Do response times matter?

The Fire and Rescue Service is data rich. It collects data on all its activities and, when it comes to incidents, the Home Office hosted Incident Recording System (IRS) is a treasure trove of data just waiting to be interrogated. We have done just that and turned our attention to response times to ask the question: ‘do they still matter?’

We are interested in this subject because we think there has been little attention paid to it in recent years. Fire and rescue services set attendance standards as part of their integrated risk management planning process. We know that the financial challenges over recent years have placed greater emphasis on the need to do more with less, but what is the importance of response times?

In this context, fire and rescue services need to consider all aspects of how they provide a response to the public they serve, from identifying and assessing risks to ensuring that resources are optimally deployed to meet demand. Response times are not just about the number on the clock.

There is an understandable emphasis on responding to fires and other emergencies in the HMICFRS inspection methodology. The summary report for the first tranche of inspections confirms that 11 out of 14 services were graded ‘Good’ in this respect, with two services requiring improvement and one judged to be inadequate.

HMICFRS observe in their summary report: ‘It is reasonable for response times to vary depending on risk, geography and demography of an area. But is it is not acceptable for services to commit to a response time and then consistently fail to meet it’.

We are aware that the National Fire Chiefs Council has established the Community Risk Programme, the first part of which asked detailed questions about the process that leads to the creation of an integrated risk management plan through a survey carried out by Nottingham Trent University. We await the results with great interest to see the extent to which attendance standards are based on sound data and analysis.

With this in mind, we take a look at the figures to see what has been happening with response times using the IRS data for all fire incidents in England over the last nine years.

There has been a well-publicised and substantial reduction in the number of fire incidents in England since 2009/10. During the same period there have also been substantial declines in the number of fire-related fatalities and casualties – important success stories for the fire sector. Figure 1 below shows the changes for five incident types over a nine-year period, noting that this includes the Grenfell Tower fire in June 2017. Excluding the Grenfell tragedy, the decline in fatalities would be 23.5 per cent and the decline in injuries 21.3 per cent.

 

 

Figure 1: Number of Fires and Fire-related Casualties by Financial Year

Data Source: Home Office Incident Recording System (www.gov.uk)

 

Decline in Response Times

However, this reduction in incident volumes has not translated into an improvement in response times. In fact, the opposite is true. While our analysis in this article focuses on the last nine years of data, we find that if we look back to 2001/02 we can see that this trend for increased response times has its origins much earlier and was more pronounced than in more recent years. The average response time for the first appliance to dwelling fires has deteriorated from just under six minutes in 2001/02 to nearly eight minutes in 2017/18.

These changes are more substantial for fires in other buildings, road vehicles and other outdoor fires (see Figure 2). It is a weakness of the Incident Recording System that there is no national data for RTC or other special service incidents, or for second appliance attendance; this restricts the ability to understand the impact of response times.

 

 

Figure 2

 

We have looked at this in more detail to see how this varies across the country. As with the fall in incident volumes, the increase to average response times can also be observed across most fire and rescue services. In 2009/10, there were five FRSs with an average first response to all fires of more than ten minutes; by 2017/18 there were 23 FRSs (half of those in England) that were above this measure. The extensive spread of the dark red in Figure 3 shows this at a glance.

 

 

Figure 3

 

This raises an interesting question: if services are taking longer to respond, but the number of fatalities and casualties has decreased, does the response time to fire incidents actually matter?

We have used IRS data for 400,000 fire incidents to examine the relationships between response times and the effects of fires at a national level. This broad approach has uncovered some clear outputs that are linked to the economic cost of fire.

Firstly, there is a correlation between longer response times and the amount of fire damage at incidents, and this is apparent in all types of fire and rescue services (see Figure 4). The relationship is particularly strong for non-dwelling fires, where an extra minute for the first response is equivalent to an additional three-square metres of fire damage to the building.

 

 

Figure 4

 

We found that there are similar relationships between longer response times and the time spent at the scene and the number of resources that are mobilised to the incident. None of these are surprising outcomes, but they reinforce the importance of a timely first response in order to mitigate the impacts of building fires. Thinking about the economic cost of fire, greater fire damage can lead to:

 

• Greater risk for firefighters at larger incidents
• Increased salary/equipment costs for FRSs
• Reduced capability to respond to other incidents
• Larger societal and individual impacts.

 

“We know that the financial challenges over recent years have placed greater emphasis on the need to do more with less, but what is the importance of response times?”

 

There is no correlation in the data between response times and the likelihood of fatalities or casualties.

Trying to understand how certain factors are influencing response times is a difficult process. It is possible to infer relationships for the effects of reduced staff numbers, lower availability of on-call resources, increased traffic volumes and fewer stations/appliances; however, correlation does not necessarily imply causation. In order to ascertain why response times have increased, we need to consider this at a local level.

Some FRSs have maintained their average response times while achieving the same levels of efficiency savings as other FRSs – it is not always the case that cuts to front-line services equate to longer response times. There are operational factors involved here too. Several FRSs have implemented new strategies for how they target specific incidents and risks across their service area. For example, prioritising incidents, supplementing on-call crews with whole time staff or taking a more dynamic approach to deploying resources on a day-to-day basis.

Since 2004, FRSs have taken responsibility for setting attendance standards within their own service area. This move away from nationally prescribed standards has led to considerable variation in the types of measures that individual FRSs use. Figure 5 shows the different approaches that FRSs take to define and measure attendance standards.

There are advantages and disadvantages associated with each of the approaches, hence the situation where all FRSs have developed their own different standards. This lack of conformity does raise questions about what is an appropriate approach and how transparent this is to the public. Indeed, this point is well made by HMICFRS who observe that a lack of consistent, comparable and good quality data makes it hard for the public to know if they are getting a good service or not. The summary report concludes: ‘This situation must improve’.

 

 

Figure 5

 

Seeking Improvement

In July 2017, NHS England introduced new national attendance standards as part of the Ambulance Response Programme. There are mean and 90th percentile targets, set according to the priority of the call. Consequently, Ambulance Service response times are in the public domain; they are reported on and monitored by the media. The public therefore has certain expectations for the attendance times of ambulances, but what would or should they expect from a fire appliance?

 

“Sorting out the definition and the measurement will go a long way to improve the national view on the data so that it is transparent and comparable for all to see”

 

We presented our work on response times at this year’s LGA Fire Conference. We were unsure of the reaction we would receive and were pleasantly surprised by the level of interest and engagement in the discussion by many principal officers in fire and rescue services.

While there was consensus that national response standards are no longer appropriate, it is clear that the range of different measures does not lead to transparency. At a minimum, FRSs should use the same start and end points for measuring response times. Greater conformity on the types of incidents and the process for reporting would also lead to greater clarity. We can see that the creation of the Fire Standards Board represents an opportunity to set a national definition of what constitutes a response time.

Response Times Matter

Returning to the question ‘do response times matter?’ we clearly think that they do. They matter because they give the public an assurance about what to expect from their local fire and rescue service based on the local risks that affect them. Sorting out the definition and the measurement will go a long way to improve the national view on the data so that it is transparent and comparable for all to see.

The statistics show that as the number of incidents has decreased and response times have increased, yet there is no correlation between response times and the likelihood of fatalities/casualties. If that is correct, then the case for improved response times becomes primarily an economic rather than life based argument.

We make this point above and think that it is timely for the government to update the economic cost of fire; it has been over ten years since the last version was published. We think that it will be useful to HMICFRS in terms of their efficiency assessments and those interested in using fire data to fully understand the impact of changes to not only response times but the whole range of fire and rescue service business.