Deadly Force: How CBC analyzed details of hundreds of fatal encounters between Canadians, police

This story is part of Deadly Force, a CBC News investigation into police-involved fatalities in Canada.

Close to five million Canadians have some form of run-in with police each year.

In the vast majority of cases, these interactions end without aggression or injury. However, about 30 times per year someone dies at the hands of police.

And while the deadly use of force by police can have a significant impact on the relationship between law enforcement and the communities they oversee, no official body in Canada collects or tracks the details of these incidents. Rather, all that exists currently is a patchwork of statistics across provinces or groups of private citizens who are trying to keep tabs on the number of victims.

In 2017, a team of CBC researchers began the process of tracking details surrounding the deaths of citizens during encounters with police by culling thousands of independent investigator reports, coroner reports, court records andnews reports, and by conducting family interviews.

The objective: to build a national database of Canadians who were killed by police, or who died following the use of force during a police intervention, and gain insight into the circumstances surrounding these fatal incidents. Over twodozen specific details of each case were identified.

• Most Canadians killed in police encounters had mental health or substance abuse issues
• B.C. has highest rate of fatal encounters with police in Canada
• Black victims overrepresented in lethal encounters with Toronto police, CBC analysis finds

The parameters set for this story were to focus on fatal encounters where police used force. The CBC database does not include in-custody deaths, self-inflicted wounds as a result of suicide or attempts to evade police, or accidental police-caused deaths (such as a traffic accident).

460 incidents, 70% involved shootings

In total, our research found more than 460 incidents spanning the years 2000 to 2017 inclusively, with nearly 70 per cent of cases involving fatal police shootings.

The remainder of cases often involved people dying in restraint after a combination of stun gun deployments, baton strikes and a physical struggle often while the victim exhibited substance abuse symptoms.

Details captured in the database include information about the victims such as their gender, age, race or ethnicity, how they died, how they were armed, whether they had mental health or substance abuse problems, and the location of the incident.

Details about the police involved were also gathered, including police departments, force used, Taser deployments, injuriesand outcomes in the justice system.

But this database is constantly changing. In many cases, the events that follow a fatal incident can take years to play out, so new details are constantly emerging as investigations are concluded, or as witnesses come forward.

It was also impossible to confirm certain details in some cases. For example, our researchers could not reliably confirm the race or ethnicity of 22 per cent of the victims, and in about 50 per cent of cases, we could not confirmthe mental health or substance abuse history of the victim.

If you have information that can help us improve on the data we have already collected,if you are aware of an incident that was not captured by our researchers or if you spot an error, please email us atdeadlyforce@cbc.ca.

Additional notes about the data

• The primary sources of information were official police reports or reports from independent police watchdogs. In many cases the only witnesses on scene were the police officers involved. In some cases, families have contested some of the findings in the official report.
• In order to assess whether certain groups in Canadian society are overrepresented in fatal interactions with police, our researchers made best efforts to assign a race or ethnicity to each victim using a collection of family statements in media reports, interviews with friends or neighbours, pictures from social media or obituaries, and through logical inference (e.g. if a victim was a band member of a First Nation community, it was assumed they were Indigenous).
• The categorization used for assigning race/ethnicity is a condensed version of Statistics Canada's visible minority groups. For example:Chinese, Japanese, Korean, Filipino and Southeast Asian were merged into "Asian" for the purposes of our analysis. Indigenous population was established using the census response for "Aboriginal identity."
• Statistics on race and ethnicity are captured every census year. To estimate the populations of these groups during inter-census years, we assumed a linear growth from census to census for provinces and census subdivisions.
• Populations are not static. Since the number (and proportion) of Canadians identifying as a member of a minority group increased significantly in many communities between 2000 and 2018(the timeframe our database covers), the average populationover that 18-year span was used to determine theannualized rate per million for each jurisdiction.

Main research for the project conducted by the CBCreference library team:

Cathy Ross, Diana Redegeld, Ginny Oakley, Kate Zieman, Patrick Mooney andApril Lavilla-Lount.

Thank you to the team at Hold Your Fire for assisting with imagery and research.