As expected, exposure varies based on agricultural activity and the types of crops grown. Potential for exposure to 2,4-D is much higher in areas where cereal crops, potatoes, and fruit are more commonly grown due to the higher application rates for these crops. The prairie provinces (Alberta, Saskatchewan, Manitoba), Prince Edward Island and Southern Quebec all had higher potential for exposure to 2,4-D than other areas. These estimates primarily consider exposure through inhalation, which is a more immediate form of exposure, but pesticides can also seep into ground water and soil and lead to subsequent exposure. The CAREX Canada estimates do not account for dietary ingestion, which is a major route of pesticide exposure, especially for young children.[6]

In addition to those living near agricultural areas, the spouses and children of pesticide applicators are more likely to be exposed (termed para-occupational exposure) compared to the general population.[7] This is because 2,4-D may be brought into homes on the bottom of shoes, increasing potential for exposure especially in young children through dermal means and hand to mouth transfer.[8] This is an important exposure route for young children because they spend the majority of their time indoors, and because 2,4-D persists longer indoors than outdoors.[8] In children younger than 3 years of age, non-dietary exposure to 2,4-D accounted for 15% of total exposure.[9]

In the Canadian prairies, 2,4-D is the most frequently detected herbicide in rainfall; it is present in up to 93% of samples collected.[10] 2,4-D may enter the atmosphere when it volatilizes during and after spray applications, when drops of 2,4-D are transported by wind currents after spraying, and when 2,4-D adheres to dust and is incorporated in clouds during the cloud forming process.[11]

Data sources

Actual pesticide application rates were not available for Canada, so pesticide usage was estimated based on crop types grown and typical application rates for these crops in North America. Data were obtained on the distribution and acreage of different crop types, 2,4-D application rates, and the number of people living in each area. Data on agricultural crops were obtained from Agriculture and Agri-Food Canada’s Annual Crop Inventory for 2016, which are digital images (raster datasets) that produce an average crop accuracy of 89%.

Data on pesticide usage came from the United States Department of Agriculture’s Agricultural Chemical Use Program and the Government of Ontario’s Ministry of Agricultural, Food and Rural Affairs. These datasets summarize how much 2,4-D was applied to crops along with the average yearly application rates. Group averages were applied if specific crop data were not available. For instance, we averaged the 2,4-D application rates for pears, peaches, and apples to estimate the application rate for “orchard fruits”, which was a crop type captured by the Annual Crop Inventory. Average annual 2,4-D application rates for all crops were obtained.

Population data from Canada’s 2016 Census of Population were used to estimate the number of people potentially exposed. Population estimates are based on the number of people living within the census subdivision (CSD).

Methods and analysis

Estimates were calculated at the CSD level (n=5,054). Specifically, we calculated the total area (in km²) of land within each CSD for each crop type. Total areas were then multiplied by the average 2,4-D application rate data for each crop. Finally, the total was summed together to obtain an estimate of 2,4-D use in kilograms for each CSD. Results were weighted in square kilometres to illustrate the kilograms of 2,4-D applied per square kilometre in each CSD. CSDs with less than 0.2 kg/km² were classified as having no exposure to account for very small agricultural activity within the CSD. Data are presented in quartiles and integrated with population data to show potential exposure and variations across the country.

Exposure among these workers occurs during the mixing, loading, and application of 2,4-D, but many more may be exposed during other, farm-related activities through dermal contact with treated crops. Farm types with the largest number of exposed workers are other grain farming, dairy cattle and milk production (where 2,4-D is applied on crops that are grown to feed livestock), and fruit farming.

Workers in industries other than agriculture, including forestry, public utilities, and landscaping, may be at risk of exposure to 2,4-D.[2]

Data sources

Data used in developing the occupational estimates for 2,4-D were collected from several sources:

• The Census of Agriculture collects detailed information about Canada’s agricultural industry, and its farm operators and families (excluding contract workers). It surveys all farm operators in Canada. For our 2,4-D exposure estimates, we used data from 2016.
• The National Agricultural Workers Survey (NAWS) is an employment based, random sample survey of crop workers in the United States that monitors employment and demographic characteristics of hired crop workers. For our estimate, we use data from 2002 to 2016, for the Midwest and Northwest states.
• Canadian and US scientific peer reviewed publications that addressed pesticide exposure in Canada and the United States.
• Grey literature including technical reports from governments and international bodies.

Prevalence Estimate Method

Methods were developed with input from a Pesticides Advisory Committee (PAC). The PAC comprised experts from Canada and the United States in exposure assessment, occupational epidemiology, and geography. The finalized methods yield estimates of the number (and proportion) of exposed agricultural workers by industry (North American Industry Classification System, NAICS) and province.

Two estimates were calculated – one based on the 2016 Census of Population and one based on the Census of Agriculture. The Census of Population and Census of Agriculture survey the working population differently and as a result, the number of agricultural workers captured by the two surveys differs. The Census of Population only collects information on an individual’s main job, and as a result may not include part-time workers, seasonal workers, or temporary workers. The Census of Agriculture collects data on all farm operators and hired agricultural workers (excluding contract workers), regardless of their employment status. For this reason, the Census of Population-based results represent a lower estimate of exposure, while the Census of Agriculture-based results represent an upper estimate of exposure.

Generally, the methods can be broken down into four steps. First, we estimated the number of workers that would fall into each NAICS farm type. For the lower estimates (based on the Census of Population), we applied the distribution of workers by farm type from the Census of Agriculture to labour force data from the Census of Population. The Census of Agriculture collects information about hired labour (and not contract workers) for specific industries; only these workers were pulled from the Census of Population and incorporated in this first calculation. For the estimates based on the Census of Agriculture, we directly extracted data on the number of workers by NAICS farm type.

Second, we multiplied the proportion of farms applying “herbicides” by the total number of workers (by farm type) to identify workers at risk of exposure. This calculation was only completed for farm types that describe crops for which 2,4-D is registered (e.g. wheat farms).

Third, we assigned likelihood of exposure (i.e. exposed, probably exposed, possibly exposed) to these workers using the methods described in the following section.

Finally, to complete the agricultural estimates, we added workers that are not captured by the Census of Agriculture (e.g. contract workers within the agricultural industry) but are at risk of exposure. These workers were identified using the Census of Population, and their prevalence and likelihood of exposure were assessed independently using CAREX Canada’s occupational approach. Because the Census of Population has less detailed labour force data than the Census of Agriculture, results are summarized for higher-level sectors (e.g. for “crop production”) instead of for specific farm types.

Exposure Likelihood Method

Workers who were identified as being at risk of exposure were qualitatively assigned exposure likelihood (i.e. exposed, probably exposed, and possibly exposed) using information on crop type, primary tasks, application methods, crop production practices, and the propensity with which pesticide residues transfer from crop foliage to skin upon contact.

The primary route of occupational exposure to pesticides is through the skin; thus, tasks with increased potential for dermal contact were ranked as more likely to lead to exposure. All workers who reported applying, mixing, or loading pesticides in the past 12 months were considered exposed, while workers completing tasks that would bring them into contact with treated crops were considered probably exposed. All remaining workers were considered possibly exposed.

Category 1: Possibly exposed

• A group of workers is put into this category if they have been identified as at risk of exposure but do not fall under Category 2 or 3

Category 2: Probably exposed

• A group of workers is put into this category if their primary task may lead to dermal contact with treated crops

Category 3: Exposed

• A group of workers is put into this category if they are involved with loading, mixing, or applying pesticides

Interpretation of the results and limitations

Both estimates (those based on the Census of Agriculture and the Census of Population) assume that if a specific farm type (e.g. wheat farms) applies herbicides and 2,4-D is registered for use on the crop, that 2,4-D is indeed applied. In actuality, there are multiple pesticide products registered for use on a single crop. Typically, farmers will choose just one or sometimes several products to apply to the crop. While we cannot assess the impact of this assumption due to limited data on 2,4-D application rate by crop in Canada, 2,4-D is a commonly used pesticide and the estimates are not expected to be significantly over-estimated.

We have assumed that farm types only grow the crops that they describe (e.g., that soybean farms only grow soybeans). However, farm types are classified based on the predominant type of production (the crop, or crop type, that contributes most to the estimated receipts). By assuming that farm types are restricted to growing crops described by their classification, the total number of workers exposed may be underestimated. The degree to which this occurs is not likely to be quantifiable due to limited data on the types of crops grown together on specific NAICS farm types.

CAREX Canada estimates that the acetaldehyde levels in outdoor air do not result in higher risk of cancer at a population level (low data quality). However, results show that acetaldehyde levels in indoor air do result in an increased risk of cancer (moderate data quality).

Canadian studies in Prince Edward Island and Saskatchewan measured indoor air concentrations of several compounds, including acetaldehyde, and found that smoking was a significant source of acetaldehyde in homes. This was the case especially when air exchange rates were lower.[3,4] Cooking with oil was also associated with increased levels of acetaldehyde indoors.[4]

Acetaldehyde, along with several other volatile organic compounds, is released from common building materials such as engineered wood products typically used in manufactured homes. This creates potential for exposure to inhabitants of these homes.[5]

A review of acetaldehyde toxicity from the Government of California published in 2008 lists several studies from the last 15-20 years where measurements of acetaldehyde were taken in residences as well as portable classrooms.[6] According to this review, indoor measurements of acetaldehyde usually greatly exceed outdoor measurements.

Acetaldehyde is a metabolic intermediate in humans and other organisms. It is also found naturally in some foods in trace amounts, especially after cooking, ripening of fruit, or fermentation (i.e. in beer and wine).[7] The largest source of exposure to acetaldehyde in the general population is technically via alcohol consumption, where it is produced during alcohol metabolism. However, acetaldehyde is not carcinogenic via ingestion.

Searches of Environment Canada’s National Pollutant Release Inventory (NPRI) and the US Household Products Database yielded the following results on current potential for exposure to acetaldehyde in Canada:

NPRI and US Household Products Database

NPRI 2015[8]
Substance name: ‘Acetaldehyde’
Category	Quantity	Industry
Released into Environment	668 t	Pulp, paper and paperboard mills, veneer, plywood manufacture, chemical manufacturing (61 facilities)
Disposed of	0.024 t
Sent to off-site recycling	None

t = tonne

US Household Products 2016[9]
Results: 13 products
Search Term	Quantity	Product Type
‘Acetaldehyde’	13	Arts & crafts adhesives (5), home adhesives (5), landscape adhesive(1); powdered roof leveler (1); automotive products (1)

Our Environmental Approach page outlines the general approach used to calculate lifetime excess cancer risk estimates and includes documentation on our mapping methods.

Data sources and data quality for acetaldehyde can be found in the PDF below.

Supplemental data – Acetaldehyde [PDF]