Date: March 2021
Program manager:
- Director, Centre for Social Data Integration and Development
- Director General, Census Subject Matter, Social Insights, Integration and Innovation
Reference to Personal Information Bank (PIB)
Personal information collected through the COVID-19 Vaccination Coverage Survey is described in Statistics Canada's "Special Surveys" Personal Information Bank. The Personal Information Bank refers to information collected through Statistics Canada's ad hoc surveys, which do not form part of the regular survey taking activity of the Agency. They cover a variety of socio-economic topics including health, housing, labour market, education and literacy, as well as demographic data.
The "Special Surveys" Personal Information Bank (Bank number: StatCan PPU 016) is published on the Statistics Canada website under the latest Information about Programs and Information Holdings chapter.
Description of statistical activity
Statistics Canada is conducting the COVID-19 Vaccination Coverage Survey (CVCS), under the authority of the Statistics Act, on behalf of the Public Health Agency of Canada (PHAC). The purpose of the survey is to gather information to help governments measure progress towards vaccination coverage goals for COVID-19, and to collect information on the knowledge and perceptions of the COVID-19 vaccines. This information will help the Canadian government to plan a response to the pandemic that addresses the health and safety needs of the Canadian population. Specifically, it will help inform decision-making about vaccine distribution and communication strategies to increase uptake.
- This voluntary household survey collects information from individuals aged 18 years and older who live in Canada's provinces and territorial capitals. Survey topics include vaccination status, reasons for non-vaccination, trusted sources of information for vaccination, knowledge and beliefs about COVID-19 vaccination and vaccines in general, and the respondent's health (whether respondents have had COVID-19 and whether they have chronic conditions that increase risk for COVID-19 outcomes). In addition, information such as age, gender, marital status, postal code, household size and composition, Indigenous identity, visible minority status, immigration and citizenship, education and income will be collected. The names of household members are collected when interviewers carry out age order selection (for sampling purposes). However, given the policy importance of vaccinations for health and for social and economic recovery from the pandemic, there is a potential need to have data on vaccination impacts in the future. Any requirements for record linkage would follow established procedures for approval from the Chief Statistician and a record linkage rationale would be provided. If approved, all record linkages would be performed within the secure Social Data Linkage Environment. Any data products released will be based on aggregated responses and processed to ensure that no individual can be identified.
The CVCS will be conducted on an ongoing voluntary basis throughout the spring, summer and fall of 2021. The first cycle of collection is scheduled to begin on March 11, 2021, with 4–5 subsequent cycles starting in April, June, July, August and September.
Reason for supplement
While the Generic Privacy Impact Assessment (PIA) addresses most of the privacy and security risks related to statistical activities conducted by Statistics Canada, this supplement describes additional measures being implemented due to the sensitivity of the information being collected. The CVCS will be collecting information on COVID-19 vaccination status and beliefs around COVID-19 vaccination, which becomes more sensitive when collected alongside personal information, such as gender identity or other demographic characteristics. This SPIA describes how Statistics Canada has accounted for the unique impact to vulnerable populations when designing and conducting this survey, and integrates relevant principles of the Office of the Privacy Commissioner's Framework for the Government of Canada to Assess Privacy-Impactful Initiatives in Response to COVID-19. Like all PIAs, Statistics Canada's privacy framework ensures that elements of privacy protection and privacy controls are documented and applied.
Necessity and Proportionality
The collection and use of aggregated responses and personal information for the CVCS can be justified against Statistics Canada's Necessity and Proportionality Framework:
1. Necessity:
The campaign to vaccinate the Canadian population against COVID-19 began in December 2020 and is expected to continue until the fall of 2021. The initial limited supplies of vaccines are being prioritized by population subgroups to ensure the most at-risk and vulnerable in our society are protected. Basic demographic information on vaccinated Canadians is recorded in provincial and territorial registries. However, the amount and type of information available can vary greatly from one province or territory to another. While registry data can help measure progress towards vaccination coverage goals, the registries may not collect detailed information about the people being vaccinated, or their knowledge and beliefs regarding vaccination and vaccine hesitancy. Health agencies need to understand why people are choosing not to get vaccinated. This information will help to develop and adjust communication strategies and education outreach programs to address vaccination concerns. With the CVCS, it is possible to analyze the reasons for vaccine hesitancy and to compare vaccination rates across provinces using the same definitions and characteristics (coherence). The CVCS was designed to provide timely results to help inform decisions on vaccine delivery, as well as communication and education about COVID-19 vaccination. It will allow health organizations and the government to measure progress towards immunization coverage goals, which ensures that Canadians are protected against COVID-19. In addition, results from the CVCS will provide Canadians with information on how the various vaccination campaigns across the provinces and territories are progressing. This can help inform their personal decisions about COVID-19 vaccination. An increased understanding of the concerns of their fellow Canadians about COVID-19 vaccination can also promote discussion and education about the COVID-19 vaccines. The survey data file, without direct identifiers, will be retained as long as required for statistical purposes.
2. Effectiveness—Working assumptions:
Due to the urgent need for the information, a short questionnaire (28 questions that will take 5–10 minutes to complete) was developed. It follows Statistics Canada's processes and methodology in an accelerated manner to produce data faster. The survey will be administered using a self-reported electronic questionnaire. A random sample of 3,000 households in the northern capitals (1st cycle) and 20,000 households in the provinces (subsequent cycles) will be selected from Statistics Canada's dwelling universe frame. Each household will receive an invitation letter with instructions on how to randomly select someone in the household (age order selection from those aged 18 and older) to complete the survey. The letter will also include a secure access code for the selected person to go online and complete the survey on Statistics Canada's secure survey infrastructure. After two weeks, interviewers will follow up with households that have not yet responded and will offer to collect the survey by telephone.
The survey will consist of 5 independent cycles (additional cycles may be added). The collection period for each iteration of the survey will be approximately one month, with the data releases planned for 3 weeks after the close of each collection cycle. All Statistics Canada directives and policies for the development, collection, and dissemination of the survey will be followed. Survey responses will not be attached to respondents' addresses or phone numbers. The data will be representative of the Canadian population in the provinces and territorial capitals and may be broken down (disaggregated)by age groups (18–59 years old, or 60 years and over) or gender at the provincial and territorial level. These age categories were chosen to measure vaccination rates for the elderly who are considered to be more at-risk of having poor outcomes if they contract COVID-19. When quality estimates are available, and the identity of respondents is protected, disaggregation by other demographic characteristics (such as ethnicity or immigration status) may be possible.
The Public Health Agency of Canada requires timely data for specific reference periods to inform ongoing decision-making regarding vaccine distribution and communication. As vaccination coverage rates are likely to increase as time passes, and attitudes and beliefs may change, there are currently no plans to combine results from multiple cycles to produce results at a more detailed geographical level, or by demographic characteristics other than those mentioned above (age group and gender). The sample size, at 3000 in the territories and 20,000 per cycle in the provinces, is designed to produce accurate estimates by province, territory by age and gender. While the questions for population groups are included, the ability to produce results will be dependent on response patterns. There may be situations where researchers would benefit from combining samples from different cycles despite changes in certain response patterns over time. In this event, after seeking appropriate approvals, any results from combined cycles would follow all data aggregation and suppression practices currently employed by Statistics Canada.
3. Proportionality:
Data on COVID-19 vaccination, and on beliefs regarding vaccines and vaccine hesitancy may be considered sensitive. The COVID-19 pandemic has necessitated the development and deployment of vaccination programs in a very short time period. Given the urgency of the situation, it is imperative that health departments and agencies have timely data to inform decisions on policies and programs. For this reason, experts at Statistics Canada, the Public Health Agency of Canada (PHAC), provincial and territorial health ministries, and committees created to manage the COVID-19 vaccine campaigns have been consulted on the content, scope and methodology of the survey. Wherever possible, questions about vaccination, or knowledge and beliefs about vaccination (and COVID-19 vaccination specifically) were adapted from existing surveys. Some of these contributors include the Childhood National Immunization Coverage Survey (CNICS), the COVID-19 crowdsourcing surveys and the COVID-19 web-panel surveys. In addition, the CVCS underwent qualitative testing and was revised accordingly. Results from the CVCS could be compared with these and other surveys (such as Canadian Community Health Survey or Canadian COVID-19 Antibody and Health Survey) to provide context when interpreting the results.
All the data to be collected are required for the purpose of the survey as described above. Careful consideration was made for each question and response category to ensure that it would measure the research questions and help inform future decisions related to COVID-19 vaccination. Experts at PHAC worked closely with Statistics Canada to determine the content of the survey questionnaire, including the demographic information that will help in the analysis of the survey results. Information such as the respondent's age, gender, ethnicity and education level, all have the potential to affect a person's access to healthcare and their beliefs about vaccination. It is important to include these factors, where possible, when analyzing survey results and progress towards vaccination coverage goals.
The sample sizes of 20,000 for the provinces and 3,000 for the territorial capitals have been assessed as the minimum required to provide quality estimates at the provincial or territorial level by age group or gender. Without substantially increasing the sample size, it is not possible to produce reliable results for smaller population groups, such as some of the more vulnerable populations. Options were explored to expand coverage beyond the territorial capitals in the north, but none were viable within the context of the pandemic and the need for timely data. The Northern Survey Governance Committee was consulted and involved in determining the sampling strategy for the north. The Centre for Indigenous Statistics and Partnerships also provided input to ensure appropriate engagement, and consideration for the interests of Indigenous peoples across the country.
There are not any immediate requirements for linking results from the CVCS to other administrative datasets, surveys or the Census. However, given the policy importance of vaccinations for health and for social and economic recovery from the pandemic, there is a potential need to have data on vaccination impacts in the future. Any requirements for record linkage would follow established procedures for approval from the Chief Statistician and a record linkage rationale would be provided. If approved, all record linkages would be performed within the secure Social Data Linkage Environment. Any data products released will be based on aggregated responses and processed to ensure that no individual can be identified.
Statistics Canada directives and policies with respect to data collection and publication will be followed to ensure the confidentiality of the data. Individual responses will be grouped with those of others when reporting results. Individual responses and results for very small groups will not be published. The grouping of results will make it impossible to identify individual respondents and reduces any potential impact on vulnerable populations or subsets of populations. Survey results will be made available in the Federal Research Data Centre (FRDC) but no personal identifiers will be included. PHAC experts, who have been deemed Statistics Canada employees, will have access to this micro data file to assist them with data validation and the analysis of survey results. Only aggregate/grouped data that has been vetted for confidentiality will be allowed to leave the FRDC. All data results are managed in accordance with Statistics Canada's security and confidentiality requirements.
The survey findings will support decision-making at all levels of government concerned with improving the COVID-19 response and ensuring the health and safety of Canadians. These benefits are believed to be proportional to the invasion of privacy and associated risks.
4. Alternatives:
Although some alternate data sources for vaccination rates and vaccine hesitancy exist (provincial and territorial registry data, the Canadian Community Health Survey, the Canadian COVID-19 Antibody and Health Survey, the General Social Survey, and the Canadian Perspectives Survey Series), they do not provide all of the data necessary to measure the progress of the COVID-19 vaccine campaign for the groups of interest to the Public Health Agency of Canada (PHAC), or inform decision-making in a timely manner. As well, none of these alternate data sources provide detailed demographic information about those choosing not to get vaccinated and the reasons why. The possibility of using provincial and territorial registry data to calculate vaccine coverage rates was considered, but there was insufficient time to put the required data sharing agreements in place and carry out the necessary linkages.
The CVCS will be conducted at regular intervals and will provide up-to-date and comparable results for the provinces and territories. It will also provide a detailed breakdown on why people are choosing not to get vaccinated and allow health authorities to address concerns and promote vaccine uptake as issues are identified.
Currently, there are no other surveys that gather the detailed information on COVID-19 vaccination required to measure coverage rates for the provinces and territorial capitals for the duration of the vaccination campaign. The CVCS was designed to be short and timely, to produce reliable and accurate results for the provinces and territorial capitals by age group or gender.
Mitigation factors
Some questions contained in the CVCS may be considered sensitive as they relate to an individual's health and their views on vaccination. However, the overall risk of harm to the survey respondents has been deemed manageable with existing Statistics Canada safeguards that are described in Statistics Canada's Generic Privacy Impact Assessment, as well as with the following measures:
Transparency
Prior to the survey, respondents will be informed of the survey purpose, allowing them to decide whether they wish to participate. The survey purpose reads:
The purpose of this survey is to collect information on national COVID-19 vaccination rates. Your feedback is important, even if you have not been vaccinated, or do not intend to be.
This survey aims to
- measure COVID-19 vaccination rates across Canada;
- find out about your knowledge and perceptions regarding COVID-19 vaccines, and about the sources of information you rely on.
The information you provide will help the Canadian government plan a response to the pandemic that addresses the health and safety needs of the Canadian population.
This information will be provided via invitation and reminder letters. It will also be reiterated at the beginning of the questionnaire. Respondents will also be informed in the invitation letter, reminder letters, and in the questionnaire itself, that their participation is voluntary before being asked any questions. Information about the survey, as well as the survey questionnaire, will also be available on Statistics Canada's website.
Confidentiality
Variables that directly identify respondents will be separated from the data files in the first stage of data processing and placed in a secure location with controlled access. Variables that might indirectly identify respondents are examined and modified as necessary in order to protect the privacy and confidentiality of respondents. Individual responses will be grouped with those of others when reporting results. Individual responses and results for very small groups will never be published or shared with government departments or agencies. Careful analysis of the data will be performed prior to the publication and sharing of aggregate data to ensure that marginalized and vulnerable communities are not disproportionally impacted.
Conclusion
This assessment concludes that with the existing Statistics Canada safeguards and additional mitigation factors listed above, any remaining risks are such that Statistics Canada is prepared to accept and manage the risk.