Commentary – Completing the picture: a proposed framework for child maltreatment surveillance and research in Canada

Introduction

The rapid spread and high morbidity and mortality of COVID-19 led to unprecedented disruption of people’s lives worldwide. The evolution and protracted nature of the pandemic created uncertainty and unpredictability. Multiple waves of infections, hospitalizations and deaths, and the associated public health restrictions, drastically altered everyday functioning for millions of people and increased the risk of mental health problems.^{Footnote 1Footnote 2Footnote 3} However the data are not always consistent; one report suggesting that during the first few months of the pandemic, psychological distress increased but that most metrics of distress returned to baseline by mid-2020 and that loneliness, life satisfaction and suicide rates remained stable overall.^{Footnote 4}

This special issue of the journal underscores the negative effects of the first year of the pandemic on Canadians’ mental health and well-being. During the first 10 months after the pandemic was officially declared, alcohol and cannabis use^{Footnote 5Footnote 6} and depression^{Footnote 7} increased, and self-rated positive mental health, life satisfaction and community belonging^{Footnote 8} declined, with no changes in suicidal ideation noted.^{Footnote 9} However, Canadians were not equally impacted. As Varin et al.^{Footnote 5} maintain, “understanding the social determinants of health is key to developing harm reduction and mitigation strategies.” Indeed, younger age, living in an urban area and having co-morbidities of mental health issues were related to an increased likelihood of negative outcomes.^{Footnote 5Footnote 6Footnote 7} Similarly, Canadian females, especially those who are caregivers of children younger than 18 years, tended to be more seriously affected, a finding that is consistent with international research results.^{Footnote 3}

While informative, these findings were restricted by the nature of the data collected, sociodemographic characteristics, a limited set of individual factors (e.g. anxiety, depressive symptoms and mastery), community factors (e.g. sense of belonging) and exposures to pandemic-related stressors. A major gap remains in our understanding of the full impact of COVID-19 on women and children, whether it created a “shadow pandemic” of increased family violence, notably child maltreatment.

In addition to disease and death, the pandemic brought social and physical isolation, financial insecurity, increases in alcohol use and mental health problems, school closures and reduced access to medical and social services, all of which contributed to children’s risk of maltreatment. Several studies support this interpretation of the situation,^{Footnote 10Footnote 11} as does anecdotal evidence of an upturn in calls to domestic violence shelters and kids’ phone helplines.^{Footnote 12} However, little empirical Canadian data exist.

This commentary reviews what is known about violence against children during the pandemic. It highlights data gaps that existed pre-pandemic, and how our failure to address them hampers our ability to mitigate the harm to children. The authors advocate ongoing surveillance and research with a focus on social determinants of health to target resources and health promotion and prevention efforts. This aligns with the priorities of the Minister of Health and the Public Health Agency of Canada (PHAC).^{Footnote 13} We suggest a multi-pronged approach to child maltreatment surveillance and research that addresses not only our lack of knowledge during the pandemic, but more broadly, shortcomings in family violence information in Canada.

Child maltreatment during COVID-19

Long before the COVID-19 pandemic, child maltreatment was recognized as a global problem that extends across the sociodemographic spectrum.^{Footnote 14} Crises such as the pandemic and related economic and social effects are potential catalysts for family violence. As the pandemic continues, information on children’s exposure to violence is emerging.^{Footnote 15} The data generally indicate an increased risk of victimization; however, findings are mixed and have fluctuated, depending on the source and time. In fact, referrals of suspected child abuse to police and child protection services have decreased worldwide—by as much as 70%^{Footnote 12Footnote 16Footnote 17Footnote 18}—although calls to the police for domestic disturbances increased by 12% in Canada.^{Footnote 19}

Trends in calls to helplines are unclear, with some reporting increases, others decreases, and half reporting no change.^{Footnote 12Footnote 19} These variations may reflect differences in stay-at-home orders, educators and other service providers no longer seeing children, and victims’ inability to safely or privately access services during lockdowns.

Based on hospital records, several studies observed an increase in abuse-related pediatric injuries.^{Footnote 12Footnote 20} For example, the Children’s Hospital of Eastern Ontario reported a greater than two-fold increase in fractures and head trauma in children younger than one year during the period September to January 2021 compared with the same period in pre-pandemic years.^{Footnote 21} This is consistent with reports from the United Kingdom.^{Footnote 22} Community surveys of caregivers of children have shown that pandemic-related stressors such as job loss, social isolation and parental distress were associated with increased emotional/psychological abuse, physical/supervisory neglect and greater use of harsh disciplinary practices.^{Footnote 10Footnote 11Footnote 12Footnote 23}

The mixed picture underscores the persistent challenges affecting the availability and quality of data on violence against children.^{Footnote 12Footnote 15} Some of these problems are related to inadequate investment in routine and longitudinal surveys and other methods of data collection necessary to estimate prevalence and incidence estimates of child maltreatment. Designs that combine cross-sectional surveillance with the ability to collect longitudinal data are needed to develop the full picture. What’s more, they can be adapted from existing approaches in research conducted with youth.^{Footnote 24} Routinely collected cross-sectional data provide prevalence estimates over time, and longitudinal data can detect changes in the characteristics of the target population at both the group and the individual level.

Administrative data systems that might provide national estimates of the incidence of reports to authorities and service providers have been found to be deficient,^{Footnote 15} with challenges related to underreporting within the welfare system^{Footnote 25} and concerns about screening tools that are currently utilized by hospital systems.^{Footnote 26} Lack of consistent, reliable data has made it difficult to understand child maltreatment related to COVID-19, to track patterns and to make plans. The pandemic exposed a major deficiency in the system—under-identification of the extent of child maltreatment in Canada. As we emerge from the pandemic, a comprehensive approach to collecting child maltreatment data is needed.

Surveillance and research in Canada

In many ways, Canada’s public health system has prioritized the issue of family violence with support across the political spectrum and from numerous federal departments and agencies.^{Footnote 27} However, more can be done.

The federal government ratified and implemented the United Nations (UN) Convention on the Rights of the Child, and offered its support to the UN Sustainable Development Goals 2030 Agenda, including Goal 16.2 to end abuse, exploitation, trafficking and all forms of violence against and torture of children by 2030.^{Footnote 28}

In 2018, Canada joined the Global Partnership to End Violence Against Children as “pathfinding country,” reinforcing its commitment to expanding political support, mobilizing additional resources and preparing practitioners to address violence against children. As part of this pledge, Canada agreed to accelerate domestic actions over three to five years.^{Footnote 28} Increased surveillance is also aligned with calls to action by the Truth and Reconciliation Commission, which highlighted the need to report on First Nations, Inuit and Métis children in care as well as the reasons for their apprehension; increased surveillance would also respond to the recommendation to monitor and assess neglect and indirectly reduce the number of Indigenous children in care through the provision of evidence to inform interventions.^{Footnote 29}

To determine if Canada is fulfilling its commitment as outlined by the UN Convention and the calls to action by the Truth and Reconciliation Commission, and to monitor progress towards achieving our Sustainable Development Goal target of eliminating violence against children by 2030, a rigorous and robust framework of surveillance and research, integrating multiple sources of information, must be in place.

Public health surveillance is traditionally defined as the ongoing, systematic collection, recording, analysis, interpretation and dissemination of data to inform and evaluate public health practice.^{Footnote 30} Surveillance data are used for monitoring, generating hypotheses and examining differences over time between subgroups, for example, by sex or province, and focusses on systems at a broader level. We also need research to test hypotheses generated by surveillance and answer specific questions around causal effect and the effects of context. Evidence from both surveillance and research can inform the design, evaluation and implementation of public health interventions and to address the questions of what works for whom, and why. This helps empower decision makers to determine effects of policies through timely and useful evidence.^{Footnote 31} Successful examples of this have been used in assessments of parenting programs in Manitoba.^{Footnote 32}

Data on family violence can be collected from a variety of sources such as child welfare agencies, hospital and police records, and population-based surveys. The data should include, whenever possible, associations with risk and protective factors at the individual, family, community and societal levels (see Figure 1).

Figure 1 - Text description

This figure illustrates a surveillance and research framework for child maltreatment adopted by the Violence Prevention Alliance. The framework takes a four-step public health approach to violence prevention to guide data collection on child maltreatment.

Step one: Define the problem.
Step two: Identify risk and protective factors.
Step three: Develop and test intervention strategies.
Step four: Assure widespread adoption.

These steps are interconnected and inform data collection from a variety of sources.

The first source includes Population-based Surveys and Cohorts, which may be comprised of the following:

• General Social Survey. 
• Canadian Community Health Survey.
• Survey of Safety in Private and Public Spaces.
• Canadian Health Survey on Children and Youth.

The second source includes Vital Statistics, Justice and Police Records.

The third source includes Child Welfare Survey and Administrative Data, which may be comprised of the following:

• Canadian Incidence Study. 
• Pan-Northern Minimum Data Set.
• Canadian Child Welfare Information System.

The fourth source includes Registries and Hospital Administrative Data.

The four sets of sources overlap to Surveillance and Research to describe:

• Person.
• Place.
• Time.

The framework includes a horizontal arrow labelled “Across time” to illustrate that surveillance and research on child maltreatment should provide information on trends over time.

Finally, the framework also includes a vertical arrow labelled “Child Maltreatment Surveillance Indicator Framework” alongside four associated risk and protective factors, beginning with Individual-level factors at the bottom, then Family-level factors, Community-level factors and Societal-level factors at the top.

A four-step public health approach to violence prevention adopted by the Violence Prevention Alliance provides a framework to guide data collection on child maltreatment. Step one includes surveillance, which provides understanding of the size and scope of the problem. Step two is research to identify risk and protective factors. Step three includes the development and evaluation of interventions. Step four is the broad dissemination, evaluation and continual assessment to ensure all components of the strategy fit with the community context and have the desired effect of preventing violence.^{Footnote 33}

Two of the opportunities for action recognized in Canada’s Roadmap to End Violence Against Children^{Footnote 28} are tied to key principles of surveillance and the Minister of Health’s Departmental Plan.^{Footnote 13} These are (1) to enhance data collection, quality and monitoring, and (2) to strengthen the evidence about what works and mobilizing knowledge.

To achieve these goals, numerous initiatives have been undertaken or planned. For example, the General Social Survey (GSS) on Canadians’ Safety (conducted every five years using a nationally representative sample of individuals aged 15 years or older) contains questions about recent victimization and about childhood physical and sexual abuse and exposure to intimate partner violence older. In 2012, the Canadian Community Health Survey–Mental Health, the 2019 Canadian Community Health Survey and the 2018 Survey of Safety in Private and Public Spaces included questions related to childhood abuse and exposure to intimate partner violence.

Statistics Canada assembles family violence information from various administrative data sources, such as police reports, youth court reporting and victims service agency data. The Canadian Incidence Study of Reported Child Abuse and Neglect (CIS) is a national initiative to collect data about children and families who come to the attention of child welfare authorities owing to alleged or suspected child abuse and/or neglect.^{Footnote 28} Several waves have been conducted—in 1998, 2003, 2008 and most recently in 2018/2019, led by the Assembly of First Nations (First Nations Incidence Study of Reported Child Abuse and Neglect/CIS).^{Footnote 28} Two relatively new initiatives include the Canadian Child Welfare Information System (CCWIS) and the Pan-territorial Minimum Data Set.

While these data sources provide valuable information, there is room for improvement. Below, we offer recommendations as steps to be taken in developing a strong surveillance and research framework in Canada.

Recommendations

1. National surveys should encompass all forms of child maltreatment. Currently, only three subtypes are included: physical and sexual abuse and exposure to intimate partner violence. Although difficult to measure, emotional abuse and neglect should also be covered. Inclusion of questions should be guided by established and comprehensive definitions and consistent measurement.
2. To fully understand associations between child maltreatment and diverse outcomes, it is necessary to include a core set of questions on maltreatment in all national surveys focussing on health and victimization.
3. The upcoming Canadian Health Survey on Children and Youth should include child maltreatment questions. The General Social Survey (GSS) asks such questions of respondents aged 15 years or older, but Canada has no national data on younger age groups or data that can be related to a number of child-related outcomes and factors.
4. It is necessary to develop specific, sensitive protocols to ethically collect data on child maltreatment while balancing the rights, dignity and safety of participants as well as the duty to report in certain instances. Much of the reluctance to ask questions about family violence stems from concerns about ensuring safety and the possibility of creating distress. However, international experience suggests that with proper protocols in place, questions can be asked.^{Footnote 34Footnote 35}
5. As recommended by the Child Maltreatment Surveillance Indicator Framework,^{Footnote 36} information on multidimensional risk and protective factors should be extracted from administrative databases and from national surveys.
6. Longitudinal data should also be collected to understand trajectories of health and well-being in individuals exposed to child maltreatment. This may be accomplished through successive waves of data collection (national surveys) or through administrative data linkages.
7. For administrative data, an established, acceptable theoretical definition of family violence is needed to facilitate the extraction of child maltreatment data across agencies. Strong collaborative links between the organizations that are responsible for collecting family violence are needed, across sectors and provinces/territories, to ensure uniform measurement to obtain a national picture of child maltreatment. Some of this work has already begun; however, continued support and investment in the Canadian Child Welfare Information System (CCWIS) are required to develop a robust infrastructure of administrative maltreatment data.
8. We need systemic changes to child welfare and more universal supports for families with a focus on primary prevention, and adequate resources to monitor implementation of interventions and what works for whom and why.

This may seem to be a daunting endeavour, but we can learn from other countries. For example, the US Department of Justice, the US Children’s Bureau, the US Centers for Disease Control and the Australia Institute for Health and Welfare gather information from multiple sources, from household surveys to official administrative data, including state-based surveillance of child maltreatment from youth aged 10 to 17 years and caregivers of children younger than 9 years old.^{Footnote 34Footnote 35}

Conclusion

Family violence is a pressing social and public health issue for all Canadians.^{Footnote 37} The COVID-19 pandemic exposed a considerable gap in the availability and quality of child maltreatment data in Canada. The lack of such data systems has not only hindered our ability to readily understand the negative outcomes of the pandemic on children and families, but also limits our ability to respond in an evidence-based manner to the needs of children and families to assist with recovery.

This commentary highlights the significant impact of the pandemic on the mental health and well-being of Canadian adults and that these effects are not equal. What is unknown is whether, and to what extent, the pandemic has created a “shadow pandemic” that placed children at greater risk for maltreatment.

International research suggests that there has been an increase in violence towards children. Even if the increases are relatively short-lived, given the multitude of long-term health consequences associated with maltreatment, the effects will be long lasting.

As we move into post-pandemic recovery and planning, we need to capitalize on our knowledge of the deficiencies within our system and adopt a new framework for surveillance and research for child maltreatment. Some of the infrastructure to collect rigorous surveillance and research data on child maltreatment exists and can be used to increase the feasibility and success of what we propose in this commentary. The success of this framework will depend on consistent commitment and investment to collect and synthesize routinely collected data from a number of sources including national surveys conducted by Statistics Canada, hospitalization/injury data, criminal and police reports as well as a national child abuse and neglect databases such as the Canadian Child Welfare Information System (CCWIS). Improving the quality, consistency and scope of child maltreatment data is critical not only to understand the extent of maltreatment and to monitor trends, but also to ensure our commitments to the Sustainable Development Goal to eliminate violence against all children. 

Conflicts of interest

Tracie O. Afifi is an Associate Scientific Editor with Health Promotion and Chronic Disease Prevention in Canada, but has recused herself from the review process for this commentary. The authors have no conflicts of interest.

Authors’ contributions and statement

AG, TOA and LT informed the concept and contributed to the design, writing and critical review of this commentary.

The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.

Acknowledgements

The authors thank members of the Child Maltreatment Surveillance and Research Working Group for the many stimulating discussions that helped inform and shape the commentary. The editorial assistance provided by Mary Sue Devereaux and Joanna Odrowaz is gratefully acknowledged.

Previous | Table of Contents | Next