Supplement to Statistics Canada’s Generic Privacy Impact Assessment related to the acquisition of child welfare data from child welfare agencies

Date: March 2021

Program manager: Director, Centre for Social Data Integration and Development

Director General, Social Insights, Integration and Innovation Branch

Reference to Personal Information Bank (PIB):

In accordance with the Privacy Act, Statistics Canada is submitting a new institutional personal information bank (PIB) to describe any personal information obtained from child welfare agencies, and provincial and territorial ministries, for the purposes of the Statistics Act. The following PIB is proposed for review and registration.

Canadian Child Welfare Information System (CCWIS)

Description: This bank describes information on child maltreatment, and family services obtained from child welfare agencies in Canada, including provincial and territorial ministries responsible for family services, Indigenous and other child welfare agencies. Personal information on children and caregivers may include first and last names, date of birth, gender, Indigenous and ethnic status, language, addresses and family demographics, as well as information related to their cases (reports made to the authorities, maltreatment types, level of substantiation, case assessments and outcomes, service provision, need for protection, placements, adoption or reunification).

Note: All direct personal identifiers such as names, addresses and data provider IDs are removed and replaced with a Statistics Canada record identifier. The personal identifiers are stored in a separate file with highly restricted access. Only the Statistics Canada identifier is kept with the information used for the database, to allow linkage between statistical files.

Class of Individuals: Children and caregivers (parents and guardians) accessing services or involved in child maltreatment cases reported to child welfare agencies in Canada.

Purpose: The personal information is used to create national statistical data on child maltreatment, and to enable statistical analyses of interactions with the system and of passage through the system, including referral to services, placement in foster care, connections to family, and reunification. Personal information is collected pursuant to the Statistics Act (Sections 3, 7, 8, 13, 22).

Consistent Uses: Subject to Statistics Canada's Directive on Microdata Linkage, information on child maltreatment may be combined with surveys or other administrative data sources for approved statistical purposes and longitudinal analysis. These data are used to: produce standardized national, provincial and territorial estimates; conduct retrospective, prospective, and life-course analyses; better understand protective and risk factors, and transfers between jurisdictions.

Retention and Disposal Standards: Information is retained until it is no longer required for statistical purposes and then it is destroyed.

RDA Number: 2007/001

Related Record Number: StatCan SSD 040

TBS Registration: To be assigned by TBS

Bank Number: StatCan PPU 813

Description of statistical activity:

With the collaboration of child welfare agencies, provincial and territorial ministries and Indigenous organizations and governments, Statistics Canada intends to acquire administrative data on child welfare in order to develop the Canadian Child Welfare Information System (CCWIS) on behalf of the Public Health Agency of Canada (PHAC). The CCWIS is a national public health information system on child welfare, and its purpose is to support nationally standardized analyses and reporting on child maltreatment; investigations and outcomes; the number of children in need of protection; and passage through the child welfare system, including referral to services, placement in foster care, connections to family, reunification, and other requests for family services. These data may be linked to other data held by Statistics Canada to better understand the family context, risk factors for child maltreatment, and long-term education, health and other outcomes for children.  Such linkages could inform policies, laws, programs and family support services to improve the lives of children and their families. Moreover, as Indigenous children are overrepresented in the child welfare system, quality national data on child welfare, including standardized provincial and territorial data, can be of particular benefit to programs and services aimed at ensuring the well-being of First Nations, Inuit and Métis communities.

Reason for supplement:

Statistics Canada's Generic Privacy Impact Assessment (PIA) presents and addresses the privacy principles and levels of potential security risks related to its statistical activities. Existing safeguards have been assessed as sufficient to address the potential privacy risks associated with the CCWIS. This project will follow standard Statistics Canada data stewardship practices to ensure the privacy of any data collected.

The purpose of this PIA supplement is to clearly illustrate the need for this personal information to achieve Statistics Canada's statistical mandate, and to demonstrate the public benefit that will be derived from the CCWIS data.

Necessity and Proportionality

The collection of personal information for the Canadian Child Welfare Information System was assessed against the four-part test proposed by the Office of the Privacy Commissioner of Canada, and aligns with Statistics Canada's Necessity and Proportionality Framework:

  1. Necessity:
    The CCWIS will fill a gap in annual national statistics, including provincial and territorial statistics, on the reporting, assessment, progression, and outcomes of children reported to child welfare authorities in Canada. Currently, there is no framework or comprehensive data source allowing for the production of national data on child maltreatment-related and other family involvement in child welfare by age, sex, Indigenous identity and ethnic status by province or territory. There are urgent calls for Canada-wide data on these issues to inform regional and national child welfare prevention and protection policies and practices. Data-informed child welfare is crucial to protect and improve the lives of many Canadian children and their families. The Truth and Reconciliation Commission of Canada and the National Inquiry into Missing and Murdered Indigenous Women and Girls, among others, are calling for data that include, but are not limited to, foster care involvement and risk factors for child placements, the need for which will be addressed by the CCWIS. The information to be contained in the CCWIS is being discussed through extensive engagement sessions aimed at identifying the needs of a broad range of stakeholders and subject matter experts in child welfare, including academics, researchers, service providers, provincial and territorial governments, Indigenous governments and organizations, the Public Health Agency of Canada and other federal partners.
  2. Effectiveness (Working assumptions):
    The high quality, timely and relevant data produced by the CCWIS will support the development of evidence-based policies and programs to improve the child welfare system. The collection of personal identifiers will help ensure high data quality by removing duplicate information and identifying repeat reporting. It will also provide opportunities to identify risk factors for child maltreatment and long-term outcomes by reviewing trends over time and outcomes of children into adulthood. Reliance on administrative data sources, as opposed to a survey approach, will reduce the burden on limited child welfare workers' resources.
  3. Proportionality:
    The CCWIS will have clear public benefits as the findings it will generate are expected to inform policies, laws, programs and family support services to ensure:
    • that children whose situations are such that they are going to need to be 'in care' for a long period of time are quickly identified and placed into stable care situations;
    • that this group of children is small, because effective supports are available within communities in need of such supports; and
    • that only children who need to be placed into care for their safety are, while children who do not need this are not wrongly placed into care.
    Future data linkage activities involving the CCWIS will need to satisfy stringent and independent approval mechanisms prior to initiation. Proportionality has been considered based on the following elements – sensitivity, ethics and design elements:
    • Sensitivity: The data collected for CCWIS is of a sensitive nature due to the elements being measured and the fact that most of the data collected concern children and child maltreatment. In addition, a significant proportion of the data to be collected will come from Indigenous children in a context where Indigenous communities are increasingly seeking greater decision-making powers in terms of the control of information on members of their communities. Moreover, the option of linking child maltreatment data with other data sources held at Statistics Canada to expand the informational value of the CCWIS, could potentially increase the sensitivity of the project. To reduce the risk of sensitive information being disclosed, the data will be processed according to current Statistics Canada best practices and in accordance with all relevant directives and policies. In particular, personal identifier variables (e.g., name, address, etc.) will be stored in a file separate from the child maltreatment data and accessible to only a limited number of employees on a need-to-know basis. Further, any proposition to link CCWIS data to other Statistics Canada data holdings will be required to undergo a stringent justification and review process in accordance with the Directive on Microdata Linkage before moving forward. Only the file with the personal identifier variables with no child maltreatment data is required for the linkage. The resulting linked data file keys are subsequently used to link the child maltreatment data with other data files – no direct identifiers are involved at this stage.
    • Ethics: Statistics Canada has engaged, and will continue to engage, with all provinces and territories, as well as with Indigenous stakeholders. Experts in the field of child maltreatment, from both PHAC and from the broader research community, have been consulted in order to ensure that any collection of data for CCWIS will be done ethically. In addition, we are adhering to transparency practices by documenting, on Statistics Canada's website, our intensions to obtain information on child maltreatment.
    • Flexibility of design elements: There are dozens of child welfare agencies in Canada, and each is under the jurisdiction of its respective province or territory. There is no national standard for collecting and storing data related to child welfare. Consequently, some jurisdictions have centralized systems, while others do not, and the data vary among provinces and territories in terms of both content and format. During a series of consultations with various stakeholders in the child welfare sector, child welfare service providers clearly advised that it would be far too burdensome for them if Statistics Canada were to request that they respond to a survey to describe a random sample of their cases. Previous methodological investigations by Statistics Canada to evaluate the sampling requirements for producing high quality estimates from a survey-based approach for the collection of data on child welfare also revealed that a high number of child welfare agencies would need to be selected for collection and that this exercise would need to be repeated throughout the 12 month reference period. Instead, it was deemed more practical for each of the jurisdictions to provide data files containing the relevant variables for every case they manage.
      Therefore, the estimated quality at the province/territory level should be sufficient to meet the policy needs.
      In terms of the content of the data acquired, Statistics Canada will request only those variables necessary to determine the national child maltreatment indicators that have been created in consultation with PHAC and other experts in the field of child welfare. Any additional variables that may be provided to Statistics Canada will be destroyed.
      The risk mitigations and the potential public benefits of this project have been deemed as proportional to the privacy considerations.
  4. Alternatives:
    A feasibility study was conducted with a broad range of child welfare stakeholders to identify an appropriate data collection approach for the CCWIS.  Through discussions with PHAC, it was determined that some child maltreatment data are being collected through the Canadian Incidence Study of Reported Child Abuse and Neglect/First Nations Incidence Study of Reported Child Abuse and Neglect (CIS/FNIS) funded by PHAC and managed by the Assembly of First Nations (AFN). While this last survey has provided great insight into child welfare, it is completed by child protection workers and there is concern about response burden. In addition, stakeholders' data needs have evolved to include the need to capture repeated interactions with the child welfare system. For example, a one-time assessment visit to a family from a child welfare worker due to a transitory situation or a mistaken concern has very different meaning than repeated visits due to chronic problems. The chronicity of child welfare involvement cannot be adequately captured by aggregate data. The need to understand trends over time, repeated interactions, and children's outcomes into adulthood requires the collection of individual level data. Potential data providers advised Statistics Canada that they would not participate in the CCWIS should the project proceed with a survey-based approach dependent on child welfare workers' participation, due to limited resources and the existing burden on these professionals. Moreover, surveying parents, guardians and children involved in child welfare investigations would not be appropriate for ethical reasons as it may harm those involved in welfare investigations or accessing services. This last approach would also encounter methodological challenges.


This assessment concludes that, with the existing Statistics Canada safeguards, the agency is prepared to accept and manage any remaining risks.

Formal approval:

This Supplementary Privacy Impact Assessment has been reviewed and recommended for approval by Statistics Canada's Chief Privacy Officer, Director General for Modern Statistical Methods and Data Science, and Assistant Chief Statistician, Social, Health and Labour Statistics Field.

Pierre Desrochers
Chief Privacy Officer
Date: March 23, 2021

Eric Rancourt
Director General,
Modern Statistical Methods and Data Science
Date: March 30, 2021

Lynn Barr-Telford
Assistant Chief Statistician,
Social, Health and Labour Statistics
Date: April 15, 2021

The Chief Statistician of Canada has the authority for section 10 of the Privacy Act for Statistics Canada, and is responsible for the Agency's operations, including the program area mentioned in this Privacy Impact Assessment. This Supplement to the Generic Privacy Impact Assessment is approved by the Chief Statistician of Canada.

Anil Arora
Chief Statistician of Canada

Date modified: