Mind the gap: the reality of reported high levels of smoke alarm ownership

Feature, By Catherine Levin, 28 April 2014

There is a wide spectrum of opinion on how far government should go to regulate for fire safety in the home and in the workplace. Under the previous Labour government, the Fire and Rescue Services Act 2004 and the Regulatory Reform (Fire Safety) Order 2005 updated and consolidated years of benign neglect in the fire and rescue service.

Red tape and fire safety are strange bedfellows. A law to compel all households to purchase and maintain a smoke alarm does not exist in the UK, although a private members bill in 1991 did attempt to do so, it was never signed into law. Building Regulations provide a framework for those building or renovating homes, ensuring that smoke alarms are appropriately used and with a building inspector engaged to sign off the build, many thousands of homes now have smoke alarms. But many more do not.

Recently DCLG published a policy paper called ‘Review of property conditions in the private rented sector’. It’s not immediately obvious that this housing paper would include matters of interest to the fire sector, but this policy paper provides a vehicle for new provisions in the Energy Act 2013 to be explored. Kudos to DCLG officials and Ministers for getting Part 6, Chapter 2, s.150 into the Act which gives the Secretary of State the legal basis for making regulations to compel landlords to fit smoke alarms and carbon monoxide alarms.

But DCLG Ministers remain unconvinced of the need to regulate and this policy paper gives the fire sector the first chance in many years to put the case for compelling landlords to do the right thing. As ever the question is about cost, but that’s hard to defend given the low unit cost of smoke alarms; here it’s more about enforcement and how practical that can be made to be meaningful. Looking at the smoke alarm ownership statistics and fire statistics a bit more closely certainly makes for a more compelling case to get those regulations in place.

No smoke alarm with many fires
Despite the latest Survey of English Housing telling us that 84% of homes claim to have at least one smoke alarm in their home, fire statistics tell a very different story about absence, poor maintenance and removal. No smoke alarm was present in one third of dwelling fires reported in Great Britain for the period 2011/12, with London reporting a higher figure than the previous year’s national average of 39% for the year 2012/13. National fire statistics also show that in 18% of dwelling fires, the smoke alarm was present but did not operate.

It is a similar story in the US, where the most recent American Housing Survey carried out by the American Bureau of Census showed a 92% ownership level. The National Fire Protection Association (NFPA) looked at fire statistics between 2005 and 2009 and found that smoke alarms were absent in 28% of reported dwelling fires and where there was a smoke alarm, it operated in only half the fires reported to US fire departments. Surveys for the Consumer Product Safety Commission carried out in the same period also found reporting at the same level.

How can it be the case that surveys from two different countries carried out by different organisations at different times all found that despite high levels of reported ownership of smoke alarms, in one third of dwelling fires, no smoke alarm was present? There is something missing here.

Researchers from Johns Hopkins University write about the concept of social desirability bias infecting survey results in recent work on smoke alarm ownership they carried out with the Baltimore Fire Department . This type of bias has been defined as the “systematic error in self-report measures resulting from the desire of respondents to avoid embarrassment and project a favourable image to others” .

The lead researcher for the Johns Hopkins University work, Andrea Gielen, said that “particularly troubling is the fact that one in three of the follow up respondents indicated greater [smoke alarm] coverage than they actually had because they knew they should” . Is this what is skewing the numbers: embarrassment?

Incentivising smoke alarm ownership
Looking at what can be done to incentivize US households to install and maintain a smoke alarm, there is a plethora of laws to wade through and understand. The State of Illinois provides an insight into how far the US has gone to regulate the world of the smoke alarm. The Illinois Smoke Detector Act covers all dwellings regardless of tenure and includes hotels, inns, motels, bed and breakfasts and lodging houses. It is highly prescriptive about the location of the smoke alarm and places the onus on the owner of the dwelling to supply the smoke alarms.

The tenant (whether the same as the owner or not) has the responsibility for maintaining the alarm. However this law only applies to dwellings constructed, reconstructed or substantially remodeled after 31 December 1987. The penalty for non-compliance is pretty tough, with a first offence (category A misdemeanor) potentially resulting in a fine of $1000 or custodial sentence of up to a year.

As has often been the case after major fires where people have died, legislators take a long hard look at the tools they have to prevent future occurrences and they create so-called ‘tombstone legislation’. This happened in Connecticut on 1 January this year, when a new Act came into force requiring working smoke and carbon monoxide alarms in certain residential buildings at the time title is transferred. This comes two years after a widely reported tragic fatal fire where five members of the same family died in Stamford on Christmas Day 2011. The home had no smoke alarms and was under renovation at the time.

Linking house sales with smoke alarm ownership is a reasonable approach to enforcement but of course only affects those homes that are sold. Connecticut’s legislation isn’t unique in the US, with many other states having similar provisions.

Enforcing smoke alarm ownership
Household insurance policies provide another avenue for enforcing smoke alarm ownership. In France the legislation has recently changed and from March 2015 all homes must have at least one smoke alarm. Where a home has a fire and no smoke alarm is present, the home insurance policies will charge occupants a 5000 Euro excess when they make a claim on the policy. This link between fire and insurance when it comes to compelling smoke alarm ownership is not one that seems to exist in either the UK or the US.

Where smoke alarm legislation is absent at a State or other municipal level, as is the case in the state of Colorado, the home of enforcement remains squarely with the fire department through fire codes (such as those set out by the NFPA and others). In most cases, the fire department cannot compel a resident to give access for inspection without at least some form of notice.

Rented homes are easier to access due to licensing requirements on landlords, but private homes remain elusive apart from when another event, such as renovation or change of use, triggers an inspection. Inspecting every home in every state is simply not practical or possible, so enforcement is patchy and likely to miss the many homes that don’t comply.

Bill Barnard, recently retired State Fire Marshal for Maryland, points out many owner occupied homes have had no changes to structure or ownership since 1975, when Maryland first required smoke alarms to be installed in some homes. It is these homes, often with older residents, which are less likely to have the more comprehensive smoke alarm coverage required by later code changes. It looks like it can take decades for code changes to really have their desired effect on a wide number of homes.

Many fire departments do not have code enforcement jurisdiction in one and two family dwellings and therefore have no means to enforce. They rely on partner agencies like social services which do enter homes to notify them about the absence of smoke alarms. This ad hoc approach adds to more formal and proactive home fire safety visit programmes. Fire departments in the US and the UK make the most of the trust that many people have in the fire and rescue service to gain voluntary access to people’s homes.

And of course there is the trigger of a fire in the neighbourhood. Barnard talked about post incident work carried out by crews, a very common approach in the US and in the UK. In Maryland, Barnard talked about crews finding 50 per cent of smoke alarms out of date and/or not operating. Replacing these with ten year sealed alarms goes some way to improve protection but depends on willingness of residents to allow access. And of course coverage is patchy at best, dependent on where the fires are happening.

Getting fire departments' message across
Sometimes spurred on by tragedy, as in the recent spate of fatal fires in Los Angeles, fire departments use the media to get the message about smoke alarms across to the public. But the media is fleeting and soon forgotten; the real task is to make the messages enduring and to get people to take responsibility for their own actions.

Los Angeles Fire Department Battalion Chief, Stephen Ruda said in the LA Times after a fatal fire on 28 January where there were no smoke alarms, “we can talk and we can write so much, but people have to act. Maybe it’s the apathy of the people.” He goes on to say that “people think it won’t happen to them”. This is, sadly, nothing new.

The spectrum of efforts set out above go a long way to counteract the apathy that Ruda talks about: varying levels of government regulation; the incremental development of fire codes and the range of penalties for non-compliance; the tool of house sales; links with household insurance and excess fees; the randomness of other agencies entering properties and noticing a lack of smoke alarms; post incident outreach and a wide range of public education locally and nationally.

When you add all of this together does it fill all the gaps? In this puzzle where all these efforts have their place, there is a gap where none of these efforts have an impact. That gap shows vulnerability and risk. It would be interesting to map these efforts across an area, perhaps a US State, and see who’s left out. Are the people having the fires, being injured or dying in fires in the gap? Closing the gap between what people say about smoke alarm ownership and what people actually do is critical.

Important role of landlords
Taking advantage of opportunities to influence policy makers through consultations, like the one in the UK that closed on 28 March, is so important to remind government that the approach to keeping people safe in their homes is multi-faceted and that sometimes regulation can be the right answer.

Here’s hoping that the UK government goes ahead and debates the draft Statutory Instrument in both Houses of Parliament to raise awareness of the need for smoke alarms and carbon monoxide detectors and the important role private landlords have to play to give their tenants early warning of fire and CO emissions.

This is all very thought provoking and makes what seemed to have been a done deal i.e. high levels of smoke alarm ownership, looks like it really hasn’t met the goals set out by policy makers. It’s hard to know how much the idea of social desirability bias really affects those filling out surveys for governments in the UK and in the US, but the level of ownership claimed just looks too high when looked at in the context of 30 per cent of fires showing no smoke alarm present.

It’s absolutely right for policy makers and fire safety practitioners in the UK and the US to focus on smoke alarm maintenance in public education messaging, but a continued effort, through whatever means, to get the smoke alarms in to all households must continue at the same time.