Review of evidence - Behavioural Factors
Behavioural Factors
The evidence identified smoking status, alcohol and/or drug impairment, having working smoke alarms, and being asleep at the time of the fire as behavioural factors relating to risk of ADF-related fatalities in ADF’s. There was no research specifically on risk of experiencing an ADF or risk of ADF-related injury in an ADF for behavioural factors, but some research was focused on a combination of factors which included those risks.
Smoking status
Behaviour relating to smoking was referenced the most among the behavioural factors, appearing in 20 of the 39 pieces of evidence used for this report. However, in the majority of the evidence (N=15) it was cited under causes of the fire, rather than as elements of people’s behaviour or factors associated with ‘who’ is at risk. For example, smoker’s materials being mishandled or discarded inappropriately were referenced as ignition sources and causes of fires, but under human factors smoking status was not mentioned. It may be that it is unclear whether victims in ADF’s were smokers themselves, or that misuse of smoker’s materials may not mean that the person was necessarily a smoker (e.g., a lighter or matches being misused or carelessly discarded does not necessarily indicate smoking status).
ADF-related fatalities
The majority of research pointed to the fact that smoking materials and/or smoking status plays a major role in ADF-related fatalities. In particular, smoking status was identified as a human risk factor (identified ‘who’ is at risk- Howland and Hingson, 1987; Warda, Tenenbein, and Moffatt, 1999; Xiong, Bruck, and Ball, 2015; 2017) and as an ignition source (Department for Communities and Local Government, 2006; Flynn, 2010; Holborn, Nolan, and Golt, 2002; Mulvaney, Kendrick, Towner, et al., 2008; Runyan, Bangdiwala, Linzer, et al., 1992), especially careless handling of smoking materials (Greenstreet Berman Ltd, 2014; Heimdall Consulting Ltd, 2005) and igniting furniture (Greenstreet Berman Ltd, 2014; Jonsson, Bonander, Nilson, and Huss, 2017; Patetta and Cole, 1990). Additionally, age was found to intersect with smoking behaviour whereby both younger and older aged ADF-related fatalities were associated with fires caused by smoking materials (Flynn, 2010; Runefors, Johasson, van Hees, 2017; Sekizawa, 2005). In both older and younger age, the additional factors identified above could impact smoking related fires. For example, misuse of smoking materials by young people, and/or igniting furniture particularly in older age groups who may also have reduced mobility, increased frailty, or other health-related impairments. Many sources of evidence suggest that although smoking materials are not the most common source of ignition in ADF’s more generally, smoking materials were the leading cause of ADF-related fatalities (Barillo and Goode, 1996; Gilbert and Butry, 2018; Home Office, 2019; Levine and Radford, 1977; Patetta and Cole, 1990; Runyan, Bangdiwala, Linzer, et al., 1992). The proportion of fires that were recorded as being caused by smoking materials ranged from 5% (Gilbert and Butry, 2018) to 25% (Barillo and Goode, 1996), with 7% of UK ADF’s in 2018/19 caused by smoking materials (Home Office, 2019). On the other hand, the proportion of ADF-related fatalities recorded as caused by smoking materials ranged from 23% (Gilbert and Butry, 2018; Home Office, 2019) to 65% (It should be noted that the figures cited in Levine and Radford’s (1977) research may be a facet of the time at which the records were taken.) (Levine and Radford, 1977). More recent figures suggest the proportion of ADF-related fatalities caused by smoking materials is likely around 25% (Gilbert and Butry, 2018; Home Office, 2019).
Fire risk, and ADF-related injury risk
There was one source of evidence, a literature review, which specified that smoking was associated with risk of experiencing a fire, as well as risk of ADF-related injury and death (Turner, Johnson, Weightman, et al., 2017).
Alcohol/Drug impairment
Alcohol and drug use (including medication) were included as behavioural factors since it is a modifiable behaviour within the individual’s control (this could include risky behaviour when using medication, whereby medication use itself may not be a modifiable behaviour for health reasons).
ADF-related fatalities
Xiong, Bruck, and Ball, (2017) identified seven characteristics most associated with fire fatality, including alcohol intoxication, or having taken drugs/medication. Alcohol intoxication, mostly recorded as ‘over legal driving limit’, was found to be a significant factor in ADF-related fatalities (Barillo and Goode, 1996; Hall, 2005; Heimdall Consulting Ltd, 2005; Jordan, Squires, and Busittul, 1999; Levine and Radford, 1977 Runyan et al., 1992; Warda, Tenenbein, and Moffatt, 1999). In particular, Marshall et al. (1998) found that alcohol intoxication played a significant role in as many as 53% of ADF-related fatalities, and Holborn, Nolan, and Golt (2002) found up to 58%, especially among men.
The intersection with gender was identified in multiple sources of evidence, particularly relating to men (Xiong, Bruck, and Ball, 2015), especially in the 40-49 age group (Sully, Walker, and Langlois, 2018), or men aged 40-60 years who lived alone and smoked (Department for Communities and Local Government, 2006). This also suggested an interaction with age, whereby the evidence showed that alcohol use was a greater risk for those aged 20-29, 40-49, 50-59 years (Sully, Walker, and Langlois, 2018), and alcohol (33% of ADF-related fatalities) and medications (12% of ADF-related fatalities) were an important factor in the 50–59-year-old age group, although this was less of a risk factor for older adults (Department for Communities and Local Government, 2014).
There was also an association between alcohol use, smoking, and risk of ADF-related fatality. Turner et al. (2017) found that alcohol or drug use was linked to ADFrelated fatalities, but more likely among smoking-related fires, especially when there were no smoke alarms or a vulnerable person lived in the property. Furthermore, Howard and Hingson (1987) found that alcohol was more likely to be a significant factor in ADF fatalities caused by smoking materials. Other research showed that 25% of ADF-related fatalities were associated with a combination of smoking and drinking, and in relation to causes of fire, blood alcohol levels were found to be raised among fire fatalities in 85% of cooking related, 60% smoking related (mostly ignition of furniture) and 39% of heating related ADF’s (Patetta and Cole, 1990). Finally, alcohol was more likely to factor in ADF-related fatalities of lone adults, or couples with no children (Department for Communities and Local Government, 2014).
Working smoke alarms
Fitting, testing, and maintaining working smoke alarms in suitable locations within a property are essential fire safety behaviours that could prevent an ADF. Analysis of UK data found that not having a working smoke alarm was associated with a 37% higher fatality rate among adults (Greenstreet Berman Ltd, 2014). Recent UK data showed that for 2019/20 incidents, 24% of dwelling fires occurred in properties with no smoke alarm, and 26% of ADF-related fatalities were in dwelling fires where there was no smoke alarm present (Home Office, 2020). This was also found
internationally, where as many as 94% of ADF-related fatalities happened in homes with no smoke alarm (Patetta and Cole, 1990), and others cited no working smoke alarm in the property as a contributing factor to fatality risk (Barillo and Goode, 1996; Jordan, Squires, and Busittul, 1999; Runyan et al., 1992). Although it is important to
note that the international research is older, and therefore technological and social advances may mean that these figures have changed over time. These figures highlight the importance of ensuring households not only have smoke alarms fitted in suitable locations, but also that residents maintain and check their working order regularly.
According to the English Housing Survey ‘Fire and Safety, 2016-17’ report, 95% of English homes reported having a smoke alarm, 90% reported that their smoke alarms were in working order, but 8% reported either not having an alarm, or an alarm that was not working, and 2% were unsure of the working order of their smoke alarm (Ministry of Housing, Communities and Local Government, 2017). These figures are based on self-reports from homeowners, and as such may not be completely accurate. Given that the previous English Housing Survey in 2014-15 identified that only 28% of respondents tested their smoke alarms regularly, and almost the same proportion (22%) reported never testing their smoke alarms (Department for Communities and Local Government, 2016) it is possible that the
reported 90% of households with working smoke alarms could be vastly overestimated.
Summary
There was no research specifically on risk of experiencing an ADF or risk of ADF-related injury in an ADF for behavioural factors, but some research was focused on a combination of factors which included those risks. This may be due to a lack of research that target’s people directly to ask about their fire-risk and home fire safety behaviours, which may make it difficult to link behaviours to incidents where the individual(s) have survived. Additionally, and particularly in cases of risk of experiencing a fire, it may be that smaller fires for which people do not call the FRS are linked to behavioural factors, and so with those incidents remaining un-reported the prevalence is not captured.
Furthermore, the interaction of behaviours with other factors such as demographics and socio-economic factors means that the behavioural predictors of fire risk may need to be considered in conjunction with other risk indicators rather than as individual targeting factors. This may be especially difficult where demographic details are not recorded in publicly accessible databases for FRS’s to use for targeting work. Additionally, it is likely that psychographic factors, such as general attitudes towards risk and fire safety would influence behaviours, and as such more research is needed to explore the potential individual human factors that relate to fire risk and safety behaviours. This also highlights the need for FRS data to capture features relevant to the people involved in ADF incidents to further refine targeting for future safety intervention prioritisation.