Kirk MD, Gregory JE, Lalor K, Hall GV, Becker N. Incidence of food- and waterborne infections in elderly people living in the community and Long-Term Care Facilities. Submitted. Emerg
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
About This Chapter
In this chapter, I analyse data on infections occurring in people ≥65 years old that doctors and laboratories reported to the Victorian health department over a ten-year period. The main reason to conduct this study was to examine whether the rates of infection were higher in people living in Long-Term Care Facility (LTCF) compared to elderly people living in the community, which contributes to answering the research questions: ‘Is the incidence of gastroenteritis and foodborne diseases higher in residents of LTCF compared to that of elderly people living in the community?’ and ‘What are the main causes of foodborne disease in elderly people living in LTCF and the community?’. The findings of this study are important to policy-makers, as funding agencies are often criticised that the incidence of disease is high amongst residents and reflects quality of care in the aged care system. This study found that residents were not at higher risk of diseases under study, except for Salmonella infections, particularly those associated with outbreaks. In addition, the reported incidence of
campylobacteriosis was lower in LTCF residents than elderly people living in the community. The paper has been submitted for publication to the journal Emerging Infectious Diseases.
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
Abstract
Elderly people may be vulnerable to food- and waterborne infections. We
calculated rates of infections reported to the Victorian health department for residents of Long Term Care Facilities (LTCF) and the community who were ≥65 years old. We used negative binomial regression to estimate incidence rate ratios, allowing for age, gender and reporting period. We analyzed 8277 infections in elderly people, finding rates of campylobacteriosis, legionellosis, listeriosis, toxigenic E. coli infections, and shigellosis to be higher in community residents, while Salmonella rates were higher in LTCF residents. Annually, 61.7 Campylobacter infections were reported per 100000 LTCF residents, compared to 97.6 per 100000 community residents (IRR 0.63, 95%CI 0.52–0.77). LTCF residents were at higher risk of S. Typhimurium (IRR 2.3, 95% CI 1.6–3.4) associated with outbreaks. 22.0% of listeriosis case-patients died. Apart from salmonellosis, LTCF residents experienced similar or lower rates of foodborne
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
118
Background
Infectious disease incidence varies with age, and elderly people are considered vulnerable to food- and waterborne infections (1-2). In many countries, elderly people unable to care for themselves live in Long-Term Care Facilities (LTCF) where they receive assistance with meals, daily living and healthcare (3). This may modify an individual’s risk of acquiring food- and waterborne infections due to different practices in food preparation and other exposures (4-5). Similarly, elderly people in the
community may have poor food safety practices leading to foodborne illness (6-8).
A variety of pathogens transmitted by food or water can infect people, including
Campylobacter spp., Clostridiumperfringens, Cryptosporidium spp., Legionella spp., and Shigella spp. and various serotypes of Salmonellaenterica (9-10). Food- and waterborne infections predominantly manifest in elderly people as gastroenteritis, but depending on the infectious agent may present as pneumonia, bacteremia and meningitis (11-12). Elderly people may become infected from ingestion of contaminated water or food, or through inhalation of contaminated aerosols in the case of Legionella spp. (13). Some infections are predominantly foodborne, while others may be acquired from infected persons or animals, or contact with contaminated environments (4).
These agents may manifest as outbreaks in facilities, leading to community concern about the safety of residents (14-15). While most outbreaks of gastroenteritis in LTCF are spread from person-to-person and are generally mild (16), foodborne
outbreaks result in higher case fatality ratios among residents (17). As a result,
regulatory agencies in many countries have mandated programs to manage food safety in facilities. To prevent legionellosis in residents, health agencies commonly provide advice about disinfection of hot water systems that may be reservoirs for Legionella
spp. (13).
Few studies have compared the incidence of infections potentially transmitted by contaminated food or water in elderly people living in LTCF and in the community. One study in the United States estimated that the lower limit of the death rate for nursing home residents from gastroenteritis of unknown etiology was 38.91 (95% CI,
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance 38.55–39.27) per 100,000 persons per year, compared to an estimated upper limit of 8.50 (95% CI, 8.47–8.53) per 100,000 persons aged 65 years or more living in the community (18). There has been little examination of the incidence among the institutionalized elderly of sporadic food- or waterborne diseases, as opposed to outbreak-associated disease.
In this study, we estimate rates of reported infection in people over 65 years old with seven different pathogens—Campylobacter, Cryptosporidium, Legionella,
Listeria, Salmonella, Shiga-toxin producing E. coli and Shigella—based on whether they lived in a government-subsidized LTCF or the community, and examine the effect of age on incidence of disease.
Methods
Infectious Disease Surveillance, Victoria
Victoria is one of six states and two territories in Australia. All doctors and pathology laboratories are required under public health legislation to report cases of notifiable conditions under the Public Health and Wellbeing Regulations 2009 (http://www.health.vic.gov.au/ideas/notifying/whatto) to the Health Department Victoria. Health Department staff members enter details about notified cases onto a database. Among the 64 conditions notifiable as at 14 April 2010, 14 were enteric diseases potentially transmitted by contaminated food or water, and one was Legionella
infection, that could potentially be transmitted by inhalation of contaminated water. Surveillance for these diseases has remained essentially unchanged in Victoria since the early 1990’s, except for cryptosporidiosis where reporting was voluntary until 2001 when notification became mandatory by law.
For this study, we analysed data on all cases of campylobacteriosis,
cryptosporidiosis, legionellosis, listeriosis, shigellosis, salmonellosis, and Shiga toxin producing E. coli (STEC) infection notified to the Victorian Department of Health in a ten-year period between 1 January 2000 and 31 December 2009.
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
120
Surveillance officers recorded whether cases were part of an outbreak or had travelled overseas. In addition, we reviewed surveillance data for all LTCF residents to identify where more than two cases of the same pathogen occurred within the same facility within two weeks that were not recorded by surveillance officers, and recoded these cases as ‘outbreak-associated’. Surveillance officers also recorded whether cases had died due to the disease, or due to other co-morbid conditions.
Data Analysis
We categorized notified cases by residential status, where a ‘LTCF resident’ was someone who had a residential address of a government subsidized LTCF, and a
‘community resident’ was someone living in a private residence in Victoria (see Appendix 1). We excluded cases without a valid address from analysis, as we assumed they were missing completely at random. We also excluded cases resident in privately funded facilities (supported residential services) catering for elderly, disabled people or those with psychiatric illness or dementia, as they were not included in the denominator of LTCF residents. We counted the annual number of food- and waterborne infections, including those due to epidemiologically important serotypes and species, in LTCF and community residents in the ten-year period. Age-specific annual denominator data for residents of government-subsidized LTCF was obtained from annual reports prepared by the Australian Institute of Health and Welfare (see:
http://www.aihw.gov.au/agedcare/index.cfm accessed on 4 August 2010) (19). Denominator data for community residents was calculated by subtracting annual age- specific estimates of LTCF residents from estimated resident populations prepared by the Australian Bureau of Statistics (see: http://www.abs.gov.au/ accessed on 4 August 2010). We calculated annual rates of notification for different diseases by residential status and compared this to the total rate of notifications for the State. We calculated case fatality rates for each disease by dividing the number of deaths due to the disease in the different groups by the number of cases where the mortality status was known for the ten-year period. To account for age differences in LTCF and community residents, we calculated age-adjusted relative risks of death due to infection with different pathogens using Mantel-Haenszel methods.
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance To estimate Incidence Rate Ratios (IRR), we used a negative binomial
regression model of the annual count of food- and waterborne infections by the period of notification (1999–2004 & 2005–2009), gender, three categories of age (65–74 years old, 75–84 years old & ≥85 years old), and residence (LTCF & community). The
number of people living in LTCF or the community in each age-group for each year was entered into the model as an offset. We used robust variance estimation suited to
longitudinal or clustered data to account for potential clustering due to outbreaks. We assessed model fit by examining the distribution of standardized Pearson residuals. To assess the effect on incidence, we repeated regression models excluding travel
associated cases and including only the first case for each known outbreak in LTCF and community settings.
The Australian National University human research ethics committee approved this study and the Victorian Department of Health approved release of the data. We analysed data using Stata Version 11.2.
Results
Incidence in ≥65 year olds
During the ten years between 1 January 2000 and 31 December 2009, there were 8534 cases of the seven diseases reported in people aged over 65 years. During data cleaning, we excluded 238 cases without a valid residential address and 19 cases living in a private institution caring for elderly or disabled persons.
In total, there were 8277 cases available for analysis, including 1.6% (132/8277) of cases living in retirement villages that were coded as ‘community residents’.
Infections in institutional and community-dwelling elderly
We observed similar or lower rates of notified infections in LTCF residents compared to community residents for all infections, apart from those due to Salmonella
(Table 1). There were no LTCF residents notified with infections due to L. longbeachae
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
122
with L. longbeachae (Crude Relative Risk 1.15, 95% Confidence Interval 1.1–1.2). The notification rate of Campylobacter spp. was lower in LTCF residents compared to community residents for all years of surveillance (Figure 1). In contrast, notifications of
S. Typhimurium displayed peaks due to occurrence of outbreaks during the study period, resulting in higher rates in LTCF residents overall (Figure 2).
Associated mortality
The case fatality rate was highest for listeriosis and L. pneumophila infection, where 22.0% and 9.1% of cases died, respectively (Table 2). The age-adjusted relative risk of death for elderly people infected with L. monocytogenes was 2.9 (95%CI 1.0– 8.6) if they lived in an LTCF compared to those living in the community. No deaths due to L. pneumophila occurred in LTCF residents. For salmonellosis, the majority of deaths occurred in case patients where death was considered to be due to other co-morbid conditions. The age-adjusted relative risk of death for elderly people infected with any
Salmonella serotype was 2.6 (0.8–8.4) if they were an LTCF resident compared to a community resident. Death status was not routinely ascertained for cases of
campylobacteriosis, although one death was recorded due to the disease along with 11 others due to co-morbid conditions. There were no deaths recorded for shigellosis or cryptosporidiosis.
Rates of food- and waterborne infections
In multivariable analysis, the incidence of S. Typhimurium was higher in LTCF residents than in community residents (Incidence Rate Ratio 2.3, 95%CI 1.6–3.4) and non-Typhimurium serotypes of Salmonella (IRR 1.4, 95%CI 1.0–1.9), and lower for
Campylobacter spp. (IRR 0.63, 95%CI 0.52–0.77) (Table 3). Incidence rate ratios could not be estimated for L. longbeachae, STECand shigellosis, as no cases occurred in LTCF residents during the surveillance period.
We observed a trend of increasing rate of reported infections for cryptosporidiosis, salmonellosis, and Campylobacter spp. over time during the surveillance period.
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance Figure 1. Notification rates for campylobacteriosis in people over the age of 65 years, by LTCF residence status, 2000–2009, Victoria, Australia.
0 20 40 60 80 100 120 140 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 R at e ( pe r 100, 000 p er so ns ) Year of Notification
Community residents LTCF residents
Figure 2. Notification rates for Salmonella Typhimurium infections in people over the age of 65 years, by LTCF residence status, 2000–2009, Victoria, Australia.
0 10 20 30 40 50 60 70 80 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 R at e (p er 100, 000 p er so ns ) Year of Notification
124
Table 1. Numbers and annual rate per 100,000 persons for potentially food- and waterborne diseases reported to the Victorian health department, by age group and residence status, Victoria, Australia, 1 January 2000–31 December 2009.
Infection People < 65 years old People ≥65 years old Total Notifications
Long-Term Care Facility
Residents
Community Residents Cases missing address or excluded facility Reported cases Rate* Reported cases Rate * Reported cases Rate* Reported cases Rate* Campylobacter sp. 50444 115.4 215 61.7 6207 97.6 206 57072 113.2 Cryptosporidium sp. 4955 11.3 7 2.0 106 1.7 3 5071 10.1 L. pneumophila/other 457 1.0 8 2.3 293 4.6 4 762 1.5 L. longbeacheae 49 0.1 0 0.0 45 0.7 0 94 0.2 Listeria monocytogenes† 46 0.1 3 0.9 70 1.1 4 123 0.2 Salmonella Typhimurium 7204 16.5 87 25.0 585 9.2 19 7895 15.7
Salmonella other serotypes 5003 11.4 44 12.6 552 8.7 20 5619 11.1
Shiga Toxin E. coli 56 0.1 0 0.0 12 0.2 1 69 0.1
Shigella sp. 845 1.9 0 0.0 43 0.7 0 888 1.8
*
—Annual rate of notified infections expressed per 100,000 persons.
†
125
Table 2. Deaths in people aged over 65 years resident in Long-Term Care Facilities and the community due to food- and waterborne infections or other co-morbid illness in Victoria, Australia, 1 January 2000–31 December 2009.
Infection Died from disease Died from co-morbid illness Unknown mortality status Number Cases Case Fatality Rate (per 100 cases)*
L. pneumophila/other 11 1 169 301 9.1
L. longbeacheae 2 1 0 45 6.7
Listeria monocytogenes 4 9 14 73 22.0
Salmonella Typhimurium 3 8 183 672 2.2
Salmonella other serotypes 0 7 177 596 1.7
Shiga Toxin E. coli infection 1 0 1 12 9.1
Shigella sp. 0 0 10 43 -
*
—Case fatality rate for the 10-year study period calculated by dividing the total number of deaths among cases with the disease over the number of cases with known mortality status.
126
Table 3. Adjusted Incidence rate ratios from multivariable model for reported cases of diseases potentially transmitted by
contaminated food or water in people aged over 65 years, gender, by time period, age group and whether the case was a Long-Term Care Facility (LTCF) or community resident, Victoria, Australia, 1 January 2000–31 December 2009.
Variable Incidence Rate Ratio (95%CI)
Cryptosporidium Listeria monocytogenes Salmonella Typhimurium Salmonella other serotypes
L. pneumophila/other Campylobacter sp.
(n=113) (n=73) (n=662) (n=586) (n=301)* (n=6387)
Gender Female reference reference reference reference reference reference
Male 0.86 (0.51–1.4) 1.5 (0.90–2.6) 0.97 (0.80–1.2) 0.97 (0.80–1.2) 2.6 (1.7–4.1) 1.2 (1.1–1.3)
Year 2000–4 reference reference reference reference reference reference
2005–9 3.7 (2.1–6.6) 1.2 (0.68–1.9) 1.8 (1.4–2.2) 1.7 (1.4–2.1) 0.35 (0.23–0.53) 1.2 (1.1–1.3)
Age-group (years)
65-74 reference reference reference reference reference reference
75-84 0.92 (0.53–1.6) 2.0 (1.1–3.5) 1.1 (0.95–1.4) 1.1 (0.86–1.3) 1.5 (0.90–2.5) 1.1 (1.0–1.2)
85+ 0.58 (0.31–1.1) 2.8 (1.2–6.4) 1.3 (0.99–1.8) 1.1 (0.80–1.4) 1.2 (0.63–2.2) 1.1 (1.0–1.2)
LTCF resident
No reference reference reference reference reference reference
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance From multivariable analysis, reported incidence rates were higher in older age categories for Listeria monocytogenes, Campylobacter sp.and S. Typhimurium infections when compared to the base age group category of 65–74 year old persons (Table 3). We did not observe a significant difference in incidence by age category for people infected with L. pneumophila, non-Typhimurium strains of Salmonella, and cryptosporidiosis.
Accounting for travel & outbreaks
During their incubation period, 1.3% (105/8227) of notified cases travelled internationally, all of whom were community residents. The proportion of community residents who had travelled ranged from 37% (16/43) for shigellosis and 8.7% (48/552) for non-Typhimurium serotypes of Salmonella, through to 0% for listeriosis, L.
longbeachae infections and STEC infections. Only 0.6% (35/6,207) of community- dwelling elderly infected with Campylobacter sp. were recorded as travelling overseas prior to infection.
During the study period, there were 42 separate outbreaks of S. Typhimurium affecting people ≥65 years old, 36 of non-Typhimurium serotypes of Salmonella, 14 of
L. pneumophila/other, 10 of Campylobacter spp. infections, and two of Shigella. In total, 2.4% (189/7913) of community resident cases were recorded as associated with outbreaks. In LTCF residents, 31% (111/364) of all cases were associated with outbreaks, including 78% (68/87) of S. Typhimurium and 15% (33/215) of
Campylobacter spp. infections. No cases of cryptosporidiosis, listeriosis, or STEC infection were recorded as outbreak-associated in either LTCF residents or community residents.
When we repeated multivariable models excluding travel-associated infections and including a single case for each identified outbreak, the incidence rate of S.
Typhimurium was similar in LTCF residents compared to community residents (IRR 0.91, 95%CI 0.64–1.29). For infections with non-Typhimurium serotypes of
Salmonella, the incidence was higher in LTCF residents than those living in the community (IRR 1.4, 95%CI 1.0–2.0). After adjusting for travel and outbreaks, the
Paper 6: Incidence of food- and waterborne infections from Victorian public health surveillance
128
incidence rate was lower for LTCF residents infected with Campylobacter spp. (IRR 0.57, 95%CI 0.48–0.68). For L. pneumophila,the incidence rate was lower for LTCF residents, but was not statistically significant (IRR 0.63, 95%CI 0.29–1.3).