FMs in the social housing sector know well the lifesaving impact of smoke alarms. The advent of BS 5893's best practice guidance will help to ensure even greater protection, says Neil Perdell
30 June 2006
There is an entire annex to BS 5839: Pt.6: 2004 devoted to the design and installation of fire protection based on a risk assessment of individual property. This strong emphasis has led many landlords in both the public and private sector to rethink how they can adequately meet their duty of care under the revised regulations.
Individual fire risk assessments of properties can be onerous and time consuming, especially for larger landlords with perhaps thousands of properties to consider. Such studies could also considerably slow down - or financially impact upon - installation programmes, thus increasing the risk factor to vulnerable tenants.
For some, rather than attempting individual fire risk assessments, the approach is now to accept the extended protection required in new build properties as best practice. Here a grade D system would be mains-powered with a back-up supply, with smoke alarms positioned in escape routes and heat alarms fitted in both the kitchen and principal habitable room. Such a move also anticipates the Building Regulations document B revision.
Obviously such systems where interconnection is required are far more complex and time-consuming to install.
Fire Statistics, United Kingdom 2003, published by the Office of the Deputy Prime Minister, is perhaps the most vital source of statistical information for the professional to draw upon. BS 5839: Pt.6: 2004 itself recommends that fire risk assessment should take into account the latest statistical information.
The statistics on fire deaths reveal immediately the importance of good smoke alarm coverage in the home. Around three quarters of all fire deaths occur in the home. Non-fatal casualties provide more evidence of just this conclusion. Eighty-one per cent of all non-fatal casualties occur in home fires - and what's particularly interesting here is that casualties are still 10 per cent higher than they were in 1993 - before smoke alarm protection became widely available.
Fire Statistics, United Kingdom 2003 also has important direct comments to make about the role of smoke alarms in all of this. The report only covers fires to which the emergency services are called (about one in every five fires) and says: "The figures reported should understate the effectiveness of smoke alarms". It suggests that the early warning given by a smoke (or heat) alarm often gives the occupier time to react and extinguish the fire by themselves. Once again, this clearly suggests that the earlier the warning, the less likely the fire is to be serious, with fewer fatalities and injuries, and less damage to property.
Smoke alarms are present in around
76 per cent of properties, so 54 per cent of all fire deaths occurred in the 24 per cent of unprotected properties. But that still leaves a large number of deaths in supposedly protected properties.
Reasons for failure
In 2003, 25 per cent of alarms failed to operate - but this masks a huge difference between properties protected by battery alarms and those with mains-power protection. Forty per cent of all battery alarms failed in 2003 compared with 13 per cent of mains-powered units. The overwhelming reason in the case of battery units was missing or flat batteries. With mains powered alarms, the single greatest reason was that the fire products failed to reach the detectors.
This could mean several things, but argues for greater coverage. An alarm in the hall may not pick up a smouldering fire in a living room behind a closed door until the fire has caught hold. Again, this is an argument for extending coverage to areas where fire is most likely to occur.
In 2003, the average casualty rate for all dwelling fires was 204 per 1,000 fires. However, this rate fluctuates drastically depending on the time of day. For example, the casualty rate for incidences between midnight and 6am leapt to 384 per 1,000. The reason is obvious: people were asleep - more vulnerable and less alert. Behind a closed bedroom door, they might not hear a downstairs alarm.
From even this very cursory look at Fire Statistics, United Kingdom 2003, it is obvious that the move towards taking the new build recommendations in BS 5839: Pt.6: 2004 as best practice is well justified. Mains-powered alarms are more reliable, and the greater coverage afforded by alarms in kitchens and reception rooms is in line with the statistical evidence of the sources of fire.
The aim is to give better, earlier warning for occupants to escape without injury or death. Greater coverage is important - but so is interconnection. Alarms sounding by themselves may not be heard, especially at night where the occupants are asleep. It's essential then that alarms are interconnected and sound everywhere in the property.
The problem is, the more alarms you have the more expensive, destructive and time-consuming hard-wired interconnection becomes. It could even be off-putting to responsible landlords intending extended alarm coverage. That's why it's so fortunate that BS 5839: Pt.6: 2004 has also opened the door to wireless interconnect technology. Greater coverage, made affordable by wireless interconnection, is truly best practice - and it's borne out by the statistics which must shape any genuine risk assessment.
Cost effective standard
Of course, as tenants change so there is a need for a fresh risk assessment depending on their ages, lifestyles, etc. The more comprehensive the initial coverage, the less need to change and - if changes really do become necessary - any good wireless interconnection system will allow the repositioning and adding of alarms with the minimum of cost and disruption.
When compared with the arduous, costly and time-consuming task of performing an adequate risk assessment on every single property and tailor-designing a system to fit, it may actually be more cost effective to install the new build grade D system as standard.
Neil Perdell is technical services manager at Aico