Vaccine coverage amongst adult Canadians: Results from the 2012 adult National Immunization Coverage (aNIC) survey

Introduction

Immunization is a major accomplishment for public health, effectively preventing infectious disease and improving the health of Canadians. Some people believe that immunizations are just for babies and children. This is not true - immunizations are also important for adults for two main reasons:

• to prevent certain vaccine-preventable diseases and to reduce the risk of morbidity and mortality associated with these diseases, and
• to protect vulnerable people who cannot be immunized, are not yet fully immunized or do not build strong immunity to vaccines.

Since 2001, the Public Health Agency of Canada (the Agency) routinely monitors national immunization coverage for select adult immunizations through the adult National Immunization Coverage Survey (aNICS). Results from these surveys are used to monitor progress towards national targets for immunization coverage, to report immunization coverage estimates to international organizations, to improve planning for pandemic influenza and to develop appropriately targeted public education strategies. Immunization coverage for the seasonal influenza, pneumococcal, hepatitis B, tetanus, acellular pertussis and varicella vaccines is measured in the general (non-institutionalized) adult population (18 years of age and older), and in specific target groups, such as seniors (65 years of age and older), adults (18 to 64 years of age) with a chronic medical condition who may be at higher risk of complications from vaccine-preventable diseases as well as health care workers, who are at a higher risk of exposure to vaccine-preventable diseases as identified by the National Advisory Committee on Immunization (NACI) ^{Footnote 1}. Additional questions are included in the survey questionnaire to assess knowledge, attitudes and behaviours (KAB) surrounding immunization as well as reasons for missed opportunities for immunization.

The aNICS is planned to be biennial. A recent cycle of the survey was conducted in 2012, and prior to that in 2001 ^{Footnote 2}, 2006, 2008 and 2010. This report details the survey results from the 2012 cycle and presents a comparison of coverage estimates from previous surveys. Caution should be exercised when viewing coverage trends over time in this report, as survey methods have evolved since 2001. This methodology is explained in the next section and study limitations are highlighted further in the report.

Methodology

Questionnaire Development

Vaccines and target groups included in the 2012 aNICS were prioritised in consultation with immunization experts from across Canada and the questionnaire for 2012 was adapted from the 2006, 2008 and 2010 aNICS questionnaires. Differences since the 2006 iteration of the questionnaire include the addition of questions on immunization history as well as KAB questions for hepatitis B, tetanus, acellular pertussis and varicella vaccines. The questionnaire also includes demographic questions pertaining to age, gender, education, employment status, household income, and country of birth. In order to answer questions in the survey, respondents were asked to recall their immunization history from memory.

Sampling

The 2012 aNICS was conducted by a contractor between August 27 and September 26, 2012. Respondents from every province and territory were selected using random-digit-dialing (RDD) and sampling was stratified by province/territory and community size. The sample was then weighted to be nationally representative using data from the 2006 Canadian Census ^{Footnote 3}. In total, 3,005 respondents completed the interview by telephone (2,755 in the general population and 250 in a health care worker oversample. Oversampling for the health care worker group is necessary because the number of health care workers obtained while sampling in the general population would be too small to yield coverage estimates with an acceptable margin of error. This is due largely to the fact that health care workers represent only a small percentage of the population. As a result, a subsample of previously identified health care workers (lists of health care professionals or health care organizations through a random selection process) was used as respondents in addition to those respondents who are health care workers and are part of the general population. Physicians and nurses were offered a monetary incentive to participate in the survey. The total cohort of health care workers based on the RDD recruitment process (n=338) and oversampling (n=250) was 588. The response rate for the general population survey was 9% and 12% for the health care worker oversample.

As in past iterations of the survey, provinces and territories were invited to purchase an additional sample (oversample) to obtain adequate power for provincial/territorial estimates. While none of the provinces and territories participated in oversampling in this cycle of the survey, this practice provides provinces and territories the opportunity to use the survey tools already developed by the Agency to estimate coverage at national level. It also provides P/T to determine coverage estimates at a provincial/territorial level. For instance, if a certain province chooses to purchase an oversample, it can buy into the survey, interviewers will conduct more interviews in that province so that the sample size in that province will be large enough that provincial coverage estimates can be calculated.

Respondents were screened during the telephone interview and categorized within the NACI-recommended target groups: non-institutionalized adult Canadians 65 years and over (seniors), non-institutionalized Canadians 18 to 64 years with chronic medical conditions, and health care workers. For the purposes of the survey, and consistent with NACI definitions, a chronic medical condition is defined as a heart condition, asthma, other chronic lung conditions, cancer, diabetes, liver cirrhosis, chronic kidney disease, immune disorder/suppression or a stroke. Chronic conditions related to the onset of pneumococcal infection also include asplenic or functionally asplenic individuals as well as the previously mentioned conditions with the exception of asthma. Approximately 400 individuals were recruited in each target group, except for the health care worker target group which required additional recruitment (oversampling) to create a larger sample size. These additional health care workers were randomly selected from lists of health care organizations for a total sample of 588 (388 from RDD process and an additional 250 from the oversampling process). Respondents not screened into a target group were included in the general adult population group.

The margin of error for the adult general population is approximately +/- 1.8%. For target groups, excluding health care workers, the margin of error is larger at +/- 5% based on a smaller sample of approximately 400 per target group. Due to the larger group size, the margin of error for health care workers is +/- 4.0%.

Statistical Analysis

While NACI recommendations are used to estimate immunization coverage, differences in delivery of programs at the jurisdictional level are considered in the analysis as required. Preliminary data analysis was conducted by the contractor and further analyses were performed by the Agency's Centre for Immunization and Respiratory Infectious Diseases (CIRID) using SPSS and SAS. SPSS version 15.0 was used to generate frequencies for each variable. Cross-tabulations of immunization status for all vaccines were produced by demographic and other variables of interest such as risk factors. National estimates for the general population were weighted by age, gender and provincial/territorial population estimates using information from the 2006 Census ^{Footnote 3}. National data for health care workers were weighted by age and region. Comparisons of 2001, 2006, 2008, 2010 and 2012 immunization coverage estimates were performed when available using student t-tests and chi-squared tests. Unless otherwise stated, only differences significant at the 95% confidence level (p<0.05) are discussed in this report. Regression analysis was conducted to examine predictors of seasonal influenza and pneumococcal polysaccharide vaccine uptake in 2001, 2006, 2008, 2010 and 2012. Crude odds ratios were determined using univariate analysis and all significant variables were included into an adjusted logistic regression model. In addition, any other variables deemed to have an important effect on vaccine uptake were also included in the adjusted model. Results from the analysis are discussed below.

Results

Bolded figures indicate a statistically significant difference with the previous survey cycle.

Figure 1: Estimated vaccine coverage among the general adult population (18-64 years) in Canada from 2001 to 2012 for the seasonal influenza, hepatitis B and HPV vaccines

Text Equivalent - Figure 1

$ The 2001 survey used a different methodology than that used since 2006 so it is difficult to compare the 2001 results with those obtained from 2006 onwards. The survey methodology used a predetermined sampling frame in 2001 and switched to random digit dialing in 2006.

* The HPV vaccine coverage estimates are not available prior to 2010.

# 2009 H1N1 outbreak led to decrease in seasonal vaccine uptake.

i) Seasonal influenza immunization

NACI specifically recommends seasonal influenza immunization for adults at high-risk of influenza-related complications, those capable of transmitting influenza to individuals at high-risk of complications, and those who provide essential community services ^{Footnote 5}. All provinces and territories offer publicly-funded seasonal influenza immunization programs to these high risk target groups. In addition, NACI recommends that all healthy adults be encouraged to receive the seasonal influenza vaccine. The following analysis reflects jurisdictional differences in seasonal influenza vaccine program delivery.

For seasonal influenza vaccine uptake, while there have been increases in the estimated coverage in health care workers (14% since 2001), the recommended national immunization coverage targets for 80% seasonal influenza immunization coverage by 2010 (as determined at a national consensus conference) for all target groups has yet to be achieved ^{Footnote 4}.

The proportion of respondents in the general population who reported receiving one dose of seasonal influenza vaccine in the year prior to the 2012 survey is 37.2% (see Table 1). This proportion which is similar to that observed in 2008 (35.8%) indicates that a return to pre-pandemic levels of influenza vaccine uptake has occurred in the general population. Among other NACI recommended target groups, seasonal influenza immunization coverage has remained relatively similar to that observed in 2008. It remains highest in seniors at 65.0% while, compared to other target groups, adults 18-64 years of age with a chronic medical condition have the lowest immunization coverage of the three target groups at 37.7%.

Predictors of seasonal influenza immunization

Factors which increased the probability of receiving an immunization for seasonal influenza include: age, occupation, having a high-risk medical condition, and receiving a recommendation from a health care professional to get immunized. Increasing age was associated with greater likelihood of having been immunized with the seasonal influenza vaccine within the past year. Due to recommendations for health care workers and volunteers to receive the seasonal influenza vaccine, respondents who were health workers were more likely to be immunized with the seasonal influenza vaccine then adults with other occupations. Individuals with high-risk medical conditions were more likely to report having been immunized against seasonal influenza and this is likely due to recommendations for high-risk individuals and possibly due to the fact that individuals with high-risk medical conditions may have increased access to the vaccine. The strongest predictor of obtaining a seasonal influenza vaccine among the general population was receiving a recommendation for the vaccine from a health care professional during a clinical visit. Interestingly, residing in a province or territory with a universal seasonal influenza immunization program was not a significant predictor of seasonal influenza immunization when adjustments for other confounders were made (e.g. education, age, occupation, etc.). Also, higher levels of education and higher household income were not associated with immunization for seasonal influenza. These predictors of seasonal influenza uptake in 2012 were similar to those observed in 2006, 2008 and 2010.

ii) Pneumococcal polysaccharide immunization (Pneu-P-23)

NACI recommends one dose of pneumococcal polysaccharide (Pneu-P-23) vaccine for seniors and those 2 years of age and older with an underlying chronic medical condition (e.g. sickle cell disease, multiple myeloma, etc.) ^{Footnote 8}. The Pneu-P-23 vaccine is publicly-funded in all provinces and territories for both of these target groups.

Pneumococcal immunization coverage remained stable for seniors aged 65 years and over between 2006 (38.2%) and 2012 (38%). In adults aged 18 to 64 years with a chronic medical condition, pneumococcal immunization coverage also remained stable between 2010 (15.1%) and 2012 (19%). Pneumococcal immunization coverage estimates are summarised in Table 1.

Predictors of Pneu-P-23 immunization

Receiving a recommendation from a health care professional significantly increased the probability of being immunized with Pneu-P-23. Also, age (65 years and older) and higher levels of education (completed university or more) are characteristics that were predictive of receiving a pneumococcal vaccine.

iii) Hepatitis B immunization in health care workers

The hepatitis B vaccine is recommended by NACI for all health care workers who could potentially be in contact with blood or blood products, those at increased risk of infection through lifestyle exposures (e.g. travel, high risk behaviours), those with certain chronic medical conditions such as hemophilias and others conditions that could lead to repeated infusions of blood or blood products, or conditions requiring hemodialysis, those with chronic liver disease or who are taking hepatotoxic drugs, persons infected with hepatitis C, people who have undergone hematopoietic stem cell transplantation, and people with other conditions for which hepatotoxic medications are likely to be prescribed in the future ^{Footnote 1}. Two or more hepatitis B doses are generally required, depending on the specific vaccine product administered and whether an adequate immune response was previously achieved ^{Footnote 2}. Routine hepatitis B immunization programs were introduced in some provinces and territories in the mid-1980s, either as infant or school-based programs ^{Footnote 7}. In addition, not all jurisdictions fund or recommend hepatitis B vaccination for health care workers. Thus, the heterogeneity of hepatitis B immunization programs in Canada limits the comparability of national estimates with jurisdiction-specific hepatitis B immunization coverage estimates.

In 2012, national hepatitis B immunization coverage was estimated at 39.7% in the general adult population. A similar proportion, 41.4%, of adults 18 to 64 yrs of age with a chronic medical condition were also immunized against hepatitis B. Approximately 64.9% of health care workers in close contact with patients received the hepatitis B vaccine by 2012, which is higher than the hepatitis B immunization coverage levels observed in 2001 when hepatitis B coverage was estimated to be 53.9% in this group.

Predictors of hepatitis B immunization

Hepatitis B immunization coverage was significantly more likely among adults 18 to 44 years of age when compared to older age groups (45 to 64 years, 65+ years). Also, higher levels of education and occupation as a health care worker or volunteer were associated with a higher likelihood of having received a vaccination for hepatitis B.

iv) Tetanus, acellular pertussis and varicella immunizations

The adult formulation of acellular pertussis is given in combination with tetanus and diphtheria as a trivalent vaccine (Tdap) in Canada. This trivalent vaccine has been in use in Canada since 1999, and while it is currently available in all provinces and territories for adolescents, its use is not yet publicly-funded for adults in every province and territory. Therefore, it was not unexpected that the baseline immunization coverage estimate for Tdap remained low at 6.7%, this has increased over time from 3.9% in 2006.

As in previous cycles of the survey, respondents in the 2012 survey were asked if they had been immunized against tetanus in the past 10 years, as per NACI recommendations ^{Footnote 1}. Tetanus immunization coverage in the general population is estimated to be 49.7%. Tetanus immunization coverage is 78.2% for those who reported being treated for a wound in the past 10 years. (See Table 2)

For varicella, among adults in the general population without prior immunity (9.1% of the general population), immunization coverage was estimated at 24.9%, 25.8% in adults with a chronic medical condition, and at 42.9% for health care workers. Among health care workers, only 32.0% had been tested for immunity to varicella prior to immunization; however interpretation of these results is limited due to small sample size.

v) Human papillomavirus (HPV) immunization

Human papillomavirus (HPV) is one of the most common sexually transmitted infections and a persistent infection may lead to the development of cervical cancer. Immunization against HPV can protect against most sexually transmitted HPV infections. NACI recommends vaccination to prevent infections from the most common types of HPV for females before the onset of sexual intercourse and for females 14 to 26 years of age who may already be sexually active, have had previous Pap abnormalities or have had a previous HPV infection. These recommendations were made in 2007. In 2012, NACI also recommended the vaccine to older women up to age 45 years and to boys and men aged 9 to 26 years. In 2012, immunization programs focused on delivering the HPV vaccine to school age girls; adults and older females wishing to obtain the HPV vaccine had to pay out-of-pocket to be immunized as it was not covered in these individuals as part of publicly-funded programs.

In the 2012 aNICS, participants aged 30 years and under were asked if they had ever been vaccinated against HPV. Approximately 12.2% of adults reported that they had received an HPV immunization.

vi) Knowledge, attitudes, and behaviours

Source of information about immunization

Approximately fifty-one percent of health care workers report having the opportunity to provide information on vaccines to their clients. Doctors and nurses provide more information about immunization than any other health care worker. Moreover, most of the information provided by health care workers is about preventative care including vaccinations (75.6%), and information on the benefits and safety of vaccines (37.3%). The most common immunization questions asked to health care workers relate to vaccine safety, side effects or risks of complications. These questions constitute nearly half (45.5%) of all questions asked of health care workers regarding vaccines.

Many Canadians (65%) prefer to obtain vaccination information from a health care worker. Media, internet and publications are the preferred sources of information on vaccines for 28% of adults.

Discussion

Results from the 2012 aNICS show that immunization coverage measured in the general adult Canadian population, and for health care workers, has improved over time for certain vaccines. However, despite some progress, further effort is still required to promote adult immunization and increase vaccine uptake. Coverage estimates for seasonal influenza among the general population of adults have remained relatively stable since 2006 at approximately 35%. Although immunization coverage for adults older than 65 years remains high, it is still below national targets. More education and health care worker participation are required to raise awareness of the importance of achieving high immunization coverage in individuals with a chronic medical condition who are at higher risk of complications. Hepatitis B immunization coverage has increased in the health care worker group since 2001. Higher hepatitis B immunization coverage observed in the younger population likely reflects the aging of the cohort with access to publicly-funded vaccine programs, as programs rolled-out from the mid-1980s onwards. Pneumococcal vaccine uptake has remained relatively stable since 2001, with immunization coverage remaining below national targets, indicating the need to improve pneumococcal vaccine promotion among the recommended target groups.

Acellular pertussis and varicella immunization questions were added in 2006 to assess baseline immunization coverage estimates in the adult population. Since 1999, the adult formulation of the acellular pertussis vaccine has been given in combination with tetanus and diphtheria as the Tdap vaccine. While the Tdap vaccine is currently available in all provinces and territories for adolescents, it is not yet routinely funded for adults. The new varicella vaccine has only been available for a short time and as varicella/chickenpox is perceived by many as a common childhood disease, vaccine uptake has been slow among adults. The immunization coverage estimates in the adult population for acellular pertussis and varicella have increased since 2006. However observed estimates still remain low and below national targets set at the 2005 Consensus Conference on Vaccine-Preventable Diseases ^{Footnote 4}.

Tetanus immunization coverage was low even though it is not a newly-funded vaccine program. Tetanus uptake was high among adults who underwent treatment for a wound. However, the question examining tetanus immunization coverage covered a ten-year period and so recall bias may have contributed to an underestimation of immunization coverage. Tetanus is a severe and often fatal opportunistic infection and low tetanus immunization coverage estimates demonstrate the need to improve efforts to increase routine tetanus immunization in the general population.

The role of health care workers in educating their clients regarding vaccines is a significant part of the immunization process within the adult population. For both seasonal influenza and pneumococcal vaccines, a recommendation from a health professional was a strong predictor of immunization uptake; thus, health professionals should be encouraged to participate in immunization promotion activities. Moreover, the results highlight the need for education among adults in identifying target groups for specific vaccines and the need to promote immunization in these vulnerable groups. Educational campaigns and programs targeting high risk groups, adults aged over 65 years and adults with a chronic medical condition have been successful in communicating reasons to be immunized and recommendations for annual seasonal influenza immunization. While more public education may be needed for influenza and other vaccines, educational programs should continue to focus on the importance of the seasonal influenza immunization to the public.

Vaccination coverage among adults in the United States (U.S.) were measured with the 2011 National Flu Survey (NFS) ^{Footnote 8}. Seasonal influenza vaccine coverage within the general population of non-high risk adults in the U.S. (18 to 49 years) was reported as 36.2%, while 62.3% of adults aged 65 years and over received an influenza vaccination during the 2011-2012 season, similar to the observed influenza vaccine uptake in Canada. Nearly two thirds (63.4%) of health care workers (19 to 64 years) were immunized for seasonal influenza. Similar to the situation in Canada, estimates of coverage for pneumococcal vaccination in the U.S. is 20.1% and 62.3% for high risk adults (19 to 64 years) and adults 65 years and older respectively ^{Footnote 9}. Some international immunization coverage estimates vary significantly from Canadian estimates. Variation in immunization coverage estimates may reflect differences in immunization program delivery as well as differences in reporting methods of vaccine uptake. Vaccine products may also differ between nations and hence may affect vaccine administration and uptake within countries. Immunization recommendations for target groups and definitions of high risk groups also vary between the U.S. and Canada resulting in differing coverage levels among the general populations. Programmatic differences due to variations in the epidemiology of the various vaccine-preventable diseases will also impact the coverage estimates observed in each country and create limitations when drawing comparisons. As such, comparisons should be interpreted with caution.

The limitations of the 2012 aNICS include a low response rate and recall bias due to respondents being asked to self-report their immunization history from memory as opposed to an immunization record. The self-reported results are likely to impact the accuracy and reliability of the coverage estimates. The low response rate obtained in this cycle of the survey does indicate that the results of the survey may not be entirely representative of the Canadian population of adults. In addition a lack of representation from First Nations populations, cell phone only households, institutionalized adults, and newcomers to Canada who may not be comfortable answering questions in French or English does limit the generalizability of the survey findings to the whole Canadian population.

The aNICS constitutes an on-going approach to measuring adult immunization coverage in Canada and monitoring progress towards national targets. PHAC will continue to work to improve immunization coverage assessment methodology and immunization coverage levels through public and professional education and outreach campaigns to promote the benefits and safety of immunization in Canada. Cross-sectional immunization coverage surveys such as the national immunization coverage survey will be used to assess immunization coverage until a fully functional network of immunization registries is in place across Canada.

Conclusion

Although immunization coverage estimates have improved for Canadian adults, immunization coverage remains below national goals. Obtaining a vaccine recommendation from a health professional was associated with higher immunization coverage. Therefore, more education and increased health care worker participation in promoting the benefits of immunization to the general adult population, seniors and those with chronic medical conditions is needed to increase immunization coverage. Various factors contribute to the coverage of certain vaccines among the general Canadian population. Inviting provinces and territories to purchase additional sample provides more power to the NICS and allows for the comparison of immunization programs within Canada and provincial/territorial immunization coverage estimates with national ones. This practice will be encouraged in future survey cycles until immunization registries containing adult populations are fully operational in all jurisdictions across the country. The next cycle of the aNICS will take place in 2014.