Every year, the Labor Department releases data on every worker death in the U.S. They’re broken down by job, cause and demographics.
Since 1992, the Census of Fatal Occupational Injury has allowed officials and advocates to target their efforts and slow the rate at which Americans are killed at work. Recently, we gained new appreciation for the popular dataset when we found, via an excellent series on “data gaps” in Toronto’s Globe and Mail, that Canada has no equivalent.
It’s one of several datasets which have thrust Canadians into the unfamiliar position of envying services provided by the U.S. government.
The U.S. didn’t always have “awesome” worker safety data, according to Peg Seminario, AFL-CIO’s occupational safety and health director. But in the early 1990s the Labor Department, states and other agencies banded together with the goal of counting every workplace fatality. They compile them from tens of thousands of state and local sources, news reports and other documents.
“It’s become a very rich data source,” Seminario said. The database includes workers’ occupation and industry as well as age, race and other demographic data. It allows regulators to identify the most dangerous jobs and which groups are at higher risk of workplace injury or death.
When unions and other advocacy groups identified a high and growing rate of fatalities among Hispanic workers in construction and other industries in the late 1990s and early 2000s, they worked with federal agencies, the Mexican Consulate and employers to educate workers on their rights and safety practices.
In 2001, Hispanic workers were injured at almost about 1.4 times the national rate. By 2017, that figure had fallen to 1.06 times the national rate.
Those kinds of targeted interventions can only happen if high-quality data are available, Seminario said.
Canada’s patchwork system of worker-safety statistics includes provincial workers’ compensation boards, as well as detailed data on the 7.8 percent of employees who work for or are regulated by the federal government.
Canadian government statistics are some of the best in the world, but oversights, quirks of history and sweeping 2012 government budget cuts have left gaps not just in worker safety, but in vaccination races, unemployment by race, and even marriage rates. These gaps in measurement can lead to gaps in research and policy, potentially allowing dangerous trends to continue unaddressed.
A country’s databases are a window into its priorities. Earlier in his career, Vanderbilt economist Christopher Carpenter would fly to British Columbia, hunker down in a basement library and analyze restricted data from a Statistics Canada survey that was one of the few in the world to ask about sexual orientation. Even today, he works with Canadian collaborators to measure underage drinking, bicycle helmet requirements and other issues that don’t have sufficient data coverage in the United States.
By providing consistent measurements of immunization rates at the state level, the U.S. National Immunization Survey allows researchers to determine the most effective state immunization initiatives.
In an analysis published this year in American Economic Journal: Economic Policy, Carpenter and University of Georgia economist Emily Lawler analyzed what happened when states required incoming middle-school students to get a booster shot for tetanus, diphtheria and pertussis (whooping cough).
The economists found that, not only were students more likely to get the booster, but they were also more likely to get vaccinated for meningococcal disease and human papillomavirus (HPV). It suggests that, once students were in the office for the required booster, doctors were able to persuade them to get other lifesaving inoculations, as well.
“We know much more clearly what works and what doesn’t in the United States for infectious disease control,” thanks to products like the immunization survey, Carpenter said.
Statistics Canada reports nationwide immunization rates. Communications chief Peter Frayne said the agency is working to expand its province-and-territory coverage with data that has been checked against provincial immunization registries. The data would be similar to the U.S. survey used by Lawler and Carpenter.
In the United States, someone who is neither white nor indigenous can be reported as Hispanic, Asian, black, Pacific Islander or other groups. In many Canadian employment reports, such groups are lumped together as “visible minorities. ”
This obscures differences in people’s’ backgrounds and employment situations. Frayne points out more detail can be found in the full Canadian census release, published twice each decade. U.S. data, published every month, shows the dynamic among racial and ethnic groups can change significantly over a five-year period.
“Statistics Canada stopped producing data on marriages and divorces at the national level in 2010 due to budget cuts,” Frayne said via email.
The U.S. National Center for Health Statistics publishes detailed marriage and divorce data in a series that can be extended to at least 1867 (incidentally when British colonies were united into a single dominion of Canada). The marriage rate has climbed slightly from its post-recession trough. The divorce rate continues to slide.
In the United States, the number of miles drivers travel in any given month can be an indicator of economic health and environmental impact. In Canada, it indicates neither — likely because it isn’t available.