Limitations of this review
Individual findings from each death were combined to allow review. The combined findings were described and reported to allow evaluation for potential patterns associated with fire fatalities in First Nation communities. When comparable findings from fire fatalities in non-First Nations communities were available these were presented to illustrate similarities or differences. However, since no complex statistical analysis was conducted there are limitations to drawing conclusions from this review.
The inability to obtain complete information over the 10-year timeline of the review is another limitation. Some information was not consistently captured, including demographic information and resources available at the time of the fire. This type of data was not part of the core information needed to complete death or fire investigations, e.g. population of First Nations communities at the time of fire.
Other sources of review included federal data which has its own limitations. The Indian Register was used to obtain estimates for the number of First Nations people living on-reserve in Ontario. These values may not reflect the actual number of individuals living in First Nations communities. The Register only includes individuals registered under the Indian Act and may not account for everyone living in the community.
Another limitation with respect to the data would be the age and status of a structure. Fire investigations by the OFM capture if a structure was built before or after 1975 (the year the Ontario building code came into effect). However, the data did not reflect whether or not a structure needed maintenance or major repairs.
The review only examined data from fatal fires. Information about non-fatal fires was not collected in this review. Factors relating to surviving a non-fatal fire could be important in informing approaches to preventing deaths. Analysis of non-fatal fires could be considered for future review.
The amount of information and the strength of the conclusions could be impacted by the limited sample size of fires (29) and fatalities (56) reviewed during this 10-year period. Findings and statistics should be interpreted carefully as the numbers could be impacted by misclassification. The statistics could also be impacted by circumstances where a significant proportion of responses to questions were unknown or undetermined. An example is the variation observed between First Nations and non-First Nations communities’ data during the time of day that a fatal fire occurred. In this review, non-First Nations communities see a continuous increase in fire fatalities from 9 p.m. to just before 6 a.m. Based upon the data available for analysis First Nations communities see an increase of fire fatalities from 9 p.m. and a decrease from 3 a.m. to just before 6 a.m. This review cannot determine if this difference is due to the sample size of the data set or other factors that affect First Nations communities during this time of day.
For the purpose of this review, First Nations communities were included into one of three groups based upon presence of road access to a service centre and involvement with a Municipal-Type Service Agreement This categorization was utilized to capture the concept of “remoteness” and accessibility to resources relevant to fire deaths which may be a potential limitation.
There were 29 fatal fires in 20 First Nations communities examined during this review. Another limitation of the review is that unique realities in each First Nations community may not have been identified, considered or reflected in the review. Examples include: population sizes ranged from approximately 30 to 12,750 and geographic land size ranged from less than 2 to 239 square kilometers. When reviewing these data, it is important to recognize the variability among the communities and to understand that multiple factors may impact each community in many different ways.