10)(11), with 80% to 90% of influenza-associated deaths occurring among people aged 65 and older (3)(4).
Influenza vaccination is currently the best method to attenuate the impact of influenza and reduce the risk of hospitalization and death among seniors aged 65 and older who live in the community, as well as among nursing home residents (9)(10)(11)(12)(13)(14)(15)(16)(17)(18). Influenza vaccination is therefore recommended for all individuals aged 65 and older in most Western countries (19). Within Canada's universal health care system, influenza vaccination is provided free of charge to this target group (20). Canadian guidelines also specifically identify nursing home residents (of any age) as candidates of publicly-funded immunization programs (20). Despite its proven effectiveness, however, vaccination coverage is still quite low in Canada among noninstitutionalized seniors, ranging from 30% to 60% (21)(22)(23)(24). Better coverage (80% on average) has been attained in nursing homes because of concerted efforts to vaccinate this target group (25).
From a public health perspective, identifying individuals who are at risk and can easily be targeted for influenza vaccination campaigns is important. Although underlying medical conditions, such as chronic lung and heart disease, increase the risk of influenza-related complications (11), individuals suffering from chronic diseases may not be easily reached, unless they visit their physician during influenza vaccination seasons. One group of seniors that could potentially be readily identified and targeted for influenza vaccination campaigns is those living in senior residencesapartments reserved for seniors who do not require nursing care but do require some assistance, such as the provision of meals.
Few studies have compared the impact of influenza on seniors living in different types of housing. Research indicates that influenza-attributable deaths are particularly high among nursing home residents as compared to seniors living in the community (26), presumably because of the presence of underlying disease as well as the increased likelihood of rapid person-to-person spreading of influenza viruses. There is a paucity of research, however, on the impact of influenza on individuals living in senior residences. Living in senior residences represents a step on a continuum ranging from living independently in the community to requiring care in a nursing home (27)(28). Senior housing residents would therefore tend to have more chronic medical conditions than their counterparts living in the community. Moreover, congregate living increases the likelihood of the spread of respiratory viruses. Compared to their counterparts living in the community, senior housing residents might consequently be more susceptible to influenza-related complications.
In sum, the purpose of the present study was to examine whether senior housing residents are at greater risk of influenza-related respiratory illnesses during influenza seasons than their counterparts in the community. Specifically, we compared hospitalizations and deaths due to respiratory conditions over four influenza seasons among seniors (aged 65+) living in the community, in senior residences, and in nursing homes.
| Methods |
|---|
|
|
|---|
Patient Characteristics
Age.--
Three age groups were created: 6574, 7579, and 80+ years old to differentiate between young-old, old-old, and oldest-old seniors. These age groups constituted about 53%, 21%, and 26% of the senior population, respectively.
Location of residence.-- Senior residences are apartment blocks specifically reserved for seniors who do not require nursing care beyond that provided through home care. Meals are provided in some senior residences. Senior residences were identified by means of a senior housing directory that provided a postal code for each senior residence in Winnipeg. Using postal code information contained in our administrative data (street addresses are not available in the dataset), we identified individuals living at these postal codes. Senior residences that shared postal codes with other (nonsenior residence) addresses had to be excluded, as we could not be sure where people lived. Overall, 111 of 130 (85%) senior residences listed in the senior housing directory were included in the present study. Thus, a small proportion of seniors who lived in a senior residence were misclassified as living in the community (<1% of community residents).
Nursing home residents were identified from administrative nursing home files. Access to nursing homes in Manitoba occurs on the basis of a standardized assessment based solely on need. Nursing home care is publicly insured (subject to a copayment assessed based on income). Community residents were defined as those individuals aged 65 and older who lived neither in a nursing home nor in a senior residence.
In 1998, 74,844 seniors (85%) lived in the community, 8723 (9.9%) lived in senior residences, and 4517 (5.1%) lived in nursing homes.
Comorbidity.-- The Refined Diagnostic Related Group classification system was used to examine whether community, senior housing, and nursing home residents who were hospitalized for respiratory conditions differed in terms of their level of comorbidity and complications. The system uses secondary diagnoses to classify cases into subgroups based on the presence or absence of substantial comorbidity or complications (31)(32).
Influenza Seasons and Comparison Period
Influenza seasons were defined as the weeks in each study year during which at least two positive tests for influenza A or B were obtained for Winnipeg residents at the provincial laboratory where all viral testing is conducted. Given this definition, influenza seasons lasted on average 15.3 weeks (range: 8 to 26 weeks). Influenza A viruses predominated in all 4 study years (33)(34)(35)(36).
Fall comparison periods were defined for each study year as a 3-month period ending prior to the influenza seasons (August to October).
Outcome Measures
Hospital admissions for respiratory conditions were examined for all of Winnipeg's acute care hospitals. They were identified among emergent/urgent medical patients (patients admitted through emergency), with medical cases identified using the Diagnostic Related Group classification system and emergent/urgent cases classified based on the admission status reported on the hospital discharge abstract. Among these emergent/urgent medical cases, patients were then identified who were admitted for respiratory conditions, using the most responsible diagnosis (the diagnosis that, upon discharge, was deemed most responsible for the patient's hospital stay). Deaths and cause of death were derived from Vital Statistics data for all Winnipeg seniors.
Diagnostic codes for respiratory conditions were chosen that have been shown to be complications of influenza in previous research (3)(4)(5)(6)(7)(8)(9)(13). They included: pneumonia and influenza (ICD-9-CM codes 480487), chronic lung disease (codes 490496), and acute respiratory diseases (codes 460466).
Analytic Approach
Hospital admission and mortality rates for respiratory conditions were determined for both influenza seasons and fall periods, and 95% confidence intervals were calculated using methods developed by Carriere and Roos (37). To allow meaningful comparisons between the influenza seasons and fall periods, all rates (and descriptive statistics) were annualized by, for example, multiplying hospital admissions during the 61 weeks of influenza season by 0.852. Age-specific crude rates (per 1000 age-specific population) were calculated for each of the three age groups (6574,7579, 80+). Rates (per 1000 population) for the different types of housing were age-gender standardized to the 1998 population aged 65 and older (using the direct method), to control for the different age and gender distributions among community, senior housing, and nursing home residents.
Two general comparisons were made: (i) hospitalization and death rates during the influenza seasons were compared to those during the fall periods, and (ii) hospitalization and death rates were compared among the three different locations of residence. The rate of a particular group that is outside the confidence interval of a given comparison group differs in a statistically significant way from the comparison group ( p < .05).
| Results |
|---|
|
|
|---|
2 = 19.6, p < .0001 for the influenza season). Moreover, nursing home and senior housing residents hospitalized for respiratory conditions were older than community residents (
2 = 139.4, p < .0001). Community residents, senior housing residents, and nursing home residents did not differ in terms of their level of comorbidity and complications.
|
|
Table 3 shows that admission rates (per 1000 population) were significantly higher ( p < .05) during influenza seasons than fall periods for all age groups for pneumonia and influenza, chronic lung disease, and acute respiratory diseases, although rates for acute respiratory diseases were low. Furthermore, admission rates were considerably higher among individuals aged 80 and older than among persons aged 7579, who had, in turn, higher rates of admission than the 6574-year-olds (during both the influenza seasons and the fall periods).
|
|
|
|
| Discussion |
|---|
|
|
|---|
Hospital admission rates were higher for seniors living in senior residences than their counterparts living in the community for pneumonia and influenza, chronic lung disease, and acute respiratory diseases. Indeed, rates of hospitalization for chronic lung disease were higher among senior housing residents than nursing home residents. Although deaths due to pneumonia and influenza and chronic lung disease were also elevated for senior housing residents, compared to community residents, rates were relatively small. Only 8% of deaths due to respiratory illnesses involved senior housing residents.
It is important to address several limitations of the present study at this point. First, although we know that influenza A (and to a much lesser extent influenza B) viruses were circulating during the periods that we identified as influenza seasons, increased hospital and mortality rates during these time periods may, in part, have been caused by other respiratory viruses, such as parainfluenza virus, respiratory syncytial virus, and adenovirus (38)(39). Given that the time periods during which these viruses are active typically overlap with influenza seasons, it is difficult to disentangle the relative impact of each virus.
Second, it was not possible to identify seniors' influenza vaccination status, as no systematic vaccination records were kept for adults in the province of Manitoba during the study years. Thus, it was not possible to examine hospitalizations and deaths for those vaccinated versus those not vaccinated. Third, we included only respiratory conditions in the present study, although research indicates that influenza can also exacerbate underlying chronic heart conditions (7)(8)(9)(13). The present study therefore likely underestimates the burden of serious morbidity during influenza seasons.
Previous research indicates that influenza vaccination is the best method to reduce hospitalizations and deaths due to influenza-associated illnesses during influenza seasons among seniors, including individuals in nursing homes (9)(10)(11)(12)(13)(14)(15)(16)(17)(18). Although Canadian guidelines recommend that all seniors aged 65 and older be vaccinated for influenza, with influenza vaccination provided free of charge, vaccination coverage has been low among noninstitutionalized seniors (about 40% to 50% in Manitoba) (21)(22). Thus, the relatively high rates of hospital admissions and deaths during influenza seasons among individuals living in senior residences, compared to their counterparts living in the community, found in the present study are unlikely to be due to differences in influenza vaccination rates. The findings suggest, therefore, that senior housing residents should be a priority for influenza vaccination campaigns.
Individuals living in senior residences are a readily identifiable population that could presumably be easily reached with mass influenza vaccination strategies, such as on-site clinics, with the potential benefits for reducing serious morbidity being substantial. Canadian guidelines do not currently specifically identify individuals in senior housing as a target group for influenza vaccination, although it is recommended that noninstitutionalized individuals aged 65 and older should be offered immunization by public health units in a variety of settings, including senior citizen centers and residences (1). The present findings clearly support this recommendation.
Similar to previous research (26), the present study also shows that hospital admissions and deaths due to pneumonia and influenza and chronic lung disease were higher among nursing home residents than individuals aged 65 and older living in the community. Given that influenza vaccination coverage is already high in nursing homes in Manitoba, this may reflect reduced vaccine efficacy in a population with a high prevalence of preexisting health problems, which would be consistent with previous research (40).
In sum, the present study adds to the literature by specifically focusing on individuals aged 65 and older living in senior residences and showing that they are at higher risk of being hospitalized for and dying of respiratory conditions during influenza seasons than their counterparts living in the community. Influenza vaccination campaigns targeted at this high-risk group could potentially reduce the burden of illness substantially.
| Acknowledgments |
|---|
Received January 3, 2002
Accepted March 11, 2002
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. E. Morley Editorial: Hot Topics in Geriatrics J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2003; 58(1): M30 - 36. [Full Text] [PDF] |
||||
![]() |
J. E. McElhaney Guest Editorial: Influenza: A Preventable Lethal Disease J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2002; 57(10): M627 - 628. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|