The Architectural Healthcare Environment and its … · The Architectural Healthcare Environment and its Effects on Patient Health Outcomes ... 3 Research methods page 4 ... architectural

For queries on the status of this document contact [email protected] or telephone 029 2031 5512

Status Note amended March 2013

The Architectural Healthcare Environment and its Effects on Patient Health Outcomes

2003

STATUS IN WALES

INFORMATION

By:

Prof. Bryan Lawson and Dr. Michael PhiriUniversity of Sheffield

In collaboration with:

John Wells-ThorpeSouth Downs HealthNHS Trust

Poole Hospital NHS Trust

NHS Estates

The Architectural HealthcareEnvironment and its Effects on Patient Health OutcomesA Report on an NHS Estates FundedResearch Project

UNIVERSITY OF SHEFFIELD

Poole HospitalNHS Trust

South Downs HealthNHS Trust

The Architectural Healthcare Environment

and its effect on Patient Health Outcomes

A REPORT ON AN NHS ESTATES-FUNDED RESEARCH PROJECT

Professor Bryan Lawson and Dr Michael PhiriSchool of Architectural Studies, University of Sheffield

in collaboration with

John Wells-Thorpe

© Copyright NHS Estates 2003

THE ARCHITECTURAL HEALTHCARE ENVIRONMENT AND ITS EFFECT ON PATIENT HEALTH OUTCOMES

1 Introduction page 2

2 Executive summary page 3

3 Research methods page 4

4 Background and the literature page 5

5 Patients’ reactions to their architecturalenvironment page 6

6 Patient health outcomes page 10

7 Architectural factors responsible for these effects page 13

8 Costs page 19

9 Opportunity page 21

Contents

1

Most hospital patients may get the personal attention ofa doctor for only a few minutes a day and slightly longerperiods of personal care from nurses and therapists.However they may remain in bed or, if they are morefortunate, sit for many hours with little to do. This may well make them even more susceptible to theenvironment and more sensitive to it. It is reasonable,therefore, to assume that their environment may be acontributory factor to their sense of well-being andactual recovery. So we ask here whether thearchitectural environment of the hospital can contributeto the treatment of patients and significantly influencetheir health outcomes. The study clearly indicates thatthe answer is “yes”. It goes on to show how this effectworks and what factors might be chiefly responsible.

This is the report of a series of three one-year fundedprojects conducted over a four-year period finishing in December 2001. The main research work wasconducted by the University of Sheffield directed byProfessor Bryan Lawson. The research group worked in close collaboration with two NHS trusts at Poole

Hospital and South Downs Health in Brighton. Asteering committee overseeing the work was chairedthroughout by John Wells-Thorpe, who originated thestudy.

The study examines the effects of the architecturalenvironment on the lives of patients and to some extentstaff in two NHS hospitals, one each in the generalmedical and the mental health sectors. We examinedpatients’ reactions to the environment and the healthoutcomes that resulted from their treatment in it. A hugeamount of data was collected in this study. This reportonly illustrates a small proportion of this data, much ofwhich is very detailed. As a result of this work we havealso established a very large database of references inthe literature which are relevant to the major questionunder investigation. This database is being madeavailable separately in electronic searchable form.

This work has already been published in part in anumber of journals and books. Further, more detailedresearch publications are planned.


2

1 Introduction

Figure 1 Interior of Poole Hospital single-bed room Figure 2 Entrance to Mill View Hospital Brighton

• Patients are sensitive to and articulate about theirarchitectural environment in hospital.

They are able to discriminate between poor and goodenvironments and to tell us clearly what they like anddislike about them.

• Patients appear to make significantly better progressin the new purpose-designed buildings than in theirolder counterparts.

In the mental health sector patient treatment timeswere reduced by about 14%.

In the general medical sector non-operative patienttreatment times were reduced by about 21%.

• There is considerable evidence that an overallimproved atmosphere and quality of life may be oneof the benefits of better places.

Patients rate both their treatment and the staff caringfor them more highly.

In the mental health wards the number of seriouscases of verbal abuse and threatening behaviourwere significantly reduced. Patients were required to spend significantly less time in secureaccommodation.

In the general medical wards patients requiredsignificantly reduced levels of class A analgesicmedication.

• Most of the architectural features apparentlyresponsible for these benefits appear to be genericplace-making features rather than hospital-specificfactors.

Patients feel very strongly about the issue ofcommunity versus privacy and have strongpreferences for either single- or multiple-bedaccommodation.

Patients in the kind of accommodation they preferappear to do significantly better than those who arenot.

Having a view of the world outside seems veryimportant.

Not only being comfortable, but having personalcontrol over their immediate environment, isimportant.

Cleanliness and tidiness are given a high priority bypatients.

• Our results may be conservative. Neither of our newward designs was ideal in terms of the architecturalfeatures described above. In particular:

Poole had less single-bed accommodation than mightbe optimal.

Views could have been improved for many patients atboth Poole and Brighton.

Noise remained a significant problem at both Pooleand Brighton.

Neither design gave patients much control over theirenvironment.

The cleaning regime at Poole was below whatpatients expect.

• Costs do not appear to be significantly higher in thenew accommodation.

Capital costs were less than the relevant benchmarkfigures for each type of building.

Service delivery costs show no significant differencesbetween old and new wards.

Over the life-cycle of the buildings we have studied,they are likely to save their respective trusts moneycompared with continuing to operate the previousbuildings.

3

2 Executive summary of findings

The main study consists of a field investigation of wardsat two hospitals, one in general medicine (Poole) andone in mental health (South Downs Health in Brighton).Both trusts had development programmes and werebuilding new accommodation which was opened duringour study. In both cases the new wards were to takepatients with the same pattern of referral and whounderwent the same treatment regimes as those in the older accommodation.

Poole Hospital Trust was refurbishing a series of existing1960s general wards. In the original wards there were 6 four-bed bays and 6 one-bed bays, with lavatories ateach end of the ward. In the refurbished unit there are16 single bedrooms and 3 four-bed bays. The newbedrooms have a clean, simple interior using naturaltimber and have en-suite bathrooms (Figure 1).

At South Downs Health NHS Trust in Brighton theoriginal accommodation for the mentally ill comprised15-bed wards in typical Victorian brick institutionalbuildings with characteristically high ceilings. These werereplaced with a new medium secure mental healthcarebuilding designed by Powell and Moya using only singlerooms and now known as Mill View Hospital Hove(Figure 2).

We therefore studied four samples of patients,consisting of a sample in each hospital in the old wardsand one in the new wards. Inevitably the samples ofpatients in the new buildings cannot have been perfectlyidentical to those in the old, but we are confident thatthey were as similar as could reasonably be hoped for inreal practice. The patterns of referral, treatment regimesand other factors were substantially the same and inmany cases the staff were also the same. Sample sizeswere approximately 140 in Poole General Hospital wherepatients typically stayed for 9 or 10 days, and about 75in the Brighton Mental Health units where patientstypically stayed rather longer, about 35–40 days. In thecase of the two samples in the new accommodation,

several months were allowed to elapse and the wholesystem to settle down after moving in before we beganour study.

In addition to the normal literature review we conducteda series of focus groups with key people as a first stepin defining the detailed investigation. Focus groupmeetings were held with groups of staff in each of thetwo hospitals. We also held a focus group with invitedhealthcare design specialists. A postal study was doneof the views of NHS trust Directors of Estates.

Together with the literature review this work enabled usto compile meaningful questionnaires to be completedby patients in our surveyed hospitals. This alsostructured the health outcome data that the trustsgathered for the same samples of patients.

The questionnaires were completed by patientsthemselves in the general medical wards and with thehelp of their carers in the mental health wards. Thequestionnaires were administered at the end of theperiod of hospital treatment.

Finally we conducted a number of studies to test thecomparability of our samples. These include the levels of staffing, treatment patterns, service delivery costs and the capital costs of the buildings. We looked at the size and timing of our samples and other suchmethodological factors that might have influencedresults. We also closely inspected the data to seewhether any unusual patterns of patient symptoms such as major infections, admissions factors and the like might be responsible for any of the results.

Various sections of the research work reported herehave already been published in journals, and the readerwishing to see more details of the data, statistical tests,levels of significance and research methodology mayfind these useful.1


4

3 Research methods

1 Lawson, B. R. and Phiri, M. (2000). ‘Room for improvement,’

Health Service Journal, 110 (5688 20:1:2000): 24–27.

Lawson, B. (2002). ‘Healing architecture,’ The Architectural

Review, CCXI: 72–75.

Lawson, B. and Wells-Thorpe, J. (2002). ‘The effect of the

hospital environment on the patient experience and health

outcomes,’ The Journal of Healthcare Design and

Development, March: 27–32.

The idea that our environment can contribute to ourwell-being is perhaps not extraordinary, and yet hasreceived relatively little attention in the literature.However, the general literature on the psychology of thearchitectural environment is growing steadily.2 In a verywidespread literature review we have found manysuggestions that this might be the case in hospitals, and this is recently well summarised by Peter Scher,3

whose suggestions are certainly confirmed by our data.However, very little hard evidence has been gathered,and almost none of it in strictly controlled comparativestudies such as this. There is remarkably little researchof this kind of a holistic nature, but a number of moredetailed studies have been published. Roger Ulrichconcluded that patients in accommodation with a view

were more likely to be released from hospital morequickly than those without.4 Interestingly, FlorenceNightingale had already suggested the importance ofthis a century earlier, simply based on her own personalobservations!5 Another study shows that sunny aspectshave a better effect than dull ones.6 Others have lookedat the organisation of space and the arrangement offurniture, for example in mental health.7 Other morerecent work has looked at the effects of music and art in hospitals.8

We shall not here review the whole of the extensivemore detailed literature which may be brought to bearon this issue. A database of that literature is being madeavailable separately.

5

4 Background and the literature

Figure 3 A ward sitting area with a sea view at Poole

Figure 4 A day space with a garden view at Brighton

2 For a full discussion of this generally and many references

see Lawson, B. R. (2001). The Language of Space. Oxford,

Architectural Press.

3 Scher, P. (1996). Patient-Focused Architecture for Health

Care. Manchester, Manchester Metropolitan University.

4 Ulrich, R. S. (1984). ‘View through a window may influence

recovery from surgery,’ Science, 224: 420–421.

5 Nightingale, F. (1860). Notes on Nursing. London, Harrison

and Sons.

6 Beauchemin, K. M. and Hays, P. (1996). ‘Sunny hospital

rooms expedite recovery from severe and refractory

depressions,’ Journal of Affective Disorders, 40: 49–51.

7 Baldwin, S. (1985). ‘Effects of furniture arrangement on the

atmosphere of wards in a maximum-security hospital,’

Hospital & Community Psychiatry, 36(5): 525–528.

8 Staricoff, R. L., Duncan, J. et al. (2001). ‘A study of the

effects of the visual and performing arts in healthcare,’ Hospital

Development, 32(6): 25–28.

In our focus group held with experienced healthcaredesign specialists we were somewhat surprised andrather depressed to encounter some views thatsuggested this study was unlikely to be successful. One member of the group was very explicit about this:

I think we waste a lot of effort asking people in thatstate (ill in hospital) what they think of the architecture.It’s vain of us and we don’t actually find out very much.

The suggestion here was that patients in hospital haveso much on their mind about their illness that they areunable to focus on the architecture and may not evenreally notice it. Our study has very clearly shown this tobe false as a generalisation. Of course there may well besome very seriously ill patients for whom this is true, butthe vast majority of our samples were sensitive to andhighly articulate about their architectural surroundings inhospital. Perhaps this is hardly surprising, for while theymay see a doctor once a day and a nurse several timesa day, they see their surroundings all day. It is worthnoting that in general, levels of expressed satisfactionwere lower for the mental health patients than for thoseon general wards. This is probably understandable giventhe nature of the problems these patients have. What is

important here is that the trends between old and newbuildings were the same in both sectors.

APPEARANCE AND OVERALL DESIGN

Patients in the newer buildings expressed significantlymore satisfaction with the appearance, layout andoverall design of their wards. When asked aboutappearance at Poole about 73% of the patients on the new wards gave them the highest possible rating,compared with only about 37% on the old wards (Figure 5). At Brighton these figures were both lower, at 41% and 20%. Clearly the designs are bothconsiderable improvements, not just in their appearancebut also functionally. When asked about the overalldesign at Poole about 65% of patients gave the highestrating to the new wards compared with only 35% on theold wards. Similarly we saw figures of 47% against 14%at Brighton.

THE IMMEDIATE PERSONAL OR PRIVATE BEDAREA

We are aware that there is a considerable debate to behad about the nature of the private accommodation thatshould be offered to patients in hospital. Some clearly


6

5 Patients’ reactions to their architecturalenvironment

Figure 5 Patients assess the appearance of the old and new wards

Poole overall appearance Brighton overall appearance

%

60

40

20

01 2 3 4 5

%

80

60

20

40

01 2 3 4 5

Old

New

Old

NewRatingRating

feel that all patients should have private rooms, andthere is an increasing tendency to provide this. We shallreturn to this question later in the report. However, herewe report that patients in general in both our hospitalsreported substantially higher levels of satisfaction in thenew wards than the old. In fact, at Poole 72% gave theirpersonal bed area the highest rating on the new wardscompared with only 38% on the old wards. Similarly, thefigures for Brighton were 51% compared with 16%. Thisimprovement holds true both for patients in single roomsand patients in multiple-bed bays (Figure 6).

ENVIRONMENTAL COMFORT

We asked patients quite detailed questions about theenvironmental quality in terms of lighting, temperature,air quality and noise. Again, in both the newer wardsthese showed improvements, but relatively small ones,and most of them either not statistically significant oronly marginally so. In fact the patients giving the highestrating for environmental comfort factors rose only to55% from 46% for temperature, for example. Thefigures for Brighton did not show significantimprovement. Taken together with more detailedquestions this data clearly indicates that our newerwards were perceived by patients as significantimprovements in spatial and visual terms but onlymarginally better in terms of environmental comfort(Figure 7). We had many complaints about noise. Ourhospitals, it seems, are generally pretty noisy places.This may well be in part exacerbated by the hardreflective surfaces that seem common. However, forthose in open wards the levels of background noise canbe hard to escape. Patients frequently mentioned thesources of noise to us in the open-ended sections of

the questionnaire. The results accord with other workdone on noise annoyance. It may not necessarily be theabsolute physical level of noise but more its meaningthat can be annoying.9 For example, patients frequentlycomplained about nurses chatting as they change overshifts at night. Such noise levels are probably very low inreal terms, but none the less annoying for that!

ENVIRONMENTAL CONTROL

Why is it the case, with all our contemporary technologyand understanding of environmental comfort and thesystems needed to deliver it, that we still find such roomfor improvement? One clue to this lies in the degree towhich patients have control over their environment. Infact the data makes very depressing reading here.Although in both hospitals patients seemed to havemore control over lighting, they report having very littlecontrol over temperature and air quality, blinds andcurtains, and noise (Figure 8). One of the most powerfulpieces of anecdotal data to support our empirical workcame from a nurse in one of our focus groups. Shesuddenly and unexpectedly became a patient in herown hospital and was for a while entirely bed-bound.She remembered vividly lying in bed with the sun fallingdirectly on her face. She could see blinds on thewindows but could not operate them from her bed.Knowing how busy nurses are, only after some time,and then reluctantly, did she press the nurse call button.After another period of 5–10 minutes a nurse came andclosed the blinds. With the inevitable capriciousness of

7

Figure 6 Patients assess the private bed areas at Brighton

9 Lawson, B. R. and Walters, D. (1974). ‘The effects of a new

motorway on an established residential area’, in Psychology

and the Built Environment, D. Canter and T. Lee, London,

The Architectural Press: 132–138.

New

60

40

20

01

23

4

5

Old

%

the English summer, the sun soon went in and now shewas left in darkness and unable to see clearly enough to read. She recalled feeling unable to bother the nurseagain and told how she “simply lay there getting angry”.

We suspect that the costs of bedside controls fornatural and artificial light and for temperature andventilation are very small in terms of the overall capitalbudget but would pay off hugely in terms of patientsatisfaction. Moreover, we would argue that the idea ofone set of environmental conditions being suitable for all, and at all times, is frankly absurd in the light of theliterature on environmental comfort. Somehow, giving

patients control is still seen as an expensive luxury on the one hand and likely to lead to managementproblems on the other. We would hope these attitudeswould be reconsidered as a matter of some priority, atleast in the specification of new buildings.

PATIENTS’ REACTIONS TO THEIR TREATMENTIN THE NEW ARCHITECTURE

We also asked our patients if they thought that thearchitectural environment had helped to make them feel better. In general again, both newer wards showed higher ratings. At Poole this was a significant


8

Figure 7 Patients assess environmental comfort

Figure 8 Patients’ level of environmental control

New

%

80

60

40

20

01

23

45

Old New

%

80

60

40

20

01

23

45

Old

Noise at Poole Temperature at Brighton

New

%

60

40

20

01

23

45

OldNew

%

50

40

20

30

10

01

23

45

Old

Noise at Poole Temperature at Brighton

improvement, with 85% on the newer ward feeling theenvironment helped them compared with only 68% onthe old ward. Again we saw lower but comparablefigures at Brighton of 68% compared with 39% (Figure 9). In both general medicine and psychiatriccare, then, patients clearly see the environment asplaying a role in their care, just as many researchers anddesigners in healthcare have believed.

However, perhaps more remarkably and interestingly, wefound that patients on the newer wards also gave higherratings to both the treatment they received and the staffwho delivered it. These differences were inevitably verysmall at Poole, since all patients showed very high levelsof approval for their care even in the old wards.However, at Brighton some 56% of patients in thenewer building were pleased with their treatmentcompared with only 39% in the old buildings (Figure 10).Similarly, doctors, nurses, therapists and ancillaryworkers at both Poole and Brighton attracted higherscores from the patients in the new buildings. Theimprovements in the doctors’ ratings are the mostdramatic of these, but again this is because patientsthink so highly of their nurses even on the older wards.None of these improvements are sufficiently large to be statistically significant and, standing alone, might bediscounted. However, taken together with all our otherdata, we see a clear overall picture of the patients in thenewer buildings being happier in their surroundings andfeeling this has helped.

So are the doctors and other staff better in the newerarchitecture, and is the medical treatment superior?Well, as we reported in the methodology section, manyif not most of the staff were actually the same people.They certainly reported to us their intention to provide

the same treatment regimes. However, of course, staffare likely to respond to better environments just as arepatients. As we shall see later, patients are significantlymore cheerful in the new wards and again staff are likelyto react to that – after all, they are only human!

It seems likely to us that we are seeing a double effecthere, albeit a small one. Patients themselves are happierin the better architectural settings and may transfer thisfeeling of well-being to their judgements of the wholesituation, including their assessments of staff andtreatment. In turn staff are more positive and indeedcommunicate this to their patients. It is a virtuous circleor upward spiral of effect.

9

Figure 9 Does the environment help you feel better? (Brighton)

New

%

40

30

20

10

01

2

3

4

5

Old

So far, we have reported empirical evidence to supportour argument about the effect of architecture onpatients, and most of it is statistically significant.However, it still remains largely subjective. Twoquestions remain. First, do the patients actually benefitfrom this in real health outcomes?; and secondly, whatare the major contributory architectural factorsresponsible?

The two health trusts involved monitored the patientsand provided us with a mass of data about theirprogress while in hospital. Measuring patient progress isnot as simple a matter as it might seem, since there aremany potential indicators, with perhaps the mostobvious being the length of treatment.

In fact many patients in both our acute hospital andmental hospital samples were released significantly morequickly from the new wards than from the old ones.Non-operative acute patients showed a significantreduction of some 21% (Poole) in treatment times, andmental health patients a reduction of 14% (Brighton;Figure 11). Those patients at Poole who underwentoperations were not released more quickly. This seemsto be due to particular circumstances relating to theirpre-operative period, which was on average almost

twice that on the old wards. As far as we can tell, thereasons for this have nothing to do with the design oroperation of the new ward.

At this point it is worth reporting that we have takenconsiderable trouble to search for other medical andcircumstantial factors that might be influencing theseresults. For example, we have checked that there wasno uneven distribution of syndromes presented at eitherPoole or Brighton. We have checked for cases of MRSAand other complications that might have causedperturbations in the data. Whilst there may be minorvariations, we have been quite unable to discover anymajor significant differences in our samples that mightexplain this variation in treatment times.

There are also other interesting indicators that furthercontribute to the picture. In our acute hospital, analgesicmedication is largely taken on demand within prescribedlimits. There was a dramatic reduction in the amount ofanalgesic medication taken by the patients on the newwards. On the newer wards the average number of dayson which Class A pain-killing drugs were administeredwas reduced by 22%. Moreover, the number of dosesapplied on these days reduced by 47% (Figure 12). Werecorded the total quantity of drugs used in each of the


10

Figure 10 Patients’ assessment of their overall treatment (Brighton)

6 Patient health outcomes

New

%

40

30

20

10

01

23

45

Old

three classes, A1 morphine, A2 oramorph and A3codeine phosphate. In fact codeine phosphate was notused on either ward, but some 55% less morphine and86% less oramorph was used on the newer ward.Although these figures sound very dramatic, somecaution should be applied to them, as the standarddeviation in the quantities was quite large. To slightlyoffset this data there was a slight increase in the amountof orally-taken Class C drugs on the newer wards.Effectively, here we see patients requesting less pain-killing medication in the new hospital ward environmentsthan in the old. We can only conclude that as a result of

their environment, patients were less aware of, or atleast less prone to complain of, their pain.

At Brighton we also had staff record the rate of progressthey thought their patients were making. In the newwards significantly more patients were judged to bemaking good rather than slow progress (Figure 13). In amental health hospital it is normal to record many itemsof patient behaviour including all instances of verbalabuse, physical violence towards others and physicalself-harm. These records were studied and the resultswere quite remarkable. Whilst the number of incidents ofverbal and physical abuse remained largely the same,

11

Figure 12 Levels of analgesic medication taken at Poole

Figure 11 Length of patients’ treatment times

New

%

10

8

6

4

2

OperationNon-op

OverallSecure

Old

New

%

0

40

20

0

Old

Length of treatment (Poole) Length of treatment (Brighton)

Refurbished wards

No

of

do

ses

16

14

12

10

8

6

4

2

0Class A Class C

Old wards

Levels of analgesic medication

their severity dropped quite significantly in the newwards. The number of instances of patients injuringthemselves was dramatically reduced by two-thirds(Figure 14). Patients who become particularly distressedand are considered a danger to themselves are normallyput for a period into seclusion in a safe room withintense supervisory care. The amount of time for whichthis was necessary was reduced by a remarkable 70%in the new unit, with an average reduction of 9 days,from 13 to 4, in a typical stay (Figure 14).

A clear and consistent picture emerges from a verycomplex set of data. Patients in the new buildings seemto spend less time in hospital and appear to feel lessphysical pain or to be psychologically calmer. In additionto making life better for patients, this must in turn makelife easier for the staff, certainly in the mental hospital.


12

Figure 13 Staff assessments of patient progress (Brighton)

Figure 14 Levels of abusive and threatening behaviour (Brighton)

New

%

80

60

40

20

0

SlowGood

Old

Patient progress

New

% 40

60

20

80

0

HighLow

Old

Verbal outbursts

New

%

60

40

20

0High

Low

Old

Threatening behaviour

Of course the new wards are just that. They are new!Some might suggest that our results may purely be theresult of this, and that as the buildings age, the effectswill disappear. We cannot be sure this will not happen,but there are some reasons to doubt that pure newnessis responsible for the effects we have recorded. First,other studies we have done show that if asked aboutarchitecture in general, most patients are more likely toprefer old buildings to new ones. In fact we wouldexpect there to be more tolerance of new architectureamongst younger patients, but they form a minority inour samples. However, secondly and much moreimportantly, the patients in our samples were able to tellus quite clearly what features of the new wards they feltmade them better places. Thirdly, we were able toanalyse our results in terms of some variation in oneimportant architectural factor, that of the design of theprivate bed space. Patients’ evaluation of this area gavethe highest overall correlation with their general levels ofsatisfaction. This has turned out to be both importantand complex, so we now devote considerable space tothis issue here.

SINGLE VERSUS MULTIPLE BED SPACES

In our initial study it appeared that the type ofaccommodation a patient is in might be an importantfactor. However, we also suspected that a moreimportant question might be whether or not patients arein the type of accommodation they preferred, but wehad not recorded this. We were also aware that asubstantial number of patients get moved, and that thisevent itself and the type of accommodation involvedmay be significant. We therefore conducted a specialnew study to examine these factors.

Two types of accommodation are considered here, all in wards in Poole General Hospital. They are single-bedspaces and multiple (always in this case four-bedded)spaces. Some 473 patients were interviewed. Of these106 (22%) were moved during their stay. They answereda questionnaire mainly consisting of five-point scale andopen-ended questions.

Of the patients who remained in one type ofaccommodation throughout their stay, 24% were insingle-bed spaces and 76% were in four-beddedspaces. The two groups of patients appear to have very

similar gender balances and age distributions and tohave stayed in hospital for roughly the same amount oftime. In fact the patients in single-bed spaces werereleased on average fractionally but not significantlyearlier than those in multiple-bed spaces.

When asked about their overall ward experience, theiroverall treatment and whether the environment helpedthem to feel better, the two groups gave almost identicalresponses. However, the group in single-bedaccommodation were significantly more impressed bytheir bed area or private space, with 71% giving it thehighest rating compared with only 33% in the multiple-bed spaces. They also rated their bathroom and toiletarea significantly more highly (48% against 26%).

PREFERENCES FOR SINGLE- OR MULTIPLE-BED ACCOMMODATION

Overall 54% of patients actually expressed a preferencefor multiple-bed space accommodation, with 43%preferring single and the rest not feeling any preference.However, this figure alone does not fully reveal thepicture. Of the patients who stayed in one type ofaccommodation, the great majority expressed apreference for it. Of the patients in multiple-bed spaces76% said they preferred them, while as many as 93% ofthe patients in single-bed spaces said they preferredthem. Whether this is because the hospital had done anexcellent job matching patients to their preferences orwhether patients simply come to like and see thebenefits of the accommodation they are in, we can onlyguess. One clue to this might be from a patient whowrote about the issue in the final open section. “The lasttwo times I was here I was in the multi-bed wards and Idid not think I would like the single room, but it was abetter environment once I became used to it.” Clearlythis patient would have previously been one of thosevoting for multi-bed spaces but this time voted for thesingle-bed space. However, there were also examples ofexactly the reverse happening.

Of the patients who were not moved during their stay and expressed a preference for one type ofaccommodation, 80% preferred the accommodationthey were in, while 20% would have preferred to be inthe other type. If we examine their assessment of theward and the experience of patients in their preferred

13

7 Architectural factors responsible forthese effects

type of accommodation, we can see that they weresignificantly happier than those patients who would havepreferred the other type of accommodation. In fact thisone factor alone appears to be the most important onein determining overall assessments.

Significantly, the group who were in their preferredaccommodation type thought the environment helpedthem to feel better, with 42% giving this the highestscore compared with only 26% of those in their non-preferred spaces (Figure 15). They described thefacilities as meeting their needs significantly better, with52% awarding the top score compared with 34%. Theyrated their overall ward experience significantly morehighly, with 78% giving it the highest rating comparedwith only 55%. They also rated their overall treatmentmore highly (81% compared with 68% giving the highestrating).

If we look to see what it was about their situation thatcaused this significantly higher assessment, we can seea clear indication. They felt their overall environmentalcomfort was higher, with 48% giving the top scorecompared with 25%. However, none of the individualenvironmental factors were actually significantly more highly rated. They did, however, feel they hadsignificantly more control over their environment. Theyawarded no higher scores for the appearance of theward or the overall design, nor did they feel they couldfind their way around any more easily. Clearly, then, theoverall design of the ward is not the factor causing theresult. However, they assessed their bedroom or privatearea significantly more highly, with 49% giving the topscore compared with 34%. They also rated the toiletand bathroom area more highly (36% against 25% for

the top score), but were no more impressed with thelounge area.

The issue, then, seems to come down largely to twofactors. It is a personal choice between privacy andcommunity, with the single-bed accommodation havingthe added advantage of offering more environmentalcontrol. Of all the extra comments patients made in their questionnaires this came up most frequently aftercomments about the staff. Two examples from eachside of the argument illustrate the choice perfectly. “Ifyou could choose the other patients, the multi-bed baywould be excellent.” This contrasts with “The single-bedroom is far better due to the nature of the illness, it’s notso embarrassing.” However, a couple of comments alsoin favour of the multi-bed spaces suggest that access tonursing staff is thought to be more frequent and easierthere. Many readers may feel, as do we, that they wouldprefer single rooms. However, we are convinced fromthis study that there are a significant number of peoplewho prefer to be in multiple-bed accommodation.However, our study may well be overestimating the sizeof this group. The two main reasons given by suchpeople are “having others to talk to and not beinglonely”, and “I am more likely to see a nurse than Iwould tucked away in a room by myself.” Whatever thereality about the latter, this is clearly the perception forsome at least.

What causes these preferences for either single- ormultiple-bed accommodation? We might expect this tobe a function of personality, but it might also be relatedto age, gender and socio-economic grouping. We haveno data on personality, but did collect data on age andgender and also domestic postcode. Gender and age


14

Figure 15 Patients in the accommodation they preferred or not (Poole)

%

40

20

60

0

Level of agreement

1 2 3 4 5

The environment helped me to feel better

Preferred

Not pref

both turn out to be no indication at all. We have beenunable to do the analysis on postcodes. Such factorsmay modify the percentage preferring private spacesand enable trusts to predict more accurately the balanceof preference.

PATIENTS WHO WERE MOVED DURING THEIRSTAY

So far we have only considered those patients in thesame accommodation throughout their stay, but some22% of our sample patients were in fact moved. Whatare their views on the factors considered above?

First, those who stayed in one place thought that theenvironment helped them to feel better significantlymore than those who were moved. They also gavehigher scores for their overall ward experience and theiroverall treatment, although these differences were notstatistically significant. Again, they thought the facilitiesmet their needs better, preferred the appearance of theward, and thought more highly of their bed or privatespace. Again, none of these differences was statisticallysignificant. Overall, however, these results fairly stronglysuggest that being moved is in itself not a positiveexperience.

WINDOWS, NATURAL LIGHT, SUNLIGHT,VENTILATION AND VIEWS

After the question of privacy versus community the nextmost frequently mentioned concern was that of views,or more often the lack of them. In our study the windowwas the most frequently mentioned building element. Itcropped up in our focus groups and was very frequently

mentioned by patients in the free response parts of ourquestionnaire. Quite simply patients, and for that matterstaff too, like to have windows and to be able to see outof them. This is perhaps best summed up by one of ourfocus group nurses. “I think there’s something desirableabout natural light and ventilation.” She was of courserepeating an opinion expressed confidently by herillustrious predecessor Florence Nightingale.10 Perhaps it is time for us to take more seriously this kind ofanecdotal evidence from those who are in daily contactwith hospitalised patients. In fact there is a growingbody of empirical work in the literature supporting theidea that both natural lighting and sunlight havetherapeutic qualities.11

The window also has the potential to offer views out ofthe building. We continue to be surprised at the lack of

15

Figure 16 A bedroom with a sea view at Poole

Figure 17 The courtyard garden at Brighton

10 Over a hundred years ago, Florence Nightingale mentioned

light and views and colour as being “second only to fresh

air” in this regard. Nightingale, F. (1863). Notes on

Hospitals (3rd edition), London.

11 Beauchemin, K. M. and Hays, P. (1996). ‘Sunny hospital

rooms expedite recovery from severe and refractory

depressions,’ J of Affective Disorders, 40, 1996, 49–51.

Study of psychiatric inpatients with depression in a hospital

in Edmonton, Alberta, Canada (n = 174). Finding – Patients

in sunny rooms stayed an average of 16.9 days, whereas

those in “dull” rooms stayed an average of 19.5 days. The

difference was consistent over all seasons during a period

of two years (October 1993–September 1995).

Jackson, J. G. (1973). ‘What we want from daylight’, Light

and Lighting, 80.

attention to this that we detect in much recent hospitalarchitecture. Again patients raised this with us veryfrequently, either commenting favourably on the viewthey had or bemoaning the lack of a view. A nurse inone of our focus groups was in no doubt about thetherapeutic value of the view. “You know that if you sit infront of a view for a long time watching the clouds youforget what you’ve been worrying about and you thinkof other things, and you know that that’s doing yougood.” Equally another nurse worried about the lack ofviews in her hospital. “You can get very depressed whenyou can’t see past your own door.” The specific need tosee the outside world, feel in touch with it and see whatis going on is stressed by all groups, but particularly bythe caring staff.12 This is no longer just a matter of suchanecdotal data. In a now well-known study Ulrich hasclaimed evidence of the actual therapeutic value of theview through hospital windows.13 Our data very stronglysupports this finding. However, we must be careful herenot to romanticise the content of the view. In fact wehave no evidence that patients in general desireconventionally beautiful landscapes, for example, andthere is certainly no demand for the view onto theubiquitous hospital courtyard, however ingeniouslylandscaped. In fact what evidence we have suggeststhat what is most desirable here is a view of the worldbeyond the hospital and of life going on in the normalway. It may well be that one of the benefits of thenatural lighting that comes through the window is theway it assists in defining the diurnal rhythm. A view itselfmay well further anchor the patient in the daily pattern oflife, with such events as the postman making hisrounds, children going to school and the evening rush-hour home. We would argue that the patient may feelless cut off from their normal world and indeed mayhave distraction and items for conversation with otherpatients. This was again expressed to us strongly by apatient carer from the mental health sector in one of ourfocus groups:14 “Patients do like to know what’s goingon outside and to be kept in touch with ordinary life . . .It’s very easy to become enclosed in your own reality

when you’re actually an in-patient, and that won’t speedyour recovery . . . You’re not really going to promoteindependence and good coping skills when you areremoving the patient from reality and the real world.”

There is one final point that came very dramatically anddirectly from our data in this study. We must not onlyhave good design principles, but also realise them wellin actual designs. Our hospital at Poole had potentiallysplendid views over Poole harbour, and they were much appreciated by patients would could see them.However, the sill heights of the windows and bedarrangements in some rooms prevented bedriddenpatients from seeing them. Such patients commentedquite angrily about this, and one can hardly blame them!Yet again the strength of their feeling reinforces theimportance of view for hospital patients.

A CONTROLLABLE AND PERSONALENVIRONMENT

We have already seen in our earlier section how littlecontrol patients said they had over their environmentalconditions. The general issue of how much controlpatients have over their immediate surroundings and theextent to which they can make them individual attractedthe next largest group of comments from patients onquestionnaires and from staff in focus groups. Onemember of the medical staff, reminiscing about alsohaving recently been a patient, summed this up verywell. “After the operation, well the second night, it wasmidsummer, and I couldn’t close the window. I couldget out of bed but I hadn’t the strength to lean over theshelf and close the window. The call button wouldn’twork. I couldn’t get the curtains closed or anything and I couldn’t switch my light off. It wasn’t until aboutmidnight when the staff came round and I was just lying there so irritated and so agitated.” This of courseraises the possibility of bedside controls for patients,commonly found for many items in the modern hotel but seldom for the bedridden patient in hospital!15 Staffcommented to us on the frequency with which patients’relatives bring in bedside fans and other devices to tryto remedy this deficiency. Even controls for staff tooperate would in many cases be a remarkableimprovement. Staff seemed to feel that such controlsand consideration were generally missing, leaving themto improvise. “I wanted a muted light and we ended uphanging a pillowcase on it, which really!” It is worthpointing out here not only the negative frustrationcaused by a lack of control and wasted staff time,


16

12 Ulrich, R. S. (1979). ‘Visual landscapes and psychological

well-being’, Landscape Research, 4: 17–23.

13 Ulrich, R. S. (1984). ‘View through a window may influence

recovery from surgery’, Science, 224: 420–421. Study of

Cholecystectomy patients (n = 46). Finding – Patients

assigned to rooms with window views of a natural setting

had shorter post-operative hospital stays, received fewer

negative evaluative comments in nurses’ notes, and took

fewer potent analgesics than patients in similar rooms with

windows facing brick wall. Outcomes – Length of hospital

stay, negative evaluative comments in nurses’ notes,

amount of potent analgesics taken.

14 Ulrich, R. S. (1979). ‘Visual landscapes and psychological

well-being’, Landscape Research, 4: 17–23.

15 Kenny, C. (1979). Evaluating acute general hospitals, in

Designing for Therapeutic Environments, John Wiley &

Sons, 1979: 309–332; also see Restuccia, J. D. (1982),

‘The effect of concurrent feedback in reducing in

appropriate hospital utilisation,’ Medical Care, 1982: 20:

46–62.

but the denial to the patient of a more therapeuticenvironment.16

Finally we return to our earlier discussion of single- andmultiple-bed spaces in this section. What patients reallyseemed to be telling us here is the frustration andunpleasantness of not being able to control your privacyand community as you would normally and naturally doin everyday life. Choosing whether your door is open,ajar or shut, for example, is something we all do veryfrequently in the domestic and work environment, andyet the ability to do this is again normally denied to thebedridden patient.

A CLEAN, TIDY AND CARED-FOR PLACE – THE BATHROOM AND TOILET

Next we have grouped together a whole series ofcomments that initially might have seemed to be aboutdifferent architectural features or service aspects. Webelieve they relate strongly to each other under theheading of a clean and tidy place that appears loved orat least cared for. Patients frequently mentioned thebathroom/toilet area to us. The overall score for thebathroom/toilet area was the lowest of all of the specificareas we asked about, and lower than for any of thegeneral questions about appearance, design and theenvironment of the ward. This was true for both old andnew wards. In this regard it is interesting to see that the ratings of the bathroom/toilet area correlatesurprisingly strongly with the general assessments of theward design. We see a correlation of 0.6 between the

bathroom/toilet scores and both the “overall design”,and “how facilities meet my needs”. There is even acorrelation of 0.4 with “does the ward environment helpto make you feel better”.

These scores suggest the quality of the bathroom andtoilet areas are highly influential in determining patients’feelings about their experience. The number of patientswho also made comments about this in the open-endedquestions supports this. Some 48% of patients madenegative comments about the issue, while 29% madepositive ones. Of the negative comments by far thelargest number were about a lack of cleanliness, whilethe next most frequent complaint was about lack ofnumbers of baths, showers or toilets. The lack offacilities obviously causes distress, as does a lack ofprivacy. This last issue was also mentioned by severalpatients as the worst feature of the ward design overall.It seems that this area is of great significance to patientsand yet rather neglected by both the design and themanagement system.

Comments on bathrooms and toilets most often raisedissues of the lack of privacy or cleanliness there. It isinteresting that cleanliness featured most highly on boththe positive and negative side of these comments, andthis again indicates that it is a strongly-felt issue withpatients. When doing this they frequently mentioned theactual process of cleaning. It seems that they are lessfocused on the problems of infection than on the qualityand frequency of the act of cleaning. We interpret this to be more of a symbolic concern. “If in a hospital theycannot even keep this area clean,” they seem to besaying, “how much do they really care for this place?”The obvious next step in their argument is to questionhow well they will be cared for in such a situation.

17

16 French, J. (1990). ‘Effects of bright illuminance on body

temperature and human performance: annual review of

chronopharmacology’, Annual Review of

Chronopharmacology, 7, 37–40.

Figure 18 Capital costs v. benchmarks and revenue costs new v. old units

Actual

£60,000

£40,000

£20,000

£0

BrightonPoole

Benchmark

New

£150

£100

£50

£0

BrightonPoole

Old

Capital costs per bed Revenue costs per bed day

In fact, in our focus groups one of the most frequentcomplaints made by designers was the lack of caretaken in maintaining and managing the buildings forwhich they are responsible. The apparently haphazardway notices and pictures are put up, left to become outof date and dirty was a common complaint. This wasalso mentioned by patients, but again more in symbolicterms.

APPEARANCE

While there is considerable consensus about the above factors, appearance is more subjective. Patientsobviously note and comment considerably on colourand decoration. In some cases at both the new wardsthese comments were favourable, but most commentsabout these matters were negative. Patients on the newward at Poole commented favourably on the presenceof pictures, but even here they did this by calling for

more! Other comments asked for wallpaper andcarpeting “to make a more homely appearance”. “Light”and “airy” were two adjectives frequently used todescribe patients’ satisfaction with the feel of the place.Similarly, “dark” spaces were not appreciated. Therather bright and strong colours in the new wards atBrighton were clearly not to the liking of a considerablenumber of patients.

There is some evidence that patients like to see a placethat they consider “homely”. From the comments made,this seems to include having variety and texture in allmatters. Variety is desirable in lighting, colour andmaterials. Having materials that are tactile and withwhich you can interact seems desirable. All these thingsrun counter to the conventional appearance of ahospital, in which smooth, cleanable surfaces anduniform fluorescent lighting are commonplace.


18

We have conducted a very extensive review of all thecosts associated with the construction and running ofthe new buildings in our sample, and compared thesewith both standard provisions and with the previousaccommodation in each case. Such comparisons arenot as easy to make as it may seem, and we haveconducted many tests in order to discover anyexcessive expenditure that may in some way accountfor the improvements shown earlier in this report. Ingeneral our conclusions are that there were noexcessive costs here either in construction or in therunning of these new facilities.

CAPITAL COSTS

Our first test is to see whether the two facilities costmore than the appropriate benchmark figures forbuilding work of their type. NHS Estates have calculatedthe benchmark (full business case) costs for similar unitsat current cost based on MIPS360 and includingallowances for “Consumerism” as detailed in the Healthcare Capital Investment document.17 We havetherefore converted these to the equivalent rates at the times of completion of our two projects, based onMIPS331.18 The Poole Hospital scheme is comparedwith a standard trauma unit of 90 beds based on HealthBuilding Note 419 and its guidance recommendation for50% single-bed rooms. The Brighton Hospital scheme iscompared with a 53-bed acute mental health unit basedon Health Building Note 35.20 Since the number of bedsat Poole (84) and Brighton (54) slightly differ from these

benchmarks we have arrived at a comparator price perbed.

On this per bed basis Poole cost 47.4% of itscomparator benchmark price and Brighton cost 77.1%(Figure 18). Poole was of course a conversion with partnew-build, and the significantly lower costs are probablylargely explained by this. The overall higher costs atBrighton compared with Poole are also largelyaccounted for by the 100% provision of single en-suitebedroom accommodation. In addition to thesebenchmarks we also checked to see that the actualcosts were comparable with other similar buildingsconstructed at around the same time.

REVENUE COSTS

These costs are largely made up from the staffing coststo operate the building and provide the service topatients. Here we have compared both Poole andBrighton with the costs actually incurred in the two olderschemes which they replace. In our calculations wehave included the costs of nursing, medical staff,administration, catering, housekeeping andmaintenance. In the case of Poole we included chargesfor therapy, pathology, blood, drugs and tests. Thesefigures were all calculated and supplied by the twotrusts (Figure 18).

On this basis the new unit at Poole saved some 3.6% ifwe calculate the costs per ward, but cost some 2.6%more on a per bed-day basis. This is due to the slightreduction from 30 to 28 beds per ward in the new unit.

By contrast, the new unit at Brighton cost some 9.5%more per ward but saved 4.7% per bed-day. This isexplained by a rather higher bed occupancy rate of97.4% in the new unit compared with 84.8% in the old.

CONCLUSIONS ON COSTS

We can find no evidence in these figures that the newbuildings are lavish or expensive provisions. Indeed, inboth cases they are below the target capital cost figuresfor buildings of their kind. Neither is there any evidenceof increased levels of recurrent expenditure which couldaccount for the improvements that patients and staffhave reported to us and the enhanced health outcomeswe have recorded in these two new schemes.

19

8 Costs

17 NHS Estates, Healthcare Capital Investment Version

2.0, 14 December 2000, NHS Estates, Leeds.

18 MIPS represents the Median Index of Public Sector Building

Tender Prices and is a price index capable of correcting for

inflation over time in the costs of such buildings.

19 NHS Estates, Health Building Note 4 – In-patient

Accommodation, Vol 1 (Options for Choice), 1997 and

Vol 2 (European Case Studies), 1998, HMSO, London.

20 NHS Estates, Health Building Note 35 –

Accommodation for people with mental illness: Part 1,

The Acute Unit, 1996, Part 2, Facilities in the

Community (in preparation), Part 3, Case Studies (in

preparation), TSO, London.

One further set of statistics is worthy of mention here. Itis often argued in the construction industry that clientsin general and the public sector in particular place fartoo much emphasis on saving design fees and capitalcosts and not enough on life-cycle costs of buildings. It is worth noting that at Poole the revenue costs willexceed the capital costs in the second year of operationand at Brighton this will happen in the first year!

Earlier we reported savings in patient treatment timefound in our data of 21% for non-operative patients(about half the patients we monitored) at Poole and 14%at Brighton. We suspect that there may be other savings

to add to this of staff absenteeism, staff turnover andrecruitment costs, particularly at Brighton, and somesavings on drugs, particularly at Poole. However, thepatient treatment time savings alone amount to annualrevenue savings of nearly £2000 per bed-year at Pooleand nearly £7000 per bed-year at Brighton. Thesecompare interestingly with the annual capital charges(assuming depreciation over 25 years and 6% interestrates) of about £2800 per bed-year at Poole and about£4800 at Brighton. Of course we recognise that inpractice such savings are hardly likely to occur, and theeffect is much more likely to be increased throughput ofpatients.


20

Taken together, all these results demonstrate theimportance and value of good design in the healthcareenvironment. This is not simply, as is so often assumed,a superficial matter, but rather one of real consequenceto the quality of life of the patients, their visitors andcarers, and the staff of our hospitals. Our report alsodemonstrates that good design alone is not enough,and that facilities must also be well managed andmaintained.

What this report also shows is that, as well as makinglife pleasanter, good design contributes significantly tothe health outcomes of patients. It can therefore have areal effect in financial terms too and, far from costing thehealth service money, good design will probably save it.

21

9 Opportunity


22

Documents

The Architectural Healthcare Environment and its … · The Architectural Healthcare Environment and its Effects on Patient Health Outcomes ... 3 Research methods page 4 ... architectural