UBC Faculty Research and Publications

Influenza immunization in Canada’s low-income population Hobbs, Jennifer L; Buxton, Jane A Jul 21, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12889_2014_Article_6884.pdf [ 563.68kB ]
JSON: 52383-1.0223682.json
JSON-LD: 52383-1.0223682-ld.json
RDF/XML (Pretty): 52383-1.0223682-rdf.xml
RDF/JSON: 52383-1.0223682-rdf.json
Turtle: 52383-1.0223682-turtle.txt
N-Triples: 52383-1.0223682-rdf-ntriples.txt
Original Record: 52383-1.0223682-source.json
Full Text

Full Text

RESEARCH ARTICLEnmigoffers the best protection from infection and disease and at Immunization statement on seasonal influenza immunizationHobbs and Buxton BMC Public Health 2014, 14:740http://www.biomedcentral.com/1471-2458/14/740been well described [6-11]. In a recent study, older age,School of Population and Public Health, University of British Columbia,Vancouver, BC, Canadathe national level is recommended for all Canadians oversix months of age [1]. Delivery of influenza immunizationprograms occurs at the provincial level. Some publiclyfunded provincial programs are universal in scope andoffer the vaccine free of charge to all residents, while others[4]. In 2007 to 2008, national influenza immunization ratesreached 31% among Canadians 12 years of age and older [5].However, immunization coverage is not equal among allsegments of the Canadian population. The influence ofsocio-economic and demographic factors such as age, gen-der, chronic disease status, ethnicity, access to health care,education and income on influenza immunization have* Correspondence: Leigh.Hobbs@oahpp.calarly among high risk groups [1,2]. Influenza immunizationincreased risk of influenza-related complications. This analysis examines if the type of main source of householdincome in low-income groups affects influenza immunization uptake. We hypothesized that individuals on socialassistance have less access to immunization compared to those with employment earnings or seniors’ benefits.Methods: Data was obtained from the Canadian Community Health Survey annual component 2009-2010. A totalof 10,373 low-income respondents (<20,000$ Canadian per annum) were included. Logistic regression, stratifiedaccording to type of provincial publicly funded immunization program, was used to examine the associationbetween influenza immunization (in the last 12 months) and main source of household income (employmentearnings; social assistance as a combination of employment insurance or worker’s compensation or welfare; orseniors’ benefits).Results: Overall, 32.5% of respondents reported receiving influenza immunization. In multivariable analysis ofuniversal publicly funded influenza immunization programs, those reporting social assistance (AOR 1.24, 95% CI1.02-1.51) or seniors’ benefits (AOR 1.56, 95% CI 1.23-1.98) were more likely to be immunized compared to thosereporting employment earnings. Similar results were observed for high-risk programs.Conclusions: Among the low-income sample, overall influenza immunization coverage is low. Those receivingsocial assistance or seniors’ benefits may have been targeted due to higher rates of chronic disease. Programsreaching the workforce may be important to attain broader coverage. However, CCHS data was collected duringthe H1N1 pandemic influenza, thus results may not be generalizable to influenza immunization in non-pandemic years.Keywords: Immunization, Vaccination, Influenza, Flu, Socioeconomic status, Low-income, Social determinants, Health,CanadaBackgroundInfluenza infection poses a significant public health burdenaffecting millions of Canadians each year resulting in20,000 hospitalizations and 4,000 to 8,000 deaths, particu-are limited to high risk groups and offer the vaccine free ofcharge only to individuals at high risk of complicationsfrom infection and/or individuals providing care for thoseat high risk [3]. Further details describing high risk groupsare provided by the National Advisory Committee onInfluenza immunization ipopulationJennifer Leigh Hobbs* and Jane A BuxtonAbstractBackground: Immunization offers the best protection froin immunization uptake among low-income individuals. H© 2014 Hobbs and Buxton; licensee BioMed CCreative Commons Attribution License (http:/distribution, and reproduction in any mediumDomain Dedication waiver (http://creativecomarticle, unless otherwise stated.Open AccessCanada’s low-incomeinfluenza infection. Little evidence describes disparitiesher rates of chronic disease put this population atentral Ltd. This is an Open Access article distributed under the terms of the/creativecommons.org/licenses/by/4.0), which permits unrestricted use,, provided the original work is properly credited. The Creative Commons Publicmons.org/publicdomain/zero/1.0/) applies to the data made available in thisAccess to immunization programs may not be uniformHobbs and Buxton BMC Public Health 2014, 14:740 Page 2 of 8http://www.biomedcentral.com/1471-2458/14/740and barriers may be distinct from those described forthe general population and other sub-populations. Twostudies in the United States assessing rates of influenzaimmunization in disadvantaged urban areas found that ac-cess to social services, health services or health insurancewere important determinants [12,13]. However, findingslikely need to be interpreted with caution in the Canadiancontext due to differences in immunization program deliv-ery within public and private health care systems [8].This analysis aims to examine the influence of the sourceof household income (employment, social assistance, orseniors’ benefits) on influenza immunization among low-income individuals, defined as individuals reporting a totalannual household income of less than $20,000, in theCanadian population and whether this differs according tothe type of publicly funded program (universal or highrisk) available in the province. Source of income may serveas a surrogate measure to identify potentially marginalizedindividuals within the low-income group. Awareness of in-fluenza immunization programs may be through the work-place, seniors’ residence immunization clinics, and morefrequent physician visits with age. Therefore, individualson social assistance may have less access to and awarenessof immunization programs compared to those with em-ployment earning or on seniors’ benefits.MethodsStudy designData was obtained from the Public Use Microdata File ofthe Canadian Community Health Survey (CCHS) annualcomponent January 2009 to December 2010. The CCHS isa cross-sectional survey that collects information on thehealth status, health care utilization, and health determi-nants of the Canadian population. Survey data was col-lected using a multistage stratified cluster design from arepresentative sample of 124,188 individuals aged 12 orolder in all 10 provinces and 3 territories. The sampleis representative of approximately 98% of the Canadianhaving a chronic disease, and having a medical doctorwere associated with influenza immunization uptake [11].However, in Ontario, unlike other provinces, no relation-ship between immunization and age and income was re-ported suggesting more equitable access to vaccineperhaps due to the universal scope of the publicly fundedimmunization program [11].While factors influencing influenza immunization in cer-tain sub-populations have been examined, little evidenceexists describing disparities among low-income individuals[12]. Higher rates of chronic disease put this population atincreased risk of influenza-related complications [12,13].population. The 2% not represented includes individualsliving on Indian Reserves, Crown Lands, those residing ininstitutions, full-time members of the Canadian Forces,and residents of certain remote regions. Further detailspertaining to the CCHS sampling methodology are re-ported elsewhere [14].Study sampleThis analysis was restricted to individuals 15 years of ageor older reporting a total annual household income of lessthan $20,000, consistent with the low income cut off esti-mated by Statistics Canada (Figure 1) [15]. Individuals notstating annual household income were not eligible forthe study sample. Individuals reporting “other” (dividends,interest, child support, alimony, other, or no income) asthe main source of household income were excludedbased on conceptual considerations. Non-valid responses,including individuals who did not state their main sourceof household income, influenza immunization status, or aresponse for confounding variables, were also excluded.Study variablesThe outcome variable, receiving influenza immunization inthe last 12 months, was constructed from the CCHS ques-tions “Have you ever had a seasonal flu shot?” and “Whendid you have your last seasonal flu shot?”. A comparisonwas made between individuals who reported influenzaimmunization in the last 12 months and those who didnot report influenza immunization in the last 12 months.Individuals who did not report influenza immunization inthe last 12 months included those immunized one to twoyears ago, two or more years ago, and those reportingnever receiving an influenza immunization (Figure 1). Theprimary explanatory variable, main source of household in-come in the last 12 months, was listed in the CCHS asthree categories: employment earnings; social assistance asa combination of employment insurance or worker’s com-pensation or welfare; or seniors’ benefits. Inclusion of con-founding variables (age, gender, education, immigration,and self-perceived health) in the adjusted multivariable re-gression analysis was based on previous study findings andconceptual considerations [6-12]. Additional variables (notincluded in the multivariable regression analysis) used todescribe the sample included chronic disease (as a combin-ation of asthma, bronchitis, emphysema, chronic obstruct-ive pulmonary disease, diabetes, cancer, and heart disease),occupation group, and reasons for non-immunization (noflu shot). Reasons for non-immunization were captured inthe CCHS as a number of questions, including “Did notthink it was necessary” and “Have not gotten around to it”.AnalysisUnivariable analysis used logistic regression to determineunadjusted odds ratios and 95% confidence intervals meas-uring the association between influenza immunization andthe primary explanatory variable, source of householdHobbs and Buxton BMC Public Health 2014, 14:740 Page 3 of 8http://www.biomedcentral.com/1471-2458/14/740income, and potential confounders. Confounding variablesshowing an association in the direction expected, based ona priori knowledge and conceptual considerations, wereincluded in the multivariable model. A large proportion ofindividuals over 65 years of age reported seniors’ benefitsas the main source of household income, thus this rela-tionship was further explored to ensure assumptions re-quired for statistical tests were met. The multivariablemodel was developed with manual step-wise entry of eachconfounder. Analysis was stratified based on provincial in-fluenza immunization program type. Provinces and terri-tories with publicly funded universal programs (defined asprovinces offering the vaccine free of charge to all resi-dents), those with publicly funded high risk programs (de-fined as provinces offering the vaccine free of charge toindividuals at high risk for complications from infectionand/or individuals providing care for those at high risk),and those that implemented publicly funded universal pro-grams during the CCHS data collection period (referred toas recent universal programs) were considered separately.Sensitivity analyses compared individuals reporting in-fluenza immunization in the last 12 months and thoseFigure 1 Number of Canadian Community Health Survey respondentwho had never received an influenza immunization toinvestigate whether or not the group reporting never re-ceiving an influenza immunization was unique. Furthersensitivity analysis explored the relationship between influ-enza immunization and source of household income witha sample restricted to individuals of low-income and loweducation (excluded post-secondary graduates). Additionalvariables (not included in the multivariable regression ana-lysis) were used to provide descriptive results (propor-tions) where appropriate.Power was sufficient overall and for each stratum (atleast 0.92, at an alpha of 0.05), with the exception of thestratum including provinces with universal programs(where power was 0.54). Therefore, results from thisstratum should be interpreted with caution. In order toproduce estimates representative of the Canadian popula-tion, probability sampling weights were constructed forthe analytic sample from Statistics Canada survey weights.The constructed probability weights were applied in allanalyses to account for the sampling methodologies usedin the CCHS. All analyses were done using SAS 9.3 forWindows (SAS Institute).s in each stage of study sample selection.Hobbs and Buxton BMC Public Health 2014, 14:740 Page 4 of 8http://www.biomedcentral.com/1471-2458/14/740ResultsStudy sampleA large number of respondents (20,524 or 18.3% of thoseover 15) were excluded from the sample since annualhousehold income was not stated. Influenza immunizationcoverage was similar among the income not stated and in-come under $20,000 categories (OR 0.92, 95% CI 0.88-0.98). However, source of income differed among theincome not stated and income under $20,000 categories.Individuals with an income under $20,000 were morelikely to report social assistance as the main source ofhousehold income. Among individuals 15 years of ageor older who reported an annual household incomeunder $20,000, an additional 951 (11.9%) reporting “other”source of income were excluded. Influenza immunizationwas similar among respondents in the “other” category.An additional 922 respondents were excluded due to non-valid responses for influenza immunization status (203),main source of household income (352) and confounders;education (285), immigrant status (62), and self-perceivedhealth (20). Respondents excluded due to missing data didnot differ significantly on influenza immunization status.The final sample size consisted of 10,373 respondents,representing 8.4% of the original CCHS sample (Figure 1).Baseline characteristicsA large proportion of older individuals were represented inthe final analytic sample. The median age of respondentswas the 50 to 54 year age category, representing 7.5% ofthe sample (Table 1). More females (59.3%) and non-immigrants (74.7%) were included. Almost half (45.7%)reported post-secondary graduation; however, 30.4%also reported not completing secondary school. Few re-spondents (8.9%) reported poor self-perceived health.Outcome and exposureAmong the sample, 32.5% reported receiving an influ-enza immunization within the last 12 months. Of thosenot reporting an influenza immunization in the last 12months, 47% reported never receiving an influenzaimmunization. The main source of household incomewas fairly evenly distributed in the sample, with 34.7% ofrespondents reporting employment income as their mainsource of household income, 29.2% reporting social as-sistance, and 36.1% reporting seniors’ benefits. Amongthose reporting influenza immunization in the last 12months, the majority (58.2%) reported seniors’ benefits,while only 18.4% reported employment income.AnalysisIn the univariable models, stratified according to type ofpublicly funded influenza immunization program, theodds of influenza immunization was higher among indi-viduals reporting social assistance (defined as employmentinsurance or worker’s compensation or welfare) and sig-nificantly higher among individuals reporting seniors ben-efits compared to those reporting employment income(Table 2). In the multivariable models, stratified by type ofimmunization program and adjusted for age, gender, im-migration, education, and self-perceived health; the oddsof influenza immunization and the variability of the esti-mates were reduced but remained elevated for both indi-viduals reporting social assistance and seniors benefitscompared to those reporting employment earnings re-gardless of the type of provincial publicly funded program(Table 2). Adjusting for chronic disease did not alter thefindings, thus was not included as a confounder in thefinal model (results not shown).Among provinces with publicly funded universal in-fluenza immunization programs, the odds of influenzaimmunization were increased for individuals reporting so-cial assistance (AOR 1.24, 95% CI 1.02-1.51) and for indi-viduals reporting seniors’ benefits (AOR 1.56, 95% CI1.23-1.98) compared to individuals reporting employmentincome. Similar results were observed in provinces withrecent universal and high-risk programs. The odds ofinfluenza immunization was increased among provinceswith recent universal programs for individuals reportingsocial assistance (AOR 1.27, 95% CI 0.91-1.76) and se-niors’ benefits (AOR 1.90, 95% CI 1.29-2.79) and wasgreatest among provinces with high risk programs for in-dividuals reporting social assistance (AOR 1.49, 95% CI1.20-1.86) and seniors’ benefits (AOR 2.09, 95% CI 1.62-2.70) compared to individuals reporting employment in-come. All 95% CIs excluded ‘1’, with the exception of theCI for individuals reporting social assistance for provinceswith recent universal programs. Within each stratum, all95% CIs overlapped but did not include point estimates,with the exception of the CI for individuals reporting se-niors’ benefits for provinces with universal programs.For the confounding variables, the odds of influenzaimmunization increased with age and were higher amongfemales in both univariable and multivariable models. Theodds of influenza immunization were also higher for thosewith poor self-perceived health, with the exception of themultivariable model for high risk programs. In both uni-variable and multivariable models there did not appear tobe a strong relationship between influenza immunizationand immigration (confidence intervals included ‘1’). Therelationship between influenza immunization and educa-tion was also less clear, without a consistent trend in uni-variable and multivariable models.For the additional variables (not included in the multi-variable regression analysis), chronic disease was higheramong individuals reporting social assistance and se-niors’ benefits (29.7% and 52.3% respectively) comparedto those reporting employment earnings (18.0%). Investi-gation of reasons for non-immunization demonstratedbeerateHobbs and Buxton BMC Public Health 2014, 14:740 Page 5 of 8http://www.biomedcentral.com/1471-2458/14/740Table 1 Characteristics of low-income sample, relationshipinfluenza immunizationOvUnweighType of publicly funded influenza immunization programthat among those that did not report influenza immu-nization in the last 12 months, the most common rea-sons for non-immunization included “Did not think itwas necessary” and “Have not gotten around to it”. Fur-thermore, rates of immunization were lowest amongindividuals reporting occupations related to trades ortransport and equipment operator, and primary industryUniversal*High risk†Recent universal‡Influenza immunization (in last 12 months)YesNoOver 12 months agoNeverMain source of household incomeEmployment incomeSocial assistance§Seniors’ benefitsAge (5 year category)Median (50-54)GenderMaleFemaleImmigrantYesNoHighest level of household education< SecondarySecondary graduateSome post-secondaryPost-secondary graduateSelf-perceived healthPoorFairGoodVery goodExcellent*Provinces and territories with universal publicly funded programs included; Ontario†Provinces with publicly funded high risk programs included; Newfoundland and La(Prince Edward Island provides the influenza vaccine free of charge but charges a f‡Provinces that implemented a universal publicly funded program during the study§Employment insurance or worker’s compensation or welfare.Characteristics of low-income (<$20,000 per annum) Canadian Community Health Stween household income and confounders, andll study sample Influenza immunization (in last 12 months)d n (n = 10373) % Yes (%) No (%)or processing or manufacturing and utilities (resultsnot shown).DiscussionAmong the low-income population, it was hypothesizedthat individuals reporting social assistance as their mainsource of household income would have less access to3279 34.9 40.2 32.34841 49.5 41.9 53.22253 15.6 17.9 14.54312 32.56061 67.51910 20.34151 47.22189 34.7 18.4 42.62516 29.2 23.5 31.95668 36.1 58.2 25.5733 7.5 6.5 8.03565 40.7 31.4 45.16808 59.3 68.6 54.91419 25.2 25.3 25.28954 74.7 74.7 74.84235 30.4 39.1 26.11589 15.0 15.3 14.9753 9.0 6.8 10.03796 45.7 38.8 49.01073 8.9 11.6 7.62355 19.4 27.1 15.73374 31.7 32.3 31.42492 25.0 21.2 26.91079 15.0 7.9 18.5, Nunavut, the Northwest Territories and the Yukon.brador, New Brunswick, Quebec, British Columbia, and Prince Edward Islandee for vaccine administration for non high-risk groups).period included; Nova Scotia, Alberta, Saskatchewan, and Manitoba.urvey sample 2009-2010, limited to valid responses (weighted analysis).hUn1412Hobbs and Buxton BMC Public Health 2014, 14:740 Page 6 of 8http://www.biomedcentral.com/1471-2458/14/740Table 2 Odds of influenza immunization for main source ofprogram typeUniversal programUnadjusted OR(95% CI)Adjusted§ OR(95% CI)Source of incomeEmployment income Reference ReferenceSocial assistance* 1.26 (1.06-1.51) 1.24 (1.02-1.51)Seniors’ benefits 3.95 (3.33-4.70) 1.56 (1.23-1.98)Age (5 year increments†) 1.21 (1.19-1.24) 1.16 (1.13-1.19)GenderMale Reference ReferenceFemale 1.97 (1.71-2.28) 1.65 (1.42-1.93)Immigrantand awareness of influenza immunization programs com-pared to those reporting employment earnings or seniors’benefits. This association, however, does not appear to exist.Among a national sample of low-income individuals thosereporting social assistance and seniors’ benefits are morelikely to report receiving influenza immunization in the last12 months compared to those reporting employment in-come, regardless of the type of provincial publicly fundedprogram. The higher odds of influenza immunizationobserved for those receiving social assistance or seniors’benefits may be a result of the fact that these individualsrepresent high risk groups targeted by immunization pro-grams. Rates of chronic disease were higher among thosereporting social assistance and seniors’ benefits (29.7% and52.3% respectively) compared to the healthier workforce(18.0%). Higher rates of influenza immunization uptakeamong individuals with chronic conditions (and over 65Yes Reference ReferenceNo 0.94 (0.82-1.08) 1.07 (0.91-1.25) 1Highest level of householdeducation< Secondary Reference ReferenceSecondary graduate 0.66 (0.54-0.78) 0.98 (0.79-1.23) 0Some post-secondary 0.39 (0.30-0.51) 0.69 (0.51-0.92) 0Post-secondary graduate 0.66 (0.56-0.78) 1.10 (0.91-1.34) 0Self-perceived healthPoor 2.52 (1.88-3.38) 1.42 (1.03-1.95) 3Fair 3.33 (2.58-4.31) 2.17 (1.64-2.86) 3Good 1.79 (1.40-2.29) 1.21 (0.93-1.57) 2Very good 1.89 (1.47-2.43) 1.62 (1.24-2.12) 1Excellent Reference Reference*Employment insurance or worker’s compensation or welfare.†Odds of influenza immunization for a 5 year increase in age (up to 80 years).§Adjusted for age, gender, immigration, education, and self-perceived health.Unadjusted and adjusted multivariable logistic regression results for odds of influenhousehold income stratified by type of publicly funded influenza immunization prosample 2009-2010 (weighted analysis).ousehold income by influenza immunizationRecent universal program High risk programadjusted OR(95% CI)Adjusted§ OR(95% CI)Unadjusted OR(95% CI)Adjusted§ OR(95% CI)Reference Reference Reference Reference.67 (1.24-2.24) 1.27 (0.91-1.76) 2.54 (2.08-3.11) 1.49 (1.20-1.86).72 (3.69-6.04) 1.90 (1.29-2.79) 8.59 (7.16-10.31) 2.09 (1.62-2.70).20 (1.17-1.23) 1.12 (1.08-1.18) 1.32 (1.30-1.35) 1.21 (1.18-1.25)Reference Reference Reference Reference.18 (1.76-2.70) 1.97 (1.56-2.48) 1.53 (1.35-1.74) 1.07 (0.93-1.24)years of age) are supported by the literature [6,11,16]. Fur-thermore, since findings were consistent for provinces withboth high risk and universal programs, this suggests highrisk groups are being immunized regardless of the type ofprovincial publicly funded program. Public health cam-paigns appear to be effective at reaching those most at riskof influenza-related complications, however overall cover-age remains low.Findings were consistent with a study among disad-vantaged urban areas in the United States where access tosocial services was found to be an important determinantof influenza immunization [12]. No comparable studiesamong low-income individuals in the Canadian contextexist. However, Canadian studies have examined the influ-ence of socio-economic indicators on influenza immu-nization and have reported an increasing likelihood ofimmunization with greater income [11,16]. Thus, a lowerReference Reference Reference Reference.55 (1.16-2.08) 1.24 (0.89-1.72) 1.09 (0.94-1.28) 0.80 (0.67-0.95)Reference Reference Reference Reference.67 (0.48-0.94) 1.14 (0.79-1.65) 0.64 (0.53-0.78) 0.99 (0.79-1.22).68 (0.47-0.96) 1.61 (1.08-2.41) 0.37 (0.28-0.49) 0.67 (0.49-0.90).62 (0.49-0.78) 1.28 (0.98-1.68) 0.39 (0.34-0.45) 0.77 (0.66-0.91).45 (2.16-5.50) 1.85 (1.11-3.10) 4.45 (3.35-5.91) 2.71 (1.99-3.68).60 (2.40-5.40) 2.11 (1.35-3.29) 4.67 (3.69-5.91) 2.66 (2.06-3.45).56 (1.74-3.79) 1.70 (1.11-2.59) 2.84 (2.28-3.55) 1.88 (1.48-2.40).50 (0.99-2.26) 1.19 (0.77-1.84) 1.78 (1.40-2.26) 1.47 (1.13-1.91)Reference Reference Reference Referenceza immunization (yes in last 12 months) associated with main source ofgram, low-income (<$20,000 per annum) Canadian Community Health SurveyHobbs and Buxton BMC Public Health 2014, 14:740 Page 7 of 8http://www.biomedcentral.com/1471-2458/14/740overall rate of influenza immunization was expected in thestudy sample. However, the rate of influenza immunizationin the low-income study sample was similar to that of thegeneral population [16,5]. While this finding is not consist-ent with the influence of income reported in Canadianstudies, a recent review stated the influence of income oninfluenza immunization has been inconsistent [10].Since the low-income employed population appears tobe healthier (lower rates of chronic disease) than thosereceiving social assistance or seniors’ benefits, coveragerates may be lowest among this population due to lessperceived need for immunization, regardless of access.Investigation of reasons for non-immunization demon-strated that among those that did not report influenzaimmunization in the last 12 months, the most commonreasons for non-immunization included a lack of per-ceived need for immunization (“Did not think it was ne-cessary” and “Have not gotten around to it”) rather thanaccess to health care services. Similar findings have beenreported elsewhere [17].Several limitations arise due to the cross-sectional andself-reported nature of the data. Despite self-reporting,confounding variables previously described as determi-nants of influenza immunization showed associations inthe direction expected adding face validity to the mainfinding. Also, influenza immunization rates were similar tothose estimated for Canadian population suggesting influ-enza immunization status was reported accurately [5,16].However, there may have been a social desirability report-ing bias where high risk individuals over-reported influenzaimmunization. The use of self-reported data led to furtherlimitations in describing income earning and incomesource. It is likely that a number of low income individualsdid not state their earnings thus were excluded from theanalysis. Furthermore, low-income individuals may moveabove and below the low income cut off as their source ofincome changes. The income decile and income sourcereported may have differed from that at the time ofinfluenza immunization. Furthermore, CCHS data was col-lected during the H1N1 pandemic influenza, thus resultsmay not be generalizable to influenza immunization innon-pandemic years [10].Defining populations with higher rates of chronic dis-ease, thus increased risk of influenza-related complica-tions, is challenging. Increased risk of influenza-relatedcomplications is likely the result of a number of socio-economic factors, including income, education, and oc-cupation, but for the purposes of this study was limitedto income. However, findings were also consistent a sen-sitivity analysis that further restricted the sample to indi-viduals with low education (results not shown).The structure of the CCHS questionnaire also led tolimitations. The survey grouped individuals reportingvarious types of social assistance into a single category.However, immunization rates may vary with type of socialassistance. Low immunization rates have been reportedamong populations with a high number of individualson welfare, such as Vancouver’s Downtown East-side [18].Due to the CCHS categorization of social assistance lowerrates of immunization among individuals on welfare couldhave been masked by higher rates among individuals onemployment insurance or worker’s compensation. Further-more, as analysis was done with the Public Use MicrodataFile, probability sampling weights constructed from StaticsCanada survey weights, rather than bootstrapped weights,were used. Bootstrapped weights provide a more precisemeasure of the variability around the estimates.ConclusionsWhen implementing influenza immunization programs it isimportant to identify populations with low coverage andrecognize the factors that affect immunization uptake. Thisstudy was unique in investigating influenza immunizationamong the low-income population in Canada and demon-strated that while overall coverage among the low-incomepopulation is low, public health efforts appear to be reach-ing high risk individuals. However there is a need, particu-larly with respect to education and awareness, to improveimmunization coverage among the low-income workforce.Studies have demonstrated the public health benefits ofwidespread influenza immunization with reduced rates ofinfluenza-related mortality and morbidity [19]. The low-income employed population may be an important groupto reach to attain broader influenza immunization coverage,thus greater protection in the community, in both provinceswith universal and high risk publicly funded programs.To further describe barriers to influenza immunizationamong the low-income population, future studies shouldinvestigate influenza immunization among the workforceby identifying occupation groups with the lowest coverage.This study found that rates of immunization were lowestamong individuals reporting occupations related to tradesor transport and equipment operator, and primary indus-try or processing or manufacturing and utilities. Furtherdescription of influenza immunization among occupationgroups could be important in designing public health ef-forts. Studies should also investigate the association be-tween influenza immunization and income source usingfiner social assistance categories (in particular a separatecategory for individuals on welfare) to determine if the as-sociation observed in this study is consistent across differ-ent types of social assistance.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsJLH conceived of the study, performed the analysis, and drafted themanuscript. JAB contributed to the interpretation of the findings andmanuscript revisions. All authors read and approved the final manuscript.AcknowledgementsSubmit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionHobbs and Buxton BMC Public Health 2014, 14:740 Page 8 of 8http://www.biomedcentral.com/1471-2458/14/740We thank Mieke Koehoorn who provided methodological advice on behalfof the School of Population and Public Health, University of British Columbia.Received: 4 April 2014 Accepted: 10 July 2014Published: 21 July 2014References1. Public Health Agency Canada: The flu shot. http://www.fightflu.ca/fight-combattre-eng.php.2. Health Canada: It’s your health - influenza. http://www.hc-sc.gc.ca/hc-ps/dc-ma/influenza-eng.php.3. Public Health Agency Canada: Public funding for influenza vaccination byprovince/territory. http://www.phac-aspc.gc.ca/im/ptimprog-progimpt/fluvacc-eng.php.4. National Advisory Committee on Immunization: Statement of seasonalinfluenza vaccine for 2013-2014. Can Communicable Dis Rep 2013, 39:http://www.phac-aspc.gc.ca/publicat/ccdr-rmtc/13vol39/acs-dcc-4/index-eng.php.5. BC Centre for Disease Control: Canadian community health survey resultsinfluenza immunization coverage. http://www.bccdc.ca/NR/rdonlyres/A650CC1E-76F3-4827-91B2-3B65BECDB168/51128/CCHS20070820110601.pdf.6. Chen Y, Yi QL, Wu J, Li F: Chronic disease status, self-perceived healthand hospital admissions are important predictors for having a flu shot inCanada. Vaccine 2007, 25(42):7436–7440.7. Vozoris NT, Lougheed MD: Influenza vaccination among Canadians withchronic respiratory disease. Respir Med 2009, 103(1):50–58.8. Quach S, Hamid JS, Pereira JA, Heidebrecht CL, Deeks SL, Crowcroft NS,Quan SD, Brien S, Kwong JC: Ethnic disparities in influenza vaccination inCanada. Can Med Assoc J 2012, 24(15):1673–1681.9. Bish A, Yardley L, Nicoll A, Michie S: Factors associated with uptake ofvaccination against pandemic influenza: a systematic review. Vaccine2011, 29(38):6472–6484.10. Brien S, Kwong JC, Buckeridge DL: The determinants of 2009 pandemica/H1N1 influenza vaccination: a systematic review. Vaccine 2012,30(7):1255–1264.11. Polisena J, Chen Y, Manuel D: The proportion of influenza vaccination inOntario, Canada in 2007/2008 compared with other provinces.Vaccine 2012, 30(11):1981–1985.12. Bryant WK, Ompad DC, Sisco S, Blaney S, Glidden K, Phillips E, Vlahov D,Galea S: Determinants of influenza vaccination in hard-to-reach urbanpopulations. Prev Med 2006, 43(1):60–70.13. Armstrong K, Berlin M, Schwartz JS, Propert K, Ubel PA: Barriers to influenzaimmunization in a low-income urban population. Am J Prev Med 2001,20(1):21–25.14. Statistics Canada: Canadian Community Health Survey Annual component.2011. http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=3226&lang=en&db=imdb&adm=8&dis=2.15. Statistics Canada: Low Income Cut-Offs. 2012. http://www.statcan.gc.ca/pub/75f0002m/2009002/s2-eng.htm.16. Kwong JC, Rosella LC, Johansen H: Trends in influenza vaccination inCanada, 1996/1997 to 2005. Health Rep 2007, 18(4):1–11.17. Chambers CT, Buxton JA, Koehoorn M: Consultation with health careprofessionals and influenza immunization among women in contactwith young children. Can J Public Health 2010, 101(1):15–19.18. Weatherill SA, Buxton JA, Daly PC: Immunization programs in non-traditional settings. Can J Public Health 2004, 95(2):133–137.19. Kwong JC, Stukel TA, Lim J, McGeer AJ, Upshur RE, Johansen H, Sambell C,Thompson WW, Thiruchelvam D, Marra F, Svenson LW, Manuel DG: Theeffect of universal influenza immunization on mortality and health careuse. PLoS Med 2008, 5(10):e211.doi:10.1186/1471-2458-14-740Cite this article as: Hobbs and Buxton: Influenza immunization inCanada’s low-income population. BMC Public Health 2014 14:740.Submit your manuscript at www.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items