I reflect on the use of social science during the analysis phase of the investigation, for the purpose of broader education on how to other organisations might improve management responses to environmental, health and safety.
On the 3rd of March 2014, the media reported that the Country Fire Authority’s (CFA) key regional firefighter training site in Fiskville, Victoria, was shut down indefinitely after a chemical residue was found on site. Specifically, ABC News reports that PFOS (perfluorooctane sulfonate), a synthetic chemical compound, was found in the “mains water stored in two large tanks on the site.” PFOS has been used primarily in Scotchgard and other non-stick materials, such as teflon, and it was previously used in firefigher extinguishing foam.
The ABC is reporting that a 2012 report deemed the site “safe” from cancer:
A report in 2012 confirmed firefighters had been exposed to chemicals through water used in hot fire training exercises. But the report also declared the facility was safe and said the risk of getting cancer was low.
This is incorrect. I was involved in researching this 2012 report as part of the Independent Fiskville Investigation (IFI). The report did not measure cancer risk and the report did not deem the site “safe” relative to cancer. Our task was to determine the chemicals used during training exercises from 1971 to 1999. The report’s recommendations stress the need to implement ongoing environmental, health and safety audits. The key focus was on changing the safety culture within the CFA.
In 2012, I was contracted as the Senior Analyst on the Independent Fiskville Investigation (IFI). The Investigation was Chaired by Professor Robert Joy, who in addition to his academic positions, has also sat on the National State of Environment Committee and was Deputy Chairman and Executive Director of the
Environment Protection Authority Victoria.
The IFI was tasked to find evidence of possible hazardous chemicals used during firefighter training, as the CFA did not have clear documentation of practices. The Investigation was commissioned in December 2011, after the Melbourne newspaper, The Herald Sun, reported that several firefighters had contracted various types of cancers and were presenting other adverse health outcomes that they believed were linked to their exposure to dangerous chemicals during their training.
The Terms of Reference for the Investigation related to historical training practices relating to chemicals used on site, from the 1971 to 1999. We were not tasked to carry out a health study; nor were we asked to comment on cancer; nor did we have the scope, resources or mandate to declare the site “safe from cancer,” as the media has erroneously reported. We did not, for example, have access to medical records of firefighters.
Rather, we were asked to examine what was known about the types of chemicals that were used during training at Fiskville and its associated regional Victorian training grounds. This includes: where they were obtained, how they were used, how they were stored and disposed, and what groups of people may have been exposed.
This information was to lay the groundwork for future health studies, to further examine the connection between past chemicals use, training and possible health risks of these chemicals. Before a broader health study could take place, the CFA opted to first have us determine what possible chemical exposure may have taken place. The IFI report was titled Fiskville: Understanding the Past to Inform the Future. It was made available online after delivery to the CFA.
The CFA publicly accepted Professor Joy’s recommendations and expressed commitment to implementing change. Recent events would suggest that something went awry. I cannot comment on why that is; instead, I want to sketch how social science can contribute to better environmental management and improved occupational health and safety (OH&S) training.
The IFI was made up of four specialist teams working under the leadership of Professor Joy. Our Executive members had expertise in environmental protection and policy. They liaised with the CFA Board and other stakeholders, as well as reviewing the research of external consultants. This included Golder Associates who carried out chemical tests of the environment and training grounds at Fiskville and other regional training sites. KordaMentha were tasked to obtain and digitise CFA documents, as prioritised by Professor Joy.
The Executive was supported by administrative staff who also aided document management.
A team of investigators carried out over 300 interviews with firefighters, managers, former CFA Board members, other volunteers and community members. They also interviewed businesses and government agencies who had trained at Fiskville, and local businesses who donated fuels and materials to be used during training. They additionally reviewed other relevant reports and materials to corroborate these accounts.
My role was to lead a team of over 20 analysts managing the data. This included the review of 30,000 documents, focused on a sub-set of 8,000 materials deemed most critical to the investigation. For example, internal communications on occupational health and safety issues (OH&S); internal OH&S policy reviews; external reports; environmental audits by governing bodies such as the Environmental Protection Authority; Board Minutes and other official documentation of training practices; as well as analysis of the 300 interviews.
I established our analysis framework to manage the various datasets, including three databases:
- Document review: targeting analysis of the high priority records using a mix of automated and manual searches. Priority materials were individually coded and passed on for in-depth qualitative analysis (that is, an evaluation of key themes found within texts);
- Qualitative analysis database: used to code high-priority documents and interviews. The IFI’s key terms of reference were developed into a content analysis framework for documents and thematic analysis of the interviews; and
- Reference management system: the final report includes around 600 direct citations that required careful documentation and cross-referencing to other sources.
In addition to me, there were three senior researchers who conducted in-depth analysis to answer the terms of reference, and to assist in researching and drafting key chapters. Three of us were sociologists and the fourth senior researcher had an undergraduate degree in science and a postgraduate degree in education. We were supported in primary coding of our datasets by 22 paralegals.
The Investigation was highly expedited, as outlined by the CFA, who first commissioned the Investigation in December 2011 immediately following the media reports. The CFA initially only allowed us three months to address the terms of reference; this was extended to six months in March 2012. We received the majority of our most relevant documents by KordaMentha in April and May – less than two months before our report deadline. We requested more time to extend the investigation but this was not granted. Professor Joy delivered the report on the 30th of June 2012. He discusses some of the key findings in this short video. I elaborate further below.
Regardless of the limited time-frame, we gathered significant evidence suggesting that training practices for the 30-year period studied had been outside best practice. We documented that CFA management had not adequately responded to changing regulations, external audits and concerns of staff.
Our investigation was complicated by the fact that up until the early 1990s, for the most part, official CFA records did not discuss OH&S issues explicitly. The terms we might expect to look for today (such as “safety”) did not appear in CFA records. OH&S reports of critical incidents (such as the burial of drums containing hazardous waste) did not appear in Board Meeting notes, although we showed that these issues had been communicated to Board members through other channels.
Interviews revealed occasional incidents of trainees being hurt during exercises, but these were not always officially documented. As the interviewees noted, CFA culture up until the 1990s was akin to a masculine paramilitary structure, where men were expected not to complain, and do as they were told. Problems were acted upon on the spot and recorded unevenly. Staff who raised concerns about OH&S were alienated, transferred or otherwise ignored by upper management.
As part of our analysis methodology, I used principles of social network analysis (SNA). SNA is a way of examining how social relationships and cultural values (such as trust and cooperation), impact on material outcomes. SNA is used to identify influential actors and determine how information, resources and ideas flow through a group, organisation or community.
We found that the official CFA hierarchy did not always match the reality of how decisions were made on-the-ground within Fiskville. We examined informal relationships of trust, such as communication flows between staff who worked closely together on site. We found that even in the mid-1990s, as more formal OH&S reviews were rolled out, managers did not understand who was responsible for acting upon environmental notices and audits. Memos went back and forth for months without action. There was no review and follow-up on environmental audits and reports. There was little trust in upper management to follow through on requests for help, but staff trusted lower-level managers on site to manage problems.
A consistent theme in the IFI recommendations was that environmental and health issues require preventative management, ongoing evaluation and follow-up review. These elements were not happening cohesively even up to the end of the period we investigated, inclusive of the late 1990s.
The IFI collected evidence about the hazardous chemicals used from the 1970s to the late 1990s and how the handling and storage of these chemicals complied with OH&S regulations. In retrospect, many of the chemicals used were unsafe, especially during the 1970s and 1980s. What we know today about hazardous agents was not known at that time, however a chronology of events at Fiskville, when set against changes in regulations, suggest that management did not adequately enforce OH&S best practices (see images below and IFI report pages 70-71; 48-49).
PFOS Use in Fiskville
While we examined several chemicals and liquids, I will focus here on PFOS, as this is reported as being the reason for the recent closure of Fiskville.
Two types of firefighter foams containing PFOS and PFOA (AFFF and AR–AFFF) were in use at Fiskville from the 1970s up to 2007. In between this time, in 2003, the Australian regulatory body, the National Industrial Chemicals Notification and Assessment Scheme, had already announced that PFOS and PFOA foams should no longer be used. The CFA disposed of stocks of these foams in 2007, four years after the national recommendation that PFOS and PFOA based foams be restricted only to essential uses and not employed in training.
Why is PFOS a Concern?
Chapter 7 of the IFI report addresses the acquisition, use and nature of foams, including PFOS. The toxicity of PFOS became an international concern during the 1990s, as research identified its “long-term health and environmental effects,” particularly given that PFOS is not biodegradable. The Organisation for Economic Co-operation and Development (OECD), which includes 30 nations such as Australia, developed a hazard assessment of PFOS in 2000.
In 2002, the world’s largest manufacturer of products containing PFOS, 3M, voluntarily stopped production of most of its its associated products. In Australia, PFOS was phased out in 2003, including in firefighter foam.
A 2006 OECD survey and workshop addressed the dangers of PFOS, but discontinuing use of existing products remained voluntary in Australia. It wasn’t until 2009 that the OECD formally moved to recommend the withdrawal of PFOS. The United Nations, the European Union and various other national governments and international groups also followed suit in subsequent years. While most organisations have moved away from using foams, a 2013 Western Australian review by Dr Jimmy Seow finds no consistency in national or international policies, but argues that firefighter foam should no longer be used.
See the “Learn More” section at the end for a summary of PFOS testing at Fiskville.
There are four types of substance exposure that can enter the body: absorption, ingestion, inhalation, and injection (IFI pp. 88-89).
- Absorption: chemicals pass through exposure to the skin. This was found to be the least common exposure route in Fiskville, due to personal protective clothing (PPC) and the nature of training exercises. To be sure, there was inadequate PPC standards at Fiskville, especially during the 1970s and up to the mid-1980s. During foam training exercises, staff report that the biggest issue was not having access to breathing apparatus, and trainees instead used their fire helmet over their faces. Nevertheless, absorption is considered the least risky route of exposure.
- Ingestion: the unintentional swallowing of chemicals that then enter the stomach. Ingestion of toxic substances and particles, including PFOS and PFOA, was found to be the most common type of exposure in Fiskville, specifically through recycled fire water from the back-up safety system. However, the impact of this exposure was significantly less hazardous than inhalation, as most exercises relied on clean, portable water during fire-fighter training.
- Inhalation: substances and chemicals are breathed in via the airways, which then pass through the lungs. Vapours from flammable liquids, gaseous products of combustion, liquid aerosols, and soot were all found to be present in firefighter work, but prolonged exposure comes from emergency responses at incidents. This is considered the most risky route of exposure.
- Injection: chemicals pass through the bloodstream via wounds to the skin. This was not found to be relevant to the firefighter training at Fiskville.
We examined the potential risks of exposure to hazardous chemicals via these routes by looking at the different groups who trained at Fiskville.
Some firefigher groups, such as those who worked on the Practical Area for Drills (PAD) were determined to be at higher risk of exposure to hazardous chemicals (pp. 90-97). In particular, PAD supervisors and operators regularly managed flammable liquids, extinguishing foams (including PFOS), combustion products and recycled firewater. They were at higher risk also because they were closer to sustained inhalation of chemicals during their regular work (that is, outside of training).
Full-time operators and supervisors would have up to eight hours of exposure during training exercises. Trainees, who only came on site for short courses would have lower cumulative frequency and were likely to have been at a lower chronic risk of exposure relative to full-time staff.
Although up to 50,000 people trained at Fiskville during the time examined by the IFI, this encompasses various types of training, not all of which involved exposure to dangerous chemicals. Professor Joy estimates that the number of firefighters who had ongoing exposure to dangerous chemicals is around 100 people. Professor Joy explains in this video.
During the period investigated, there were three acute incidents of exposure that the IFI verified and documented. The first was in 1977, when an officer was exposed to chlorine gas during a training exercise. He was given oxygen and taken to hospital, where he spent one week.
The second incident was a fire in the drum storage area in 1982. Six barrels combusted. One officer was temporarily overcome by fumes from a leaked substance; however this incident was raised with management years later as evidence of broader health risks. The affected drums were buried, but others remained as they were. Drum storage did not improve until the late 1990s.
The third incident occurred in 2002. While it was outside the period of the investigation, it was included in our analysis as an independent contractor was exposed to fumes after inadvertently unearthing buried drums whilst operating heavy machinery. Drum burial on site was one of the key elements of our investigation, as evidence of unsafe chemical management practices that occurred during the period under investigation.
Officers’ concerns about drum handling, storage and disposal dragged out for two decades. The first set of drums containing hazardous waste were buried in late 1979 or early 1980. This was in response to corrosion of the barrels which omitted unpleasant vapours. Up to 100 drums were buried on site at Fiskville.
The second burial involved up to 30 drums after the 1982 fire. There had been previous attempts to have the drums removed by a professional waste company. In January 1982, a Fiskville officer sent a sample to be tested, revealing the contents were toxic and could not be used during training. The company declined the move due to the toxicity. After six drums caught fire in December, they were buried on site.
The third burial involved the remaining 100 drums, and occurred sometime between 1983 and 1986. One interviewee reports that the drums were scheduled to be moved, and some may have been exhumed and buried in the nearby golf course, but the IFI investigators could not corroborate this further.
A fourth burial happened around the mid-1980s, involving an estimated 120 to 400 new drums.
There were two further drum movements, as one of the existing burial sites was extracted in 1991. Seventy-five drums were removed under the supervision of the CFA Chairman and in accordance to legal and regulatory requirements. In 2002, due to the contractor’s accidental unearthing of one of the existing drum burial sites, 56 drums were removed from Fiskville. While they were transported under EPA guidelines, it appears that no environmental clean up was engaged, which is in breach of environmental regulation.
The management of drum storage and disposal highlights both the inconsistency of following environmental regulations, and the uneven responses by the CFA Board and upper management. Without adequate responses, officers on the ground made poor decisions to bury drums, in lieu of official support.
I was tasked with analysis of the CFA Board and Management’s knowledge of these potential training, health and environmental risks during the 30 year period (Chapter 9 of the report).
Our team established a chronology of key documents and reports, showing that the CFA Board and managers consistently failed to act on recommendations of consultants, internal OH&S reviews and external environmental audits.
Concerns about materials and substances were raised in the early 1970s, but formal documentation of these concerns are found from March 1980. Officers and lower-level managers attempt to communicate with higher level managers that chemicals and disposal of materials were out of step with the Environmental Protection Authority and other governing bodies at the national and international level.
In the early 1980s, officers communicate concerns about the flammable materials delivered by outside companies via unsafe drums. It had long-been a common practice for nearby businesses to drop off flammable materials for firefighters to use during training, as an alternative to businesses having to dispose of these materials themselves. Following the drum fire in December 1982, Fiskville staff repeatedly contact the CFA’s higher-level managers about the drum sites. This includes direct communication passed to the CFA Board Chairman, who eventually responds to a union request to deal with the drums in 1987. He elects not to take action. A subsequent document is sent to the Chairman in 1990 with similar inaction. It was not until 1991 when a new Chairman is made aware of the situation, that 75 drums were removed by a waste management company. Still, the remaining drums remain unearthed.
In between all this were various other environmental reports, including the “Fiskville Master Plan,” which was first proposed to the CFA Board in April 1987. This was almost a decade after the then-Officer in Charge proposed to upgrade facilities at Fiskville. This included upgrading the water system, in response to various complaints by residents and officers (full-time staff lived on site with their families). The OIC, now the Commanding Officer in the late 1980s, engaged a contractor to begin plans to improve Fiskville. The Plan would have addressed environmental concerns and significantly improved safety outcomes. After several delays and reiterations, the Plan was rejected by the Board in 1991 due to reallocation of funds.
The 1990s saw several more reports and reviews once again raising OH&S and environmental issues present at Fiskville. The most significant changes followed the Dangerous Goods, Occupational Health and Safety report and the Environmental Audit report, both in 1996. Further environmental audits and reports followed, with recommendations to improve the Flammable Liquids Training PAD. Construction was completed in 1999.
What does this history tell us about management at Fiskville? Our report clearly demonstrates that the Board and Fiskville management not only received consistent formal reports and complaints by staff to address OH&S matters, but that higher management consistently failed to act on these concerns. There are three broad reasons why this seems to have eventuated.
- Lack of strategic planning: For the majority of the period of our investigation, the CFA Board did not review and audit health, safety and environmental issues.
- Narrow focus on safety responsibilities: The Board’s focus on safety was viewed firmly with respect to dangers in the field. In the Australian bushfire context, firefighter operations are highly perilous. Unfortunately, this issue received more attention than the safety during training.
- Inflexible organisational culture: The mentality at the time was to create a “realistic” training environment that pushed men beyond their limits and exposed them to additional risks. The CFA excels in adapting firefighter techniques, but it has not been a learning organisation with respect to safety and risk assessment. Traditional gender norms were used to punish trainees into compliance, rather than inviting open and responsive dialogue about health and safety issues.
The Linton fires was an impetus for rethinking firefighter safety, however this did not necessarily translate into changes of practices during hot firefighter training.
Additionally, corporate knowledge of training practices was lost as staff left. Training lessons did not follow a standard guide, meaning that subjective notions of safety and risk governed how firefighters were treated during exercises.
Professor Joy delivered ten recommendations. These were, in brief and contracted:
- That soil and groudwater quality be assessed, particularly where fuel storage tanks and flammable liquids PAD are currently placed, and where they were historically located
- Further investigations of water in and discharging from Lake Fiskville. This includes the recommendation to “investigate potential sources of PFOA and PFOS discharges to Lake Fiskville and discharging off site, if the potential risk of adverse impact on downstream human health receptors is
found to be unacceptable;”
- Independent inspection of electrical transformers
- Follow up study of possible health effects related to exposure of training materials
- Implement procedures to protect health of staff potentially exposed to waters and sediments of Dams 1 & 2, and manage future potential risks
- Further assessment of drum burial sites and historical landfills within Fiskville. This includes a walking path constructed on top of the historical landfill 1
- Management of historical landfill 2 and exhumation of drum burial sites should await results of aforementioned site tests
- Conduct further targeted investigation of other training sites.
Why does the report recommend further tests and studies? Because within the short time frame of the investigation, we were bounded the Terms of Reference. The IFI report was intended as a preliminary report outlining the next course of action.
In July 2012, the CFA responded to the IFI report and accepted all of its recommendations.
A number of health and environmental studies have taken place since Professor Joy delivered the IFI report in mid-2012. These studies follow his recommendations, but are unrelated to the IFI.
In June 2014, the the Cancer Council examined cancer risk experienced by former Fiskville trainees. In November 2014, Monash University examined mortality and cancer incidence amongst firefighters. Also in response to Professor Joy’s recommendation, water tests were regularly being undertaken at Fiskville. In July 2014, the EPA declared Fiskville posed a low and acceptable risk for trainees and visitors. Clearly, this advice has been overturned in recent days.
I’m not privy to the events that led up to the recent closure of Fiskville, nor have I had knowledge, access nor input into CFA work practices or policies since 2012. Instead of speculating on the closure, I want to summarise a few key lessons from the IFI report on how social science can inform environmental and OH&S policy, planning and evaluation.
Lessons from Social Science
For most of the period under review by the IFI, the CFA Board did not adopt modern and evolving enterprise risk management practices. While on-site officers and managers pushed for changes, there was no systemic process to ensure compliance with OH&S regulations and environmental requirements.
The IFI report of 2012 outlined that the CFA should establish a clear strategic plan to address environmental and OH&S issues. This includes a plan-do-review model that assesses historical and present-day safety risks, and acts upon existing and outstanding recommendations.
Of specific sociological interest is that the OH&S and environment culture within the CFA was dictated by traditional Anglo-Australian values of masculinity. Trainees were placed in unnecessarily risky situations involving foam, smoke and other materials.
The IFI involved the input of different specialists. I have discussed how social science was used in the analysis of thousands of documents and hundreds of interviews. The task of organising the various datasets and evaluating the information was undertaken using social theory and methods, including social network and qualitative analysis. The concepts of gender norms and organisational culture also shaped how our analysis identified gaps in our information, and how we made sense of risk and management decisions.
A consistent theme in my research and consultancy is that gender inequalities and lack of cultural diversity adversely impact on innovation. Increasing women’s leadership positions, incorporating new perspectives, and addressing communication in a reflexive manner are all conducive to improving workplace culture.
A restrictive, masculine style of management compounded the environmental and health damage within Fiskville during the period under review. Personnel were discouraged from voicing concerns, and requests for better support went continually unanswered. Having a more open and inclusive management style, where the organisation draws on alternative modes of masculinity (less regimented thinking, for example), may have better ensured that concerns were heard and acted upon. Actively incorporating participation and leadership from women and minority groups is another way to improve strategic thinking.
Moverover, OH&S and environmental issues require preventative planning, documentation, review and evaluation. Ignoring staff’s health and safety concerns over a 30-year period had disastrous consequences for the CFA. As a volunteer organisation, the health and trust of members is paramount to growth.
While it’s encouraging that the safety of trainees, workers and visitors will be better protected by the closure of what has now been deemed an unsafe site, it is disappointing that persistently mismanaged events over decades led to this point. The CFA is an important Victorian organisation. They appeared to be responding to the IFI’s recommendation by implementing further studies and environmental testing. Hopefully in future, the CFA will draw further lessons about reflexive organisational management.
It would be especially fruitful for similar organisations to move towards a preventative model OH&S policy and environmental planning, with a more reflexive, diverse and inclusive style of management.
Learn more: PFOS Chemical Testing
Chemical tests and analysis gathered by Golder Associates suggest that most analyses of the surface water and sediment from Lake Fiskville were well below drinking water quality. The lake is not used for drinking water and as a result, Golder Associates concluded that the risk posed by this water source was low (IFI, p. 9).
The exception was levels of PFOS and PFOA (perfluorooctanoic acid). Golder Associates noted that levels of dioxins and furans (PCDD/Fs) exceeded Canadian environmental guidelines, but they note this “does not necessarily demonstrate evidence of an adverse impact.” Regardless, Golder Associates recommend ongoing assessment of the ecological condition of the lake. The IFI adopted this measure in its recommendations to the CFA (see Chapters 6 & 12).
A 2010 report by Wynsafe (not associated with the IFI), tested for PFOS and PFOA in the primary pump and Dam 2 at Fiskville. These sources of treated firewater supplied backup safety hoses and, as a result, may have been a source of exposure of dangerous chemicals. The Wynsafe report detected levels above drinking water levels, however they conclude that the daily intakes were “several hundred times lower than the recommenced Tolerable Daily Intake (TDI) for both PFOS and PFOA even based on the highest result” (IFI, p. 84).
Further assessment by Golder and Associates in 2012 finds that PFOS and PFOA were “between two and three orders of magnitude below the TDI for the waters in Dams 1 & 2.” They also conclude that dilution of PFOA and PFOS concentrations in the nearby Moorabool River Catchment following discharge from Lake Fiskville is
“likely to result in reduced exposure concentrations compared to that reported in Lake Fiskville. In addition, the exposure concentration may be further reduced via mechanism of environmental fate and transport.”
Golder and Associates nevertheless identify that further water tests continue, especially of residual PFOS that may have moved downstream. The IFI endorses this as one of its recommendations to the CFA.