project report · 2018-07-22 · project report preparation of housing health and safety rating...

Project R

eport – Preparation of H

ousing Health and S

afety Rating S

ystem G

uidance (Version 2)O

DP

M

housing

Project Report

Preparation of

Housing Health and

Safety Rating System

Guidance (Version 2)

Project Report

Preparation of Housing Health and Safety

Rating System Guidance (Version 2)

November 2004

Safe & Healthy Housing Research Unit, Warwick Law School

Office of the Deputy Prime Minister: London

The Office of the Deputy Prime MinisterEland HouseBressenden PlaceLondon SW1E 5DUTelephone: 020 7944 4400Web site: www.odpm.gov.uk

© Queen’s Printer and Controller of Her Majesty’s Stationery Office, 2004.

Copyright in the typographical arrangement rests with the Crown.

This publication, excluding logos, may be reproduced free of charge in any format or medium forresearch, private study or for internal circulation within an organisation. This is subject to it beingreproduced accurately and not used in a misleading context. The material must be acknowledged asCrown copyright and the title of the publication specified.

For any other use of this material, please write to HMSO Licensing, St Clements House, 2-16 Colegate, Norwich NR3 1BQ. Fax: 01603 723000 or e-mail: [email protected].

Further copies of this publication are available from:

ODPM PublicationsPO Box 236WetherbyWest YorkshireLS23 7NBTel: 0870 1226 236Fax: 0870 1226 237Textphone: 0870 1207 405E-mail: [email protected] online via www.odpm.gov.uk

Printed in Great Britain on material containing 75% post-consumer waste and 25% ECF pulp.

November 2004

Reference no. 04 HC 02608(b)

CONTENTSACKNOWLEDGEMENTS IV

CHAPTER 1Introduction 1

Background 1This Project 2

CHAPTER 2Tasks in the Development 4

Guidance 4The General Information and Guidance 4The Hazard Profiles 6

Worked Examples 7Software 9Paper Scoring Form 11

CHAPTER 3Recommendations 13

General 13Public Release of Version 2 14Training 15

ANNEXRevisions Suggested for HHSRS Version 2 16

Acknowledgements

The project was commissioned by the Office of the Deputy Prime Minister (ODPM) and wascarried out by a team from the Safe and Healthy Housing Research Unit of the Law School,University of Warwick.

Work developing the associated software was carried out by the University of WarwickInformation Technology Services. Thanks also go to Ged Burden, Mel Cairns, Viv Mason andMark Nelson who provided useful comments on the software and paper scoring form.

The project team are grateful for the co-operation and assistance of staff of the ODPM, inparticular Dr Christina Golton and Kevin Roper.

Stephen BattersbyRachel CourtRichard MooreDavid Ormandy (project manager)April 2004

CHAPTER 1

Introduction

Background

1.01 Following a review of the Housing Fitness Standard in 19961 and consultation on thedevelopment of a methodology to grade the severity of housing conditions2, theGovernment commissioned work to develop the Housing Health and Safety RatingSystem (HHSRS or Rating System). The objectives of this work were to:

l Devise a logical and practical means of assessing and grading dwellings from ahealth and safety perspective, such means being capable of replacing the currentHousing Fitness Standard.

l Develop an electronic dwelling hazard survey program for use with hand-heldcomputers.

l Make recommendations for the interpretation of the results, including banding andaction levels.

l Assess the implications of a hazard based assessment system for enforcement andmake suggestions for options for action.

1.02 In July 2000, the government released Version 1 of the HHSRS. The release includedGuidance (Version 1), Report on Development, and a Housing Survey and ScoringProgram for Palm OS hand-held computers3. At this stage the System was notincorporated in any formal assessment procedures. However, the Government hadannounced its intention to replace the Housing Fitness Standard with the HHSRS. Thisrelease was a further stage in the development and consultation process, allowingpotential users to become familiar with the System and its application.

1.03 To complement the Guidance (Version 1) and promote consistency of assessment, workwas commissioned to develop illustrations of housing hazards with detailedexplanations of the assessment using the HHSRS. These Worked Examples werepublished in April 20014.

1

1 See Controlling Minimum Standards in Existing Housing (1998) LRI, Warwick.

2 See Health and Safety in Housing: Replacement of the Housing Fitness Standard by the Housing Health andSafety Rating System (2001) DETR, London.

3 Housing Health and Safety Rating System: The Guidance (Version 1) (2000) DETR, London; Housing Healthand Safety Rating System: Report on Development (2000) DETR, London; and Housing Health and Safety RatingSystem: Programme (Version 1) (2000) DETR, London.

4 Housing Health and Safety Rating System: Worked Examples (2001) DETR, London.

1.04 In late 2000, the government commissioned three separate projects on the HHSRS:

l An evaluation of potential users’ reaction to Version 1.

l A study into the application of the Rating System in houses in multiple occupation(HMOs).

l The refining and updating of the statistical evidence supporting the System.

1.05 Reports from these three projects were published in early 20035.

This Project

1.06 In March 2003 the government commissioned the Safe and Healthy Housing ResearchUnit of Warwick Law School to develop Version 2 of the HHSRS. The tasks involvedwere limited to:

l Production of a revised version of the HHSRS Guidance incorporatingrecommendations as required by the Office of the Deputy Prime Minister (ODPM).

l Design of a paper scoring form.

l Revision and testing of electronic scoring programs (based on the Version 1 surveyprogram) for use with two hand-held operating systems and on desktop PCs.

l Production of a desktop management program to store and handle data from thehand-held scoring programs.

l Production of a Handbook to support the scoring and management programs.

l Revision and updating of the existing Worked Examples and the production ofadditional Worked Examples.

1.07 A major stipulation made by ODPM was that the core principles of the HHSRS were toremain unchanged. These included:

l the two step approach of first judging the likelihood and then judging the outcomes;

l the formula used to generate the HHSRS ratings; and

l the list of hazards to be covered.

1.08 Other tasks were also suggested by the Warwick team, but not included in this project.These included additional research into the potential cocktail effect, and establishing aprocess for the development of a library of Worked Examples.

Project Report: Preparation of Housing Health and Safety Rating System Guidance (Version 2)

2

5 Evaluation of Version 1 of the Housing Health and Safety Rating System – final report (2003) ODPM, London;The Application of the Housing Health and Safety Rating System in Houses in Multiple Occupation (2003)ODPM, London; and Statistical Evidence to Support the Housing Health and Safety Rating System (Volumes 1, 2 and 3) (2003) ODPM, London.

1.09 This report describes the work involved in the development of Version 2 of the HHSRS.It first reports on the revision of the Guidance; second the revision and production ofthe Worked Examples, then the development of the software and handbook, and finallydiscusses the paper scoring form. The report concludes by putting forward somerecommendations for future developments.

1.10 Although these various aspects of the development are considered separately, each, of necessity, depended and related to the others to ensure that the final results wereconsistent and complementary.


3

CHAPTER 2

Tasks in the Development

Guidance

2.01 The Guidance provides a brief introduction and background to the HHSRS, theprinciples underlying the HHSRS, advice on assessing conditions, and how to scorehazards. As well as the general information, the Guidance contains profiles of theindividual potential housing hazards, including the potential for harm (e.g. how eachcan affect health and/or safety), a summary of the statistical evidence, and the housingfactors which can increase or reduce the threat to health.

2.02 Before commencing the revision of the Guidance, all the comments andrecommendations from the three associated projects6, the projects team’s notes ofdetailed problems, and the queries received by the ODPM through the website ande-mail help-line were collated, reviewed and their implications considered.

2.03 Revision of the Guidance fell naturally into two parts – first, the general informationand guidance, and then the profiles for the individual hazards. Revision of the generalinformation primarily involved taking account of recommendations previously made7.Revision of the profiles took account of the recommendations, and also included up-dating the statistical information based on the Statistical Evidence project8, up-datingrelevant research findings, researching and up-dating the Ideal, and providing advice on the assessment for each hazard.

The General Information and Guidance

2.04 There were several drafts of the general information and guidance. The first was anupdate of the 2000 Version 1 of the Guidance, taking into account the variouscomments and recommendations and expanding those areas where clarification seemednecessary.9

2.05 As it was based on the Version 1 Guidance, this draft used the same terminology andincluded a detailed description of a suggested survey procedure as well as explainingthe Rating System and the assessment procedure. The inclusions of the suggestedsurvey procedure had been included in Version 1 Guidance as an electronic surveyprogram had been released as an integral part of Version 1. The survey procedure tookthe user through the stages of that program as well as giving guidance on housingsurveys generally.

4

6 Evaluation of Version 1 of the Housing Health and Safety Rating System – final report (2003) ODPM, London;The Application of the Housing Health and Safety Rating System in Houses in Multiple Occupation (2003)ODPM, London; and Statistical Evidence to Support the Housing Health and Safety Rating System (Volumes 1, 2and 3) (2003) ODPM, London.

7 See Evaluation of Version 1 of the Housing Health and Safety Rating System – final report (2003); and TheApplication of the Housing Health and Safety Rating System in Houses in Multiple Occupation (2003).

8 See Statistical Evidence to Support the Housing Health and Safety Rating System (Volumes 1, 2 and 3) (2003).

9 See Annex A for the list of comments and recommendations and the action taken.

2.06 In addition to those comments and suggestions raised in the Report on Evaluation ofVersion 1 and the Report on the Application of the HHSRS to HMOs, several otherissues were clarified. These were issues that had been noted by the project team inmeetings, training sessions, or found in various publications. They included making it clear that the assessment was of the likelihood of an occurrence over the twelvemonths following the survey, that ‘vulnerability’ was limited to age groups, thatdisrepair could frequently lead to hazards, and that the HHSRS assessment was only the first factor in the process of deciding whether enforcement action was appropriateor necessary.

2.07 This first draft was submitted to ODPM for comment in July 2003. A second draft wasprepared taking account of the various comments and suggestions received. This draftstill included the suggested survey procedure, but as an Annex. It also included anextended list of examples for the four Classes of Harm. This second draft was preparedby the end of September 2003. By this time, a draft Housing Bill had been published,and this included provisions on enforcement action based on the assessment ofconditions using the HHSRS.

2.08 In December 2003, the Housing Bill was introduced in Parliament, and to coincide withthis, the Government issued a consultation paper on the proposed guidance on theenforcement procedures. To inform the Members of Parliament debating the HousingBill and as reference material for the consultation process, a further draft of HHSRSGuidance was prepared which incorporated the Hazard Profiles. This was clearlymarked as an unfinalised draft, and was made available on the ODPM website.

2.09 A finalised draft of the Guidance was prepared, and after a final proof-reading andcheck, was submitted to ODPM by the end of March 2004. The general explanation,information and advice in this final version covered the background, theory andprinciples underlying the HHSRS, the terminology used, a general overview of RatingHazards, advice on assessing conditions, and supplemental advice on assessingconditions in multi-occupied buildings. Annexes provided information on responsibilityfor hazards, a suggested survey procedure, examples of health outcomes for the Classesof Harm, the Hazard Profiles, and selected references.

2.10 As the software developed for Version 210 only covered the rating/scoring of hazards(with no provision for recording information on conditions noted during a survey), thesuggested survey procedure in the final version of the Guidance was in general terms.The survey procedure Annex also included a suggested paper based Scoring Form11.

2.11 The expected shelf-life of the Guidance is uncertain. It was therefore necessary toinclude reminders to users that they should ensure they keep up-to-date with currentresearch and developments.


5

10 See paras 2.29 – 2.40

11 On this see paras 2.41 – 2.44.

The Hazard Profiles

2.12 The Profiles of Hazards in the 2000 Version 1 Guidance were given in rank order basedon risk – i.e., from those that were most prevalent and resulted in the greatest severityof harm, to the least severe and most infrequently occurring hazards. While thisarrangement was logical, the reason was not necessarily immediately apparent or user-friendly. After considering various options, the one chosen for Version 2 was to arrangethe Profiles into four groups reflecting the nature of the threat to health. These groupswere then sub-divided according to the nature of the hazard.

2.13 The grouping and sub-division of hazards adopted for Version 2 was as follows:

A – Physiological RequirementsHygrothermal conditionsPollutants (non microbial)

B – Psychological requirementsSpace, security, light and noise

C – Protection against infectionHygiene, sanitation and water supply

D – Protection against accidentsFallsElectric shocks, Fires, Burns and scaldsCollisions, cuts and strains

2.14 As well as being consistent with the spirit of the HHSRS, this approach indicated thatthe Rating System covered all the significant potential threats to health and safety whichmay be found in housing.

2.15 Although the order was changed, the Hazards for Version 2 are the same as for Version1, with minor amendment to the hazard titles. The format for each Profile followed asimilar pattern to that in Version 1. However, the structure for the Profiles wasexpanded to include a description and definition of the Hazard, its potential for harm (including the statistical averages), the causes of the hazard, preventativemeasures and the Ideal (the current optimum to avoid or minimise the Hazard), acheck-list of matters which may affect the likelihood and/or outcome, and advice onthe assessment of that Hazard.

2.16 A major change from Version 1 was in the statistical information given. First, for Version1, the base population for the average likelihoods differed for each Hazard. Forexample, for the Hazard of Damp and Mould Growth etc. the base was “the populationof people living in dwellings that are damp or have defective ventilation or heating”;while the base for Entry by Intruders was “the population of people living in dwellingswith insecure doors and windows”. This approach was criticised as confusing12. So, forVersion 2, the national averages were given using as a base the age range of thepopulation most vulnerable to that particular hazard. This age group was specified(although for some Hazards, where no age group is more vulnerable than others, theaverages relate to the total population).


6

12 See in particular Recommendations Nos. 22, 23 and 24, Annex A.

2.17 For the December 2003 draft, detailed information was not available on the basepopulation used in the calculations for some statistical averages, including those where the averages were unchanged from Version 1 during the Statistics project. Thisinformation was obtained from ODPM, and, in January 2004, the base recalculated toensure consistency for all Hazards.

2.18 The change of the base for the statistical averages had a major consequence for theaverage likelihoods given. Any differences in the averages which may have resultedfrom up-dated and refined statistics used in the Statistics project13 would be relativelyminor had the base population remained the same for Version 1 and Version 2.However, the change of the base resulted in a dramatic difference in the averagelikelihoods for several Hazards between those given in the two Versions. Therecalculation in January 2004 of the remaining statistics also meant there were somedifferences between the December 2003 draft and the final Version 2 Guidance.

2.19 An additional change from Version 1 was in relation to the type and age of dwellings.For Version 1, only a limited number of averages were given. For Version 2, thestatistical averages given were for up to eight different ages and types of dwellings, aswell as the averages for ‘all dwellings’. However, a further change from the December2003 draft was to identify those statistical averages where, because of the small numberof cases, the confidence level for the different types and ages of dwellings was low. Forthese, the statistical averages were given for all flats, or in some cases, all dwellings.

2.20 Also given in the Version 2 Profiles was the strength of the evidence for the statisticalaverages including an indication of where there might be over or under estimation.

2.21 The published research evidence for each Hazard was reviewed and updated. However,as the expected shelf-life of the Guidance is uncertain, references to current standardswere kept to a minimum.

Worked Examples

2.22 A separate complementary project carried out after the development of Version 1 wasthe preparation of Worked Examples to illustrate the assessment of particular Hazardsusing the HHSRS14. These were particular well received15, and the ODPM wished toexpand the number of Examples, as well as update the existing Examples.

2.23 Based on the Version 1 Examples, the format for the Version 2 Worked Examples wasredesigned. Side 1 remained very much the same as the original, but including theVersion number and date. References to the Fitness standard on Side 2 wereunnecessary and removed leaving more space for information on the justification for the assessment if needed. Also, as the statistical evidence had been refined16, it waspossible on Side 2 to show more accurately the relevant national averages for thelikelihood and the outcomes. The amended layout and format for the WorkedExamples, though giving more information, was designed to appear less cluttered,particularly on Side 2 (see Figure 1).


7

13 See Statistical Evidence to Support the Housing Health and Safety Rating System (Volumes 1, 2 and 3) (2003)ODPM, London.

14 Housing Health and Safety Rating System: Worked Examples (2001) DETR, London.

15 See for example Recommendations Nos. 10, 15 and 114, Annex A.

16 By the Statistics project – see Statistical Evidence to Support the Housing Health and Safety Rating System(Volumes 1, 2 and 3) (2003) ODPM, London.

Figure 1 – Side 2, Typical Version 2 Worked Example

2.24 Each Worked Example shows, on Side 2, the average likelihood and average outcomesfor the relevant age group for the particular Hazard illustrated. After 48 of the originalWorked Examples were put into the new layout and format, the revised and up-datedVersion 2 averages were entered. The examples were then scored, reviewed, and theWorked Examples finalised.

2.25 One change to the basic HHSRS principles proposed to ODPM by the project team wasa minor amendment to the Likelihood Scale. This was to insert a range of 1 in 2 and, toretain the same number of cells (16), to delete the lowest cell leaving 1 in 5600 as theleast likelihood in the Scale. This change was accepted by ODPM and made throughoutthe HHSRS components.

2.26 The procedure for preparing the Worked Examples followed that adopted for theoriginal project. One team member prepared the revised Worked Example and reviewedthe score. This was then circulated to the other team members. It was checked andproof-read, and any ambiguity or disagreement was highlighted and reasons given. As these were revised Version 1 Examples, there was relatively little discussionnecessary between the team members to reach agreement and finalise them.


8

2.27 Eighteen new Worked Examples were prepared, again following a similar procedure. Asthese were new Examples, there was more debate over each, and, in those cases wherethere was disagreement, the Example was discussed until consensus was reached andthe Example could be finalised. However, in the event there was no wide variation inthe scores and any discussions tended to centre around the description of the Hazardand the justification for the scoring, rather than the scores themselves.

2.28 The introduction and explanation of the Worked Examples were revised, and, after afinal proof-reading and check, the Worked Examples were submitted to the ODPM bythe end of March 2004.

Software

2.29 For Version 1 an electronic survey program for Palm OS hand-held computers wasdeveloped17. There was considerable criticism of this program, particularly the surveyelement and of the lack of any Desktop PC data handling program18.

2.30 In the light of the evaluation of Version 1, ODPM decided that, for Version 2, only thescoring element of the program would be developed. This scoring element was to bedeveloped for both Palm OS and Microsoft PocketPC19 hand-held computers and forDesktop PCs. In addition, the ODPM required a Desktop PC data handling programto be developed which would handled and display the results of rating hazards at adwelling in a user-friendly form.

2.31 The structure of the Scoring program was to include the address of the dwelling andinformation about the type of dwelling, the surveyor, and individual screens for scoringeach Hazard.

2.32 Two major improvements were planned for the Scoring element of the program. First,the layout of, and information shown on, the scoring screens would mimic as close aspossible that contained in Side 2 of the Worked Examples (see Figure 2). This wouldgiven consistency across the HHSRS components.


9

17 See Chapter 4, Housing Health and Safety Rating System: Report on Development (2000) DETR, London.

18 See Recommendations 45-94, Annex A.

19 PocketPC is the replacement name for Windows CE.

Figure 2 – Typical Scoring Screen from HHSRS Scoring Program

2.33 Second, a separate scoring screen would be available for each age and type of dwellingfor which averages were given. These screens would be linked to the dwelling dataentered at the start of the program, so that, for example, if the dwelling was a 1920-1945 house, only the scoring screens showing the appropriate averages for that age and type of dwelling would be available for scoring.

2.34 Since 2000, when Version 1 was released, there had been considerable developments in hand-held computers and in the operating systems used. While the structure of thescoring element of the program was to remain very similar, the new operating systemsmeant that the best option was to build the programs from scratch using newdevelopment tools.

2.35 The PocketPC version was built first. As with the original survey program, the aim wasto build a satisfactory structure initially and then to finalise the content. Once oneprogram had been built satisfactorily, the other programs – the Palm OS and theDesktop PC program – could be built to mimic it.


10

2.36 Once the PocketPC program was virtually finalised, it was tested at a workshop withfour end-users who already had experience of using Version1 of the HHSRS. The userswere each given hand-held computers with the new Scoring program loaded. Theywere asked to try it out scoring several dwellings from information they provided. Toget views on the ease of operation, no instructions on the use of the program weregiven, other than how to switch on and start the program. They were asked to notedown any problems they encountered and comments or suggestions.

2.37 Only very minor comments and suggestions were made by those attending theworkshop, and these were easily taken into account for the final build. The final buildwas thoroughly checked to ensure that all the averages shown in the scoring screenswere consistent – i.e. they were the same as the averages shown in the WorkedExamples and in the Guidance.

2.38 Before building the Palm OS and Desktop scoring programs, a Desktop Data Handlingprogram was built. The program was designed to allow instructions (addresses) to bedownloaded onto handheld computers, for completed scoring reports to be uploadedon to the Desktop PC, and the reports to be reviewed and printed in a reader-friendlyformat. The Data Handling program was also able to name and identify individualhand-held computers.

2.39 Once the Data Handling program was finalised, the Scoring program for Palm OS hand-helds was built. These two programs mimicked the form and structure of the PocketPCprogram, but again, each had to be thoroughly checked for accuracy. Once finalised, all the programs were put onto a CD.

2.40 A handbook was prepared giving instructions on loading the programs and on gettingstarted, entering and saving data, and using the Data Handling program.

Paper Scoring Form

2.41 As well as testing the survey program, the workshop was used to trial proposals for a paper scoring form.

2.42 There is no standardised form to record dwelling inspection details, and it seemedneither necessary nor useful to attempt to design a standard HHSRS survey form. Also, in the light of the poor reception of the survey element of the Version 1 Surveyprogram, it is unlikely that any such form would be well received. However, for those preferring to use a paper system rather than an electronic method of recordingassessments, a HHSRS Scoring Form could be used in association with any existing form or a note-book.

2.43 Various designs and layouts of a Scoring Form were prepared and discussed within theteam. The aim was to mimic, so far as possible, the layout and structure of the WorkedExamples and the Survey Program scoring screens. It also seemed useful to includespace to record notes to support the judgments made. Versions of the form wereprepared for trial at the workshop, and those attending were asked to try them andgive their preference.


11

2.44 The preferred version was finalised and checked. (Figure 3 shows the heading and firsthazard score chart. The hazard score chart is repeated four more times, but could berepeated more often if necessary.) This final version was included in the Guidance forcopying or adapting.

Figure 3 – First Part of HHSRS Scoring Form


12

CHAPTER 3

Recommendations

3.01 Several recommendations for future development have been made in the three postVersion 1 reports20 – see Annex A, Recommendations 98 – 131. The additionalrecommendations summarised here, although not strictly part of the project, are madefor the immediate and long term future of the HHSRS.

General

3.02 The HHSRS was originally proposed and developed as a non-specific means ofassessing housing conditions. Its development for ODPM has been geared towardsenforcement purposes to be associated with housing legislation. We believe that there isconsiderable potential for the principles of the HHSRS to be developed for much wideruse. For example, it could be adopted in the assessment of housing for grant-aidedworks, the assessment of the immediate housing environment, and the assessment ofproposed new buildings, as well as in promoting health and safety awareness foroccupants.

3.03 The development of the Rating System has included reviewing and analysing statisticalevidence to provide information and support for those applying it. The review ofreported research shows that there are several areas where additional research, as wellas the updating of existing statistics, could usefully strengthen the HHSRS and informthe work on the relationship between housing and health generally.

3.04 One issue which is raised on a recurring basis is whether the Hazard Scores at adwelling can be totalled to give a single score. Superficial investigation of this duringthe development of Version 1 indicated that it was not possible to provide a simple andpractical way of generating a meaningful single score. Adding probabilities (likelihoods)is a complex issue in itself, and there are additional complications involved in addingHazard Scores. However, the issue continues to be raised and it would be useful toexplore possible practical methods of totalling individual hazard scores to confirm ornot whether this would give meaningful results.

3.05 A similar and related issue is the potential cocktail effect of Hazards, that is whether a certain combination of hazards at a dwelling would increase the overall threat. Forexample, are the threats from Fall hazards increased in a dwelling which also suffersfrom the Hazard of Excess Cold? Although this issue has not been widely raised, itappears likely and should be investigated further.

13

20 Evaluation of Version 1 of the Housing Health and Safety Rating System – final report (2003); The Application ofthe Housing Health and Safety Rating System in Houses in Multiple Occupation (2003); and Statistical Evidenceto Support the Housing Health and Safety Rating System (Volumes 1, 2 and 3) (2003).

3.06 Worked Examples have been very well received21. Although 66 have been prepared,more Examples would be useful, particularly ones demonstrating the national averagesfor each Hazard and ones illustrating borderline cases. One option would be to invitelocal authorities and others to submit Examples in a standard format to be scored by anapproved panel. This would produce a much wider range of Examples and wouldinvolve end-users in the process. This approach would be on-going, and would leadcomprehensive library of Worked Examples.

Public Release of Version 2

3.07 There continues to be much debate about the HHSRS, particularly since it became clear that the Housing Bill introducing the associated enforcement regime was beforeParliament. As well as the original Version 1 Guidance and survey program (which weresupplied to every local authority) there are currently22 at least seven HHSRS publicationsavailable on the ODPM website, including Version 1 Worked Examples and theDecember 2003 draft of the Version 2 Guidance. The presence of both Version 1 andVersion 2 documents makes for confusion and misunderstandings.

3.08 We believe and recommend that the finalised Version 2 Guidance, the Version 2 WorkedExamples and the Scoring and Data Handling programs should be made available assoon as possible, regardless of the progress of the Housing Bill in Parliament As well as enabling authorities to familiarise themselves with Version 2, this would help thoseauthorities who have already used the HHSRS as a key element within their privatesector housing policies required by the Regulatory Reform Order, and for decidingpriorities for assistance. It would also provide an opportunity for individual localauthorities to resolve any local difficulties associated with their IT systems.

3.09 We also believe and recommend that it would be useful for this Project Report and theReport on the Development of Version 123 to be similarly made available. The latterReport helps to explain what was considered in the original development and whycertain options were discarded and others chosen; the former helps to explain thechanges between Version 1 and 2.

3.10 With these documents available, any debate should be centred around Version 2, ratherthan a confusion between the two Versions. (Although Version 2 has been developed as a means of assessment associated with a proposed enforcement regime, the HHSRSitself is not referred to in the Housing Bill, and therefore its release before thecompletion of the Parliament procedure should not be seen as prejudging Parliament.)

3.11 As well as releasing all of Version 2 components generally, the HHSRS Scoring programand the source code should be released to software developers servicing localauthorities. This would allow developers to incorporate the program(s) into theirsystems, so that reports from the data handling program can be imported into existingprograms. It would also allow developers to build survey programs if they consideredthis to be appropriate.


14

21 See for example Recommendations Nos. 10, 15 and 114, Annex A.

22 That is at April 2004.

23 Housing Health and Safety Rating System: Report on Development (2000), DETR, London.

3.12 So far as the hard copy of the guidance is concerned, it would assist in its day-to-dayuse and for quick reference, if each of the Hazard Profiles (in Annex D of the GuidanceVersion 2) could be easily identified by the reader. This could be done by ‘thumb-marking’ each Profile, by using different coloured paper for each Profile, and/or bydifferent headers and footers for each Profile.

3.13 For Version 1 of the HHSRS, the Worked Examples were made available on the ODPMwebsite and on CD. The Version 1 HHSRS Survey program was released on a diskette at the same time as the Guidance Version 1. For Version 2 it is recommended that allcomponents are made available on a single CD and on a dedicated HHSRS website. As well as the Guidance Version 2, the Worked Examples, the HHSRS Scoring Program,and the Paper Scoring Form, this site and the CD could include the HHSRS Version 2Guidance for Landlords.

3.14 Although a straight-forward and non-technical Guidance for Landlords has beenprepared, we also recommend that leaflets of two or three sides should be published.These could provide a very simple outline of the basic principles of the HHSRS andwhere to obtain further information.

Training

3.15 A key concern for local authorities is the training of their staff in the use of the HHSRSto assess housing conditions. This is particularly so in relation to enforcement, whendecisions made may be subject to challenge. This is an issue which has been raised bythe Chartered Institute of Environmental Health. Also the Committee on the Office ofthe Deputy Prime Minister during its pre-legislative scrutiny of the draft Housing Bill,stressed the importance of training for local authority staff, and ODPM has recognisedthat funding for such training should be made available.

3.16 It is estimated that around 2,500 – 3,000 local authority staff will need training. To meetthis need, a training model should be developed which would deliver effective trainingwithin a limited timescale and which meets normal value for money criteria. In additionto local authority staff training in the use of the HHSRS will be necessary for lecturers atcolleges providing degrees in environmental health and commercial training provider.

3.17 As the HHSRS is intended to be the national prescribed method for assessing housingconditions, it is suggested that ODPM should decide, as a matter of urgency, thecontent and format of such training. This approach would help promote consistency of application.

3.18 We also suggest that ODPM should consider the most appropriate training strategy. Thiscould be by providing training for potential trainers, or providing a system to accredittrainers and courses.


15

ANNEX

Revisions Suggested for HHSRS Version 2

The following schedule gives in the first column, a summary drawn-up by the ODPM ofthe recommendations and suggestions made by:

l Evaluation of Version 1 of the Housing Health and Safety Rating System – final report(2003);

l The Application of the Housing Health and Safety Rating System in Houses in MultipleOccupation (2003); and

l Statistical Evidence to Support the Housing Health and Safety Rating System (Volumes1, 2 and 3) (2003).

The second column gives ODPM’s response to the recommendations and suggestions,the third the project team’s comments, and the fourth column gives details of theaction taken.

16

Recommendations, ODPM Responses, Project Team Comments and Action Taken

Guidance Version 2

Recommendations ODPM Response Project Team Comment Action Taken

Surveyors do not have the informationneeded (the dose – response relationship)to assign a specific probability to a hazard.(Eval 4.26)

A range of information is available to aidsurveyors. There is a clear need to haveimproved guidance and a more detailedtraining pack to support surveyors andexplain how to apply the information that isavailable.

There never has been any requirement thata dose – response relationship is relevantto the HHSRS assessment. The HHSRS isnot a predictive system but is based onassessing potential threats arising fromdeficiencies (see para 3.14). There issufficient evidence, summarised in theGuidance, to assess risk.

Specific probability (and outcomes) are notrequired – only a range. This is now explained.

See Profiles in Annex D.

See also Note following para 1.4, and 2ndpara page 45.

See paras 2.24-2.27, para 3.15, Box 3page 15, para 3.20, Box 5 page 17,Figure 3 page 24, Figure 4 page 25, andScoring programs.

4

Hazards that occur at more than one pointin a House: We find the Worked Examplesnot entirely clear on this point – theguidance should show clearly how toaggregate them, or they should berecorded separately. Our preference wouldbe to record hazards separately. (Eval 7.47)

We will improve guidance and provideWorked Examples to show how to dealwith this issue.

As 1.

Note – Preference for recording hazards inseparate locations separately is notappropriate as it goes against the principleof the HHSRS of assessing the dwelling asa whole and scoring a hazard for thewhole dwelling.

As 1.3

Guidance should make clear how theyshould combine the effects of faults invarious locations into likelihood. It shouldbe made clear whether they are looking atthe combined likelihood or the worst caseonly. (HMO 1.3.2)


As 1. As 1.2

Problem – Difficulty of quantifying risksderived from a number of different hazardsin different locations. Recommendation –more guidance needed. (HMO 1.2)


Recommendation adopted. See para 3.16, Box 4 p16, para 3.22, Box6 p18, and para 4.08. (Version 2)

See also Worked Examples 1.1, 1.2, 1.3,2.1, 2.2, 2.3, 2.4, 3.1, 4.1, 5.1, 7.1, 8.1, 8.2,8.3, 10.1, 12.1, 13.2, 15.1, 15.2, 20.1, 20.2,20.3, 21.1, 22.1, 22.2, 22.3, 23.1, 24.1,24.2, 24.3, 25.2, 26.1, 26.2, and 29.2.

1

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

17

We believe that users are right to believethat they lack the evidence or guidance toallow them to make highly specific numericassessments of risk – indeed paragraph7.17 of the HHSRS V1 Guidance confirmsthat there is limited evidence linkingindividual features of buildings to injuries orhealth outcomes. We believe that theconcerns of potential users about legalchallenge on the numeric precision of theoutputs of the system have somefoundation. (Eval 7.6)


As 5.

Detailed inspection should providesufficient evidence to support surveyor’sjudgment. Rating is only one factor to betaken into account in determining action.

As 5.

See para 3.27.

7

The system demands that users make aseries of judgements about theseriousness of a wide range of hazards,and their potential outcomes, taking intoaccount other secondary hazards. Theprincipal difficulty identified by users todate is that they feel they lack theinformation they would need to assign theprobability and the spread of harms to thevery specific numeric values which thesystem demands they choose. (Eval 7.5)


As 5.

HHSRS uses numbers to represent thesurveyor’s judgments.

As 5.

See para 3.04.

6

The precision of the outputs of the systemappears somewhat arbitrary and spuriousto many users, who are aware that indecision making in the field, they aremaking a professional judgement abouthow a particular situation might differ fromthe average, as for most hazards, there isno data which would allow one to statethe actual risk of an occurrenceassociated with the particular set offeatures which they observe in the field.(Eval 7.8)

We note that the system can be seen as‘arbitrary’ and that we need to addressthis perception. We propose revising theguidance to emphasis more the HazardBands and the ranges of likelihood andharm outcome and less emphasis on thespecific values selected.

Specific probability (and outcomes) are notrequired – the figures used only representa range. This is now explained.

See paras 3.25-3.27, para 4.19, and para4.28.

(See also Enforcement Guidance andlegislation.)

5

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

18


The HHSRS V1 Guidance is markedly lessuser-friendly than the later-issued WorkedExamples, but it does contain muchuseful, and indeed essential material – itsets out the underlying basis for thesystem. We suggest that the illustratedmaterial in the Worked Examples areessential in assisting new users to graspthe principles of the system, but there ismaterial in the Guidance which is neededto supplement the Worked Examples. Inour view the two documents should beintegrated. A revised Guidance Manual forusers should be based on the successfulformat and layout of the WorkedExamples, but should incorporate for eachhazard the material on the ideal, relevantfeatures and matters to be taken intoaccount. (Eval 7.17)

Version 2 research will cover considerationof structure of guidance and how best topackage it to include testing of its ‘userfriendliness’. Worked examples will beupdated or added to where necessary.

Options discussed. To be decided by ODPM.10

More intensive training using a wider rangeof worked examples would enable moreaccurate selection of a scale point foreach hazard. (Eval 4.26)

We propose to improve the guidance andworked examples to improvebenchmarking and calibration.

More Worked Examples have beenproduced.

A training strategy has beenrecommended.

All Version 1 Worked Examples have beenrevised and updated. 18 new WorkedExamples produced.

9

Although an enforcement situation wherethe officer struggles to ‘prove’ a 1,000+score is unlikely to fail if enforcementlegislation incorporates discretionarypowers, local authorities wouldundoubtedly feel more comfortable with asystem which allowed them to say that theprobability of an occurrence was in aspecified range, rather than a singlenumber. We would also wish to see therange of options to be considered inrelation to the probability and spread ofharms reduced in number.

We propose revising the guidance toemphasise more the Hazard Bands andthe ranges of likelihood and harm outcomeand emphasise less the specific valuesselected.

As 5. As 5.8

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

19


It may be desirable to reduce the numberof scale points to reduce the potential forvariance. (Eval 4.26)

We acknowledge this view. However theconsequence of this action is that a singlemove up a band in the likelihood scale forexample will result in a Hazard Band jumpto 2 or 3 bands.

As 12 and 13. No action.14

Alternatively, users would find acceptable asystem which used a five point nonnumeric scale similar to those used inmany house condition surveys. (Eval 7.9)

We acknowledge this view. However theconsequence of this action is that a singlemove up a band in the likelihood scale forexample will result in a Hazard Band jumpto 2 or 3 bands.

This option was considered and rejectedduring the development of V1 (see paras5.61-5.74 of Report on Development(2000)).

No action.13

In presentational terms, it would bepreferable if the scoring methodology wereto be simplified and rationalised so that itno longer demands that EHOs quantifyrisk with a degree of precision whichcannot be justified by the evidence,and the use of a range, rather than asingle score would meet that objective.(Eval 7.10)

We acknowledge this view. However, suchan approach is not a rating system andwould therefore not be consistent withgovernment policy and earlier consultation.

We propose revising the guidance toemphasise more the Hazard Bands andthe ranges of likelihood and harm outcomeand emphasise less on the specific valuesselected.

Scale points replaced with ranges on thelikelihood and output scales. As 5 above.

See also options considered duringdevelopment of V1 (see paras 5.61-5.74of Report on Development (2000)).

As 5.12

The HHSRS V1 guidance manual requiresclarifying to remove the ambiguity ofcategorisation of faults into hazards. It maybe that there are too many overlappingcategories. (Eval 4.26)

We will revise HHSRS guidance to clarifythis point.

Recommendation adopted, although it isnot accepted that there are too manyoverlapping categories.

See paras 4.06 and 4.0711

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

20


Interviewees were concerned that amenityprovision in HMOs would be more difficultto justify as only likelihood of 1 in 1resulted in Band A-C hazards for thosehazards dealing with basic amenities. Thisis an issue that is potentially easily settledby provision of worked examples whichjustify such likelihood. It may be that thesewould have to depend on issues of stressand mental health rather than physicalharms as it is very unlikely that majorphysical harms would arise frominadequate ratios of bathrooms etc.Recommendation – Worked examplesshould be provided to cover hazardsexacerbated by sharing by multiplehouseholds and these should include Fire,Hot surfaces, Crowding and space,Personal Hygiene, Sanitation. (HMO 6.2.4)

We propose to improve the guidance and worked examples to improvebenchmarking and calibration. We do not propose the retention of specifiedphysical standards.

As 15. As 15.

Of the 65 Worked Examples, 4 deal withFire, 2 of which are in multi-occupiedbuildings; 3 deal with Hot surfaces, 1 ofwhich is in multi-occupied building; 2 dealwith Crowding and Space; and 3 deal withPersonal Hygiene, 2 of which are in multi-occupied buildings.

17

In the absence of specified physicalstandards, worked examples should beprovided to cover items of concern. (HMO1.3.2)

We propose to improve the guidance and worked examples to improvebenchmarking and calibration. We do notpropose the retention of specified physicalstandards.

As 15. As 15.16

A series of worked examples be related tokey points on the scale and these shouldinclude several examples for the averagehazard and the first Band C hazard.Worked examples have been one of thesuccesses in explaining the system. Whilesome worked examples could beproduced from the case study materialthese will be insufficient for the numbersenvisaged. The production of the workedexamples should be a separatecommission. (HMO 1.3.2)

We propose to improve the guidance and worked examples to improvebenchmarking and calibration.

More Worked Examples have beenproduced.

All Version 1 Worked Examples have beenrevised and updated. 18 new WorkedExamples produced.

15

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

21


Populations on which averages based varyand are not explicit. Recommendation –Normalise statistics (HMO 1.2)

The revised statistics for Version 2 will benormalised for the dwelling population.

As 22. As 22.23

Statistics based on sub-groups should benormalised for the whole population. (HMO1.3.3)

The revised statistics for Version 2 will benormalised for the dwelling population.

Recommendation adopted. For all hazardsit is the national averages for a specifiedage group of the population (if any) living inall dwellings of a stated age group.

See 1st para page 46, and the ‘Potentialfor Harm’ section for each Profile in AnnexD.

22

A comprehensive reference work on thestatistics underpinning the system shouldbe produced. (HMO 1.3.3)

We will publish detailed statistical evidencebase based on research currently beingundertaken by University of Warwick.

As 19. As 19.21

Access to Up to Date Source Data forRisk Statistics: Statistical data for thevarious hazards will need to be updatedfrom time to time. Authorities who arepreparing their enforcement cases willwant to have easy access to the most upto date information, which makes clear thedate, source and method of capture of thedata. (Eval 7.43)


As 19. As 19.20

Some of the statistics felt to be weak.Recommendation – Comprehensivereference work on the statistics wouldincrease acceptance of system (HMO 1.2)


Recommendation adopted, and Statisticsresearch published.

Also, Guidance contains main references –see Annex E.

19

HHSRS seen as more time-consumingand puts greater strain on resources thancurrent fitness standard. Recommendation– Guidance to clarify if full assessmentrequired in all cases to determine action.(HMO 1.2)


Recommendation adopted. See Chapter 5 and Annex B paras B11-B20.

18

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

22


Defining vulnerable groups by age ignoresgroupings based on health. (HMO 1.1)


Not accepted. One of the main principlesof HHSRS was that vulnerability would beby age alone. To make this clear the term‘vulnerable age group’ has been adopted.

See paras 2.30 and 2.31.27

Interviewees felt there was little or noguidance on how to take account ofoccupancy in HMOs, there is the addeddimension of multiple households.Recommendation – Worked examplesshould be produced to support theprocess described in the previous sectionfor HMOs. (HMO 6.2.1)


Guidance clarified in relation to multi-occupied buildings generally, and fire,crowding and sharing of amenities.

See Chapter 5, Annex B paras B10 andB18, and Annex D Profiles for Crowdingand Space, Noise, Domestic Hygiene,Food Safety, Personal Hygiene, and Fire.

See also Worked Examples 17.1, 17.2,24.1, and 24.2.

26

There was particular concern from someinterviewees on how they should assesslikelihood and spreads for fire safety inHMOs. Recommendation – Workedexamples should make as explicit aspossible how likelihood and spreads arebuilt up from consideration of the faultspresent. (HMO 6.2.2)

We will review and provide additionalworked examples where necessary.

Recommendation adopted. See Annex B paras B11-B20, and AnnexD paras 24.21, 24.30 and 24.36-24.38.

Of the 65 Worked Examples, 4 deal withFire, 2 of which are in multi-occupiedbuildings.

25

The main issue raised was the need forseparate HMO statistics which have beenprovided. Some interviewees were criticalof the small sample that was used in theresearch which is the basis of the firesafety statistics. A further criticism wasthat some of the statistics were notapplicable to HMOs as they were drawnfrom other dwelling types although theonly case quoted was the hazard forasbestos. Recommendation – Furtherresearch be considered with regard to firesafety in HMOs. (HMO 6.2.5)

We will publish detailed statistical evidencebase based on research currently beingundertaken by University of Warwick. Thiswill include evidence where available forHMOs

As 22.

The Statistics Research project produced,where sample size permitted, averages for4 age groups of houses and 4 offlats/HMOs.

See 3rd para page 46, and the ‘Potentialfor Harm’ section for each Profile in AnnexD.

24

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

23


Worked examples should be produced toillustrate this process and should be basedon dwellings occupied by non-aged basedvulnerable groups in multiple occupatione.g. hostel for alcoholics. (HMO 7.1)

It would be practically impossible toproviding additional data for such aminority group. However, we will revise theHHSRS guidance to provide informationon what action to take in this type ofsituation.

As 27. As 27.29

Vulnerable Groups: There are confusingreferences in the Guidance andConsultation Document about vulnerablegroups – and in particular the importanceof visitors. It is inconsistent to argue thatthe possibility of visitors should be takeninto account, but that enforcement may bedeferred if no member of the VulnerableGroup is resident. A number of potentialusers of the system appear to havebecome confused by the suggestion thatenforcement might be deferred if nomember of the vulnerable group is residentin the house, and have drifted towardsrating houses in relation to the presentoccupants. (Eval 7.49)


As 27.

Enforcement issues not relevant toHHSRS Guidance.

As 27.28

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

24


The additional worked examples should beproduced to illustrate:

– how hazards can be scored when theyarise from multiple faults scatteredthroughout the HMOs;

– how hazards which are particularlyassociated with HMOs should bescored in a variety of different types andsizes of HMO; and

– how hazards which are exacerbated bybuildings being shared by multiplehouseholds should be scored (HMO7.2).


Clarification given – it is the dwelling whichis assessed, not the building (house)containing the dwelling. Effect of sharingamenities etc clarified.

See 1 and 26.31

The guidance should be amended tomake explicit the process by which non-aged based vulnerable groups andoccupancy by more than one householdcan be considered. This may be bestachieved by amending Chart 5 P31 andsections 7.09-7.17 (To Score a Hazardand Spread of Outcomes) and inserting anadditional flowchart to describe thescoring process. An alternative would beto leave the existing Guidance unchangedbut provide supplementary guidance onHMOs. This would mean that the issue ofnon-age based vulnerable groups wouldnot be dealt with. A third alternative istherefore to amend sections 7.09-7.17 todeal with vulnerable groups but leave theissue of occupancy to supplementaryguidance. (HMO 7.1)


As 27. See also 26.

Those using HHSRS should use theirinformed professional judgment to assessimplications of conditions for thosevulnerable other than by age.

As 26 and 27 30

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

25


Users were not clear on whether to enterone record per accommodation unit.Those that adopted this procedurebecame confused about where to enterinformation relating to shared facilities.(HMO 6.2.3)

We will revise HHSRS and HMO/HHSRSspecific guidance to clarify this point.

As 35. As 35.36

There is no guide to HMO surveyprocedure and it is unclear how usersshould enter information on an HMO.(HMO 6.2.3)


Recommendation adopted.

(In part a software issue.)

See Chapter 5 and Annex B paras B11-B20

35

Guidance should be written on surveyprocedure advising that hostels andpurpose built HMOs with shared amenitiesare best dealt with by grouping units intoclusters around shared amenities. Whilethis results in several records per HMO it isa practical method of making large HMOsmore manageable to survey. (HMO 6.2.3)

We agree with this in principle but adetailed survey is needed for enforcementpurposes.

Clustering and sampling will be covered inthe revised in guidance as will thedevelopment of a paper version.


This approach not accepted as it isagainst the HHSRS principle of assessinga dwelling or dwelling unit. However, re-surveying/assessing shared rooms/areas isdealt with.

See Chapter 5 and Annex B paras B11-B20.

See para 5.07 and Note in Annex B page39.

34

A separate guidance note should beproduced covering the application ofHHSRS to HMOs. This will draw on theoriginal guidance, plus the amendments,clarifications and additions identified duringthis research. The guidance should beillustrated with examples, and the mostup–to-date statistics.

The HMO guidance should suggest howthe HHSRS might be used to inform anauthority’s HMO strategy. (HMO 7.2)

HMO guidance will be provided as part ofthe Version 2 guidance. The contractorswill need to decide on whether the HMOguidance is separate from Version 2 orintegrated and present argument for this.

Enforcement guidance is seen as the bestplace to tackle guidance on and LAsstrategy for HMOs.


Options for publication / presentationproposed.

See Chapter 5 and Annex B paras B11-B20. Where relevant guidance is given inthe Profiles in Annex D (see for example,para 2.22).

33

Need for more guidance on what is meantby ‘average’ hazard. Recommendation –training. (HMO 1.1)

The revised statistics for Version 2 will benormalised for the dwelling population andadditional worked examples would beconsidered.


Training strategy proposed.

See para 4.14 and Note page 23, andProfiles in Annex D.

32

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

26


Instrumental Readings: The Guidanceshould give a clearer indication of the roleof instrumental readings for those hazardswhere this is appropriate. (Eval 7.48)


This is only relevant to Radon. For all otherhazards, the need for, and use of,instrumental readings as supportingevidence is a matter for professionaljudgment.

See Annex D, Table page 78.40

In self contained converted flats it seemslogical to score each flat as a record. Itwould then be necessary to use one of therecords to store information on the sharedparts and the external survey. (HMO 6.2.3)


As 35. As 3539

Purpose Built Flats: There would appear tobe the need for a section of the Guidanceto deal with the surveying of purpose builtflats and common areas of flats,particularly with regard to fire safety, whichhas to be looked at for the building as awhole. This could be dealt with in HMOguidance, provided that it is made clearthat purpose built flats are included in thatguidance. (Eval 7.39)


As 35. As 35.38

Such a system could work for housesdivided into bedsits, shared houses,houses let in lodgings. For large purposebuilt HMOs with shared facilities andhostels it might be more manageable togroup units which share common facilitiesinto single records. (HMO 6.2.3)


As 35. As 35.37

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

27


A number of authorities pointed outsituations where worked examples werethought to need improvement/addition, forexample in relation to structural stability,asbestos, contaminated water etc. Oneauthority pointed out that some of theexamples – for example in relation torefuse storage and electrical wiring – arevery extreme, and they believed theanswer for the overcrowding example tobe wrong. (Eval 7.40)


Additional Worked Examples have beenproduced and V1 Worked Examplesrevised and updated.

(See Project Report for explanation of howExamples were scored.)

Additional Worked Examples produced onAsbestos, Water Supply, Electrical, Refuse,and Structural Collapse.

43

Excessive Cold: Some case studyauthorities felt that landlords needed to begiven greater clarity about what heatingprovision was needed or about minimumSAP ratings – the phrase ‘means ofeconomically heating the whole of thedwelling’ in the Guidance was felt to beinsufficiently clear. The subsequentdescription of the type of heating systemrequired in para 1.17 of the Guidanceappeared to suggest that some form ofcentral heating system was required, butwithout explicitly stating so. (Eval 7.37)


Clarification given. However, need foradditional expert advice is a matter forprofessional judgment.

See Annex D, paras 2.25 and 2.26.42

A worked example should be provided tocover the use of asbestos insulationboarding in HMOs. It should consider thegreater risk of disturbance brought aboutby heavy wear and tear associated withHMOs as well as potential for damagearising from aberrant behaviour. (HMO6.2.5)


Additional Worked Examples have beenproduced. This suggestion wasconsidered.

See Worked Example 4.2.41

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

28


Program, Hardware, Format and Paper based Version

An improvement in the software installationprocedure and the quality of theaccompanying documentation would alsogive significant benefits. (Eval 5.102)

We will consider the need for this type ofguidance as part of Version 2.

Recommendation adopted. Installation, upload and download anddata handling programs developed.Handbook produced.

47

The guidance be amended to makeclearer that the core information thatneeds to be collected is for the user todecide and not all fields are compulsory.(HMO 1.3)


Handbook produced and programs totallyrevised.

Handbook advises what very limitedinformation required.

46

Because users quickly becamediscouraged by the difficulties with thehandheld computers and the absence of ausable output, the system has not in facthad substantial testing in use. However,those who have used it have drawn ourattention to a number of minor errors andomissions – generally omissions of acategory in the pull down menus. Thesewill be reported separately to the Office.(Eval 7.34)


Survey element of program eliminated.Issue no longer arises.

Data handling program developed andrevised and much reduced scoringprograms have been developed.

45

Some of the statistics were found to bevery surprising – for example, the fact thatcarbon monoxide had a 0% probability ofa Class 1 outcome, when ergonomics hada 0.1 % risk of a Class 1 outcome. Thesefigures may be a result of the nature of thestatistical sources used, but they aredifficult to reconcile with the averageperson’s perception of the nature of therisk of carbon monoxide poisoning. (Eval7.41)


The examples quoted occurred becausethe base population varied for eachhazard. The base population is now thesame for every hazard.

For all hazards in V2 it is the nationalaverages for a specified age group of thepopulation (if any) living in all dwellings of astated age group. (See also Annex D para6-1.04.)

44

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

29


In many authorities, time which wasneeded to secure understanding of thesystem as a whole has been taken up withtrying to master the use of the handheldcomputer, with difficulties exacerbated bythe lack of computer program guidancemanual, and in many cases, an unsuitablemachine. If time and money are at apremium, it is more important for users to devote it to mastering the principles of system rather than the handheld. (Eval 7.21)

In hindsight we agree that this could havebeen handled more effectively. As part ofVersion 2 there will be an improveddissemination and marketing strategyfocusing on HHSRS principles and notform of delivery.

Policy matter – options proposed fortraining and dissemination. A paper basedscoring form has also been devised, sothe use handheld technology is notessential

ODPM decision.50

A comprehensive program manual shouldbe provided covering installation and use.(HMO 1.3)


As 47. As 47.49

Users were discouraged and demotivatedby the way in which the system wasreleased to users. The Guidance andsoftware were released together, but theGuidance did not offer an adequateintroduction to the software, which was, in effect, released without an instructionmanual or user guide. If continued use isto be made of the software it will need amanual, which deals with installation,uploading and presentation. Theinstallation of the software could easily be automated to a much higher degree.(Eval 7.16)


As 47. As 47.48

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

30


On the subject of data analysis, thereseems little point in issuing the handheldsystem without an accompanyingcomprehensive database to receive andreport on the data in both standard anduser-definable formats. This should becommissioned alongside the modificationsto the handheld program. (Eval 5.105)

We will seek as part of Version 2 toproduce standardised output reports.

Recommendation adopted. Data handling program developed.54

HMOs are difficult to survey using papermethods and the Palm is an addedcomplication. (HMO 6.2.3)

As part of Version 2 there will be animproved dissemination and marketingstrategy focusing on HHSRS principlesand not form of delivery. In addition, wewill be providing Palm, Windows CE andpaper versions of the HHSRS.

As 50. As 50.53

The computer system was the singlebiggest objection to the use of the systemin HMOs but most of the objections areagainst the use of the Palm computer andprogram rather than the system itself.

As part of Version 2 there will be animproved dissemination and marketingstrategy focusing on HHSRS principlesand not form of delivery. In addition, wewill providing Palm, Windows CE andpaper versions of the HHSRS.

As 50 As 5052

With the benefit of hindsight, one canquestion the wisdom of combining a newsystem with a new technology, thussharply raising the investment in timeneeded to master the system or to trainothers – the combined difficultiesundoubtedly heightened perceptions thatthe system was unworkable. There are realdoubts as to whether a hand heldcomputer system is the most effective wayof recording the type of information whichlocal authorities need for enforcementpurposes in relation to fitness. (eval 7.26)

In hindsight we agree that this could havebeen handled more effectively. As part ofVersion 2 there will be an improveddissemination and marketing strategyfocusing on HHSRS principles and notform of delivery.

As 50. As 50.51

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

31


Problem – Impossible to transfer keyinformation from Palm to PC. Format ofdata not user friendly. No reportingfacilities on PC software.

Recommendation – Amendments tosoftware. (HMO 1.2)


As 54. As 54.58

No usable output records from thepackage, other than for authorities whoprepared their own front end – a tinyminority. (Eval 7.24)


As 54. As 54.57

The consequence of the absence of auser friendly reporting format was that theEHOs who had been asked to run thesystem in parallel with existing fitnessassessments became discouraged by thenature of output currently produced by thesystem and did not continue to use it. Theintroduction of Version 2 with a ‘front end’which will allow authorities to utilise thedata they have collected, will in our viewmake a significant difference to theacceptability of the system. (Eval 7.13)


As 54. As 54.56

The program was issued without a ‘frontend’, as it was argued that each localauthority might wish to customise theoutput for their own systems. However,authorities have been very reluctant todevote resources to developing their ownindividual ‘front ends’ for a system whichhas not yet been formally introduced inlegislation. Our case study worksuggested that most authorities need touse the data for the same purposes, andthere was overwhelming support for astand report format. (Eval 7.12)


As 54. As 54.55

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

32


The program is reasonably well designed,but it is not, as shall be shown, not whollyuser friendly in its present form. It wouldbe helpful if it linked faults with hazardsmore directly. Certain hazards which mustbe considered at a whole house level, notroom by room, such as excessive coldand the risk of fire, appear to be difficult torecord. (Eval 7.11)

Feasibility of this shall be considered aspart of the Version 2 guidancedevelopment and implemented wherepossible and changes made to theHHSRS programme.

Survey element of program eliminated. Asno faults recorded in program, issue nolonger arises.

No action.62

There is a need to improve theconsistency of data entry and the formviews of the data as it is being logged, sothat the user gets a greater sense of theflow of information and has a clear mentalview of the whole picture and the part of it that they are currently addressing. (Eval 5.100)



No action.61

Either faults should be linked to drop down lists of remedies (to form basis ofspecification of remedial work) or a similarfacility should be provided in the PCSoftware. (HMO 1.3)


This is not a practical option. Theappropriate remedy will depend on theform of construction, not the faultcontributing to the hazard. Determining theappropriate remedy is part of the role ofprofessional judgment of the person whosurveyed the dwelling.

No action.60

A user-friendly PC interface using an MSAccess database or Excel spreadsheetshould be provided. This should becapable of amendment by the user to suit their needs. (HMO 1.3)


As 54. As 54.59

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

33


If scoring and banding is to be retained, amixed paper/computer system would bemost appropriate for most users, whowould record the faults observed in thedwelling using notebooks, speciallydesigned forms, sketch plans andphotographs as appropriate, and then usea handheld or a PC back in the office todo the scoring. A minority may prefer tocontinue entirely with the handheldcomputer software. (Eval 7.31)

A palm, Windows CE and paper versionand focus on HHSRS principles and notform of delivery will be delivered as part ofthe overall version 2 guidance.

Recommendation adopted. Paper Scoring form devised – see AnnexB pages 41 and 42.

67

Palm screen difficult to read in someconditions, H&S implications. (HMO 1.1)

Not a HHSRS issue. However,developments in hand-helds havedramatically improved screen visibility.

No action.66

It may also be beneficial to remind theuser of any key items which have not beenentered before leaving the property. Thiswould help to ensure that a balanced andusable set of data has been collectedwithout the need for time consuming andexpensive revisits. (Eval 5.101)


As 63. As 63.65

It would be helpful if the softwareprompted user to consider the ‘matters to be taken into account’ for the varioushazards when a fault is recorded. (Eval 7.51)

This recommendation will be addressed aspart of the Version 2 development.

As 63. As 63.64

Software Issues: If the handheld computersystem is to be used, consideration shouldbe given to making entry of basic data foreach house – which users find frustrating –optional. (Eval 7.50)


No longer relevant (because of revised andshortened program).

No action.63

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

34


Handheld computers can generateeconomies and efficiencies when they are used for surveys where bulk dataprocessing is likely to be needed – theyare for example very appropriate forsample house condition surveys. It wouldappear that this is seldom likely to be thecase with HHSRS, where it is the accuracyand degree of detail in the individualsurvey which is important. EHOs arehappy to write up their notes afterwards in the office – it gives them greaterownership and confidence in their reports and evidence. (Eval 7.22)

A Palm, Windows CE and paper versionwill be delivered as part of the Version 2.

As 67. As 67.70

It is recommended that the revisedguidance includes a HMO surveyprocedure using a paper form, althoughscoring should continue to be undertakenusing electronic methods (using Palm,Palm emulator or spreadsheet). (HMO 7.2)


As 67. As 67.69

The complexities of HMOs are such thatthey are best surveyed using papermethods. A survey procedure using apaper form should be devised. The formshould be fairly simple (relying on freehandnotes to deal with the complexities) butshould include a hazard list as an aidememoire to prompt surveyors to recordfaults relating to all hazards. The formshould not attempt to emulate the Palmprogram and should not include a scoringsheet. Scoring would be done using thePalm or on a PC using a spreadsheet orthe Palm emulator. The form could bedeveloped from a local authority’s currentpractice. (HMO 6.2.3)


As 67. As 67.68

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

35


It is further recommended that theprogram be split into the hazard scoringmodule, which could not be altered, andthe survey and administration modulewhich should be freely available to amendaccording to the wishes of the user. It issuggested that the Office encourage usergroups probably based around thoseauthorities with common softwareproviders to further develop thesemodules. This may well be the mostrealistic way of adding a scheduling facilityto the software (likely to be a major task).(HMO 1.3)

A Palm, Windows CE and paper versionwill be delivered as part of the Version 2.This is likely to result in a hazard scoringmodule as proposed.

Recommendation adopted. Only scoringmodule to be developed.

Scoring programs developed for Palm OS,Pocket PC and Desktop PCs.

74

Looking to the future, it is generally felt thatalthough the program with the suggestedmodifications will do an adequate job, itsoverall quality and ease of use could beconsiderably improved by rewriting to rununder Windows CE. (Eval 5.104)

A Palm, Windows CE and paper version willbe delivered as part of the Version 2.

As 72. As 72.73

Many authorities prefer a Microsoftplatform. (HMO 1.1)


Recommendation adopted. (NB –PocketPC is now the Microsoft platform.)

Scoring program for Pocket PCdeveloped.

72

A paper based system which uses ahandheld computer or a PC for scoring isthe preferred solution, with the option ofhandheld computers for the minority whoprefer to use them. If a handheld system isto be offered, there may be advantages inmoving to a Windows CE system –advantages which could not have beenforeseen at the time the system wasdeveloped. However, these advantageswould have to be set against theinvestment which authorities have made in Palm machines. (Eval 7.23)


Recommendation adopted. Scoring programs developed for Palm OS,Pocket PC and Desktop PCs. Also paperScoring form designed – see Annex Bpages 41 and 42.

71

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

36


In bedsits, shared houses and houses letin lodgings, information should be enteredon a single record. An additional page orpages should be added for occupancyinformation to be entered against eachdwelling. A page should also be added foreach room to allow lettings using eachroom to be recorded. The score sheetshould be amended to allow hazards to berecorded against one unit, a selection ofunit or all units. (HMO 6.2.3)


As 75. As 75.76

One of the most basic issues to addresswith regard to survey procedure is how toenter information on rooms/circulationspace which are shared. Rooms used bythe bedsit are currently entered with thebedsit. This leads to potential forduplication and confusion as information isscattered over several records. It would bemore logical to name each unit, enteroccupancy details, and then keep all thisinformation in a single record. This wouldhowever require amendments within eachroom page to allow units using the roomto be recorded. It would also require thescore sheet to be amended to allowhazards to be assigned to one unit, aselection of units or all units. (HMO 6.2.3)


Survey element of program eliminated asnot seen as helpful. Surveys can beundertaken and recorded in largely thesame way as currently. Issue no longerarises.

No action.75

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

37


Following on from this, whether or not asketch plan facility is incorporated into theprogram, we would see one as essentialfor applying the system to HMOs.Guidance on and when to use the facilitywould be required. (HMO 7.2)

This will be investigated as part of theVersion 2 development.

As 75. As 75.81

A sketch plan facility should beincorporated into the program. (HMO6.2.3, HMO 1.3)

This will be investigated as part of theVersion 2 development.

As 75. As 75.80

It is also recommended that a facility toclone records be included in the system.By adopting the above procedure ofclustering units it should also be possibleto clone groups of records. This could beused to speed up surveys of large HMOswhere there are repeating layouts as is(sometimes found in hostels). (HMO 6.2.3)

Feasibility of this shall be considered aspart of the Version 2 guidancedevelopment and implemented wherepossible and changes made to theprogramme.

As 75.

Also, Desktop PC scoring program makescloning possible.

As 75.79

If the department is serious about verylarge HMOs being inspected using thePalm some form of cloning information islikely to be necessary. (HMO 6.2.3)


As 75. As 75.78

If the Office wishes to allow for authoritiesto use the full Palm system for HMOs,it is recommended that a number ofamendments to the program are made toallow information on more than one unit tobe recorded within a single record (seeSection 7). Additional guidance would berequired on how this information should berecorded. (HMO 7.2)


As 75. As 75.77

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

38


Problem – Number of minor problems withsize of fields etc. Recommendation –Amendments to software. (HMO 1.2)

This recommendation will be addressed as part of the Version 2 development.


No action.86

The program should be amended (orguidance should be given) to allowdeletion of hazard bands and recall oflikelihoods and spreads. (HMO 1.3)


As 84. As 84.85

Practitioners found it difficult to deleteoriginal score when they wished to make a correction. Recommendation – training.(HMO 1.1)

This recommendation will be addressed as part of the Version 2 developmentincluding amended guidance.

Recommendation adopted. Programs developed allowing scores to bedeleted and/or reviewed.

84

Limited advice on equipment selection sothat many authorities were working withcomputers which were not the mostappropriate for the purpose. (Eval 7.24)

The ODPM does not have the capacity totest different machines as this might leaveus liable to complaints of endorsement.

Recommendation adopted. Included in Software Handbook.83

Environmental Health Officers need highlyaccurate and detailed evidence about thehouses which they survey. They need tobe able to create a record which mayinclude plans and sketches, photographs,and very detailed notes. It will be difficultfor them to assess many houses forexcessive cold – the most significanthazard – without preparing a SAP rating. It follows then that they need a supportingmodule – or a system which canaccommodate all of these matters.(Eval 7.14)


As 75.

Also, for all other hazards, the need for,and use of, instrumental readings assupporting evidence, or the need to takeadditional expert advice, is a matter forprofessional judgment.

As 75.82

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

39


The use of ‘shortcuts’ be promoted. (HMO 1.3)


Survey element of program eliminated.Issue less of a problem, if at all.

No action.93

Facilities for note taking should be mademore widely available throughout theprogram. (HMO 1.3)


Recommendation adopted. Scoring programs developed to allow fornotes anywhere within the program.

92

Graffiti system too slow for extensive note-taking. Recommendation – Use of short-cut facility (HMO 1.2)


Survey element of program eliminated.Issue less of a problem, if at all.

No action.91

Specific amendments should be made toinclude frequently occurring items in thedrop down menus. (HMO 1.3)



No action.90

Treatment of Insulation and otherExcessive Cold Issues in Software: Severalauthorities found it difficult to identify howthe software recorded insulation – thereappeared to be no provision for recordingattic spaces which is where the principalinsulation for the house is installed. (Eval7.38)



No action.89

Address and UPRN fields should be madelarger and provision should be made totransfer these from PC software. (HMO1.3)


Recommendation adopted. Data handling and scoring programsdeveloped, allowing addresses and basicinformation to be transferred to hand-helds.

88

One interviewee felt that seeing the displaychange as likelihoods and spreads arechanged influences decisions.Recommendation – Consider providingoption to switch off display. (HMO 1.2)


Option considered, but not adopted. No action.87

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

40


Marketing and Dissemination

The presentation of the system to the laypublic is one of the areas which should beaddressed in Version 2. It is believed that itis possible to explain the results in astraightforward every day language,provided that no attempt is made toexplain the complex calculations whichunderlie the score given to the property.(Eval 7.4)

This is something that we will look at inrelation to the dissemination andmarketing strategy for Version 2. Onepossible approach to this is for thepublication of a leaflet explaining the basicprinciples of the HHSRS that can be givento householders.

ODPM commissioned development andpreparation of Guidance for Landlords.

Guidance produced.96

Users have been asked to master a newtechnology at the same time that theyhave been asked to grasp a conceptuallydifficult new system which needsconsiderable training to make userscompetent and confident in assessing allthe hazards. The limited surveyor variabilityanalysis carried out suggested that there isa need for detailed training, benchmarkingand calibration for the full range ofhazards. (Eval 7.20)

In hindsight we agree that this could havebeen handled more effectively with greaterconsistency in timing of the relevant partsof the guidance. A marketing anddissemination plan for all elements forVersion 2 of guidance and more workedexamples to improve benchmarking andcalibration have been built into thespecification.

Additional training support will beconsidered through the process of Version2 to ensure consistency with the revisedguidance and HHSRS.

Options proposed. ODPM decision.95

It is further recommended that the sourcecode for the program be made available todevelopers wishing to develop a programto be used on Microsoft Handheldplatforms. (HMO 1.3)

The source code has been made availableto users wishing to use it with certainconditions attached. This will continue tobe the case.

Not part of research project. ODPM decision.94

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

41


Future Development

Court cases would be bogged down byappeals as HHSRS too subjective.Recommendation – training. (HMO 1.1)

It is difficult to predict what will happenwith the legal process in relation to therating system until it is live and has hadappeals against it. What we must ensure is that we continue to provide robustevidence on areas such as variations ofresult to reduce the chance of appealsoccurring. We are considering carrying outadditional research on the sources of thevariation in surveyor assessments in 2003.

Point not accepted, there is no evidencenor indication that this system will besubject to more appeals than the currentsystem associated with the fitnessstandard.


No action.

ODPM decision.

100

It is believed that users are right to believethat they lack the evidence or guidance toallow them to make highly specific numericassessments of risk – indeed paragraph7.17 of the Guidance confirms that there islimited evidence linking individual featuresof buildings to injuries or health outcomes.It is believed that the concerns of potentialusers about legal challenge on the numericprecision of the outputs of the systemhave some foundation. (Eval 7.6)

There is a clear need to have improvedguidance and more detailed training packto support surveyors and explain how toapply the information that is available. Thiswill be considered in Version 2 of theguidance. Improvements to the trainingpackage currently provided will need to beconsidered to ensure consistency with anychanges made in the revised guidance.

Specific probability (and outcomes) are notrequired – only a range. This is nowexplained.

Detailed inspection should providesufficient evidence to support surveyor’sjudgment. Rating is only one factor to betaken into account in determining action.

See paras 3.25-3.27, para 4.19, and para4.28.

(See also Enforcement Guidance andlegislation.)

See para 3.27.

99

A feasibility study to determine ways ofmaking the statistical evidence moredirectly relevant to housing conditions.(Stat 7.21)

Although this is not something that will beincluded into Version 2 of the guidance, asmall scale research project to look atspecific hazards could be an option.Further discussions will be required toidentify hazards to target.

Not part of V2 project. ODPM decision.98

Landlords will have difficulty getting togrips with the system. Recommendation –Training needed for landlords, particularlysmaller landlords. (HMO 1.2)

As above we consider the production of aleaflet for landlords explaining the basicprinciples of the HHSRS and the theirduties in relation to them to be a goodway forward.

As 96. As 96.97

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

42


The accuracy of the HHSRS statistics relieson having accurate population estimates atpostcode level. The publication for the 2001Census of enumeration district data,including a breakdown of population byage, should substantially increase thereliability of any postcode populationestimates for two main reasons.

Firstly, the 2001 Census will provide up-to-date primary information on the agedistribution of the population in the wholeof England and Wales. In contrast, allcurrently available information is either wellover 10 years out of date being reliant onthe 1991 Census, incomplete being basedon sample surveys, and/or heavily relianton modelling techniques, these beinglargely undisclosed in the case ofcommercially available data. (Stat 7.05)

The development of the HHSRS is seen asa continuous process taking account morerobust data as it becomes available. Thedata that is being recommended for use isnot yet available. As soon as this data isavailable, we will consider carrying out thisrecommendation.

Not relevant to V2 project. ODPM decision.103

Users would also like support withdeveloping appropriate specifications forremedying the types of faults identified byHHSRS. The nature of many of the faultsidentified is different from those identifiedunder the Fitness Standard. The training ofEHOs may need review to ensure thatofficers feel competent to draw upappropriate specifications for this widerrange of hazards. This will require a review ofthe content of full and part time professionalcourses and there may be a need for CPDseminars on this issue. The CIEH will havean important role to play here. (Eval 7.15)

We agree that changes to the trainingregime will be required to take account ofVersion 2 guidance. However in terms ofthe HHSRS a suggested remedy for thehazard is a separate issue from theprinciples of providing a judgement basedon the hazard itself. We will liaise withtraining providers and professional bodiessuch as CIEH on matters such as these toensure that they are aware of the issuesand can adjust or recommend changes totraining they already provide.

Determining the appropriate remedy is partof the role of professional judgment of theperson who surveyed the dwelling, and willdepend on the form of construction.

This is a matter for colleges/universitiesresponsible for training for EHPs and forthe EH Education Board.

No action.102

Appeals – the ‘subjectivity’ of the systemwhich results in such specific scores wasexpected to lead to a large number ofappeals. (HMO 1.3.4)

See above (item 100). As 100. As 100.101

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

43


A repeat of the safety statistics usingpopulation estimates from the 2001Census and HASS data for the 5 years,1997 to 2001, once these are available.(Stat 7.21)

See item 103. As 103. As 103.104

Secondly, because of their significantlysmaller size, enumeration districtpopulations can be related to postcodesmuch more accurately than ward orpostcode sector based populations.Compared with an average of some 2,330dwellings in a local authority ward and2,730 dwellings in a postcode sector, anaverage enumeration district contains onlysome 130 dwellings. The averagepostcode in the HASS catchment areascomprises just over 18 dwellings. This isslightly larger than the national average of14 dwellings, but postcodes of between30 and 60 dwellings are common in urbanareas. (Stat 7.05)

For the above reasons, we would stronglyrecommend that the HHSRS statistics berecalculated on the basis of the 2001Census enumeration district populationdata, once this data becomes available.(Stat 7.06)

Not relevant to V2 project. ODPM decision.103(cont)

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

44


24 It is understood from the Census Customer Services Department in Fareham that such data will not be officially published for the whole of England and Wales until “the end of2003”, but it is possible that the required data will be available to Government Departments before then.

Improving Training: Training should focuson benchmarking, to achieve greaterconsistency between surveyors – this isessentially a kind of calibration exercise,where a more consistent approach isdeveloped to agreeing the numericprobability to be assigned to certaincommon situations. (Eval 7.44)

Issues of training will be passed on to therelevant professional and training bodiesfor consideration in line with Version 2guidance. We are considering carrying outadditional research on the sources of thevariation in surveyor assessments in 2003.

As 108. As 108.109

Surveyors should have additional trainingto ensure better consistency in identifyinghazards. (Eval 4.26)

We are considering carrying out additionalresearch on the sources of the variation insurveyor assessments in 2003.

Training strategy proposed. ODPM decision.108

There were a number of issues raisedwhich were judged to be perceivedproblems caused by lack of familiarity withthe system and misunderstandings ormisinterpretations. In general thereforethey would be addressed through trainingand experience.

Amendments made to Version 2 of theguidance will provide a greaterunderstanding of the processes involved.In additional a greater focus on targeteddissemination and marketing will ensureusers are aware of the relevant informationat their disposal.

Training strategy and disseminationoptions proposed.

Guidance clarified generally.

ODPM decision.

107

Surveyors do not feel confident with theirestimates. Recommendation – training.(HMO 1.1)

Additional research will be commissionedlater in the year to identify sources ofsurveyor variability and providerecommendations that will feed intoVersion 2.

No part of V2 project. No action.106

System is too subjective and there is toomuch variability in its application.Recommendation – training. (HMO 1.1)

Additional research will be commissionedlater in the year to identify sources ofsurveyor variability and providerecommendations that will feed intoVersion 2.

The HHSRS (like housing standards beforeit) relies on professional judgments,primarily qualitative, and so will be, tosome extent, subjective. Consistency willdevelop and Worked Examples will assistin this.


No action.105

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

45


Worked Examples Library: A number ofauthorities suggested that a library ofworked examples, which would be addedto over time would be very helpful – a webbased version would be particularlyappropriate. Some authorities had alreadystarted to collect examples. (Eval 7.42)

We agree that this would provideadditional benefits to the overall package.The successful contractor for Version 2guidance will be asked to review thisapproach and present options for itsimplementation.

Options proposed, and although a positivesuggestion it was not otherwise, part of V2project.

No action.114

A consistent way of recording the severityof injuries and illnesses is required, whichwould enable classes of harm to be readilydetermined. Such a standard classificationsystem, possibly based on the ICD systembut providing more information on theseverity of the condition, could then beused in all Government funded surveys inwhich information on injuries and illnessesis collected. Future research. (Stat 7.20)

This can be considered for futuredevelopment in conjunction withdiscussion with DTI and others with aprofessional interest in this area.

Not part of V2 project. No action.113

Training for other Property Professionalsand Landlords: There is likely to be aneed/demand for training for otherproperty professionals and landlords. (Eval 7.46)

Issues of training will be passed on to therelevant professional and training bodiesfor consideration in line with Version 2guidance.

As 108. As 108.112

As well as guidance, it is recommendedthat separate training on the application ofthe HHSRS to HMOs is promoted by theDepartment. (HMO 7.2)

Issues of training will be passed on to therelevant professional and training bodiesfor consideration in line with Version 2guidance.

As 108. As 108.111

Training will also need to cover theapproach to the less frequently occurringhazards, where there is little depth ofexperience of using the system. (Eval 7.45)

We agree that training on surveys forenforcement purposes should cover allhazards as identified in the guidance.Information on all hazards is available andshould be considered. We will pass thisinformation on to the relevant professionaltraining bodies.

As 108. As 108.110

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

46


The HASS accident records contain aparticularly large proportion of unspecifiedor only partially specified data. It isrecognised that there are inherent difficultiesencountered in collecting accident data inthe Accident and Emergency Units.Obviously, the first priority of the victim willbe in getting early treatment, and that of thehospital staff will be in diagnosing andadministering the appropriate treatment.Asking and answering details about thecircumstances of the accident will rarely beseen as a major priority by either side.However, in the interests of reducing thenumber of future accidents in the home, webelieve efforts are required to improve theamount and quality of the HASS data. Howthis might best be done could be thesubject of a small research project in itsown right. However, the validation andanalysis of the 1997, 1998 and 1999 HASSsamples carried out for this project hasprovided a number of pointers. (Stat 7.08)

We will continue to liaise with DTI andseek improvements in the usability ofHASS data in the context of the HHSRS.

As 115. As 115.116

The analysis of the Home AccidentSurveillance System (HASS) was based ondata for the three years 1997 to 1999, thesebeing the years with the best post codeddata currently available. However, despiteaveraging the statistics from a number ofyears, for less frequent hazards such as fallsbetween levels and electrical hazards, thesample sizes are still generally too small tobe accurately broken down by both agegroup and type and age of dwelling. Infuture, it is recommended that the safetystatistics be repeated using at least five yearsof fully validated HASS data25. (Stat 7.07)

The development of the HHSRS is seen asa continuous process taking account morerobust data as it becomes available. Thedata that is being recommended for use isnot yet available. As soon as this data isavailable, we will consider carrying out thisrecommendation.

Not relevant to V2 project and the matterof updating data used in the guidance is apolicy issue for the ODPM.

No action.115

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

47


25 In 2003, the HASS was terminated and no further accident data collected. Archive data is now held by the Royal Society for the Prevention of Accidents.

Missing data on inpatient days alsosuggests that the HASS data is generallycollected at the time of admission to the A& E Unit or soon afterwards and is notnecessarily revised when the full details ofthe injuries are known. Checking andfinalising the HASS forms just before theperson is discharged, when the accidentvictim may be more receptive to answeringdetailed questions on the circumstances ofthe accident, might help to improve thequality of the information collected,particularly that concerned with hospitaloutcomes and in-patient days. (Stat 7.10)


As 115. As 115.118

HASS clerks should be reminded of theimportance of the data for housing policyand preventative medicine. This projecthas shown that while there is a generallygood response rate for the basicinformation on the age of the victim andthe medical details of the injury and body-part affected, insufficient attention seemsto be given to collecting the ancillary, non-medical information on the circumstancesof the accident, such as on the roomlocation, the dwelling type and person’sactivity at the time. (Stat 7.09)


As 115. As 115.117

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

48


As well as due to missing data, the HASSfigures also significantly underestimate thenumber of non-fatal accidents in the HASScatchment areas due to the loss of cases toneighbouring A & E units. The definitivesolution to the determination of the trueaccident rates in the HASS catchment areaswould be to use the national hospitalepisode statistics to determine all theaccidents from the postcodes in thecatchment areas that were treated at othernon-HASS hospitals in the same period ascovered by the HASS data. This is a largepiece of work involving a larger accidentsample and was outside the scope of thisproject. However, this might be one way ofincreasing the reliability of the HHSRSstatistics in future. Generally, such ananalysis could also confirm the accuracy ofDTI’s national accident estimates. (Stat 7.11)


As 115. As 115.121

A study of the potential improvements inthe collection and validation of HASS data.(Stat 7.21)


As 115. As 115.120

Although this approach requires moredevelopment, further surveys of this kindmay be needed to cover other frequenthazards. However, it might be possible toachieve the same results, more economically,by extending the HASS questions on thephysical aspects of the accident and byincluding comparable questions in the EHCSto determine the incidence of particularphysical conditions in the housing stock. Atthe very least, a feasibility study is required todetermine how this issue might best beaddressed. (Stat 7.17)


As 115. As 115.119

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

49


To improve the HHSRS safety statistics,the mortality data needs to be made morecompatible with the HASS data. While itmay not be feasible to extend the recordof all fatal accidents, extending the recordin line with the HASS data, at least, for alldeaths occurring in the HASS catchmentareas is recommended. With respect tofatal accidents, the DTI’s current review ofthe Home Accident Deaths Database(HADD) is welcomed. (Stat 7.15)

We will continue to liaise with DTI andseek improvements in the usability ofHASS and HADD data in the context ofthe HHSRS.

As 115. As 115.124

Depending on the year, up to a fifth of allHASS postcodes were found to be mis-punched or mis-formatted and these hadto be corrected to maximise the samplethat could be matched to other data. Thequality of the HASS postcode data couldprobably be improved substantially byusing an experienced punching agencyand adopting the common commercialprocedure of double punching whentransferring the HASS records to a datafile. It is understood from DTI that casualstudent labour is currently used for thistask. (Stat 7.12)


As 115. As 115.123

The determination of HASS catchmentareas and the loss of cases to other A&Eunits using hospital episode data. (Stat7.21)


As 115. As 115.122

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

50


Determining Classes of Harm: For thespecific purposes of the Housing Healthand Safety Rating System, thecategorisation of other HASS questionscould also be improved. For example,none of the questions on the healthoutcomes, the type of injury, body partaffected, hospital outcome and inpatientdays give a particularly good idea of thetrue severity or class of health outcomes.(Stat 7.18)


As 115. As 115.126

Defining the Hazards: Relating the HASScategories for the accident mechanism,article involved, the activity and location ofthe accident to specific hazards in thehome, as covered by the HHSRS, hasproved tenuous in many cases. That said,it is still possible with the HASS data todistinguish certain critical attributes of ahazard, for example, to distinguish falls oninside stairs from falls on outside stairsand steps. Such an analysis shows thatthere is a far stronger link between non-fatal accidents and housing type and agein the case of inside stairs than for outsidestairs and steps, where such factors asthe topography of the site may be moreimportant. (Stat 7.13)

We will continue to liaise with DTI andseek improvements in the usability ofHASS and HADD data in the context ofthe HHSRS.

As 115. As 115.125

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

51


There is a particular need to improve thestatistical evidence for houses in multipleoccupation (HMOs). The HASS, generalmortality and Fire Brigade data should alldistinguish accidents and fires in HMOs,using standard common definitions ofmultiple occupation, possibly based onthose used in the EHCS. (Stat 7.04)

We will continue to liaise with DTI, FireDirectorate and others and seekimprovements in the usability of variousdatasets in the context of the HHSRS.

As 115. As 115.129

Data relevant to non-fatal classes of harmvaries substantially between differentdatabases – e.g., for the HASS data,classes of harm have been determinedfrom variables giving the type of injury, thebody part affected, the nature of treatmentand the number of in-patient days.However, for the Fire Brigade data, themost relevant information concerned adifferent categorisation of type of injury, thecircumstances of the injury and thenumber of people killed or injured.Moreover, none of the variables or theircombination was wholly compatible withthe BRE’s classification of injuries intoclasses of harm26. (Stat 7.19)


As 115. As 115.128

In the final statistics distinctions in theharm outcome are lost due to the need toalso consider fatal accidents, as providedby the mortality data held by NationalStatistics (ONS). These data arecategorised according to the externalcauses codes of the InternationalClassification of Diseases (ICD) andprovide significantly less detail of hazardsthan is possible from the HASS data. (Stat 7.14)

We will liaise with ONS and seekimprovements in the usability of Mortalitydata in the context of the HHSRS.

As 115. As 115.127

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

52


26 Raw, G., et al., A risk assessment procedure for health and safety in buildings. 1999, BRE: Watford.

While it may be unrealistic to expect thevariables in commercial surveys to bestandardised, Residata, for example,would have been of even greater use inproducing the HHSRS statistics had itdistinguished converted from purpose builtflats and provided dwelling age bandsexactly comparable with those used in theODPM’s national housing surveys,particularly the EHCS. (Stat 7.03)

Realistically, our priority is focusing ondatasets within central and localgovernment as a starting point.

As 115. As 115.130

A major advantage of the HHSRS is that itprovides a clear focus for further work inthis field and an important part of thisproject has been to identify andrecommend improvements in the statisticalevidence that can be made in future years.It is likely that in future, the health andsafety statistics will be further updated andthat there will be other projects relatinghealth statistics to other data sources.Consequently, it would be helpful ifstandard protocols for recording theinformation could be adopted. Whilecommercial companies may not followsuch protocols, it would seem appropriatefor these to be followed by the variouscentral, regional and local governmentdepartments. (Stat 7.02)


As 115. As 115.130

Project R

eport: P

reparation of H

ousing Health and

Safety R

ating System

Guid

ance (Version 2)

53


Product code 04 HC 02608(b)

Project R

eport – Preparation of H

ousing Health and S

afety Rating S

ystem G

uidance (Version 2)O

DP

M

The Housing Health and Safety Rating System is a means ofidentifying faults in dwellings and of evaluating the potentialeffect of any faults on the health and safety of the occupants orvisitors. This report highlights and describes the changes in thedevelopment of Version 2 guidance and its associated elements.

project report · 2018-07-22 · project report preparation of housing health and safety rating...

Documents