Dossier: “Kids Are Dying”: A Statistical Critique of Canada’s Online Harms Bill Rhetoric Alexandra Kitty

Executive Summary

In June 2026, the Canadian federal government introduced legislation under what it rebranded from the “Online Harms Act” to the “Digital Safety Act,” proposing a social media ban for children under 16. Culture Minister Marc Miller publicly stated “kids are dying” to justify the urgency of the legislation. This dossier examines that rhetorical claim against actual child mortality data in Canada, analyzes the evidentiary basis for attributing those deaths to social media, and contrasts the legislative priority with the documented and well-evidenced determinants of child death, most prominently, poverty, housing insecurity, and food insecurity, especially in Indigenous and northern communities.

Part 1: What Is the Actual Legislation?

Canada’s Bill C-63, introduced in February 2024 under the Trudeau government, died on the order paper when Parliament dissolved for the 2025 federal election. The Carney government committed to reintroducing online safety legislation, initially floating a ban for children under 14, then settling on under 16. The legislation, tabled in June 2026, proposes:

A ban on social media platforms for users under 16
Age verification requirements for platforms
A new Canadian Digital Safety Regulator
Requirements to remove child sexual abuse material and self-harm-promoting content
Platforms that comply with safety standards may apply to allow younger users back with oversight

The bill follows Australia, which became the first country to implement a social media ban for under-16s in December 2025, with similar measures under consideration in Denmark, France, Greece, Spain, and New Zealand.

In Canada, a legal expert described a parallel Senate bill as a “trojan horse” that, in its most expansive form, would apply age restrictions to anyone under 18, well beyond any comparable international regulation, and would allow courts to issue blocking orders against platforms that fail to comply.

Part 2: What Are Children Actually Dying From?

The Baseline: Child Mortality in Canada Is Very Low and Declining

Overall child mortality rates in Canada are among the lowest in the world and have declined significantly over decades. The country’s under-5 mortality rate stands at approximately 4–5 deaths per 1,000 live births, down from hundreds per 1,000 in the 19th century. For school-age children (ages 5–14), death rates are in the single digits per 100,000 children per year.

In any given year, only a small fraction of Canadian children die at all. This baseline context is entirely absent from political rhetoric invoking child mortality to justify social media legislation.

Infants (Under 1 Year)

The leading causes of infant death in Canada are:

Perinatal conditions (prematurity, complications of labour/delivery, birth asphyxia)
Congenital malformations and chromosomal abnormalities (congenital heart defects, neural tube defects; approximately 5% of all infants with a congenital anomaly die in their first year)
Sudden Infant Death Syndrome (SIDS), approximately 3 babies die per week in Canada from SIDS; peak mortality occurs at 2–4 months of age
Infection

From 2015 to 2020, approximately 1,700 infants under age 1 died per year in Canada, with roughly 1 in 15 of those deaths occurring during sleep. None of these causes are related to social media.

Ages 1–14: Accidents, Cancer, Congenital Conditions

Across Statistics Canada’s leading-cause tables, the same three causes dominate for children aged 1–14:

Accidents (unintentional injuries): motor vehicle crashes, drowning, suffocation/choking, falls. A Parachute Canada analysis found 183 unintentional injury deaths among children aged 0–14 in 2012 nationally, at a rate of 3.2 per 100,000, itself a 30% reduction from a decade earlier.
Cancer (neoplasms): brain cancer accounts for approximately 41% of all cancer deaths in the 1–14 age group.
Congenital malformations and chromosomal abnormalities

Ages 10–14 and 15: Where Suicide Enters

Suicide emerges as the third leading cause of death in the 10–14 age group, after accidents and cancer. For youth aged 15–34, suicide has been the second leading cause of death for over two decades.

Canada’s youth suicide rate (ages 10–19) averaged 5.01 per 100,000 from 2010–2018, above the global average of 3.77 per 100,000. Notably, Canada is reportedly the only country among 35 studied where girls aged 10–14 have a higher suicide rate than boys in that age group.

However, these are still very small absolute numbers within an already low overall mortality rate. Suicide among children under 16 is not a mass mortality event, it is a concentrated tragedy that demands precision, not sweeping platform bans.

Part 3: The Social Media–Child Death Link: What the Evidence Actually Shows

Correlation vs. Causation

The claim that social media is killing children conflates correlation with causation in ways that peer-reviewed literature does not support. The key evidentiary picture:

The Mental Health Research Canada (MHRC) national survey of 27,000+ respondents found that specific negative online experiences, cybervictimization, persistent social comparison, are associated with worse mental health outcomes, not general access to social media.
Cybervictimization is associated with a threefold increase in suicidal ideation.
Students spending 7+ hours daily on screens show significantly higher rates of anxiety, depression, and suicidal thoughts. However, this is a correlational finding; heavy screen use and poor mental health share many confounders, including poverty, social isolation, and pre-existing conditions.
The Canadian Paediatric Society itself acknowledged “evidence gaps on the long-term impacts” and called for action on the basis of “enough red flags,” not definitive causal proof.
Jonathan Haidt’s widely cited The Anxious Generation concedes his own data shows only a 15% correlation between social media use and mental health decline.
The Information Technology and Innovation Foundation (ITIF) analysis of the Canadian context concluded that “banning social media will not change the rates of depression, anxiety, or other mental health challenges among teenagers” and that “the evidence doesn’t support the premise.”
The Science Media Centre’s expert commentary states: “The current research does not support the usefulness of banning kids from social media.”
A meta-regression of problematic social media use (not general use) and depression found a moderate but statistically significant correlation (r = 0.273). The distinction between problematic use and any use is critical, and is routinely erased in political messaging.

Social Media Is Not a Coded Cause of Death

Social media is not a recognized category in Statistics Canada’s leading-cause-of-death tables (Table 13-10-0394-01). It is not a coded cause of death. Where suicide is the proximate cause of death, social media may in some individual cases be a contributing factor, but the mortality data does not support characterizing it as a systemic killer of Canadian children.

What the Amanda Todd Case Illustrates

Carol Todd, whose daughter Amanda died by suicide in 2012 following online sexual exploitation and cyberbullying, is a frequent witness before parliamentary committees and was cited again in February 2026 hearings. Amanda’s death is genuine and tragic. But it illustrates a specific, documented harm, sexual exploitation and sextortion of minors, that is already illegal and was addressed in Bill C-63’s core provisions (mandatory removal of CSAM and non-consensual intimate content). Using one extraordinarily high-profile case involving criminal activity to justify a blanket social media ban for all children conflates distinct phenomena.

Part 4: The Real Killers: Poverty, Housing, and Food Insecurity

The most robust documented determinants of child mortality in Canada are socioeconomic, not digital. This is the context systematically absent from the online harms debate.

Nunavut: A Case Study in Structural Child Death

Nunavut’s infant mortality rate is approximately 21.4 deaths per 1,000 live births, compared to a national average of 4.8. In 2016, the rate rose to 17.7 before improving slightly. To put this in perspective: you are more likely to die as a baby in Nunavut than as a 70-year-old in the rest of Canada.

The causes are well-documented and entirely material:

Extreme poverty: In 2021, Nunavut had Canada’s highest child poverty rate at 35.8% for children under 18, and 43.2% for children under 6, both the highest of any province or territory.
Food insecurity: Food insecurity in Nunavut is four times the national average. Indigenous families above the poverty line in the provinces are still twice as likely to be food insecure as non-Indigenous families.
Housing overcrowding: In Nunavik, 60% of children under 15 grow up in overcrowded homes. Overcrowding drives infectious disease transmission, including tuberculosis.
Tuberculosis: Nunavut’s tuberculosis rate has been recorded at 62 times the Canadian average. In 2024, the TB incidence rate for Inuit living in Inuit Nunangat remains catastrophically elevated. TB disproportionately affects children and youth in these communities.
SIDS elevated by structural factors: In Nunavik, SIDS accounts for approximately 28% of infant deaths. Risk factors include bed-sharing (driven by housing overcrowding), parental smoking, and limited access to prenatal care, all conditions shaped by poverty and underfunded services.
Bronchiolitis: In Nunavut, 31% of Inuit infants were hospitalized for bronchiolitis in their first year of life, and 42% for lower respiratory tract infections, rates far above national averages, driven by overcrowded housing and limited healthcare access.

For youth living in Inuit Nunangat from 1994 to 2008, the mortality rate for Inuit children was 5 times higher than for other Canadian children, and the suicide rate was 30 times higher.

First Nations and Indigenous Children Nationally

37.4% of First Nations children living on reserve lived in poverty in 2021, more than three times the rate for non-Indigenous children (10.8%).
23.7% of all Indigenous children (on and off reserve) lived in poverty.
Poverty is a significant risk factor for premature death regardless of immigration status, with those in the most deprived areas showing the highest all-cause mortality rates.
Canada’s National Advisory Council on Poverty’s 2025 report notes that 17.5% of the Indigenous population living off-reserve lives below the poverty line, compared to 9.9% of the non-Indigenous population, and calls for immediate action.

Poverty as Health Determinant: The Academic Consensus

Public Health Canada identifies income and social status as the #1 determinant of health, ahead of all others. A University of Toronto/ICES study found that those living in the most deprived areas had approximately 60% higher rates of all-cause and premature mortality. For children, poverty is associated with higher incidence of disease, hunger, dental problems, lower immunity, and lesser access to palliative and preventive care.

Part 5: The Rhetorical Mechanics of “Kids Are Dying”

The Statement and Its Source

Culture Minister Marc Miller stated “kids are dying” in justifying the urgency of online harms legislation in June 2026. The claim builds on a pattern of advocacy testimony:

“If we don’t do something, we are going to lose more Canadian kids, more Canadian youth and more Canadian young adults to online harms” (Carol Todd, parliamentary committee, February 2026).
“We have been waiting 1,600 days for government to table the online safety act. Every day that passes causes kids immeasurable harm” (Advocates for child safety, February 2026).
Children First Canada CEO Sara Austin: “We do not need more studies and we do not need to debate whether social media is a problem for children. We already know the answer.”

How the Rhetoric Functions

The “kids are dying” framing works by:

Using true but contextually stripped facts. Children do die; suicide is the second leading cause of death in youth 15–34. Neither fact implicates social media as a systemic cause of child mortality.
Conflating mental health distress with death. There is evidence that heavy, problematic social media use correlates with anxiety and depression. Correlation with mental health distress is not the same as causing death.
Erasing poverty and structural determinants. The children at the highest documented risk of death in Canada, Inuit infants in Nunavut, First Nations children on reserve, are dying from causes rooted in poverty, housing, food insecurity, and inadequate healthcare. These communities have no significant social media mortality profile. They have a structural inequality mortality profile.
Treating a blanket access ban as a precision tool. Even researchers sympathetic to concerns about youth mental health, including the Canadian Paediatric Society, acknowledge that outcomes depend on specific online experiences, not general access. A blanket ban addresses neither specific harms nor root causes.
Obscuring the political convenience of the issue. A social media ban has popular support, one poll found broad public backing. It creates a visible, actionable policy signal that does not require confronting the expensive, politically difficult work of reducing Indigenous child poverty, building housing in Nunavut, or funding northern healthcare.

The Opportunity Cost

Parliamentary committee time, public attention, and legislative resources devoted to social media bans are not being devoted to:

Nunavut’s infant mortality rate, which has been four times the national average for years
First Nations child poverty rates of 37.4% on reserve
TB rates 62 times the national average in Nunavut
Food insecurity four times the national average in northern Indigenous communities
Overcrowded housing that kills children through respiratory disease and SIDS

These are the conditions under which Canadian children are, in fact, dying. None of them involve a phone.

Part 6: What Effective Policy Would Look Like

Researchers and critics of the ban approach, including the ITIF, the Science Media Centre, and independent academics, converge on a different framework:

Platform design accountability: Targeting addictive design features (infinite scroll, notification architecture, engagement-maximizing algorithms) rather than access per se
Mandatory transparency reporting by platforms on safety metrics and harmful content volumes
Precision prohibitions: Strengthening enforcement of existing laws against CSAM, sextortion, and non-consensual intimate images, the concrete harms with documented mortality links
Digital literacy and resilience programs in schools
Regulation of specific high-risk interactions (e.g., direct messaging by adults to minors) rather than blanket platform bans
Funding the social determinants of health in high-mortality communities: housing, food security, prenatal care access, tuberculosis control

A 12-year-old witness before the Canadian Heritage Committee in February 2026 put it plainly: “Kids will find workarounds. If you ban, they will probably protest.”

Conclusion

The claim that Canadian children are dying because of social media, and that a social media ban will stop them dying, is not supported by national child mortality data. Children under 16 in Canada die primarily from accidents, cancer, congenital conditions, and, at much higher rates than the national average, from the structural consequences of poverty in Indigenous and northern communities. Social media is not a coded cause of death. The evidence linking general social media access (as opposed to specific harmful experiences) to mortality is weak, correlational, and contested by multiple independent reviews.

Canada’s most urgent child mortality crisis is located in Nunavut and on First Nations reserves, where infant mortality rates run three to four times the national average, tuberculosis rates are 62 times the national average, and child poverty rates exceed 37%. Those children are not dying from TikTok. They are dying from underfunding, overcrowding, food insecurity, and generations of structural neglect.

Legislation that addresses the former while ignoring the latter is rhetorically effective, politically convenient, and epidemiologically dishonest.

Additional Research by Perplexity