Carbon Monoxide Poisoning and Structure Fire Prevalence in British Columbia

Vancouver skyline

Current knowledge, data resources, and engagement strategies to mitigate technical systems risks

Courtney van Ballegooie, Wai Ho Choy, Abhijit Pandhari, and Ido Refaeli prepared this report as part of the 2018 consulting project contracted to Acada Consulting in partnership with Technical Safety BC.

For more information about this project, please contact:

Project Lead
Courtney van Ballegooie - Consultant
Email: cballegooie@bccrc.ca

Project Team/Authors
Wai Ho Choy - Consultant
Email: choywaiho1@gmail.com

Abhijit Pandhari - Consultant
Email: abhijitpandhari@gmail.com

Ido Refaeli - Consultant
Email: idorefaeli@gmail.com

Courtney van Ballegooie - Consultant
Email: cballegooie@bccrc.ca

DISCLAIMER

The information and views set out in this publication are those of the author(s) and do not necessarily reflect the official opinion of Technical Safety BC. The author(s) do not guarantee the accuracy of the data included in this study. Neither Technical Safety BC, the author(s), nor any person acting on their behalf may be held responsible or liable for the use which may be made of the information and recommendations contained herein.

© Acada Consulting Inc., 2018. All rights reserved. Reproduction is authorized provided the source is acknowledged.

Acknowledgments

We want to take this opportunity to thank the BC Office of the Fire Commissioner for their cooperation and help in providing raw ignition data from the last five years (2013-2017). We also want to thank Soyean and her team including Doris, Abraham, Jeff, Anja, Frank, Bo, Sue, and Gina who have wholeheartedly contributed their time, enthusiasm, and encouragement into this project. We could not have done this without you and we are very grateful to have such an open-minded and approachable client that is supportive of the young talent here in BC.

Who are we?

Acada Consulting is a graduate student-initiated business that began in 2018. Acada Consulting originated from the Graduate Consulting Program (GCP) through the Graduate Management and Consulting Association (GMCA). GCP provided a unique opportunity for graduate students and postdoctoral fellows to apply their critical thinking and communication skills to solve real-life business problems for companies, government agencies, and nonprofit organizations in BC. After the successful completion of the original 3-month GCP project, which ended in a successful client report and presentation, the four GCP members founded Acada Consulting in order to continue their pursuit of providing graduate students the opportunity to apply their skills in a business-oriented environment.

Table of Contents

Executive summary

  1. Introduction: Carbon Monoxide Poisoning and Structure Fire Background
  2. Objective  
  3. Timeline
  4. Module 1: Identifying Sources, Contributing Factors, & Data Collection Strategies
    1. Abstract
    2. Objectives
    3. Results
      1. Main Sources of CO and Fire Incidents
      2. Factors Contributing to CO Incidents
      3. Key Data Collection Strategies for CO
      4. Sources and Data Collection Strategies for Structure Fires
    4. Methods
      1. Identifying Sources and Contributing Factors of Hazards
      2. Verifying Main Sources and Contributing Factors
      3. Delineating Collection Strategies
      4. Refining and Compiling Data
  1. Module 2: Carbon Monoxide Preliminary Data Gathering and Outreach Implementation
    1. Abstract
    2. Objectives
    3. Results
      1. Preliminary Results
      2. Analysis
    1. Conclusion
    2. Methods
      1. Survey Design
      2. Survey Administration
      3. Data Analysis
  1. Report Conclusions and Future Directions 
Appendix

A: Supplementary Tables and Figures
B: Categorized Structure Fire Incidents
C: Structure Fire Raw Data
D: Survey Questionnaire
E: Survey Questionnaire Raw Data

Executive summary

Technical Safety BC’s Research & Analytics Team strives to combine primary and secondary data in order to better predict and monitor technical safety equipment. This, in turn, will result in more improved and automated systems that will prevent incidents and injuries from occurring. One of the main primary sources of data that Technical Safety BC collects is incident investigations that can be found in their annual incident report forms.  These reports allow Technical Safety BC to better understand the root causes of technical incidents. They then share the information and findings with clients, stakeholders, and the public. Although many incident investigations are complex and multifaceted, their investigations lead to lasting changes within the safety environment of British Columbia (BC). This can, for example, be seen in Technical Safety BC’s-led boiler investigation in 2016 which resulted in a change of global standards in the safety certification of boilers. 

Our first meeting with the Research & Analytics team of Technical Safety BC revealed that they would like to acquire more primary and secondary data sources to use in the identification of high risk and low compliance groups/regions. Our challenge, therefore, was to collect information about incidents which pose the highest risk to groups/regions within BC and identify the source of the problem. Over the span of 3 months, we divided this research task into three defined steps: 1) Identify the sector of technical safety which causes the most serious injuries, 2) Determine the major root causes why the injuries occurred, and 3) Distinguish potential strategies to better identify the high risk, low compliance groups. 

Our findings indicated that gas and electricity were the sectors that had the highest serious injuries reported by means of carbon monoxide (CO) poisoning and fire incidents, respectively. Reports within Technical Safety BC as well as literature found within Canada and other regions of the world indicated similar technical causes of CO poisoning. The main, non-vehicular sources of CO poisonings were related to heating equipment, such as furnaces, boilers, and water heaters.  Although contributing factors of CO poisoning are not as uniformly documented, a key theme of improper ventilation was identified in most provinces and countries investigated. During our investigation, we identified 5 key sources of information: Emergency Department Records, Inpatient Hospitalization Records, Death Certificates records, Poison Center Records, and Newspaper and Other Media Sources that provided the majority of data regarding CO incidents. 

Parallel to the CO poisoning investigation, fire incident investigation revealed (1) Of the 14,875 total recorded ignition cases from 2013-2017, up to 1,535 of the incidents with a known cause were within the jurisdiction of Technical Safety BC, (2) refrigeration equipment and portable generators, made up less than 5% of the 1,535 incidents, (3) the majority of permanent electrical distribution equipment ignition incidents were due to non-aluminum, permanent electrical wiring, (4) heating equipment made up the majority of ignition cases when only considering incidents that were caused by equipment within Technical Safety BC’s jurisdiction, and (5) there was no increasing or decreasing trend when looking at the number of incidents over the 2013-2017 time period. 

Next, we evaluated successful data collection and communication strategies regarding CO and fire safety in Canada. We identified four successful campaigns that had well documented methods and results. These included Take the Pledge-Sound the Alarm Campaign, Targeted Residential Fire Risk Reduction led by the Surrey Fire Department, Residential Structure Fires by BC Coroners Service, and Reducing Carbon Monoxide Risk in the Home by Technical Standard and Safety Authority. Based on the data and communication approaches found within the previously listed campaigns, we designed and conducted a survey within the Greater Vancouver area. Key findings included (1) the majority of persons either do not own or are unaware if they own a CO detector. Additionally, those who do not own a CO detector are uninformed as to where they are able to obtain a CO detector, (2) technical equipment maintenance may not be occurring as frequently as recommended, (3) some CO emitting appliances were more easily identifiable than others, and (5) future communication routes were preferentially identified as email, bill inserts, and social media. 

Given the potential applications of the data highlighted in this report, we recommend that Technical Safety BC pursues partnerships with prominent CO and Fire Safety Authorities in BC in order to gather and disseminate technical safety information. In doing so, Technical Safety BC would not only identify regions within BC that have a high risk and low compliance in CO and fire safety but also better understand the knowledge gaps in the technical safety environment with respect to CO poisoning and structure fires.

Introduction

Carbon Monoxide Poisoning: Characteristics & Prevalence in BC 

Total CO Hospitalizations In Canada By Province

CO is a poisonous gas due to its interference with the body’s ability to transport oxygen to cells and organs. CO is colourless, odourless, and tasteless. 1,2 CO’s adverse effects depend greatly on the level and duration of exposure. At low concentrations, symptoms of CO poisoning present flu-like symptoms: headaches, nausea, fatigue, shortness of breath, impaired motor functions, and dizziness. In higher concentrations or prolonged exposure, symptoms can include dizziness, chest pain, poor vision, difficulty thinking, loss of consciousness, convulsions, and even death. Although CO exists naturally in the atmosphere in small quantities, it is also a combustion by-product of carbon-based fuels such as oil, coal, wood, gasoline, propane, and natural gas. CO is also produced when the combustion of a carbon-based fuel is incomplete, usually due to insufficient oxygen during the combustion process. Consequently, it is vitally important to have properly ventilated areas for household equipment, such as boilers, furnaces, and heaters, in order to prevent the buildup of CO.1

Although mortality and hospital admission rates for unintentional nonfire-related carbon monoxide poisoning have decreased steadily in Canada over the past decades, BC still sits in second place for the highest total number of CO hospitalizations and has a higher than average annual rate than that found across Canada. 2,3 Unsurprisingly, CO poisonings in a residential/home environment have had the smallest magnitude of decrease when it came to mortality and hospitalization rates. 2 In fact, when missing data or unspecified data was removed from the analysis, 75% of CO-related hospitalizations occurred as a result of CO poisoning originating in a residential/home environment.3 This is likely due to approximately 60% of Canadians not having a CO detector in their homes.2 Although there is no legislation in BC requiring the ownership of a CO detector, it has been shown that CO detector ownership saves lives and has the ability to reduce CO-related deaths.3

The epidemiology of the mortality and morbidity from CO poisoning in Canada has begun to receive more attention over the last few years in an attempt to identify risk factors for those who may be in danger of CO poisoning.2 Notably, in terms of the total number of deaths, 34.8% occurred between the ages of 25 and 44 years old, and an additional 40% occurred between 45 and 64 years old. (Cohen, 2017) Males in Canada had the highest rates of carbon monoxide mortality and admission to hospitals. This could be due to men potentially having more opportunities to handle combustion appliances inappropriately or without appropriate ventilation.2 Lastly, CO poisoning mostly occurred from September to April with peaks during the winter period.2 This phenomenon is likely due to the blocking of proper ventilation during snow storms as well as the improper use of generators during power outages. 4 

As CO poisoning risk factors become more discernible in BC, it is within Technical Safety BC’s interest to identify the technical equipment involved in such incidents in order to mitigate and predict them. 

Structure Fire: Characteristics & Prevalence in BC 
Annual Fire Report Causes Of Fire 2014

Structure fires are fires that involve the structural components of residential and/or commercial buildings. These exclude fires such as chimney fires, vehicle fires, wildfires, and other outdoor fires. From 2002 to 2013, in BC alone, a total of about 7,500 fires occur annually, costing roughly $300 million and claiming about 35 lives each year.5 Over half of all fires are through accidental means with the top causes of fire being cooking equipment, matches and lighters (not for smoking), smoking material, electrical equipment, heating equipment, appliances and equipment, and internal combustion.5 Technical Safety BC has a stated interest in identifying where electrical equipment, heating equipment, and appliances and equipment caused fires are occurring across BC. 

Figure 2. Annual Fire Report, Causes of Fire 2014

Identifying groups at elevated risk of structure fires is well documented in comparison to CO poisoning in Canada. Across Canada, the groups at the highest risk of experiencing a residential structure fire incident include children under the age of 6, older adults over the age of 64, and those who are socio-economically disadvantaged..6,7 Interestingly, unlike other provinces, very young children in BC were indicated as not being at a higher risk for structure fire deaths.8 Older adults are at higher risk of dying from a residential fire due to their inability to hear and/or respond to an activated smoke alarm. The presence of a physical disability may be another contributing factor for older adults. Residents in low socio-economic areas are also at greater risk, primarily due to their tendency not to have a working smoke alarm in the home.6,7 Additionally, structure fire victims were shown to be primarily males, having a 1.5 times elevated risk relative to females. Persons of Aboriginal identity had a 4 times higher rate of residential fire death and were younger, on average, than non-Aboriginal victims. While the Northern region of BC had the lowest number of deaths (average of 3.8 deaths per year), it also had the highest rate of death (13.3 per million). Fraser, on the other hand, had the highest average number of deaths (approximately 7.6 deaths per year, tied with Vancouver Island) but the lowest rate of death (5.0 per million). Single-family dwellings account for the vast majority of fire incidents as well as injuries and fatalities, typically as a result of smoke inhalation or carbon monoxide poisoning.5-7 With so many well-defined characteristics, multiple fire prediction models have been generated such as the Dynamic Risk-Based Framework for Redesigning the Scheduling of Fire Safety Inspections and the Targeted Residential Risk Reduction Model. 9,10 

It is within Technical Safety BC’s interest to identify the technical equipment involved in structure fire incidents by utilizing data from agencies that are recognised to collect detailed structure fire incident data. This, in turn, will allow Technical Safety BC to temporally and spatially monitor technical equipment caused structure fire incidents.

Objectives 

What is the goal of this project? 

The goal of this project was to collect information about incidents which pose the highest risk to groups/regions within BC and identify the causes for low-compliance. Over the span of 3 months, we divided this research task into three defined steps: 1) Identify the sector of technical safety which causes the most serious injuries, 2) Determine the major root causes why the injuries occurred, and 3) Distinguish potential strategies in mitigating the risk of the high risk, low compliance groups.

Timeline 

What did we do? 

We used the first month to identify the sector of technical safety which causes the most serious injuries. We then did a provincial and global sweep of the literature to verify the root causes of CO poisoning. At this time, provincial fire data over the last five years (2013-2017) was being collected and processed. The second month consisted of identifying successful fire or CO campaigns and data collection methods across Canada with a focus on ones located in BC. In the third month, we used the information we gathered from the first two months and constructed a survey regarding technical systems awareness with respect to CO. This survey was then sent out around Vancouver and was analyzed. 

Throughout the three-month project, the project team met at least once a week for 2 hours. We also met with our BCSA client (~2 meetings/month) to update them regularly on our progress and to get their feedback on the direction of our work. 

Module 1: Identifying Sources, Contributing Factors, & Data Collection Strategies for CO and Structure Fires

Abstract

In order to understand CO and Structure Fires across BC, one must first understand and identify the sources and contributing factors of CO poisoning and ignition cases. Therefore, the first task is to locate these sources and contributing factors; the key sources for identification were the following: 1) Technical Safety BC’s annual incident reports, 2) Annual Statistical Fire Reports, 3) literature documenting investigational reports both within Canada and globally. It should be noted that countries outside of Canada will have different regulations regarding their technical equipment and will, therefore, impact the contributing factors of CO poisoning. However, we have used these sources to better understand whether similar contributing factors occur elsewhere, even if they are in dissimilar proportions.

In summary, we have attained the following:

  1. Located the two most significant contributors to major and severe hazards, electrical (797 incidents) and gas (134 incidents), accounting for 82% of all major and severe hazards.
  2. Identified key contributing factors for structure fires and CO poisoning
  3. Provided recommendations in data collection and partnerships to obtain more information regarding structure fire occurrences and CO poisoning
  4. Described our approach for collecting the data 

The value in this data comes from the better understanding it will develop regarding the sources and mechanisms of action of CO poisoning and structure fires. This module will also highlight the potential partnerships that Technical Safety BC could form in order to collect a greater body of CO poisoning and structure fire data. Therefore, the aim of this module it to, provide a glimpse into the sources and mechanisms of action of CO poisoning and structure fires within BC as well as to describe collaborations which would allow Technical Safety BC to better understand and monitor CO and structure fire incidents.

1.1 Objectives   

This first module will cover the key sources and contributing factors of CO poisoning and structure fire incidents as well as how we can better obtain data regarding these cases in BC. The main deliverables for this module will include:

  • a list of ignition cases based on location, year of incident, and ignition cause
  • a list of data collection methods including information that can be found within each source
1.2 Results
Table 1: Number of As-Found Hazard Assessments Recorded in 2016; Severe and Major cases
Hazard Ratings Amusement Devices Boilers, Pv, Refrigeration Electrical Elevating Devices Gas Passenger Ropeway Railways Asa Total
Severe 0 18 77 3 41 0 4 4 147
Major 7 97 720 42 93 1 14 15 989
Total 7 115 797 45 134 1 18 19 1136

Table 1. Indicates the number of Severe & Major As-Found Hazards within Technical Safety BC’s Jurisdiction in 201611 

1.2.1 The two key contributors to major and severe hazards regarding technical equipment incidents in BC include electrical and gas technical equipment. These comprise 82% of all major and severe hazard incidents.

Our initial investigation using Technical Safety BC’s annual reports discovered that 82% of all severe and major hazards are due to electrical and gas technical equipment in BC, 2016 (Table 1). Electrical incidents that had an identified cause were mainly due to fire in both severe, major, and moderate cases (Table 2-A). Gas incidents were comprised mostly of CO exposure in the severe and moderate cases (71.4% of all severe and moderate cases). The single major gas incident was due to a propane leak. Two fire incidents were also identified as a cause of a gas incident in both a severe and moderate case (Table 2-B). Technical Safety BC’s annual reports help to identify the sectors that have the most severe and major incidents as well as the incidents’ causes. The causes of the incidents should be further investigated to identify the technical equipment involved and the mechanism behind the incident.11 

Table 2-8: Severe, Major and Moderate Gas Incidents in 2016 with Identified Cause

Rating

Quantity

Description

Severe

2

CO Exposure

 

1

Fire Incident

Major

1

Propane Leak

Moderate

3

CO Exposure

 

1

Fire Incident

Table 2-A: Severe, Major and Moderate Electrical Incidents in 2016 with Identified Cause

Rating

Quantity

Description

Severe

4

Fire Incident

 

l

Component Failure

Major

3

Fire Incident

 

1

Arc Blast

Moderate

7

Fire Incident

 

1

Damaged Component

 

1

Arc Flash

 

1

Contact with Energized Equipment

Tables 2-A and -B. Indicate the number of Severe, Major, and Moderate As-Found Hazards within Technical Safety BC’s Jurisdiction in 2016 with an identified cause 11

1.2.2 Factors contributing to CO incidents often involve poor ventilation, lack of equipment servicing, equipment failure, and installation issues.
Factors Contributing To CO Incidents In 2016

Figure 3. Factors Contributing to CO incidents in 2016 provided by Technical Safety BC 12

Technical Safety BC has published a case study of CO incidents that were reported to and investigated by safety officers in BC between 2007 and 2015. (BC Safety Authority Case Study, 2016) Key findings from their investigation include: 1) of the 84 incidents, there were nine fatalities and 161 non-fatal injuries, 2) most CO-related incidents occurred from October to April, 3) the majority, 63%, of CO incidents occurred in a single family, primary residential environment, 4) furnaces, boilers, and water heaters accounted for 81% of confirmed CO cases, and 5) key contributing factors to CO incidents included ventilation issues, service/maintenance and/or equipment replacement, equipment or component failures, and installation issues.12 

It should be noted that Technical Safety BC does not regulate all fuels that produce CO; therefore, the occurrences recorded in the case study involving fuels not regulated by Technical Safety BC were not considered in the publication. Additionally, the review could only publish cases that were reported to Technical Safety BC and investigated by safety officers. In order to obtain a better understanding of CO sources and their contributing factors, we performed a literature search on publicly available data depicting the CO poisoning landscape in Canada and globally. 12 

  British Columbia
2007-2015
(n: 84)
Quebec
2005-2010
(n:851)

U.S.
2005-2015
(n:1,795)

U.K. & Ireland
1986-2011
(n:244)

Iran
2007-2009
(n:328)

Main Sources

Furnace (38%), Boiler (23%), & Water heater (20%)

Motor vehicles (30%), Heaters (28%), Motor appliances & tools (14,6%), & Fires (12.3%)

Gas furnaces, Water heaters, & Clothes dryers

Central heating & Boilers (48%), portable heater {15%), & Fitted fire (10%)

Water heaters (59%), Heating devices (25%), and boilers (7%)

Contributing Factors

Ventilation (26%), Servicing/maintenance(19%), Equipment/component failure (18%), & Instillation Issues (17%)

Ventilation (35%), Equipment/component failure (14.5%), Misuse (10%), Servicing/ maintenance (6%), & Installation issues (6%) Human error (54%), Equipment failure (40%) Blocked or leaking exhaust systems (majority) Installation issues (55·65%), Misuse  (0-30%) Defective device (<20%), Confined space (<10%)

Table 4: Sources & Contributing Factors of CO Incidents in Canada & Globally

A cross-Canada search for comprehensive and reliable publicly available reportings of CO incidents was carried out as described in the methods section. During this search, it was noted that most provinces do not have well defined CO incident reports that are publicly available. Part of the reason behind the non-uniform reporting of CO is due to CO incidents being reported to a number of different authorities. These can include: multiple health authorities, poison control centers, government agencies, and non-government technical safety regulatory agencies. Additionally, those reported to health authorities frequently contain personal identifiers, thereby breaching the patient’s confidentiality if improperly disclosed. This data, therefore, could not be used until all identifiers are removed. In Quebec, however, the adoption of new Ministerial Regulations for the Application of Public Health Act in 2003 requires mandatory reporting of health concerns that have a chemical origin (MADO-chemical) without the inclusion of personal identifiers.13 Their CO sources and contributing factors are summarized in table 4. It should be noted that, similarly to Technical Safety BC, the majority of cases occurred in a residential environment, including garages/residential workshops (61%).12,13

Table 4. Compiled sources and contributing factors of CO incidents both within Canada & Globally

Next, we searched globally for countries that had well documented incidents of CO poisoning. A summary from each country is as follows:

  • United States – Unintentional, non-fire-related CO poisoning is the second most common cause of non-medical poisoning deaths in the United States. Human error and equipment failure (a failure of process vessels, storage vessels, valves, pipes, pumps, or other equipment that allows the release of hazardous substances) are the primary contributing factors for most of the CO incidents accompanied by an injury (53.9% and 39.9% respectively). Many of the unintentional CO poisoning incidents occurred inside the home as a result of negligence and complacency regarding the maintenance of the most common sources of CO, such as gas furnaces, water heaters, and clothes dryers. Negligence of proper maintenance was therefore indicated as human error. Similar to Technical Safety BC findings, CO poisonings happen year-round but peak with increased use of generators, alternate warming appliances, and power tools after storms and disasters that cause power outages, mostly during the colder winter months. 14,15
  • United Kingdom & Ireland – Comparable to Technical Safety BC findings, the majority of incidents (50%) occurred in the home, along with the highest reportings of fatality (62%). Although the main sources of CO were boilers, portable heaters, and fitted fires (118, 37, and 24 incidents respectively), the incidents that caused the greatest number of fatalities were boilers, portable heaters, and water heaters (51, 29, and 16 incidents respectively). Cumulatively, central heating and water boilers comprised 48% of the total 244 incidents investigated. Once again, the autumn and winter months were indicated as having the highest number of incidents for similar reasons as those mentioned above. 16
  • Iran – The two largest studies which identified sources of CO poisoning conducted in Iran, as well as a meta-analysis of CO poisoning cases, all concluded that the major sources of CO poisoning included conventional heating and cooking appliances. 17-19 The study by Dianat and Nazari in particular identified the main sources of unintentional CO poisoning as water heaters (59%), heating devices (25%), and boilers (7%) and the root causes of incidents were faulty installation, defective devices, confined spaces/improper ventilation, and misuse. Interestingly, the mechanism underlying the cause of the incident varied greatly by the source of CO. For example, cooking devices had approximately 30% of their incidents caused by misuse while boilers had no cases attributed to misuse. Additionally, and as similarly documented by Technical Safety BC, most CO-related incidents occurred from October to April with the largest increase during the winter months. 17 

Although there are many aspects of each of the countries that cannot be generalized for use in British Columbia such as differences in their range of fuels, types of appliances, age distribution of the population, average education level, public policies of the country, and so forth, some trends can still be identified. One such trend is that the main sources of CO poisoning come from central heating. Other trends would include the time of year incidents occur, as well as similar contributing factors of CO poisoning. Although this module went into depth identifying one province and three countries which had comprehensive and reliable publicly available data regarding CO poisoning sources and mechanisms, studies from other countries such as Turkey,20 Australia, 21 Switzerland,22 Denmark, 23,24 Belgium, 25 France, 26 Taiwan, 27 and South Korea 28 have had similar findings to those stated above. 

CO-related incidents have gained worldwide attention and are now being studied in many countries. The true magnitude of CO-related injury and mortality, however, is predominantly unable to be evaluated due to different reporting schemes, lack of detailed reporting, and potential inaccuracy of the data within and between countries. In an attempt to clarify the magnitude and mechanisms of CO-related injury and mortality, the World Health Organization (WHO) conducted the largest known survey specifically for CO data collection from all 53 Member States of the WHO European Region from 1980 to 2008. Of the 53 members, 28 countries reported CO-related injuries and death but only 11 countries had more detailed interpretations on the causes of the deaths. These countries included Andorra, Austria, Bosnia and Herzegovina, Czech Republic, Germany, Hungary, Malta, Republic of Moldova, Slovenia, Sweden, and Switzerland. Their major conclusions are as follows: 1) men comprised the overwhelming majority of CO-related deaths (71%), 2) 55% of CO deaths were unintentional and 36% were due to accidental exposure such as faulty heating systems and gas appliances, unvented coal or wood combustion, misuse, etc., 3) 19% of deaths were due to structure fire related occurrences, 4) there was an increase in mortality in the elderly aged 65 years or older, and 5) the vast majority of cases occurred within the home. Overall, the data and annual CO-related death rates strongly suggest that CO is a serious public health challenge and poses similar death rates (2.2 per 100,000 people) as HIV/AIDS, skin cancer, and alcohol abuse. CO poisoning is a challenging public health issue to tackle due to the lack of monitoring and reporting, and limited available statistics, but outreach campaigns have the ability to create a lasting impact on the CO monitoring policies and residential environment. 29 

1.2.3 Key Data Collection Strategies for CO
Table 5: Key Data Collection Sources for CO Poisoning
  Emergency Department Records Inpatient Hospitalization
Records
Death Certificate Records Poison Center Records
Age, Sex, and Race Age & Sex Only Yes Yes Yes
Cause of Exposure Some Some Some Yes
location of Exposure No No Yes Yes
Date of Exposure Yes Yes Yes Yes
Confirmed Exposure Yes Yes Yes No
Data Available in a Standardized Database Yes Yes Yes Yes
When Data are Available Quarterly Quarterly Quarterly Real-Time

Table 5: Key data collection strategies for CO poisoning as detailed in Bekkedal et al., 2006 30 

Some of the most common disease and hazard surveillance tools for monitoring public health include: emergency department records, inpatient hospitalization records, death certificate records, poison center records, and newspapers. 30-34 A summary describing the features of the data that can be obtained from each is shown in table 5. It should be noted that no single source of CO poisoning data is able to determine the true burden of CO poisoning as indicated by the minimal overlap of CO incidents between the sources.30 Some constraints of the CO poisoning data sources are as follows:

  • Restricted availability of health records
  • Time and resources required to format the data
  • How often the data is updated/released
  • Confidence in the accuracy and representativeness of the data

With respect to the available CO poisoning data resources, there are pros and cons to each. Restricted availability of health records is most prevalent in emergency department, inpatient hospitalization, and death certificate records due to patient confidentiality. This restricted availability can be seen in their delayed release due to the time it takes to anonymize records.  Despite this anonymization, many records are still unavailable to the public. The time and resources required to format Poison control data, as a third party accessing the data, is one of the greatest concerns with Poison control data. Although the data is available in real time, it is in an audio recorded format which requires transcription to extract the features of interest.35 Additionally, the restricted access to poison control data is of concern. Unlike the US, Canada, excluding Quebec, does not produce an annual poison control center report (e.g., reports, advisories and alerts) on routine poison control calls. 35 Alternative methods that do not rely on external authorities and/or sources will be explored in module two.

1.2.4 Sources and Data Collection Strategies for Structure Fires
Ignition Cases Caused By Permanent Electrical Distribution Equipment From 2013 2017

Figure 6. Ignition Cases Caused by Permanent Electrical Distribution Equipment from 2013-2017

After our initial investigation in 1.2.1 that revealed fires as one of the most common causes of major and severe hazards, we performed a literature search and identified key agencies that documented fire reports in BC. From the interactions with the BC Office of the Fire Commissioner, we obtained raw data on all structure fire incidents that had occurred from 2013 to 2017, their ignition code identifiers, and the location code for each of the incidents (Appendix A). Key insights from the data included:

  • Of the 14,875 total recorded ignition cases from 2013-2017, up to 1,535 of the incidents with a known cause were within the jurisdiction of Technical Safety BC
  • Refrigeration equipment and portable generators, under the appliances and equipment section, made up less than 5% of the 1,535 incidents
  • The majority of permanent electrical distribution equipment ignition incidents occurred with non-aluminum, permanent electrical wiring (Figure 6)
  • Heating equipment made up the majority of ignition cases when only considering incidents that were caused by equipment within Technical Safety BC’s jurisdiction
  • There was no increasing or decreasing trend when looking at the number of incidents over the 2013-2017 time period (Figure 7)
Ignition incidents by source from 2013 to 2017

To conclude, we suggest that Technical Safety BC forms a partnership with the BC Office of the Fire Commissioner in order to collect structure fire incident information within their jurisdiction in real time. Although this data was based on previously documented incidents, it demonstrates the type of information Technical Safety BC would be able to obtain, for example (1) ignition source, (2) location of ignition incident, and (3) time of incident. Utilizing such information would allow Technical Safety BC to identify trends over time, location, and incident source.

Figure 7. Ignition Incidents by Source from 2013 to 2017

1.3 Module Conclusions & Future Recommendations

Collecting information regarding the sources and mechanisms of action of CO poisoning and structure fires is the first step in understanding the BC landscape. As collaborations and partnerships change, there is a need to update the method of data collection to better integrate the newly available data. Therefore, the aim of this module is to provide a glimpse into the sources and mechanisms of action of CO poisoning and structure fires within BC, as well as to describe collaborations which would allow Technical Safety BC to better understand and monitor CO and structure fire incidents.

Within module 1, it was demonstrated that gas and electrical equipment are the major contributors to severe hazard incidents. Gas and electrical incidents were further broken down by identifying the causes of both. It was revealed that the majority of incidents were caused by fires and CO. Next, we then identified the factors contributing to the occurrence of the incidents. For CO in particular, due to the lack of available data, we performed a cross Canada and global literature search. This search, including the documented 84 cases within BC, highlighted the need to better monitor and understand residential use and operation of central heating units, such as furnaces, boilers, and water heaters. Lastly, we described collaborations which could be formed, as well as the resultant combined data availability by each, which would aid Technical Safety BC in their monitoring endeavors.

In conclusion, the systematic identification of the most numerous and severe incidents with a subsequent description of the incident’s sources and contributing factors is the first step in better understanding the CO poisoning and structure fire landscape with the future intent of formulating a mitigation strategy.

1.4 Methods
1.4.1 Summary of methods

In summary, we identified the sources and contributing factors of severe and major hazards within the jurisdiction of Technical Safety BC. We were then able to verify the main sources and contributing factors through both a literature search and using publicly available fire statistics data. The literature search was performed by searching key terms using google search engines as described below, while the fire statistics data was extracted and refined utilizing tools such as Tabula and Excel. While currently an unautomated process, this data has the potential to gain automation in both its extraction, through partnerships or scheduled release dates, and formatting.

Summary of data collection method

Figure 8. Summary of data collection method

1.4.2 Identifying Sources and Contributing Factors of Severe and Major Hazards

To identify the sources and contributing factors of severe and major hazards within Technical Safety BC’s jurisdiction, we first sorted through data which was publicly available through their website (www.technicalsafetybc.ca). Navigating their webpage proceeded as follows:

  • Clicking the “Safety Data” tab on the website’s main banner
    1. Selecting “State of Safety Report”
      1. Using the dropdown menu to choose “State of Safety 2016” and clicking “Go”
        • Choosing “Download Full Report” on the right-hand side under the “State of Safety At A Glance” banner
      2. Scrolling to the bottom of the webpage and selecting “Carbon Monoxide Incidents (2007-2015)” under “Multi-Year Incident Case Studies”

From this search, we were able to identify:

  • That gas and electrical sectors had the greatest total number of major and severe incidents
  • Of those major and severe incidents, CO and fire made up the majority of the hazards
  • Specifically, for CO, contributing factors included ventilation, service, maintenance, and/or equipment replacement, equipment or component failures, installation issues, unqualified persons performing regulated work, unsafe use of equipment, poor air to gas ratio, and downdraft created by wind.
1.4.3 Verifying Main Sources and Contributing Factors

To verify main sources and contributing factors, we 1) performed a literature search and 2) reached out to experts in the field with the hopes of the formation of a collaborative partnership. These two actions occurred in parallel in order to uphold the proposed timeline.

To identify sources within Canada, we used the following keywords in google and google scholar search engines:

  • “[Canada or name of province/territory] Carbon Monoxide (or CO) poisoning”, “[Canada or name of province/territory] Carbon Monoxide (or CO) related-deaths”, “[Canada or name of province/territory] Carbon Monoxide (or CO) hospitalization”, “[Canada or name of province/territory] Carbon Monoxide (or CO) injuries”, “[Canada or name of province/territory] Carbon Monoxide (or CO) burden”, “[Canada or name of province/territory] Carbon Monoxide (or CO) sources”, and “[Canada or name of province/territory] Carbon Monoxide (or CO) contributing factors”

It should be noted that only government published data and scholarly journals were considered. The government and scholarly journals were then compiled and the data of interest was manually extracted by reading each data source. This process was then repeated to include global populations by removing the statement [Canada or name of province/territory] from each search entry. Structure fire data was acquired in the same fashion but had the term Carbon Monoxide (or CO) or Carbon Monoxide poisoning replaced with the term structure fire (or fire).

Agencies that were contacted via email and/or by phone included:

  • British Columbia Injury Research and Prevention Unit including researchers in the Surrey Fire Department
  • BC Poison Control Center
  • BC Coroners Service
  • Hawkins-Gignac Foundation for CO Education
  • BC Office of the Fire Commissioner

These engagements resulted in three responses, including one which informed us that CO poisoning was not a priority of that particular agency at this time. From the two agencies that voiced a positive response, the BC Office of the Fire Commissioner and the Surrey Fire Department, we were able to obtain a more detailed record of the sources and contributing factors of structure fires. The formatted data can be seen in Appendix A.

1.4.4 Delineating collection strategies

Once we had identified the major sources and contributing factors of CO and structure fires, we extracted collection strategies from the sources found in section 1.4.3. This part of the module was only pursued for CO poisoning as the partnerships that were created with respect to structure fires were able to provide us with the necessary data regarding structure fires across BC. The sources listed in 1.2.3 were later considered in module 2 when further identifying campaigns/methods in collecting information with respect to CO poisoning.

1.4.5 Refining and Compiling Data

Raw data provided by the BC office of the Fire Commissioner from 2013 to 2017 was compiled and refined by excluding fires that were caused by sources outside of Technical Safety BC’s jurisdiction as well as merging the types of fires over the five-year period for each of the cities/districts/communities.

Module 2: Carbon Monoxide Preliminary Data Gathering and Outreach Implementation

Abstract

After the initial agencies were contacted for collaboration, detailed in 1.4.3, we identified other methods of CO data collection including census data, surveys, meetings/think-tanks of experts, and Internet of Things (IoT) devices for CO monitoring. After a thorough literature search, a survey was chosen as a data collection method due to (1) the non-specificity/granularity of census data within cities and communities,6,7 (2) the availability of literature detailing previously held meetings/think-tanks of CO experts with their subsequent conclusions and,36 and (3) the installation, cost, and number of anomalies that could be collected within the given timeframe for IoT devices.37,38 Technical Safety BC’s ability to communicate and disseminate technical information to those who need it is limited and makes (1) raising technical safety awareness and (2) implementing safety measures in residential areas that are found to be at high risk, a challenging task. In this module, we report findings that were derived from the dissemination of a CASPER-based survey questionnaire that was primarily formatted from three successful data collection and awareness campaigns in Canada, including “Take the Pledge-Sound the Alarm”, “Reducing Carbon Monoxide Risk in the Home”, and “TSSA CO Research Executive Summary”. 33,39-43  In doing so, we hope to (1) design a questionnaire that Technical Safety BC will be able to use to generate a sufficient amount of data over time, (2) disseminate the survey and analyze the initial data thereby generating pilot results in order to better tailor the survey before further use.

Briefly, our findings indicate that:

  1. The majority of persons either do not own or are unaware if they own a CO detector. Additionally, those who do not own a CO detector are uninformed as to where they are able to obtain a CO detector.
  2. Technical equipment maintenance may not be occurring as frequently as recommended.
  3. Some CO emitting appliances were more easily identifiable than others.
  4. Future communication routes were preferentially identified as email, bill inserts, and social media.
2.1 Objectives

This second module will cover the outreach strategy that was employed to identify characteristics of persons unaware of the danger of CO as well as how we can better obtain data regarding technical equipment safety and practices in a residential setting. The main deliverables for this module will include:

  • a CASPER-based survey questionnaire formatted from successful campaigns
  • initial insights and analysis of preliminary survey data
2.2. Results
2.2.1 Preliminary Results:

The survey was divided into the following four sections:

  • Section 1: Demographic Information (Q# 1-7; Figure 10)
  • Section 2: CO Awareness (Q#8-13; Figures 9 & 11)
  • Section 3: Knowledge About CO Sources (Q#14-17; Figure 9 & 11)
  • Section 4: CO Detectors & Maintenance (Q#18-27; Figure 11)
  • Section 5: Acquiring Further Information (Q#28-36; Figure 11)

For a complete list of the survey questionnaire, see Appendix A.

In Section 1, respondents were asked to fill out basic demographic and identifying information (Figure 10). In Section 2, respondents were asked a series of questions aimed at gauging their awareness level of the properties of CO (i.e., its color and odor), its risks and negative effects on health, and the symptoms of CO exposure, as well as comprehension of depressurization and its causes. Using this section of the survey, we were able to correlate respondents’ knowledge about CO with basic demographic data to identify a subpopulation characterized by lower rates of education and lack of CO knowledge (see our analysis of Group 2).

Existing Appliances With The Capacity To Emit CO Tracked

In Section 3, respondents were asked about the potential sources of CO in their homes. This section was, in fact, remarkable in its ability to help us track the potential sources of CO in residential neighborhoods in Vancouver (Figure 9). Lastly, Sections 4 and 5 contained questions about homeowner practices and acquiring further information (Figure 11). Specifically, respondents were asked questions such as whether they own a CO detector, whether they know where to acquire one, whether their detector is battery-powered or wall-plugged, how often they schedule a maintenance, etc. Additionally, respondents were asked if they are open to receiving more information about CO, and if so, by what method of communication.

Fig. 9: Existing appliances with the capacity to emit CO tracked. The type of CO-emitting appliance was tallied and plotted as a histogram showing the relative prevalence of each type of appliance in respondents’ homes in Vancouver, BC.

Existing appliances with the capacity to emit CO tracked

Fig. 10: Respondents’ Demographic Data. Demographic information on (A) homeowner status, (B) highest level of education, (C) current location of residence, and (D) age was collected from 42 respondents in Vancouver, BC. The data were plotted as pie charts for each demographic, and each of the possible response types is provided to the right of their respective panel.

Respondents Demographic Data

Fig. 11: CO detectors in respondents’ homes. Respondents’ answers to questions regarding (A) whether they own a CO detector, (B) whether they know how to acquire a detector, (C) how frequently they check their detector‘s functionality, (D) how often they schedule a maintenance  for any technical appliance, (E) if they would like to receive more information, (F) their preferred method of communication with Technical Safety BC.

2.2.1 Analysis:

We began our analysis with tenants that have lived in their current homes for 5 years or less as this demographic encompassed the bulk of respondents. Of the 42 survey respondents, 37 (88%) were tenants. 28 of the 37 tenants (76%) lived in their homes for 5 years or less, had an approximate male-to-female ratio of 17 to 10, and showed no preferential correlation with any specific neighborhood in Vancouver, BC (Group 1). Of this subpopulation, 100% ranged between ages 18 to 35, 79% (22 of 28) were Caucasian, and 89% (25 of 28) had a University level education (undergraduate and/or graduate).

Sample Group 1 demonstrated knowledge about the risks and negative health effects of CO. For example, 100% were correct in stating that CO is a colorless gas and 82% were correct in stating that it was odorless. Furthermore, the majority of Group 1 were aware of the health effects of CO (100%), the minimum exposure time that will elicit symptoms (89%), and the many home appliances that can potentially emit poisonous CO gas (100%). This subpopulation understood even more complex concepts related to CO with 19 of 28 (68%) demonstrating a correct understanding of the mechanism of depressurization and its causes.

79% of Group 1 have never scheduled a maintenance checkup for technical appliances in their home, with the remaining 21% stating only sporadically scheduling maintenance or solely when a particular piece of equipment is malfunctioning. On average, however, individuals belonging to this sample group do concede that maintenance should be scheduled regularly. Shockingly, only 25% of this subpopulation own at least one CO detector, while the remaining 75% either do not own one or are unsure of its presence in their home. Some Group 1 respondents displayed knowledge of whether their detector is wall-plugged or battery-powered. Strikingly, 86% of Group 1 do not regularly check their detectors for functionality. Encouragingly, however, they demonstrated a clear knowledge of how to respond to a CO leak in the event that a detector sounds the alarm.

Perhaps the most informative insight is the fact that 50% of this subpopulation were not aware of how to acquire a CO detector, and 100% were not aware that they can get CO-related information from Technical Safety BC. Furthermore, 64% of Group 1 were open to receiving further information from Technical Safety BC on how to mitigate the technical risks associated with CO exposure either via email, social media, or by receiving inserts in their bills. Lastly, when asked to rate their self-perceived level of knowledge about CO risks, this group rated itself 5.7 out of 10 on average, with 1 being the least knowledgeable and 10 being the most knowledgeable. Similarly, when asked how concerned they are about being exposed to CO within their lifetime, they rated themselves 4.5 out of 10 on average, with 1 being the least concerned and 10 being the most concerned.

Because Group 1 were relatively young and exhibited higher levels of education (89% had a university-level degree), we focused next on a new subpopulation characterized by relatively lower levels of education (Group 2) and compared their CO literacy to that of Group 1. We filtered the data to select for only the respondents whose highest achieved levels of education were high school, community college, or neither (i.e. blank) leaving us with a sample size of 5 respondents ranging from 18 to 65 years of age, of which 40% were male and 60% were female. Before analyzing the data, we find it prudent to mention that any conclusions made based on this sample group should be taken with a caveat as the sample size precludes us from being able to draw any statistically significant insights. Nonetheless, we feel it is important to perform a mock analysis on this sample in order to demonstrate the usefulness of surveys in gathering data over the long-term.

Group 2, similar to Group 1, demonstrated a basic understanding of CO, as 100% of this group were correct in stating that it is a colorless gas and 80% correctly stated that it is odorless. However, when asked what is the minimal amount of exposure time that will elicit symptoms, 40% of respondents gave incorrect answers demonstrating a knowledge gap regarding the risks and health effects of CO in lower-educated cohorts. Concerning more challenging concepts, 80% did not understand depressurization and the variables affecting it. However, we were encouraged to note that this group of respondents did demonstrate an adequate understanding of the sources of CO in the home, with 100% demonstrating a comprehensive understanding of the types of appliances that can emit CO. We further found that 80% of this subpopulation either does not own (n = 3) or is unsure if they own (n = 1) a CO detector. Similar to Group 1, 100% of group 2 respondents were not aware that they could acquire information about the risks of CO from Technical Safety BC. Furthermore, they were open to receiving more information about CO via email, social media, or through inserts in their bills.

2.3 Module Discussion and Conclusions

In this module, we deployed a CASPER-based survey questionnaire aimed at understanding respondents’ general awareness about CO risks. Although a larger sample size is required to draw statistically significant conclusions, our current data demonstrates the potential utility of a survey method in gauging CO awareness.

The majority of respondents in this survey fell under a similar umbrella of demographics; they were mostly educated tenants living in Vancouver between the ages of 18 and 35 (Group 1). Indeed, we found our survey population to be lacking diversity and any conclusions made in this report should be taken with the caveat that it may be biased towards one respondent population. Future iterations of this survey should be administered to more diverse populations so that the results may be more representative of the general Vancouver, BC population as well as including other cities and communities of interest.

Our first major finding shows that there is a significant difference between Group 1’s and Group 2’s CO literacy (Fig. 12; P = 0.0058). Group 1 (highly educated) demonstrated strong knowledge of the health risks and side effects of CO, as well as awareness of the gas’s molecular characteristics and the concept of depressurization. Conversely, while Group 2 understood the basic characteristics of CO, they did not accurately characterize depressurization and the health risks associated with CO. While the Groups 1 & 2 exhibited these significant differences in means, it is important to appreciate that the correlation coefficient, R-squared, equaled 0.1811, indicating that the model explains some, but not all of the variability of the response data. To increase the robustness of this model, it would be prudent to repeat the survey and increase the number of respondents in Group 2 to enable a larger sample representation in the analysis.

Comparison of CO Literacy

Fig. 12: Comparison of CO literacy between groups 1 and 2. Subjects were asked 10 questions (see Qs #8-17, Appendix D) to test their knowledge of the characteristics of CO, the risks of CO exposure and the potential sources of CO in the home. A score out of 10 was tallied, and an unpaired student’s t test was conducted to test for a statistically significant difference between the two groups. Data is reported as mean +/- SD.

Our second major finding was that neither Group 1 nor Group 2 often schedule maintenances. In fact, they only do so when a piece of equipment is likely to malfunction. Because chronic exposure to CO often goes unnoticed, this can potentially be life-threatening. Knowing that 88% of our respondents are tenants, and that neither the tenant nor the landlord is held accountable by law to conduct regular maintenance of household appliances, perhaps Technical Safety BC can instill an active mechanism that incentivizes regular maintenance checkups in high-risk homes. Alternatively, an IoT strategy that involves collecting data from household appliances in high-risk areas may also be a viable method to track the level of safety in residential housing.

Perhaps the most interesting finding was that a large portion of respondents from both groups did not know where to acquire a CO detector. Furthermore, 100 % of respondents were unaware that they can acquire information about CO safety from Technical Safety BC. We find this to be a point of great return on investment for Technical Safety BC. Indeed, it has been shown that the benefit to cost ratio of supplying CO detectors to homes in need was about 7:1. This calculation, performed by Iqbal et al.44, involved dividing the total benefit (a calculation which incorporated the total medical cost averted, total non-health sector cost, and total deaths averted) by the total cost (annual CO detector cost x number of households). Therefore, we believe that if Technical Safety BC were to fund the provision of CO detectors for at-risk households in BC, this would dramatically decrease the risk of CO incidents in the lower mainland demonstrating a high return on investment.

To conclude, this survey helped identify potentially novel vulnerable populations at risk for CO poisoning, as well as track the sources of CO in single, detached homes in the lower mainland. Although surveying methods are not an efficient mode for big data generation in the short term, we believe that building on this survey questionnaire and administering it over the long term will help Technical Safety BC garner large, meaningful datasets over time.

2.4 Methods
2.4.1 Summary of methods

Summary of methods

In summary, we have designed, administered, and analyzed a CO survey that has collected information regarding (1) demographics, (2) CO awareness, (3) knowledge of CO sources, (4) CO detectors and maintenance, and (5) additional understanding and further communication. The method for designing, administering, and analyzing the proposed CO survey is briefly described in the following section.

Figure 10. Summary of survey implementation method

2.4.2 Survey Design

We designed our survey using the Community Assessment for Public Health Emergency Response (CASPER) toolkit guidelines developed by the Centers for Disease Control and Prevention, Atlanta, GA. CASPER questionnaires are excellent rapid assessment tools used by public health officials and emergency managers to quickly assess the needs of their community.42,43 We deployed this method because it was specifically developed to be administered to individual household members within a community. The survey questions themselves were primarily formatted from three successful data collection and awareness campaigns in Canada, including “Take the Pledge-Sound the Alarm”, “Reducing Carbon Monoxide Risk in the Home”, and “TSSA CO Research Executive Summary”.36,39-41 These questions were then altered using the CASPER toolkit in order to (1) use more neutral and/or better-received wording and (2) better comply with an online survey format.

2.4.3 Survey Administration

We administered the survey to 42 participants residing exclusively in Vancouver, BC. The participants were chosen based on accessibility to Acada Consulting Inc. We did not discriminate by age, gender, level of education, or any other demographic, though the survey was administered at random to the personal contacts of Acada Consulting Inc. via Google Forms. Therefore, the respondents assessed by this survey are likely not a representative sample of the Vancouver, BC population.

2.4.4 Statistical Analysis

Statistical analysis was performed using either Excel or Prism 7 software. For Fig. 12, an unpaired student’s t-test was conducted using Prism 7 and the data reported as mean +/- SD. Pie charts reported in this module are representative of the raw data, and hence no statistical analysis was necessary.

Report Conclusions and Future Directions

In this report, we have identified the sectors that had the most severe and major hazards.  We then determined how the accident had occurred through Technical Safety BC reports. Next, we discovered the major sources and mechanisms of action that lead up to the occurrence of an event. And, finally, we identified partnerships as well as autonomous actions that Technical Safety BC could take in order to better understand the CO and structure fire landscape in BC. Together, Modules 1 and 2 describe a method that can be used to (1) identify and understand the causes and mechanisms of action of severe and major hazards, and (2) engage communities and organizations in order to generate a greater body of knowledge regarding CO and structure fire occurrence.

In conclusion, we recommend that Technical Safety BC reach out to the agencies identified in Module 1 which they feel would have a mutual benefit in forming a collaboration. Additionally, the methods described in Module 2 will allow Technical Safety BC to obtain more tailored data regarding specific technical equipment information they wish to gather from residents. These complementary data partnerships and acquisition tools allow Technical Safety BC to gather both statistically significant and personalized data on each community and/or subpopulation of interest, thereby creating a more comprehensive understanding of the technical safety environment with respect to CO poisoning and structure fires. We also recommend that Technical Safety BC facilitate further survey dissemination methods through social media and/or geofencing efforts in order to acquire information on communities/areas of interest.45

We sincerely thank Soyean Kim and the rest of Technical Safety BC for this incredible opportunity.

References

  1. Government of Canada. (2017). Prevent Carbon Monoxide Poisoning. Retrieved from https://www.canada.ca/en/health-canada/services/air-quality/indoor-air-contaminants/keep-carbon-monoxide-out-your-home.html
  2. Lavigne, E., Weichenthal, S., Wong, J., Smith-Doiron, M., Dugandzic, R., & Kosatsky, T. (2015). Mortality and hospital admission rates for unintentional nonfire-related carbon monoxide poisoning across Canada: a trend analysis. CMAJ Open, 3(2), E223-30. https://doi.org/10.9778/cmajo.20140122
  3. Cohen, I., Garis, L., Rajabali, F., & Pike, I. (2017). Carbon Monoxide Poisoning Hospitalizations and Deaths in Canada. Retrieved from https://cjr.ufv.ca/wp-content/uploads/2017/10/Carbon-Monoxide-2017-Final-.pdf
  4. Johnson-Arbor, K. K., Quental, A. S., & Li, D. (2014). A comparison of carbon monoxide exposures after snowstorms and power outages. American Journal of Preventive Medicine, 46(5), 481–486. https://doi.org/10.1016/j.amepre.2014.01.006
  5. England, E., Nieman, T. (2002-2013) Annual Statistical Fire Report. Retrieved from https://www2.gov.bc.ca/gov/content/safety/emergency-preparedness-response-recovery/fire-safety/fire-reporting/annual-reports
  6. Garis, L., Hughan, S., Maxim, P., & Tyakoff, A. (2016). Targeted Residential Fire Risk Reduction A Summary of At-Risk Aboriginal Areas in Canada. Retrieved from https://www.surrey.ca/files/Canada Aboriginal (FINAL RESEARCH STUDY) - Targeted Oct 2016.pdf
  7. Garis, L., Hughan, S., & Mccormick, A. (2016). Canada-Targeted Residential Fire Risk Reduction-formatted. Retrieved from https://www.surrey.ca/files/Canada-Targeted Residential Fire Risk Reduction-formatted.pdf
  8. BC Coroners Service. (2007). BC Coroners Service statistical report - Residential Structure Fire Deaths in BC, 2007-2011. Retrieved from https://www2.gov.bc.ca/assets/gov/public-safety-and-emergency-services/death-investigation/statistical/fire.pdf
  9. Garis, L., & Clare, J. (2014). Dynamic Risk-Based Framework for Redesigning the Scheduling of Fire Safety Inspections. Retrieved from https://www.surrey.ca/files/RiskAssessmentToolUFVResearchNoteFullReport.pdf
  10. Garis, L., Hughan, S., & McCormic, A. (2016). Targeted Residential Risk Reduction Model: A Summary of At-Risk Areas in Canada. Retrieved from https://cjr.ufv.ca/targeted-residential-fire-risk-reduction-summary-risk-areas-canada/
  11. British Columbia Safety Authority. (2016). State of Safety Report. Retrieved from https://www.technicalsafetybc.ca/sites/default/files/sos-2016_toc_may15.2017.pdf
  12. British Columbia Safety Authority. (2015). Carbon Monoxide Incidents (2007-2015). Retrieved from https://www.technicalsafetybc.ca/sites/default/files/sos2016_casestudy_co_final.pdf
  13. Mercier, M., Arcand, R., Canuel, M., Laplante, O., Palmieri, S., Venables, H., Bherer, L., Deshaies, P., Lebel, G., & Sanfacon, G. (2011). Surveillance des maladies à déclaration obligatoire au québec maladies d’origine chimique ou physique rapport des maladies et intoxications d’origine chimique déclarées. Retrieved from http://publications.msss.gouv.qc.ca/msss/document-000513/?&txt=2012
  14. Mukhopadhyay, S., Hirsch, A., Etienne, S., Melnikova, N., Wu, J., Sircar, K., & Orr, M. (2018). Surveillance of carbon monoxide-related incidents - Implications for prevention of related illnesses and injuries, 2005-2014. The American Journal of Emergency Medicine, 0(0). https://doi.org/10.1016/j.ajem.2018.02.011
  15. Sircar, K., Clower, J., Shin, M. K., Bailey, C., King, M., & Yip, F. (2015). Carbon monoxide poisoning deaths in the United States, 1999 to 2012. The American Journal of Emergency Medicine, 33(9), 1140–1145. https://doi.org/10.1016/j.ajem.2015.05.002
  16. Fisher, D. S., Bowskill, S., Saliba, L., & Flanagan, R. J. (2013). Unintentional domestic non-fire related carbon monoxide poisoning: Data from media reports, UK/Republic of Ireland 1986–2011. Clinical Toxicology, 51(5), 409–416. https://doi.org/10.3109/15563650.2013.786833
  17. Nazari, J., & Dianat, I. (2011). Characteristics of unintentional carbon monoxide poisoning in Northwest Iran-Tabriz. International Journal of Injury Control and Safety Promotion Article, 18(4), 313–320. https://doi.org/10.1080/17457300.2011.589006
  18. Hosseininejad, S. M., Aminiahidashti, H., Goli Khatir, I., Ghasempouri, S. K., Jabbari, A., & Khandashpour, M. (2018). Carbon monoxide poisoning in Iran during 1999-2016: A systematic review and meta-analysis. Journal of Forensic and Legal Medicine, 53, 87–96. https://doi.org/10.1016/j.jflm.2017.11.008
  19. Sheikhazadi, A., Saberi Anary, S. H., & Ghadyani, M. H. (2010). Nonfire carbon monoxide-related deaths: a survey in Tehran, Iran (2002-2006). The American Journal of Forensic Medicine and Pathology, 31(4), 359–363. https://doi.org/10.1097/PAF.0b013e3181f23e02
  20. Akköse, Ş., Türkmen, N., Bulut, M., Akgöz, S., İşcimen, R., & Eren, B. (2010). An analysis of carbon monoxide poisoning cases in Bursa, Turkey (Vol. 16). Retrieved from http://applications.emro.who.int/emhj/V16/01/16_1_2010_0101_0106.pdf
  21. Australian Bureau of Statistics. Causes of Death, Australia, 2016. (2017). Retrieved from http://www.abs.gov.au/ausstats/abs@.nsf/0/47E19CA15036B04BCA2577570014668B?Opendocument
  22. Gujer, H. R. (1982). Accidental carbon monoxide poisoning caused by incomplete combustion of liquid gases. Soz Praventivmedizin, 27(1), 39–42. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/7072383
  23. Taudorf, K., & Michaelsen, K. F. (1983). The danger of carbon monoxide poisoning from gas water heaters. A study of 124 systems and their users. Ugeskrift Laeger, 145(46), 3593–3598. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/6649148
  24. Thomsen, J. L., & Kardel, T. (1983). Accidents caused by gas water heaters. Fatalities and a non-fatal case. Ugeskrift Laeger, 145(46), 3598–3600. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/6649149
  25. Milis, L., & Lagasse, R. (1989). Carbon monoxide poisoning in the Brussels metropolitan area. Home survey technics and proposals for action. Arch Belg, 47(1–4), 24–28. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/2610566
  26. Gajdos, P., Conso, F., Korach, J. M., Chevret, S., Raphael, J. C., Pasteyer, J., … Chastang, C. (1991). Incidence and Causes of Carbon Monoxide Intoxication: Results of an Epidemiologic Survey in a French Department. Archives of Environmental Health: An International Journal, 46(6), 373–376. https://doi.org/10.1080/00039896.1991.9934405
  27. Hung, D.-Z., Deng, J.-F., Yang, C.-C., & Jen, L.-Y. (1994). The Climate and the Occurrence of Carbon Monoxide Poisoning in Taiwan. Human & Experimental Toxicology, 13(7), 493–495. https://doi.org/10.1177/096032719401300707
  28. Kim, Y. S. (1985). Seasonal variation in carbon monoxide poisoning in urban Korea. Journal of Epidemiology and Community Health, 39(1), 79–81. http://dx.doi.org/10.1136/jech.39.1.79
  29. Braubach, M., Algoet, A., Beaton, M., Lauriou, S., Heroux, M.-E., & Krzyzanowski, M. (2012). Mortality associated with exposure to carbon monoxide in WHO European Member States. Indoor Air, 23, 115-125. https:// org/10.1111/ina.12007
  30. Bekkedal, M., Sipsma, K., Stremski, E. S., Kristen, Malecki, K. C., Anderson, H. A. (2006). Evaluation of Five Data Sources for Inclusion in a Statewide Tracking System for Accidental Carbon Monoxide Poisonings. Wisconsin Medical Journal, 105(2), 36-40. Retrieved from https://www.wisconsinmedicalsociety.org/_WMS/publications/wmj/pdf/105/2/36.pdf
  31. Clifton, J. C., Leikin, J. B., Hryhorczuk, D. O., & Krenzelok, E. P. (2001). Surveillance for carbon monoxide poisoning using a national media clipping service. The American Journal of Emergency Medicine, 19(2), 106–108. https://doi.org/10.1053/ajem.2001.19999
  32. Creswell, P. D., Vogt, C. M., Wozniak, R. J., Camponeschi, J., Werner, M. A., & Meiman, J. G. (2017). Improving Outreach and Surveillance Efforts Following a Large-Scale Carbon Monoxide Poisoning in Wisconsin. Journal of Public Health Management and Practice: JPHMP, 23(5), S67-S71. https://doi.org/10.1097/PHH.0000000000000583
  33. Ghosh, R. E., Close, R., Mccann, L. J., Crabbe, H., Garwood, K., Hansell, A. L., & Leonardi, G. (2015). Analysis of hospital admissions due to accidental non-fire-related carbon monoxide poisoning in England, between 2001 and 2010. Journal of Public Health, 38(1), 76-83. https://doi.org/10.1093/pubmed/fdv026
  34. Verrier, A., Daoudi, J., Gourier, C., & Salines, G. (2010). Carbon monoxide poisoning surveillance: a French Environmental & Health surveillance system integrated in preventive policies. Injury Prevention, 16, A263-A289. https://doi.org/10.1136/ip.2010.029215.936
  35. Durigon, M., Elliott, C., Purssell, R., & Kosatsky, T. (2013). Canadian poison control centres: preliminary assessment of their potential as a resource for public health surveillance. Clinical Toxicology, 51, 886-891. https://doi.org/10.3109/15563650.2013.841182
  36. Thorne, S., Butte, G., & Hailey, S. (2017). Saving Lives from a Silent Killer: Using Mental Modeling to Address Homeowners’ Decision Making About Carbon Monoxide Poisoning. In Mental Modeling Approach (pp. 195–221). New York, NY: Springer New York. https://doi.org/10.1007/978-1-4939-6616-5_12
  37. Yang, J., Zhou, J., Lv, Z., Wei, W., & Song, H. (2015). A Real-Time Monitoring System of Industry Carbon Monoxide Based on Wireless Sensor Networks. Sensors, 15(11), 29535–29546. https://doi.org/10.3390/s151129535
  38. Pfeifer, K. B., Flemming, J., Sinclair, M. B., & Blair, R. G. (2007). Detection of Carbon Monoxide (CO) as a Furnace Byproduct Using a Rotating Mask Spectrometer. Retrieved from https://prod.sandia.gov/techlib-noauth/access-control.cgi/2007/070667.pdf
  39. Peterborough Fire Services. (2013). Take the Pledge Sound the Alarm. Retrieved From http://www.peterboroughfireservices.ca/home.htm
  40. Technical Safety Standards Association. (2002). Reducing Carbon Monoxide Risk in the Home. Retrieved from https://www.tssa.org/en/about-tssa/reports-and-policies.aspx
  41. Technical Safety Standards Association. (2002). TSSA CO Research Executive Summary. Retrieved from https://www.tssa.org/en/about-tssa/reports-and-policies.aspx
  42. Centers for Disease Control and Prevention (CDC). (2012). Community Assessment for Public Health Emergency Response (CASPER) Toolkit. Retrieved from https://www.cdc.gov/nceh/hsb/disaster/casper/docs/cleared_casper_toolkit.pdf
  43. Centers for Disease Control and Prevention (CDC). (2012) Developing CASPER Questionnaires. Retrieved from https://www.cdc.gov/nceh/hsb/disaster/casper/docs/DevelopingQuestionnaires_20160224_508.pdf
  44. Sircar, K., Clower, J., Shin, M. K., Bailey, C., King, M., & Yip, F. (2015). Carbon monoxide poisoning deaths in the United States, 1992 to 2012. Am J Emerg Med, 33, 1140. http:// dx.doi.org/10.1016/j.ajem.2015.05.002
  45. Smith, J., Zheng, X., Lafreniere, K., & Pike, I. (2018). Social marketing to address attitudes and behaviours related to preventable injuries in British Columbia, Canada. Injury Prevention: Journal of the International Society for Child and Adolescent Injury Prevention, 24(Suppl 1), i52–i59. https://doi.org/10.1136/injuryprev-2017-042651
  46. Cover Image retrieved from: http://thegreenmama.com/wp-content/uploads/2014/08/vanc.jpg

Appendix A: Supplementary Tables and Figures
See attached PDF. 

Appendix B: Categorized Structure Fire Incidents
See Excel sheet 

Appendix C: Structure Fire Raw Data
See Excel sheet 

Appendix D: Survey Questionnaire
See attached PDF. 

Appendix E: Survey Questionnaire Raw Data
See Excel sheet