1

Evidence-­‐based  approaches  for  cri2cal  care  design  

EVICURES  Project  Seinäjoki      –      May  27,  2015  

Roger  Ulrich,  PhD    

 

Center  for  Healthcare  Architecture  Chalmers  University  of  Technology,  Gothenburg    

       

Evidence-­‐based  design  (EBD)  is  a  process  for  the  explicit  use  of  current  best  evidence  from  research  and  pracIce  in  making  decisions,  together  with  an  informed  client,  about  the  design  of  each  individual  project.  

 

       Hamilton  &  Watkins  (2009).  Evidence-­‐Based  Design  for  Mul5ple  Building  Types.  New  York:  Wiley  

Evidence-­‐based  design  (EBD)  

•  It  makes  compelling  sense  to  use  the  best  available  evidence  to  inform  the  design  of  a  hospital  or  other  healthcare  facility  that  will  be  used  for  many  years  and  for  which  so  much  is  at  stake.    

Evidence-­‐based  design  (EBD)  research  

•  Pa2ent  safety    (infecIon,  falls,  errors)    

•  Other  pa2ent  outcomes    (such  as  pain,  length  of  stay)  

 

•  Staff  outcomes    (work  saIsfacIon,  retenIon,  effecIveness)  

 

•  Costs  of  healthcare  

More  than  2,500  strong  studies  link  the  hospital  physical  environment  to  outcomes  in  following  major  areas:  

2

•  Improving  building  design  is  centrally  important  to  improving  healthcare  quality.  

     

Much  research  supports  this  general  conclusion:  

Best technical practice

Art of architecture

EBD  

OBJECTIVE     SUBJECTIVE    

Amount  of  EBD  knowledge  is  fairly  small.  But  it  carries  weight  with  decision-­‐makers  and  medical  professionals  because  most  of  the  knowledge  relates  to  priority  issues  such  as  safety,  and  can  be  linked  to  cost  savings.  

 

(Hamilton  &  Ulrich,  2012)  

Effects  of  noise  on  outcomes  and  costs  in  cri2cal  care  

Hospital  noise  levels  are  far  higher  than  recommended  values    

� Noise  sources  are  too  numerous  and  too  loud  

 

� Surfaces  are  sound  reflecIng    

There  is  growing  evidence  that  noise  worsens  pa2ent  and  staff  outcomes    

3

Noise  worsens  outcomes  

•  Reduces  oxygen  saturaIon  in  infants    

•  Elevates  blood  pressure,  respiraIon    

• Worsens  paIent  sleep    

•  Erodes  emoIonal  well-­‐being    

•  Increases  staff  work  pressure,  strain,  faIgue,  burnout  

 

• Worsens  speech  comprehension  

"Influences  of  noise  on  pa2ent  and  staff  outcomes  in  coronary  cri2cal  care"    

(Blomkvist, Theorell, Ulrich, Eriksen, Hagerman & Rasmanis, 2005)

STUDY  

•  Pa2ents:  adults  (94)  diagnosed  with  acute  myocardial  infarcIon  in  a  coronary  criIcal  care  unit  in  a  Stockholm  hospital  

•  Interven2on:  AcousIcs  were  improved  by  periodically  changing  ceiling  Iles  from  sound-­‐reflecIng  to  sound-­‐absorbing  Iles  

•  Findings:  During  good  acousIcs  paIents  slept  be\er,  had  less  physiological  stress  and  lower  incidence  of  re-­‐hospitalizaIon  

(Hagerman,  Rasmanis,  Blomkvist,  Ulrich,  Eriksen,  and  Theorell,  2005.    Interna5onal  Journal  of  Cardiology)     STUDY  

•  Staff:  nurses  (36)  who  were  specialists  in  cardiology  and  worked  regularly  in  the  coronary  criIcal  care  unit  

•  Findings:  During  good  acousIcs  staff:  ®  experienced  lessened  work  demands  ®  increased  workplace  social  support    ®  improved  quality  of  paIent  care  ®  be\er  speech  intelligibility  

(Blomkvist, Eriksen, Theorell, Ulrich and Rasmanis, 2005. Occupational and Environmental Medicine)

4

Growing  and  serious  challenge  for  hospitals  in  all  countries:      

Mul2-­‐drug  resistant  infec2ons  

The  post-­‐an2bio2c  era  is  here    

MRSA  prevalence  in  Europe  2010  

<  1%  

>50%   46%  

<  5%  

There  are  strong  reasons  for  expec2ng  risk  from  resistant  infec2ons  will  increase  in  Finland  and  other  Nordic  countries  •  Hospital  inpa2ents  are  ge_ng  sicker,  more  vulnerable  and  immune-­‐compromised  

 

•  New  resistant  infec2ons  appear  in  other  countries.  Nordic  ci2zens  travel  abroad  a  lot  and  bring  back  ESBL  and  other  serious  infec2ons.  

 

But  many  hospitals  are  older,  designed  before  resistant  infec2ons  became  problem.  These  have  few  single  rooms,  even  in  cri2cal  care.  

 

Study:    Conver2ng  a  cri2cal  care  unit  to  single  rooms  reduces  infec2on        

•  Study  site:    25-­‐bed  criIcal  care  unit  before  and  aber  renovaIon  to  100%  single  rooms  

• Main  findings:    

w  C.  difficile  decreased  43%    

w MRSA  decreased  47%    

w Overall  average  length  of  stay  decreased  10%    (all  pa2ents  in  intensive  care)  

(Teltsch  et  al.  2011,  Archives  of  Internal  Medicine)  

5

EBD  for  reducing  infec2ons    in  the  post-­‐an2bio2c  era    

•  100%  single  rooms  with  private  toilets  (important!)  

 

• Alcohol  hand-­‐rub  dispensers  located  near  bedside,  toilet,  other  accessible  loca2ons  

 

• Handwashing  sinks  placed  in  prominent  loca2ons  near  staff  movement  paths  

 

EBD  for  reducing  infec2ons  -­‐-­‐  con2nued  

•  Surfaces  that  facilitate  cleaning    

•  Clean  air  and  segrega2on  of  airflow  direc2on  

• Good  facili2es  for  cleaning  bedpans    

• Design  and  maintain  water  system  at  proper  temperature  and  pressure  

 

Hamilton,  K.  and  Ulrich,  R.,  2008,  in  Core  Topics  in  Cardiothoracic  CriBcal  Care.  Cambridge  University  Press.  

Single-­‐bed  vs.  Mul2-­‐bed  Pa2ent  Rooms  (Ulrich,  2004)  Single   Mul2-­‐bed  

Healthcare associated infections Medical errors Falls Staff observation of patients Staff-patient communication Confidentiality of information Presence of family Patient privacy and dignity Avoid mixed-sex accommodation Dignity for end-of-life patients Lower noise Sleep quality

6

Pain Patient stress Daylight exposure Patient satisfaction Patient choice of hospital Staff satisfaction Staff work effectiveness Reducing room transfers Adapt to handle high acuity Managing bed availability Initial construction costs Operations and whole life costs

Single-­‐bed  vs.  Mul2-­‐bed  Pa2ent  Rooms  (Ulrich,  2004)  Single   Mul2-­‐bed  

Design  for  good  visibility  of  pa2ents  from  nurse  sta2ons  to  improve  safety,  reduce  mortality  

•  Study:    664  ICU  pa2ents  admijed  to  single  rooms  that  were  either  visible  or  not  visible  from  nurse  sta2ons.    

 

•  Mortality  was  higher  for  severely  ill  ICU  pa2ents  assigned  rooms  that  nurses  could  not  see  into  from  sta2ons.  

1022

CHEST Original ResearchCRITICAL CARE MEDICINE

Original Research

Many ICUs are rife with legends of differential survival among ICU rooms. These perceptions

raise the possibility that ICU architecture may infl u-ence the clinical outcomes of ICU patients. Architec-tural design of health-care facilities can infl uence patient safety 1 ; however, there are no data that deter-mine whether patients cared for in ICU beds that are poorly visualized from a central nursing station have outcomes that differ from those admitted to rooms

with greater visibility to nursing staff and physicians. Multiple strategies have evolved to improve patient safety in ICUs, but we are unaware of any strategy that considers the impact of ICU design on patient outcomes.

In the present study, we compared clinical out-comes among patients assigned to medical ICU (MICU) rooms with unimpeded visibility of patients from a central nursing station to patients assigned to rooms with poor visibility from a central nursing sta-tion. The objective was to determine whether patient visibility correlates with mortality and/or various sec-ondary clinical outcomes.

Material and Methods

MICU Description and Physical Layout

The MICU at New York Presbyterian Hospital/Columbia Uni-versity Medical Center is an organizationally closed ICU staffed by four fi rst-year and four second-year internal medicine residents who work with a pulmonary/critical care fellow and a pulmonary/

Background: Architectural design of health-care facilities can infl uence patient safety; however, it is unknown whether patient outcomes are signifi cantly affected by ICU design. Methods: Six hundred sixty-four patients admitted to the medical ICU (MICU) of Columbia University Medical Center during 2008 were included in this retrospective study. Patient outcome measures, which included hospital mortality, ICU mortality, ICU length of stay (LOS), and ventilator-free days, were compared based on random room assignment. Rooms that were not visible from the MICU central nursing station were designated as low-visible rooms (LVRs), whereas the remaining rooms were designated as high-visible rooms (HVRs). Results: Overall hospital mortality did not differ among patients assigned to LVRs vs HVRs; however, severely ill patients (those with Acute Physiology and Chronic Health Evaluation II scores . 30) had signifi cantly higher hospital mortality when admitted to an LVR than did simi-larly ill patients admitted to an HVR (82.1% and 64.0%, n 5 39 and 75, respectively; P 5 .046). ICU mortality showed a similar pattern. ICU LOS and ventilator-free days did not differ signifi cantly between groups. Conclusions: Severely ill patients may experience higher mortality rates when assigned to ICU rooms that are poorly visualized by nursing staff and physicians. CHEST 2010; 137(5):1022–1027

Abbreviations: APACHE 5 Acute Physiology and Chronic Health Evaluation; HVR 5 high-visible room; LOS 5 length of stay; LVR 5 low-visible room; MICU 5 medical ICU

Relationship Between ICU Design and Mortality David E. Leaf , MD ; Peter Homel , PhD ; and Phillip H. Factor , DO, FCCP

Manuscript received June 26, 2009; revision accepted November 9, 2009. Affi liations: From the Department of Medicine (Dr Leaf), Columbia University College of Physicians and Surgeons; the Department of Biostatistics (Dr Homel) and the Division of Pulmonary, Critical Care, and Sleep Medicine (Dr Factor), Beth Israel Medical Center, and Albert Einstein College of Medicine (Dr Factor); New York, NY. Correspondence to: Phillip H. Factor, DO, 7 Dazian, Beth Israel Hospital, 1st Ave at 16th St, New York, NY 10003; e-mail: pfactor@chpnet.org © 2010 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( www.chestpubs.org/site/misc/reprints.xhtml ). DOI: 10.1378/chest.09-1458

© 2010 American College of Chest Physicians at St Jude Childrens Research Hospital on May 6, 2010chestjournal.chestpubs.orgDownloaded from

Single  room  designed  for  high  visual  access  

Research  summary:    Effects  of  NATURE  in  health  facili2es  

•  Nature  reduces  stress    

•  Reduces  pain    

•  Lessens  anger/aggression    

•  Increases  sa2sfac2on  

7

Cri2cal  care  unit  at  Legacy  Good  Samaritan  (1996)    

Portland,  Oregon    

Architecture:  Tom  Sagerser  Landscape  Arch:  Walker  Macy  

Single  rooms  improve  staff-­‐pa2ent  communica2on,  support  family  presence  and  privacy  

• Design  pa2ent  rooms  to  support  family  presence.  

• Most  family  of  cri2cal  care  pa2ents  are  very  stressed.    20%-­‐40%  of  family  members  of  longer  stay  pa2ents  develop  PTSD.  

 

• Provide  comfortable  wai2ng  areas  and  other  spaces  with  pleasant  distrac2ons  that  help  family  cope  with  stress.  G  

Garden  for  family  of  criIcal  care  paIents  

•  Lowers  depression  (improves  emoIonal  well-­‐being)  

 

•  Reduces  pain    

•  Increases  staff  sa2sfac2on  

 

Providing  access  to  daylight  improves  outcomes,  can  lower  costs  

8

Combining  EBD  with  changes  in  cri2cal  care  opera2ons  or  care  processes        

• Most  architectural  changes  to  healthcare  buildings  involve  many  design  factors.  

 

• Architectural  changes  can  support  or  make  possible  new  care  processes/opera2ons.  

 

­  ImplicaBon:    a  project  should  begin  by  rethinking  care  processes,  then  designing  to  make  possible  the  new  processes.  

 

“First  design  the  care  organizaIon,  then  design  the  building.”  

 

                                         D.  Kirk  Hamilton  

1.  Acuity-­‐adaptable  coronary  criIcal  care  at  Methodist  Hospital,  Indianapolis  USA  

 

2.    Care  process  +  design  changes  in  neonatal  intensive  care  unit  (NICU)  at  Karolinska  University  Hospital,  Huddinge  

IllustraIng  impacts  on  outcomes  of  ‘bundle  interven2ons’consisIng  of  integrated  clusters  of  several  care  process  +  architectural  changes  

Project  examples   Priority  goals  of  acuity-­‐adaptable  coronary  cri2cal  care  project  

•  Reduce  pa2ent  transfers  to  other  rooms  and  units  

 

•  Reduce  medica2on  errors  

•  Reduce  falls    

•  Reduce  costs    

•  Increase  pa2ent  and  family  sa2sfac2on    

(Methodist  Hospital,  Indianapolis,  1999)  

9

Acuity-­‐adaptable  coronary  cri2cal  care      (Methodist  Hospital,  Indianapolis,  1999)  

•  Opera2ons  changes:    acuity-­‐adaptable  staffing  model  +  culture  that  supports  family  presence  and  involvement.  

 

•  Design  component:    single  rooms  equipped  to  permit  care  acuity  to  flex  up  or  down  according  to  the  condiIon  of  the  paIent.  

Acuity-­‐adaptable,  single-­‐bed  coronary  cri2cal  care    Methodist  Hospital,  Indianapolis    

Design:    BAS  Life  Structures  

Family  zone  

Equipment  for  higher  acuity  

w  Transfers  reduced  90%  compared  to  old  unit  with  mulI-­‐bed  rooms  

 

w  Saves  $5  million  per  year  (2004)    

w MedicaIon  errors  reduced  70%  

Annual  medica2on  error  index  (errors/pa2ent  days)  coronary  cri2cal  care    

0

2

4

6

8

10

12

1997 1998 1999 2000 2001

Med Errors

More  errors  

Move  to  new  unit  with  single  acuity-­‐adaptable  rooms  

Source: A. Hendrich (2004). In Keeping Patients Safe: Transforming the Work Environment of Nurses. Quality Chasm Series, Institute of Medicine

Old  unit  with  mul2-­‐bed  rooms  

Decentralized  nurse  sta2ons  with  good  visibility  of  paIents      

Design:    BAS  Life  Structures  

Acuity-­‐adaptable,  single-­‐bed  coronary  cri2cal  care    Methodist  Hospital,  Indianapolis  

10

0

1

2

3

4

5

6

7

1997 1998 1999 2000 2001

Move  to  new  unit  with  single  family-­‐centered  rooms  and  decentralized  nurse  sta2ons  

More  falls  

Source:    A.  Hendrich  (2004).  In  Keeping  Pa5ents  Safe:  Transforming  the  Work  Environment  of  Nurses.  Quality  Chasm  Series,  InsItute  of  Medicine.  

Old  unit  with  mul2-­‐  bed  rooms,  centralized  nurse  sta2on  

Methodist  Hospital  study:    pa2ent  fall  index  (falls  per  100  pa2ent  days)  

source:    Hendrich,  Fay  &  Sorrells,  2004  

-­‐50%  

Nurse  ajri2on  Number  of  nurses  employed  

Has  acuity-­‐adaptable  coronary  cri2cal  care  proven  to  be  successful?        

•  Yes  and  no.    

•  The  care  model  has  been  adopted  by  several  hospitals.  In  most  cases  outcomes  have  improved.  (Fewer  paIent  transfers,  falls,  medicaIon  errors.)  

 

•  However,  achieving  acuity-­‐adaptable  staffing  has  proven  to  be  difficult,  and  this  has  limited  adopIon  of  the  care  model.      

Combining  design  and  care  process  changes  in  neonatal  intensive  care  (Karolinska)  

Goal:    implement  family-­‐centered  (couplet)  care  to  improve  outcomes    

•  Make  it  possible  for  each  family  to  be  in  separate  room  with  their  preterm  infant  beginning  shortly  aber  delivery,  and  24/7  through  discharge.  

Project  example  2:  

11

Prior  to  implemenIng  new  “couplet  care”  model,  the  Karolinska  NICU  emphasized  family-­‐centered  care  with  early  skin-­‐to-­‐skin  contact  and  bonding.    

•  But  infants  were  in  mulI-­‐incubator  rooms  with  very  li\le  space  for  family.  

 

•  Mothers  (C  secIons)  were  assigned  paIent  room  in  another  unit.  Aberward  family  stayed  in  hotel.  

Karolinska  NICU  (neonatal  intensive  care)    

   

Possible  mechanisms  by  family-­‐centered  NICU  care  may  improve  outcomes  (Lillesköld  &  Westrup,  2011)    

•  Minimized  separaIon  of  mother/father  from  infant      

•  Early  skin-­‐to-­‐skin  contact      

•  Early  bonding  and  parental  involvement    

•  PosiIve  effect  on  breasoeeding    

•  Parents  feel  more  confident  caring  for  their  child;  more  secure/confident  at  discharge  from  hospital  

 

•  SImulaIng  workplace:    challenging,  inspiring  

Parent bedroom

Incubator

Nurse station

Level 3 unit

Level  3  unit  

12

Research  slides  from:    Lillieskold,  S.  &  Westrup,  B.  (2011)      

Research  slides  from:    Lillieskold,  S.  &  Westrup,  B.  (2011)      

13

 •  Reduced  approx  25%  in  level  3    •  Reduced  12%-­‐15%  in  level  2      

Infant  Mortality