This type of alarm also has advantages and disadvantages. It detects smouldering fires more quickly and is not as prone to ‘nuisance’ alarms during cooking so is less likely to be deactivated by owners and could lead to fewer deaths as a result. It does not contain radioactive material which may make it a more appealing option for some households. However, this kind of alarm is less quick to respond to fast flaming fires where the smoke particles are less thick so these fires could gain a firm hold before being detected. It is also more expensive, making it a less attractive purchase. In addition, as well as smoke, the beam in the detector could also be interrupted by the entry of dust, steam or insects into the device. So whilst it is less susceptible to nuisance alarms, these are still a possibility and still could result in the alarm being deactivated and not being reactivated later, with potentially fatal consequences.
I will now analyse data from the Fire and Rescue Service. Some may argue that this data is very reliable as those that collected it, fire fighters, are government employees and can therefore be trusted to tell the truth. However, there are a number of problems using this data. Firstly, fire fighters are often acting under pressure and therefore the data may not be precise. The evidence left behind at a serious fire is often ambiguous and open to interpretation so the data that results is not hard data but the result of interpretation of remaining evidence. It is possible, that in order to emphasise the dangers of not having fire alarms or tampering with fire alarms, those collecting and interpreting the data could have manipulated it for the best of reasons or even unconsciously. The types of fire and the types of alarm (other than whether they are mains or battery operated) are not given. The data is also based on fires attended by the fire and rescue service. Any fires involving alarms where no emergency call was made to the service are not recorded so the figures reported should understate the effectiveness of smoke alarms. That is, if a smoke alarm is working correctly it will provide the occupier with an early warning of fire or smoke and they may put out the fire themselves and not call out the fire service. Findings from the 2004/05 Survey of English Housing (SEH) suggest that the fire and rescue service were called to just over a fifth of all domestic fires.
Various research shows that the proportion of households with a smoke alarm has increased (see the table on the next page) so it is certainly the case that the data in this report does not accurately reflect the effectiveness of smoke alarms.
This table, constructed using data from the UK Fire Statistics Report (Table 2.3) shows a general trend of increase in the percentage of homes with working smoke alarms between 1988-1994 (there was no data for 1991). Thereafter there is a levelling off and even a slight decrease between 2000-2004 (there is no data for 2003). This suggests that ownership of smoke alarms is plateauxing at about 80% despite widespread publicity and campaigns to raise awareness of the life saving benefits of having alarms. It is this 20% of households, therefore, which would benefit from the passing of legislation to make the ownership of a smoke alarm compulsory.
As the table on the next page shows, there has been a general increase in the number of fires in which a smoke alarm raised the alarm from 25% in 2000 to 32% in 2004- and this is despite the fact that ownership of smoke alarms has levelled off in this period. The table compares the number of fires in which a fire alarm was present and raised the alarm, with those in which a fire alarm was either absent, or was not functioning and did not raise the alarm. From this table you can see that an increasing number of fires IN WHICH THE FIRE AND RESCUE SERVICE WERE INVOLVED are detected by alarms, whilst the number of households with non-functioning or absent alarms is decreasing. This would seem to indicate a greater awareness of fire alarms and fires generally, since the total number of fires is also decreasing.
Here is the same data in the form of a chart:
1=2000
2=2001
3=2002
4=2003
5=2004
In the table below I separate the statistics for fires in which a smoke alarm was:
- present and raised the alarm
- present, operational but did not raise the alarm
- present but not operating
- absent.
This table reveals a general decrease in the number of fires in which a smoke alarm was absent and a slight increase in the numbers of fires that were detected by a working smoke alarm. Bearing in mind that there was decrease of about 15% in the total number of fires in this period the picture is still an improving one in terms of the number of fires detected by a working alarm. The numbers of fires not detected by an operational alarm, or an alarm that was operational but did not raise the alarm, remains fairly constant. This suggests that alarms have either not improved in their reliability over this period or that similar numbers of people are either not maintaining their alarms correctly or are deactivating them due to nuisance alarms.
The trends I have described above are slightly more easy to observe in the pie charts below which show the presence and operation of smoke alarms as a percentage of the fires that occurred in two years, 2000 and 2004
When expressed as a percentage, the number of fires occurring in which an alarm in absent is on the decrease, whilst the number of fires detected by a working alarm is on the increase. This, combined with the fall in the number of fires generally does show a greater awareness of fire safety precautions in general-though the number of people failing to maintain their alarm correctly remains more or less constant and the reliability of alarms in general has does not appear to have improved.
Perhaps a more meaningful way to examine the effectiveness of smoke alarms is to examine the deaths and casualty statistics resulting from fires in which a fire alarm was absent or not working, compared to those in which a smoke alarm successfully raised the alarm.
From the table below I can see that between 2000-2004 the picture of the total number of fatalities is a mixed one, with an increase in 2002 and 2003 and a decrease in the intervening years. The increase was also matched by an increase in fires in which the alarm was raised by a working alarm, especially in 2003. This is surprising since you would expect the fact that more fires were detected by an alarm to lead to a fall in fatalities, rather than a rise. To help to explain this anomaly it would be helpful to know what type of alarms were involved. For example, households could have had optical alarms installed which failed to detect chip pan fires quickly enough to avoid them spreading.
Here is the same data in the form of a chart:
1=2000
2=2001
3=2002
4=2003
5=2004
The chart below shows the number of non-fatal casualties resulting from fires depending on the presence and operation of smoke alarms.
Surprisingly, this chart shows the opposite of what you might expect-an increase in the number of non fatal causalities where a smoke alarm WAS present AND raised the alarm, and quite a sharp decrease in the number of causalities were a smoke alarm was absent. This trend is also clear in the two pie charts below which take statistics from the beginning and end of the period and express them as percentages.
The fall in non-fatal causalities where a fire alarm was absent might partly be explained by the decline in the number of fires generally over this period. Perhaps the increasing use of flame retardant materials in furnishings has also contributed to this figure. Despite the anomalies, however, what is clear is that fires that are not detected by working alarms, or which occur in dwellings without alarms, produce far, far higher fatality rates. This is because working smoke alarms tend to shorten the discovery time as you can see in this chart. In 2004, 64% of dwelling fires where a smoke alarm raised the alarm were discovered in under 5 minutes. However, where a smoke alarm did not raise the alarm only half (51%) of all dwelling fires were discovered in under 5 minutes.
Conclusion
Having considered two different types of smoke alarm I would say that BOTH types of smoke alarm should be made a compulsory purchase for UK homes.
This is because, after looking at the tables, I have found that the average number of fatalities per 1000 fires is 3.2 where a fire alarm is present and working. However, the average number of fatal casualties per 1000 fires is 8 where a fire alarm is absent or not working. This demonstrates that when a smoke alarm is absent the number of fatalities more than doubles.
The presence of a fire alarm clearly warns the occupiers that there is a fire. The smoke alarm will go off for two reasons depending upon the type of alarm system. In a radioactive smoke alarm, the smoke will enter the ionization chamber and break the circuit. In an optical smoke alarm, the smoke will enter the sensing chamber and disrupt a light beam which then activates the alarm. Having both alarms, rather than one, would counteract the shortcomings of each, since each responds more quickly to different kinds of fire. The radioactive smoke alarm is better to detect fast flaming fires and the optical smoke alarm is best placed to detect smouldering fires.
I would also advise that any smoke alarm that is installed should be connected to the mains since the main reason for failure of a smoke alarm is removal of batteries, perhaps following nuisance alarms. This happens most often when alarms are fitted too near the kitchen in which steam is inevitable. I would advise that the alarms are placed on the ceilings in the hallways and upper corridors of a home, away from a bathroom if possible so that steam does not enter the alarm.
There would be a number of benefits of all households in the UK applying the methods I have suggested. Obviously there would less fatalities. Also there would be huge savings in the public sector. For example, if UK homes used my suggested methods, there would be less injuries, less hospitalisation and less ambulance call out. Also if fire alarms were installed carefully and maintained properly, people would be alerted to fires quickly and be able to put them out while the fire is in its early stages. Less fire crews would be called out and would be free to deal with bigger and more serious incidents. Insurance premiums would also go down as damage would be confined to fewer items which could ultimately offset the cost of installation of the alarms.
However, there are a number of arguments against making smoke alarms compulsory. Firstly, in this present time the government is making many cuts. To make smoke alarms compulsory would probably mean subsidising their cost or even making them free to certain groups, such as old age pensioners. In addition, the present government is Conservative and is not keen on interfering too much in peoples’ lives so it is unlikely to support legislation which takes away peoples’ freedom of choice. Furthermore, how would you enforce such legislation? There would have to be inspections to ensure that homes were adequately protected and that owners were maintaining their equipment correctly. The trends I have observed above suggest that there will always be some people who do not maintain their alarms correctly, or will deactivate them – possibly due to nuisance alarms or just to use the batteries in another device. In addition, the reliability of smoke alarms does not seem to be improving.
Despite these arguments my opinion is still that smoke alarms should be compulsory. This is because fires in private homes in which smoke alarms alerted the residents are associated with far lower fatalities. Furthermore, the effectiveness of smoke alarms is grossly underestimated since the statistics only count those fires in which the fire service was called, estimated only to be one fifth of all fires. This is because the ownership of smoke alarms is resulting in many people being alerted to fires so quickly the fire is confined to the first item to be ignited and people can deal with the results themselves. I believe that if statistics for ALL fires were available, the case for everyone having smoke alarms would be as clear as everyone having to wear a seat belt.
Glossary
Alpha particles: the nucleus of a helium atom, carrying a positive charge of 2c. Its proton number and neutron number are both 2 so it is a very stable particle
Americium-a radioactive element. This means it is unstable and is trying to stabilise itself by emitting particles.
Atom-the smallest part of an element that can take part in a chemical reaction.
Current-a flow of electricity
Electron-an elementary particle that has a negative charge.
Resistor – a piece of electrical apparatus possessing resistance
Appendix
GCSEscience.com
www.explainthatstuff..com/smokedetector.html
Queensland Fire and Rescue website
Wilshire Fire and Rescue Service website
Fire Statistics United Kingdom 2004
www.explainthatstuff..com/smokedetector.html
Queensland Fire and Rescue website
Wilshire Fire and Rescue Service website
Page 37 Fire and Rescue Service Data
NB: The fire service estimate that over four fifths of fires are detected early enough for owners to deal with them themselves and do not involve the actions of the Fire and Rescue Service and so are not included in these statistics
The ratio of fatalities in fires with working alarms compared to those in fires where there is no working alarm or no alarm at all does not take into account the fact that four fifths of fires in dwellings are not even included in the statistics since the presence of a fire alarm alerted the occupiers to the fire so quickly that the services of the Fire and Rescue service were not even required. The efficiency of smoke alarms is therefore vastly underestimated.