The BBC take up the story

BBC East Midlands Today covered the issue on Tuesday 12th November:

2022 vision

Derbyshire Fire & Rescue Service (DFRS) have a laudable aim: "to make Derbyshire a safer place to live, work and visit." But their plan for how the service will look in 2022 is hard to reconcile with this aim.

I have sympathy for the Service - they are being asked to deliver the same level of protection on markedly less money. My concerns are that the report is not candid about the compromises that this involves, and that if one projects current performance in despatching crews of Retained firefighters (who will assume an even greater role in the planned regime), there is real cause for concern that some incidents will see a very slow response time.

A very poor measure of success

The plan focusses on a single measure of success: the proportion of fires that the Service can reach within 10 minutes of a 999 call. The current target is 80%; the report proposes to cut this to 70%.

While this is one useful measure, it is chilling that it is the only one used. It is a bit like the NHS setting a four-hour target for treating the 70% easiest-to-treat A&E cases - it begs the question of what happens to the other 30%.

We know from the NHS and elsewhere that resource-limited management tend to "manage to the target", and that means that if your needs are simple, then you fare well, but that if you aren't going to score a "hit" in the statistics, you risk experiencing considerably worse service.

Example:  BT used to set just one target for clearing phone-line faults - the % that were fixed within two working days of report. The number was good and improving, yet there were too many complaints about really long delays. Why? Because supervisors would ring round their engineers each lunchtime, and divert those on long jobs destined to miss their target onto quicker jobs that could be turned round within the target. The trickier jobs could take weeks to complete. Eventually BT had to add a second measure (of those jobs taking over a week to fix) to correct this behaviour.

Without a focus on average time to reach ALL fires, there is a real danger that the hard-to-reach locations get badly served. DFRS are never going to hit the 10-minute target for the more remote villages, so when they come to design the network, there is no difference in the 10-minute success rate whether residents face a 15-minute wait or a 30-minute wait. If that is the only statistic being managed, that is worrying.

Fewer Stations, further apart: slower response

The plan envisages fewer fire stations, further apart. We are invited to see this as a logical response to a drop in the number of fires. In a rural area like the Derbyshire Dales, this simply isn't true. Fewer fires may mean that you can manage with fewer appliances, as the chances of having two overlapping fires in a particular area reduce to the point where we have to accept that it would be bad value to keep extra machines against those rare overlaps. But fewer fires does not change the need for a network of fire stations able to send appliances quickly. If my house catches fire, it is little consolation to be told that mine was an increasingly rare occurrence. If there are fewer stations further apart, then response times will go up. Perhaps inevitable, but we should expect more candour on this.

New stations planned for locations which may be problematic

Matlock and Wirksworth fire stations are proposed to close, with a replacement in or near to Cromford, ideally on the A6. This will have one employed crew, and one Retained (part-time) crew.

Three problems here: Cromford is part of the Derwent Valley World Heritage Site, which will make it harder to find a site, and so increase the risk that you have to put the station where you can get the land, rather than where you would ideally put it.

The second problem with Cromford is that it is a small village, and appears to have relatively few people working there. The 2011 census showed 1,466 people, but with a marked shortage of the 25-44 agegroup which must surely be key to Retained (part-time) firefighter recruitment - they have 35% fewer of that group than the national average. So they are more like a village of 950 people. (To corroborate this, the primary school had just 35 pupils at the last report). And that is fairly small when you are looking for a crew of 17 retained firefighters, some of them needing to be in the village during the working day. And you can't recruit from just anywhere, either: the Retained Firefighter response time really limits you to people living and/or working pretty close to the fire station.

The third problem is that in order to get an acceptable travel-to-Matlock time, the station needs to be on the A6, which bypasses the main part of Cromford village, so reducing the number of potential Retained Firefighter recruits still further.

Demanding (or unrealistic?) expectations of Retained Firefighter recruitment

Retained firefighters are part-timers. When there is a fire, they are contacted via an "alerter" (a pager, I presume) and then have five minutes in which to:

• Stop what they are doing (or, at night, wake up and get dressed)
• If on their own, lock up their house or shop; if looking after children, handing these over to someone else
• Get to the Fire Station (if using a car, this includes the time to de-mist the windscreen in winter, and presumably time waiting at traffic lights)
• Get dressed in protective gear
• Board the Fire Appliance and get it moving

Just to emphasise - the plans assume that Retained firefighters do all that in five minutes. And that means you have to work/live pretty close to the station.

It may be realistic to recruit and maintain a sizeable Retained crew if your station is in the centre of a town or large village, especially if there is plenty of local employment. Hathersage and Bradwell aren't enormous communities but both manage to turn out a Retained Firefighter crew more often than the average Derbyshire achievement. But - especially in an area with reasonably heavy traffic - you haven't got a very wide catchment area from which to recruit your firefighters. And if (as in Cromford) you are planning your station to be at one end of the village, the other side of long-phase often-congested traffic lights, then your catchment area would appear extremely small. So I am concerned that Cromford will not manage to recruit and retain the Retained crew of 17 people assumed in the plan.

Although I know the area less well, I have a similar worry about the station proposed to go on the Hathersage Road in Bamford. Again, it is right at one end of a modestly-sized village without any obvious large employer close enough to be relevant. Look at the aerial view on Google maps and there seem to be precious few houses or business premises nearby, and looking at census data for the area, the age profile is again older than average.

A poor track record of availability of existing Retained Firefighter crews

The main report doesn't seem to mention problems with having sufficient Retained firefighters available to turn out when called, but the background document gives some strong clues, albeit with the information dotted around in its 400-odd pages.

I have pulled out the "% Availability 2012-2013" figure for each Retained crew mentioned. Some of these are retained crews supplementing employed crews at the same fire station; more commonly, they are the only crew at a station.

The results alarmed me - just under 75% average availability. See my other post for the data.

Again and again, the background document explains poor availability with the phrase "This is primarily due to staffing vacancies, difficulties in recruiting and staff availability". 

This is really worrying when the Authority is proposing to have fewer stations, further apart, with more of them relying on Retained crews.

There may be even worse news to come - if I can get the source data from DFRS (see note below). What could that worse news be? That there may be times of the week (or year) when there is a considerably lower availability across multiple Retained crews. I don't know the rules or arrangements, but it isn't unthinkable that it could be harder to turn out a crew on a Saturday night or during the school holidays. The already-worrying availability data could be hiding some systematic problems - only access to the source data will reveal this.

No evidence to reassure us that this leaner system will be resilient

The new plan has less resilience than at present. If the retained crew at Buxton can only turn out 12% of the time, that may not be disastrous if the area currently has a generous provision of crews and other stations nearer than current conditions demand. 

But the new plan will have no such redundancy or over-provison. The new stations appear to be mapped to provide "adequate" cover (70% of fires reached within 10 minutes, down from 80%) assuming that they are located in the ideal locations and are at all times fully crewed.

With the provision pared to the bone, there is much less resilience in the event that a crew can't turn out: it will be further to the next station. And even if every station managed 75% availability at all times, that still means a 6% chance that neither of TWO adjacent retained crews would be able to turn out. And nearly 2% risk that none of THREE stations would be able to respond. If (as I fear) there are times of the week and year when availability dips simultaneously across many retained crews, then these risks could be dramatically higher.

Conclusion

While DFRS seem to have done a good job with a mapping program to work out the optimal pattern of fire stations within the resources available, they appear to have overlooked the need to refine that layout so that stations using Retained crews are in locations where the Service can be confident of recruiting and maintaining enough Retained firefighters to give near-100% availability.

Given the current poor performance in turning out Retained crews when required, and the increased role for Retained crews in the new plan, the main report should surely have addressed this issue explicitly.

In the absence of any engagement on this issue by the Service, I worry that we are being asked to consent to a fundamentally unrealistic plan.

---

Note: My grief in getting data from DFRS
Although I asked for information as an FOI request, DFRS took three weeks to decide that they couldn't do this within the £450 cost limit they are allowed to impose - and then offered no more than a meeting with an officer, which I declined as I need data to analyse. Shortly after the BBC started asking about the issue, DFRS changed their mind and have offered to see how far they can go in providing me with some data: I will add a new posting on this as and when I succeed.

Retained Fire Crews - how often are they available?

I extracted the following data from the background document  to the consultation on the proposed 2022 reorganisation of fire services in Derbyshire.

There is no clear description of how they define availability, but I presume that this measures the number of times they were successful in mustering the required size of crew in the required time, as a proportion of all fires when they would like to have sent that crew.

Station	% availability from Retained crew 2012-2013	Page in document
Buxton	11.62%	144
Swadlincote	40.68%	214
Ashbourne 2+	42.20%	172
Glossop	43.25%	134
Melbourne	51.69%	214
Clowne	52.10%	261
Clay Cross 2+	60.30%	251
Long Eaton	62.97%	227
Matlock*	63.63%	162
Shirebrook	72.40%	261
New Mills	74.01%	125
Ilkeston	75.65%	227
Whaley Bridge	81.41%	125
Hathersage	83.27%	153
Alfreton	84.19%	204
Staveley	86.55%	239
Crich	86.86%	204
Ripley	89.97%	204
Heanor	90.03%	204
Bradwell	90.79%	153
Chapel-en-le-Frith	91.08%	125
Duffield	95.54%	194
Ashbourne 1+	97.40%	172
Belper	97.65%	194
Bolsover	97.70%	261
Clay Cross 1+	99.82%	251
Dronfield	100.00%	251
Average	74.92%
* Special case as retained crew coverage is from full-time firefighters living in Fire Service accommodation next to station
+Two on-call appliances at same station: presumed to be % of time they can turn out one and both engines
Page 204 says that Alfreton Retained (84%) has 'above average availability when compared with the wider Service'
No data provided for Bakewell (and perhaps others)

Source: "Fit to Respond for Consultation Reference October 2013" - Derbyshire Fire & Rescue Service