38th Meeting of the
Fire & Rescue Statistics User Group
held on
27th July 2013, 1:15 pm – 4:00 pm in
Eland House, Victoria SW1

List of attendees

Charlotte Cheal	FPA
Chris Bigland	LFB
Dave Sibert	FBU
David Townsend	International Fire Investigators and Consultants
David Wales	South East Fire Investigation Group
Dennis Davis	Fire Sector Federation
Gavin Sayer	DCLG Fire stats
Kirsty Bosley	FRSUG Chair Scottish Government – Justice Analytical Services
Nazneen Chowdery	FRSUG Secretary, DCLG
Riana Smit	LFB
Sheila Pantry	Fire Information Group UK (FIG UK)
Simon Flood	South West FRSs and Family Group 4
Steve Emery	English Heritage
Tony Paish	World Fire Statistics Centre
Richard Hall	FireBrake – Wales
Heidi Jones	DCLG
Robin Bryant	Canterbury Christ Church University
Graham Ellicott	Fire Industry Association
Terry Edge	BIS
Steve Kelly	DCLG

1. Chair’s Introduction

Kirsty Bosley welcomed everyone to the meeting.

2. Apologies

Bob Bantock, Steward Kidd, Dave Berry, Julia McMorrow, Daniel Walker-Nolan, David Townsend, Rob Gazzard, Ron Nairn

3. Minutes of the previous meeting

Dennis pointed out that his comments regarding non-FRS input into the IRS review had not been fully recorded on the minutes. Some text was agreed and will be inserted. Otherwise the minutes were approved and can now be published on the FRSUG website and the Knowledge Hub (if access to the Knowledge Hub can be recovered).

4. Matters arising

No points were raised, most matters to be covered under specific agenda items.

5. FRSUG proposed action plan 2013 including

a. Website update and other publicity

Sheila Pantry proposed a couple of suggestions to improve the contents. She requested members to let her know about any specific journal which they thought would be useful for the wider community. She can put the link to the website. She also asked to be kept informed of the date of upcoming events. Any changes in professional career or contact details can also be updated through the website.

b. Fire Information Group UK – 25th Anniversary report back

Sheila Pantry said that the Group hosted an excellent event and they received a lot of appreciative comments. She said that a number of interesting topics were presented and are now available on the FIG website (www.figuk.org.uk) for members to read. FRSUG members expressed their appreciation of Sheila’s work to organise such a valuable and well run event.

c. Dissemination activities – distribution and membership update

Kirsty Bosley said that CFOA had not been in the membership list but David Wales and Simon Flood both represent CFOA groups and said that CFOA are planning a research group. The lead of that research group would probably be an appropriate representative on FRSUG. Heidi also suggested Stuart May to attend the FRSUG meeting.

d. FRSUG action plan

Kirsty Bosley said that she would like to review the on-going action plan to set out what the group should do and how it can be taking forward. She asked the group to consider issues that they felt could usefully be added to the plan. While FRSUG may not take forward the activities, it will be useful to monitor them and retain them on the action plan agenda item while they are active.

Graham Ellicott suggested that it would be useful to get some insight into the fire false alarm incidents, in particular the number of incidents due to failure of technology. There was some discussion about incidents that are recorded as false alarm due to equipment failure, but where it is not actually the case. The issue was noted and added to the action plan.

David Wales added that research works produced locally, for example within FRSs, are not usually publicly accessible. To get best value from this kind of work it needs to be shared and discussed.

David Sibert said that circulation of relevant presentations to the FRSUG and possible endorsement by the group could be useful, though this would need some objective tests to identify and evaluate the benefits of research. He felt FRSUG could provide a forum to bring together diverse national interests.

Richard Hall sought to find a way to share information about deaths and injuries prevented through Community Fire Safety. He has three data collection forms and would like to know what other FRSs collect in the area and the reasons behind it.

Sheila Pantry said that members would be interested in research work carried out at the university level. They may take particular interest in the Greenwich University research on evacuation. She added that Health and Safety laboratories in Buxton have a wealth of research materials. She offered to approach some of her contacts to see if they would be willing to deliver a presentation to FRSUG.

Actions: Sheila to approach HSE and academic contacts with a view to their presenting at future FRSUG meetings.

Kirsty to build on the presentation she gave at the FIG seminar to inform FRSs about FRSUG.

All members to consider and share what data is collected locally and what research is carried out.

Kirsty to start a proposal about developing the action plan to send round and develop by correspondence

e. Published since last meeting

Gavin Sayer informed that latest release of Fire Statistics Monitor series was published on 26th June. It covers incident data up to the end of March 2013. The release was accompanied by 38 spread sheet tables and about two-thirds are fire and rescue services area based. He added that new diagrams like brick and bubble charts are used for the first time to present the data. This has been appreciated with positive feedback from non-experts and he welcomed comments from FRSUG members.

David Wales had recently attended the Interflam conference and noted that two volumes of proceedings of the event on evacuation were published, including papers from Prof Ed Galea and Dr Vito Babrauskus.

Action: Gavin and David to send the associated links for inclusion on the FRSUG website.

6. Economic cost of fire

David Sibert gave an update on Fire Sector Federation work on the issue. He reported that the federation was focusing on two aspects:

1. Develop the existing cost of fire model and
2. Consider the value of the fire sector to the community.

He added that DCLG had been planning to update their model using 2010 data, though this work had been postponed and further timescales are now being determined. The Federation work will have to take into account the impact of new pension calculations.

David further said that after the January meeting a scoping paper was sent out widely and six institutions have come forward and expressed their interests in the work. On 23rd July the Federation is carrying out interviews to identify an organisation, possibly in partnership, to carry the work forward.

Action: David and Dennis to provide feedback at the next FRSUG meeting.

7. Scottish report

Kirsty Bosley reported that, with the new single service in Scotland (the SFRS), data would be published at eight fire and rescue services level for the last time this year – Fire and Rescue Service data in August and Incident data in October. The SFRS is now organised into three ‘hubs’ and 17 Local Senior Officer areas of one or more local authorities. Scottish Government will then hold a consultation to decide how the data should be published, possibly local authority level or LSO area.

She also pointed out that the Scottish published data has been incorporated into performance indicators. Some publication and data alignment issues are likely to arise with the single service: where data is reported by the SFRS appropriate processes are needed to ensure SG and SFRS data is not contradictory.

Scottish Government has been looking into a discontinuity in casualty reporting between the FDR1 and IRS reporting system. Since 2009-10 fire casualties in Scotland have been increasing, which is not in keeping with all other fire trends, and so an internal investigation is being carried out to establish whether this is related to the transition to IRS data.

HMFSIS asked about research on Home Fire Safety Visits and their value. A short review of the existing analysis revealed varying success rates and Scotland will be holding a seminar discussing the issue with the new fire board later in the summer.

Dennis Davis pointed out that recently a lot of changes to the IRS questionnaire have been made. Introducing these changes in a step by step process would bring a better result. He informed that policy colleagues are organising a seminar on fire safety guidance which would be of great benefit to FRS community.

Action: Kirsty to report back from the HFSV seminar to the next FRSUG meeting.

Action: Dennis to bring the report from policy colleagues

8. DCLG update

Mark Dunn will give an update on Sir Ken Knight review. Members may take interest in and they may have issues on the review.

Action: Gavin will ask colleagues who are still here (as Mark Dunn left DCLG in July) on plans for Cost of Fire

9. IRS

Heidi talked about the IRS review, pointing out that the system was originally designed in 2005/06 and technology has moved on. Direct data entry via the IRS web forms is currently used by 50% of FRAs while the rest use their own MIS with an IRS interface. She added that as part of the review work they are aiming to replace the current system and introduce extra functionality, but keeping the same look and feel as the current IRS.

She further said that they are focussing on three strands of work –

1. getting approval for IT improvement
2. defining what will be collected in future and
3. writing the specification

Heidi added that DCLG is working with CFOA about re-phrasing the queries correctly. She said that over three hundred views on a range of issues were up on the website. An IRS review group, led by Stuart May of CFOA will review them and make their decision. She further said that the website would change during summer and new website with correct functionality would be launched. This included

1. specification of the new system including new requirements on how IRS can be improved
2. addition of local datasets which are comparable
3. changing the way the guidance is structured
4. allowing the overlap of datasets on map
5. looking at Dashboard and use of IRS in mobile phone
6. Refinement of FSEC

There was a discussion on the data sharing issue. Gavin Sayer said that users will see a dashboard by using a specific link and can share data. He said that there are two issues which need to be paid attention (i) sharing data with wider community and (ii) dealing with sensitive information while sharing.

Simon Flood asked about reporting tools from the new data. Heidi said they have a developer working on a system that would allow users to build their own reports with access to all of the data.

David Sibert asked if FRA can flag up any user issue. Heidi Jones replied that they may pilot some of the changes as they’re introduced and this would allow issues to be raised before the system is fully implemented.

10. IRS from the FRS point of view – Simon Flood

Simon Flood demonstrated the IRS system and associated issues allowing FRSUG members who were not familiar with IRS data entry an opportunity to better understand the system.

The key points of the presentation were

• Problem with the current system: it does not allow carrying out any revision to the pre-populated information. Therefore, it may not be possible to derive meaningful data.
• The system’s inbuilt validation checks are not straight forward
• Possible improvements with regard to various options associated with each question
• Access of IRS data to wider community
• For a fire, there are 10 web pages of data entry, though the screens change according to the incident type, with fewer questions asked about false alarms for example.
• Data that is collected by Control are pre-populated
• There was discussion about some data fields and, in particular, around the IRS property type and various other sources of gazetteer and also about making appliance types consistent. These issues are being addressed in the IRS review

The presentation was greatly appreciated and provided valuable insight into the data collection.

One of the FRSUG members asked if non-FRS members could get access to the system. Gavin said that it wasn’t possible to access it generally, but he was willing to organise a demonstration at Hemel Hempstead if required.

Jeremy asked about error checking on data input and also whether the data around “Area of damage” could be improved for analytical purposes. He also raised the issue about identifying baby products that have been involved in fires. Recent analysis in support of legislation struggled to isolate this data from published records. Gavin suggested contacting DCLG in this kind of instance, where the data may be available in free text records, but would not be routinely published.

11. Analysis of electrical fire data – David Wales and Robin Bryant

This analysis started with a review of IRS data and deciding how to break down the data and take positive action from it. The crime team from Canterbury Christ Church University carried out quantitative and qualitative analysis of around 300 data records to try and establish confidence and understanding from the data.

The key points of the presentation were

• Canterbury Christ Church University analysed Kent Fire and Rescue Services incident data covering the period April 2009 to July 2012 with particular focus on fires with an electrical aspects and electrical malfunction
• Fires were categorised into Fire with an Electrical Aspect (FEA) which covered any fire where an electrical aspect is involved. While Electrical malfunction (EMF) is a subset of FEA where fires caused by malfunction, failure or similar reason. The latter would more commonly be recognised as having an electrical cause.
• A Poisson distribution was fitted to the FEA and EMF data, with actual numbers of fires per day checked against expected. A Chi-squared goodness of fit test showed a significant result (p<0.01) suggesting that the fires could be predicted from the sample. However, the result can not be justified because of small sample.
• A trend-line was constructed using the monthly FEA data which showed that the number of FEAs declined over the study period.
• While the numbers of all fires were reducing, the numbers of electrical fires were stable, therefore the proportion of EMF is increasing.
• Data mining techniques were used to draw out aspects of human behaviour in electrical fires.

The emerging recommendations of the analysis include

• developing a national definition of an ‘electrical fire’,
• considering the impact of human behaviour, and therefore education on electrical fires
• considering how to combine data for example to identify any common themes, such as certain manufacturers likely to be involved in fires.

12. UK Timber Framed Association report – Dave Berry

This report was postponed as Dave Berry was not able to attend the meeting. But he sent the following note for members: (i) The UKTFA have made some proposals to the IRS Working Party for changes to the IRS to properly identify the different ranges of Timber Frame Structures and (ii) The members of the UKTFA have agreed that the information provided to the CFOA Timber Frame Site Notification should be made available to HSE.

13. AOB

Gavin Sayer said that the department was planning a ‘statistics user day’ in the autumn where the Head of Profession would give an introductory speech.

Terry Edge said that BIS just finished the review of furniture safety (fire) regulations.

14. Date of next meeting

The next meeting will be held in October 2013. Nazneen will notify the members about the possible date via Doodle poll.