of 38
8/8/2019 Hospital Guide 2010
1/38
NAME
FIRST
MIDDLE
LAST
HOSPITAL GUIDE 2010
WHATMAKESAGOODHOSPITAL?
TEST RESULTS
Contents: TEST RESULTS
NAME:
PATIENT: NHS HospitalsBIRTH DATE: 2010DOCTOR: FOSTER
8/8/2019 Hospital Guide 2010
2/38
Contents
4 Editors letter
5 How good is my hospital?
6 Introductory message
7 Trusts of the year
10 Measuring mortality - akeysteptoensuringquality
18 Stroke - excellenceacrossacarepathway
22 Orthopaedics - excellenceamongcareteams
24 Urology - excellenceinoperations
26 Patient safety - thefoundationforquality
32 Efciency - howqualitysavesmoney
34 Patient experience - notjustamedicalmatter
36 References
37 Acknowledgements
38 About Dr Foster
38 Our methodology
8/8/2019 Hospital Guide 2010
3/38
Hospital Guide 2010 4
Itis10yearssincetherstDrFosterHospitalGuidewaspublished.Insomewaysmuchhas
changedin10years,butinotherwaysnotenoughhaschanged.Adecadeagowehaddata
onallhospitaladmissions,fromwhichwecomparedmortalityratiosandothermeasures.
Thatdataisstillourmainsourceofinformation.Wehavenoprimaryandcommunitycaredata,noprivatesectordataandnodatathatshowswhathappenstopatientsover
thewholecourseoftheirillness.Excitingly,theCoalitionGovernmentseemscommitted
tonallyaddressingthisissue,andthe2010guideisinpartacaseformoreandbetter
information.MoreonthisfromRogerTayloronpage6.
TheHospitalGuidehasalsochangedoverthepast10years,althoughsomeconstants
remain.WecontinuetopublishHospitalStandardisedMortalityRatios(HSMRs)but,in
additiontothis,havenowintroducedtwootherwaysoflookingatmortality.Youcannd
theresultsonpages16-17.
Forthe2010guidewehaveteamedupwithleadingcliniciansandanalyststoshinethe
spotlightonthreeareasimportanttomanypatients:stroke,orthopaedicsandurology
(seepages18-25).Andwehavereturnedtothethornysubjectofsafety.Thepublicity
aroundlastyearssafetyindextooksomebysurprise,butraisedawarenessoftherisks
facingpatients.Thisyearwelookbacktoseewheretherehasbeenimprovementand
whereproblemsremain.The2009guidepromptedsomechanges,includingaDepartment
ofHealthtaskforceonmeasuringmortalityandnewrulesaroundthereportingofsafety
incidents.However,westillhavesomewaytogotogetreliabledataaboutadverseevents.
Wearealsotryingoutsomenewwaysofpresentinginformationonourwebsite.Visitors
towww.drfosterhealth.co.ukcannowspecifywhichaspectsofpatientexperiencemattermosttothemandthenndoutwhichhospitaltrustsperformbestontherelevantcriteria.
Asever,thanksmustgotoallthosewhohavehelpedmakethisyearsguidecometolife,
especially the experts whose commentaries and opinions you will nd throughoutthe
report.ThankyoualsotothoseindividualsineachNHStrustwhocoordinatedactivity
aroundtheHospitalGuide,notleastinrespondingtoourannualsurvey,towhich99per
centoftrustsreturneddata.
Thechallengewesetourselvesistoproduceareportwhichisaccessibleforpatientsand
thepublicandvalidforcliniciansandmanagers.Thisguidehasbeen10yearsinthemaking
andwehopeyounditstimulatingandinformative.
Editors letter
Alex Kafetz
8/8/2019 Hospital Guide 2010
4/38
Hospital Guide 2010 5
What we can tell you:
the good news
1.Deathsinhospitalcontinuetofall,dropping7percentbetween2008/09and
2009/10incrudeterms.Seepage11.
2.ThegapbetweenthehighestandlowestHospitalStandardisedMortalityRatios(HSMRs)
hasnarrowed,witheightfewertrustsHSMRsabovetheexpectedrange.Seepage10.
3.Safetystandardshaveimproved,withhigherratesofcompliancewithsafetyalerts
andbetterreportingoferrors.Seepage27.
4.Airedaleisoursmalltrustoftheyearforaremarkablefourthtime,withverygood
performanceinclinicaloutcomes,safetyandpatientexperience.Seepage9.
5.RoyalFreeHampsteadandIpswichHospitalhavewonlargeandmediumtrustofthe
year,whileEastKentHospitalsisrecognisedforthersttimeasfoundationtrustofthe
yearwithexcellentoutcomesinarangeofclinicalareas.Seepages7-9.
What we can tell you:areas of concern
1.Variationsinmortalityratiospersist,with19trustshavinghighHSMRs.Seepage10.
2.Fourtrustshavehighratiosforthedeathsaftersurgeryindicator.Twoofthesetrusts
alsohavehighHSMRs.Seepages12-13.
3.Ratesofemergencyreadmissionsvarywidely,asdorevisionsandmanipulations
followingcommonoperations,wherethreetrustshavehighrates.Seepages22-23.
4.In2009/10over27,000potentialadverseeventswererecordedinhospitaldata.
Thisisalmostcertainlyanundercountduetoinconsistentrecording.Seepage30.
5.Standardsinthetreatmentoflife-threateningconditionssuchasstrokeandbroken
hipsvarywidely.Manytrustsfallshortofbestpractice.Seepages18-23.
What we cannot tell you
but would like to know
1.Howmanypeoplesufferpotentiallylife-threateningbloodclotsfollowingtreatment?
DespitebeingaDepartmentofHealthpriority,thisinformationisnotbeingrecorded
properly.Seepage30.
2.Thequalityofcareforpatientsafterleavinghospital.Informationaboutcommunity
andprimarycareservicesforpeoplewithlong-termconditionsisnotavailablefor
analysisinthewaythathospitaldatais.Seepage19.
3.Thelevelofmedicalerrorstakingplace.Recordingisinconsistentandtrustswithhigh
ratesofadverseeventscanoftenbebestatkeepingaccuraterecords.Seepages30-31.
4.HowNHScareforcommonproceduresinprivatehospitalscompareswiththecare
giveninNHStrusts.Seepages22-23.
5.Moredetailedinformationabouthowindividualclinicalteamstreatpatients(heldon
databasessuchastheNationalJointRegistryorCancerRegistry).Seepage23.
How good is my
hospital?
8/8/2019 Hospital Guide 2010
5/38
Hospital Guide 2010 6
Introductory message
Roger Taylor
Co-founder of Dr Foster
The aims of the rst Dr Foster Hospital Guideweresimple. We wanted to put more
informationandmorepowerintothehandsofpatientsandthepublicbygivingthem
anindependentviewonwherehealthcarewasworkingandwhereitwasnot.Wewanted
toendtheofcialmonopolyondataaboutNHSperformance.Wewantedtoseeapublicdebateaboutwhatitlooksliketohavequalityinhealthcare.
Inthedecadesincetherstguide,DrFosterhasgrownfromasmallpublishingcompany
intoabusinessthatworkswithmosthospitalsinEngland.Thedebatehasalsoshifted:
everyone now accepts that delivering good healthcare means measuring performance
andbeingtransparentaboutthatmeasurement.Thereis muchmoreinformationtoday
aboutclinicaloutcomesandaboutwhatpatientsthinkabouttheirservicesthanthere
was10yearsago.
Wehaveseenimprovementsasaresult.Thewidevariationsinhospitalmortalityratioshave
narrowedsincewerstpublishedthedata,andthereisgreaterfocusonimprovingclinical
outcomesandsafety.Inthisguideyouwillndmanyexamplesofwhereimprovementsin
qualityhavebeendrivenbybetterinformation.Buthowhavewedoneonouroriginalaims
ofgivinggreaterpowertopatientsandhavingamoreopendebateaboutquality?Thefact
is,thereisstillalongwaytogo.
Athirdofpatientsstillsaytheyarenotsufcientlyinvolvedindecisionsabouttheircare.
Asapatient,itisstilltoodifculttondoutaboutthetreatmentoptionsavailableto
you, the standards ofcareyou should expect and whether ornot the service you are
receivingmeetsthese.TryndingoutwhattoexpectfromyourGPandotherlocalservices
followingadiagnosisofdepression.Tryndingouthowthatcompareswithbestpractice.Tryndingoutwhatotherpatientsthinkabouttheseservices.Youwillnotgetfar.TheNHS
collectsvastamountsofdatabuttoolittleofitisturnedintousefulinformation.Where
informationisavailable,itisrarelyprovidedtopatientsorthepublicinwaysthathelp
themmakedecisions.
Inthisguidewehavetriedtoanswerafewquestionsabouthospitalcare:wheredoes
itappearthatstrokecareisdeliveredwell?Whichhospitalswouldwerecommendfor
treatmentofpelviccancers?Wheredohighmortalityratiosraisequestionsaboutcare?
Buttherearemanymorequestionswherewewouldliketogiveanswersbutcannot.
TheCoalitionGovernmenthasrecognisedthattheNHScannotbemanagedfromWhitehall.Itiscommittedtogreatertransparencyandgivingpatientsmoresayoverwhathappensto
them.Wewelcometheseambitionsandhopethat,beforeanother10yearsareup,wewill
beabletoaddresssomeoftheunansweredquestionsinthisguide.
8/8/2019 Hospital Guide 2010
6/38
Hospital Guide 2010 7
East Kent Hospitals University NHS Foundation Trust
Trustsof the year
Each year Dr Foster celebrates the achievements and successes
of the NHS by naming our foundation, small, medium and large trusts
of the year, as well as one overall winner. This year we have related
the awards to the Coalition Governments Outcomes Framework.1
On the next pages
See the best performing trustsin each category
8/8/2019 Hospital Guide 2010
7/38
Hospital Guide 2010 8
Preventing premature death Result Trust score National average
Thesearefourmortalityratios,
comparingtheactualnumbersof
deathswithourestimates.100isthenationalaverage.Lowerscores
aredesirable.Seepages10-17.
HospitalStandardisedMortalityRatio(HSMR) 79 100
Basketoffivestandardisedmortalityratios 81 100
Deathsaftersurgery 89 100
Deathsinlow-riskconditions 77 100
Quality of life despite long-term conditions Result Trust score National average
Infocusingonstrokewehave
selectedsixindicatorsfollowing
patientsalongahospitalpathway.
Seepages18-21.
Strokepatientsscannedonthesameornextday 54% 47%
Thrombolytictreatmentwhenappropriate 6% 3%
Pneumoniaduetoswallowingproblems 3% 5%
Dischargehomewithin56days 78% 73%
Readmissionswithin28days 114 100
In-hospitalmortality 71 100
Helping recovery from ill health or injury Result Trust score National average
Wehavemeasuredtrustsacross
orthopaedicsandurology,looking
atreadmissionsandoperations
whichneedtobedoneagain.
Seepages22-25.
Re-doratesfortransurethralresectionoftheprostate 6% 5%
Kneerevisionsandmanipulationswithinoneyear 0.04% 1%
Hiprevisionsandmanipulationswithinoneyear 1% 5%
Hipreplacementreadmissions 118 100
Kneereplacementreadmissions 124 100
Hipfractureoperationswithintwodays 71% 67%
Hipfracturestandardisedmortalityratio 89 100
Positive experiences of care Result Trust score National average
Alltrustsarefocusingonthese
vequestionsfromthenational
patientsurveyandtheycan
receivenancialrewardsfor
performingwell.Seepages34-35.
Sufficientlyinvolvedincaredecisions? 68% 70%
Staffavailabletotalktoaboutworries? 57% 59%
Enoughprivacywhendiscussingcare? 80% 81%
Medicationside-effectsexplainedpre-discharge? 49% 45%
Givenacontactforpost-dischargeconcerns? 74% 74%
Safe environment and avoiding harm Result
Wehaverevisitedanumberof
measuresofpatientsafetythat
werehighlightedinlastyears
HospitalGuide.Mostofthe
informationisfromoursurvey.
Seepages26-31.
Trusthasaboardleadforpatientsafety? 4
Patientsafetyisonboardsmonthlyagenda? 4
Inpatientswithtrackandtriggersystemsinplace? 100%
Trustcomplieswithselectedsafetyalerts? 4
Allsurgicalpatientsgivenclot-preventiondevices? 4
Patientsrisk-assessedforbloodclotsonadmission? 31-60%
Reportedrateofsafetyevents?
East Kent Hospitals University NHS Foundation Trust
Scorecards for all trusts are available at www.drfosterhealth.co.uk/hospital-guide
Key
Exceedsexpectation
Meetsexpectation4Yes
8/8/2019 Hospital Guide 2010
8/38
Hospital Guide 2010 9
Our2010overalltrustoftheyearisEastKent
Hospitals University NHS Foundation Trust,
which has shown dedication to putting the
patient rst. In fact, across all the winners
thereisacommonthemeofdeliveringservices
closertothepeoplewhoneedthem.Thistype
of reorganisation has been proven to benet
patientsand,inthesetougheconomictimes,
savemoney.
On the previous page,the scorecardfor East
Kent gives anoverviewof thecategories and
datathatweusedtodetermineourtoptrusts.
These measures are explored in more detail
throughouttheguide.ScorecardsforallEnglish
acutetrustscanbefoundonourwebsite.
Foundation trust of the year
and overall winner
East Kent Hospitals University
NHS Foundation Trust
Asdemonstratedbyitsscorecard,EastKent
HospitalsUniversityNHSFoundationTrusthas
becomeourwinningtrustthisyearbyperforming
consistently well across our chosen criteria.
Indoingso,ithassucceededinmeetingthe
particularchallengesofbeingoneofthelargest
trustsinthecountryanetworkofthreedistrict
generalhospitals,twocommunityhospitalsand
several satellite sites serving a population of
750,000people.
ItsmottoofPuttingpatientsrsthasbeena
driving force behind recent changes to make
keyspecialistservicesavailablelocallyandina
timelymanner.Newtechnologyisalsohelping
toimprovecare,notleastinthehyper-strokeservicewheretheinnovative,award-winninguse
oftelemedicinehasenabledthedevelopment
ofa24/7service.Thetrustoffersaroutine
radiologyservicesevendaysaweek,ensuring
thatallpatientsreceivethescansatthetime
theyneedthem,andrecentlybecamethelead
trust for delivering primary angioplasty for
heartattackpatientsacrossthewholeofKent.
In commenting on the award, chief executive
StuartBainexplainedthatthesuccessofthe
trustcanbeputdowntoacombinationofvery
highambitionbytheboard,goodplanning,and
the dedication of the 7,000 staff who always
gotheextramilefor theirpatients.Headded,
Safety and effectiveness have been the key
drivers in directing our investment in service
changeandthisisreectedinourexceptionally
lowHSMRandgoodinfectionpreventionrates.
Large trust of the year
Royal Free Hampstead NHS Trust
RoyalFreeHampsteadNHSTrustisknownfor
itspioneeringsurgery,beingtherstinEurope
toofferkeyholemastectomyandoneofthefew
centresofferingkeyholesurgeryforpancreatic
cancer.Itevenhasethicalapprovaltoperform
whatwouldbethecountrysrstfacetransplant.
Butdespitethecomplexityofitscaseload,itsmortality ratios have been among the lowest
inthecountry for manyyears, and this high-
qualityperformanceisrecognisedinbecoming
DrFosterslargetrustoftheyear.
Thisachievementisdowntotheeffortsofour
workforcewhoaimtoofferthebestclinicalcare,
thebestpatientexperienceand whopioneer
newandeffectivehealthcareapproaches,said
DavidSloman,thetrustschiefexecutive.
Reectingitseffortstoputpatientsatthecentre
ofeverythingitdoes,thetrustisparticipating
in a programme to improve staff-to-patient
interactions,aswellasprovidingcarecloserto
patients homesthrough its network of Royal
Freeclinics.Inanotherinnovativemove,ithas
been the rst acutetrustto appoint a public
healthlead.
Medium trust of the year
Ipswich Hospital NHS Trust
IpswichHospitalNHSTrustisavibrantgeneral
hospital that provides a range of services on
site,many of themspecialised. These include
vascular surgery, spinal surgery, radiotherapy
and gynaecological cancer surgery, which are
capable ofbeingprovided toa population of
morethan500,000.Italsooffersmidwiferyin
thecommunity,andindeedisworkingclosely
withlocalGPstofurtherincreasetherangeof
communityservices.
Inadditiontobeingoneof10hospitalsintheUK
leadinganationalprogrammeofimprovements
fororthopaedicpatients,thetrustpridesitself
on having low rates of healthcare-acquired
infectionsandhasbeenrapidlyreducingrates
ofavoidableharmsuchasfrompatientfallsand
pressuresores.
ChiefexecutiveAndrewReedsaid,Iamvery
proudthat Ipswich Hospital has beennamed
medium-sizedtrustoftheyear.Everyoneatthe
hospitalplaysapartinthequalityofcarewe
provide,but itis aparticular accoladeto our
doctors, nurses and all clinical professionals
whohavemaintainedanunrelentingfocuson
thesafetyandexperienceofourpatients.
Small trust of the year
Airedale NHS Trust
Supportingmorethan200,000peopleacross
Yorkshire andLancashire, Airedale NHSTrust
has been our best performing small trust on
severalpreviousoccasions.
WeareextremelypleasedtowintheDrFoster
awardagainforthefourthtimeinveyears,
especiallyasitisinthesameyearthatwe
achieved foundation trust status, explained
BridgetFletcher,formerdirectorofnursingand
nowchiefexecutive.2
It isnot only a fantastic achievementfor our
staffbutalsorecognitionforthehardworkthey
doeverydaytomakesureweprovidehigh-
quality,safe,compassionate,personalcarefor
allourpatients.
The trust provides services from the main
hospitalsiteandalsofromcommunityhospitalsandhealthcentres,aswellastoanumber
of prisons throughout England through its
pioneeringtelemedicineservice.Overthenext
12monthsitisplanningtoinstalltelemedicine
equipmentintosomepatientshomes,nursing
homes,GPpracticesandotherremotelocations.
Theaimistoprovide integratedhealthcare to
patientsintheirownhomesorasclosetohome
aspossible.
To categorise trusts,we used the Healthcare
Commissionsdenitions(basedonthenumber
of beds perhospital). Foundation trusts are
automaticallyinthatcategory.
Full results are available at www.drfosterhealth.co.uk/hospital-guideTRUSTS OF THE EAR
8/8/2019 Hospital Guide 2010
9/38
Hospital Guide 2010 10
10 of the 19 are new additions
102 meet expectation
26 exceed expectation
2 of the 19 have been high for 6 years
19 are below expectation
ofa
ll147
trusts
Trust mortality
An overview of HSMRs in 2009/10.
Measuring
mortalitya key step to ensuring qualityHSMRs are decreasing
across the NHS. Only 19 of the
147 hospital trusts now have
signicantly high HSMRs,
compared with 27 last year,
whereas 26 trusts have HSMRs
that are signicantly low,
down from 32 a year ago.
The overall improvement
suggests greater consistency
across trusts, both in terms of
data-recording and perhaps inthe quality of care.
8/8/2019 Hospital Guide 2010
10/38
Hospital Guide 2010 11
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Preventingpeoplefromdyingduetoillnessor
injuryis,perhaps,themostfundamentalaimof
healthcare.Sometimesitisnotpossible.Butby
alwaysdeliveringthebestcare,itispossibletoreducethechancesofdeath.Lowermortality
ratiosareonemarkerofgoodqualitycare.
Thisguide compares themortality ratios at
Englishtrustsaftertakingintoconsiderationthe
differencesinthepatientstreated.Wedothis
tosee ifthe number ofpatients who survive
followingtreatmentisinlinewiththenumber
wewouldexpect,giventheircondition.
In this section we look at threemeasures,
comparing thenumber ofdeaths atthe trust
withthenumberweestimatewouldhappenif
mortalityratioswereinlinewiththenational
average. This takes into account a patients
diagnosis, age,admission method and other
characteristics.Ifatrusthasthesamenumberof
deathsasestimated,wegiveascoreof100.Ifit
has10percentmoredeaths,wegiveascoreof
110,orfor10percentfewerdeathsascoreof90.
TheHSMRisoneofthemostcommonlyused
measuresofoverallmortalityfortrusts.Itlooks
atthoseconditionswhichaccountforthevast
majorityofdeathsinhospital(80percent).
This yearthere continues tobe a variation
in mortality ratios,rangingfrom 18 percent
higherthanexpectedto28percentlowerthan
expected.Justbychancethereisboundtobe
somedisparity. Weidentifytrusts as having
highorlowHSMRsifthevariationisextreme;
inotherwords,thelikelihoodofitoccurringby
chanceislessthanoneinathousand.
Whenatrusthasahighmortalityratio,we
cannot be sure of the reason why; itmaybebecauseofinaccuratedataoraresultof
particularly unusual circumstances at that
trust.However,itisausefulscreeningtoolthat
warrantsinvestigation,andwebelievethatthe
publicshouldbemadeawareofit.Thisyears
HSMRresultscanbeseenonpage16.
Nosinglemeasurecan tell thewholestory,
soitisimportanttolookatmortalityandthe
outcomesoftreatmentinmanydifferentways.
MORTAIT
1 Hospital StandardisedMortality Ratios (HSMRs)
Hospital Standardised Mortality Ratios (HSMRs), or death rates as they
have become known in the media, are an important outcome measure for
patients. Imperial College London developed HSMRs in the 1990s.
However, it was the 2001 Bristol Inquiry report, which recommended
transparency of hospital data for patients, that acted as a catalyst for DrFoster to rst publish them.1 Dr Foster was responding to the lack of clinical
information available to the public during the Bristol Inquiry.
In April 2010 the Department of Health stated, A high HSMR is a trigger
to ask hard questions. Good hospitals monitor their HSMRs actively and
seek to understand where performance may be falling short, and action
should not stop until the clinical leaders and the board at the hospital
are satised that the issues have been effectively dealt with.2 I strongly
support this position and welcome the continued publication of HSMRs by
Dr Foster and NHS Choices (www.nhs.uk).
Putting the data into the public domain is an essential way to focusclinicians and managers on investigating outcomes. High HSMRs, together
with concerns in other measures of mortality, prompted the Healthcare
Commissions investigation at Mid Staffordshire NHS Foundation Trust.3
It is possible that, without the alarm being raised, the problems that were
found could have continued unrecognised by the system.4
Professor Sir Brian Jarman is emeritus professor at Imperial College London
Why the
HSMR resultsmatter
Prof Sir Brian Jarman
E
P
E
R
T
O
P
I
N
I
ON
8/8/2019 Hospital Guide 2010
11/38
Hospital Guide 2010 12
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Youwillseearangeofthesethroughoutthe
HospitalGuide.Withinthischapter,inaddition
toHSMRs,wealsolookatmortalityintwo
otherways:mortalityinhigh-riskconditionsanddeathsaftersurgery.
Itisimportanttolookatmortalityforspecic
conditions(seealsopages19and23),especially
those where treatment can have the biggest
impact.HerewehaveselectedfromtheHSMR
abasketofveconditionswhichaffectalarge
number of people and where mortality is
acceptedasanindicatorofthequalityofcare
inhospital.Bycreatingabasketofconditions,
we can identify variations in mortality more
effectively than if we only look at diagnoses
individually. Restricting thebasketmeanswe
canbe more focusedinmeasurementthan
withtheHSMR.Theconditionsinourbasket
areheartattack,pneumonia,stroke,congestive
heartfailureandbrokenhips.
Theresultscanbeseenonpage17.Theyincludesometrustswhoseoverallmortalityratiosarelow
orasexpected,butwhichnonethelesshaveareas
ofhighmortality.Individualtrustsmaydecide
tomonitortheirperformanceusingdifferent
basketsrelevanttothecaretheyprovide.
Conventionallyknownasfailuretorescue,this
isthersttimethisindicatorhasbeenusedin
theUK,thoughitiscurrentlypublishedinthe
US.5ImperialCollegeLondonandKingsCollege
LondonhavedevelopeditalongsideDrFoster.
MORTAIT
DID OU KNOW? Eight hospital trusts do not have a policy
to notify GPs on the death of a patient.
Sadly, some deaths in hospital are inevitable. Much of the difference in
mortality ratios between hospitals has little to do with differences in the
quality of care that people receive. Instead it is related to the sort of people
who are treated and how vulnerable they are. Measures like the HSMR try
to account for this using statistical techniques, but no statistical adjustment
can ever be perfect.
The ratio of avoidable deaths among surgical patients with treatable
complications gives another way of exploring how a hospital performs,
one which relates to a specic group of people and which offers some
advantages. For people undergoing surgery, the chance of developing a
complication such as bleeding or pneumonia depends very much on their
age, underlying conditions and other factors. But while complications are
often a result of patient characteristics, a hospitals ability to successfully
treat it is strongly related to the quality of care provided. Staff must be
vigilant and act promptly to ensure the right treatment is given.
By looking at the ratio of death only among those people who experience
complications, this indicator allows for the fact that some hospitals will treat
more people who are at risk of complications than others. The indicator is
intended to show how well they perform once the complications occur.
Hospitals performing poorly on this indicator should consider whether they
have proper systems in place for identifying and responding to patients who
deteriorate after their operation.
Professor Peter Grifths is director of the National Nursing Research Unit
at Kings College London
Why look atdeaths after surgery?
Prof Peter Grifths
E
P
E
R
T
O
P
I
N
I
ON
2 Mortality in high-riskconditions
3 Deathsafter surgery
8/8/2019 Hospital Guide 2010
12/38
Hospital Guide 2010 13
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Theindicatorlooksatsurgicalpatientswhohad
asecondarydiagnosissuchasinternalbleeding,
pneumoniaorabloodclot,andsubsequently
died. Eitherthe patients hadthiscondition
alreadyortheydevelopeditasaconsequence
ofthesurgery.Intheformercase,operating
onapatientinthesecircumstancesmayhave
increasedtheriskofdeath.
Deathamongthisparticulargroupofpatients
will sometimes be inevitable. But trusts with
highratiosshouldmakesurethatappropriate
proceduresareinplacetominimisetheriskof
deathfollowingsurgery.
Acrossthe147acutehospitaltrustsinEngland,
fourhavesignicantlyhighratiosfordeathsaftersurgery,twoareperformingsignicantly
better, and the rest are performing within
theestimatedrange.However,thereislarge
variationinperformance,withratiosfrom26
to179.AswithHSMRs,theresultsmaybe
affectedbytheaccuracyoftheunderlyingdata.
Againtheresultsaredisplayedonpage17.
This measure uses a very different approach
fromtheHSMR,sotruststhathavehighratios
onbothofthemeasuresUniversityHospitals
Birmingham NHS Foundation Trustand Hull
andEastYorkshireHospitalsNHSTrust will
wanttounderstandthepossiblecauses.
Measurementsofhospitalmortalityratiosare
only as good as the data they are based on.
Hospitaltrustsarerequiredtodocumentin
detail the care they provide topatients usingdenedsystemsofcoding.Onthewholethis
codingisrobust,buttherecanbesomevariation
betweentrusts.
Inrecentyears,becauseofthecontinuedfocus
onmortalityratios,sometrustshavereviewed
thewaytheycodepatientsandhaveincreased
thenumberidentiedasbeinginpalliativecare.
Ifthesepatientsdie,theyhaverelativelylittle
effecton the trusts HSMR,becausedeath is
theexpectedoutcome.
Afocusonmoreaccuratecodingiswelcome.
However,wehaveseengreaterinconsistency
inthewaythattrustsarecodingpalliativecare
The safety of a patients stay in hospital depends on the level of accuracy invested in
the monitoring, recording, measuring and decision-making around crucial changes
in their vital signs. Any delay in picking up patient deterioration can have an obvious
and tragic human cost. Also, it often means a return to critical care, which incurscosts and prevents trusts from using resources efciently.
Trusts need to know their standardised mortality ratios so that crucial work can be
done to improve them. The predominantly nurse-led critical care outreach teams
have been integral to improvements in reducing complications, speeding recovery
and enabling a quicker discharge for patients. This must not detract, however, from
the need for wards to have correct stafng levels and accurate skill-mixes; we must
ensure the best care at all times.
The Royal College of Nursings training packages and nursing practice principles
are giving staff the expert help they need to guide their actions before patients get
worse. The obvious key to success is empowering nurses to work closely with otherhealthcare professionals to get basic care right.
Dr Peter Carter is chief executive and general secretary of the Royal College of Nursing
A view from the front lineby the Royal College of Nursing
MORTAIT
4 Is the dataaccurate?
Deaths after surgery
Anoverviewoftrustsresultsfor2009/10.
4high
2low
141asexpected
all147trusts
8/8/2019 Hospital Guide 2010
13/38
Hospital Guide 2010 14
Full results are available at www.drfosterhealth.co.uk/hospital-guideMORTAIT
DID OU KNOW? All hospital trusts now use the World Health
Organisations Safer Surgery Checklist.
patientsinrecentyears.Sometrustsaremore
likelythanotherstocodeparticular patientsas
palliativecareand,asaresult,theirHSMRsare
lowerthantheywouldbeifalldatawascodedinthesameway.
OnaveragethiscanreduceatrustsHSMR
byuptoaboutvepoints.Therefore,inthe
interests of transparency, we have started
topublishthepercentagesofdeathsateach
trustwhicharecodedaspalliativecare.These
rangefromlessthan1percentinsometrusts
toover 40per cent inothers, with 45.5 per
centatBasingstokeandNorthHampshireNHS
FoundationTrustand44.5percentatMedway
NHSFoundationTrust .
Transparency around outcomes depends on
hospitals coding information accurately and
consistently.Withincreasingfocusonaccurate
measurementofoutcomes,itisessentialthat
clear guidelines are issued on howpatients
shouldbeclassied,andthathighstandardsof
data-recordingaremaintained.
The real value ofmeasuring mortality is inprompting trusts to take practicalactions that
helpdeliverbettercare.Herearesomeexamples.
Forming a successful action plan
TamesideHospitalNHSFoundationTrust
Tamesides HSMR has improved from high
toasexpectedin2010.Accordingtomedical
director Tariq Mahmood, this is theresult of
the development and systematic application
of a detailed mortality action plan with the
full support of the trust chief executive, the
endorsementoftheboardandtheinvolvement
ofclinicalandmanagerialstaff.
Thetrusthasbeenfocusingoncontinuingto
improveclinicalcare,suchasbysignicantly
increasingthenumberofitsconsultantsand
nursingstaff,andbyenhancingthecriticalcare
outreachteamandintensivetherapyunit(ITU).
Ithasalsobeenimplementingcarebundlesfor
certainhigh-riskconditions.
Goodclinicaldocumentationunderpinseffective
clinicalcareandappropriatecoding,andthereby
affectstheHSMR.Tofacilitatethisthetrusthas
undertakenaperiodofeducationandtraining
forbothitsclinicalandcodingstaff.Inaddition,
itconductsregularreviewsofallunexpected
deathsinhospitalinordertoidentifyanyissues
withcaremanagementordocumentation.
Qualityend-of-lifecareisimportantforensuring
adignieddeathinanappropriatesetting,soit
isvitaltoincreaseawarenessofthesubjectand
haveanintegratedapproachacrossprimaryand
secondarycare.Thisalsohelpstoreducethe
5 How are trustsresponding?
Patients need meaningful information, delivered in everyday language, so that they
can make an informed choice of hospitals and services. However, it is essential that
the information is accessible, readable and clear. We hear from patients phoning
our helpline about how confused they are by the medical terminology used when
delivering information, or that the information is in a format, such as online, that they
nd hard to access. It is not an informed choice if patients are unable to engage or
access the information presented to them.
Although mortality ratios are an important measure for patients when comparing
hospital services, they do not tell the whole story of quality of care in a hospital and
do not apply to large areas of care. To get to the heart of hospital care, patients
need other information: infection rates and the staff-to-patient ratio for a ward or
department, and the performance and outcomes for consultants and their teams.
The information revolution is a key part of the proposed changes to the NHS
outlined in the White Paper, but it is essential that this data is meaningful and
truly representative of the quality of care in hospital wards and departments. The
information needs to be easily accessible through a range of media, consistent
between hospitals, up to date and explained in plain English, avoiding the need for
complex statistics that leave patients confused.
Katherine Murphy is chief executive of The Patients Association
What else dont we know?by The Patients Association
8/8/2019 Hospital Guide 2010
14/38
Hospital Guide 2010 15
Full results are available at www.drfosterhealth.co.uk/hospital-guide
HSMRasmorepeoplearesupportedindying
outsidehospital,ifthatistheirchoice.
Thetrust hasbeenvisitingother organisations
withlowHSMRsandisalsoamemberofthe
NorthWestReducingMortalityCollaborative
(seebelow),whichhelpstounderstandand
addressthereasonsforparticularlyhighHSMRs
acrosstheregion.
Regional cooperation to improve care
NorthWestReducingMortalityCollaborative
Ninetrustsinthenorth-westhavebeenworking
togethercloselyoverthepastyeartobringabout
improvementsinpatientmortalityratios.Thisisallpartofawiderstrategytoraisestandardsof
carethroughoutthearea.
FacilitatedbytheAdvancingQualityAlliance
(AQuA), this breakthrough collaborative is
onlypartwaythroughitsrstyear,butallthe
trustshavealreadymadeimprovements.Each
onehasmadeacommitmenttoseeafallinits
HSMRofatleast10pointsovera12-month
period.Some,suchasTamesideHospitalNHS
Foundation Trust and Royal Bolton Hospital
NHSFoundationTrust,havesignicantlyreduced
theirratiosfromahighstartingpoint.
Thecollaborativeinvolvesa series of learning
events,bothforfront-lineteamsandhospital
executives. In between the eventsthey test
out ideas for improvementsand measure the
impactofthechangestheyaremaking.Sixsets
ofinterventionsarebeingpursued:
Reducingharm,suchasbytacklinghealthcare-
acquiredinfectionsormedicationerrors.Usingcarebundlestoensurethateverypatient
hasthemosteffectivecare,everytime.
Improvingthecareofdeterioratingpatientsby
spottingwarningsignsearlyandactingquickly.
Improvingend-of-lifecaretogivepatientsand
familiesmorechoiceandcontrol.
Ensuringeffectiveleadershipandmanagement
fromboardsthroughtofront-linestaff.
Tacklingallarisingissuestodowithcodingand
dataanalysis.
AQuA,whichissupportingtheinitiative,isa
membership organisation funded by primary
caretrustsandacutetrustsinthenorth-westto
improvequalityandspreadbestpractice.
Improving HSMRs through cultural changeSouthLondonHealthcareNHSTrust
SouthLondonHealthcareNHSTrustwascreated
inApril2009fromthemergerofthreehospitals.
Itiscurrentlyintheprocessofmajorcultural
changes:mergingdepartments,reconguring
services,re-engineeringpatientpathwaysand
introducingmodernworkingpractices.
Thetrusthasrecentlyintroducedthefollowing,
whichitbelieveswillhaveapositiveimpacton
itsHSMR:
Board-levelfocusonsafetyasthetrustgoes
throughthisperiodofchange.
The introduction of new models of care,
specicallytheintroductionanddevelopmentoftheacutemedicalunitmodelofcare.
Newlydevelopedprocesses,ledbythemedical
director, for regular analysis of the Dr Foster
data, internal investigation of any alerts, and
review of all low-risk deaths. These reviews
bring front-lineclinicians into the process,
enhancingpartnershipwithcoders.
Insistence that thecoding of deaths isonly
undertakenbythemostseniorclinicalcoders,
as well as introducing an internal quality
assuranceprocess for thecoding of deaths
priortosubmission.Asaresult,theproportion
oflow-riskdeathswhichonanalysisrequire
re-codinghasdecreasedfrom45to14percent
onthemostrecentaudit.
MORTAIT
Why measuring
mortalityis important
Avoiding unnecessary deaths is an important objective for health services in
all countries. People should not die early where medical intervention could
make a difference. As far back as 1863, Florence Nightingale recognised
that uniform hospital statistics would enable us to ascertain the relative
mortality of different hospitals.
Initial interest in standardised mortality data was muted, but some in the NHSrecognised that HSMRs could help trusts to identify where improvements
needed to be made. The recent association of persistently high HSMRs with
shockingly poor clinical care in a few trusts has focused fresh interest on
the use of mortality statistics in local accountability arrangements.
Important caveats need to be made. In particular, there is no gold standard
or single indicator which can be deemed as having most power in discerning
good or poor quality care. As with most indicators, its use for all audiences
is subject to caution. It is not appropriate to use HSMR data for league
tables of hospitals. Any inferences drawn from HSMR data should be
corroborated (or investigated) by other comparative measures before
conclusions are drawn about the quality of care.
Further development and understanding of the use of standardised mortality
statistics across the NHS will bring greater quality and consistency in terms
of the way we monitor mortality associated with hospitalisation. This will be
of benet to the public at large.
Dr Robert Winter is medical director at NHS East of England
Dr Robert Winter
E
P
E
R
T
O
P
I
N
I
O
N
8/8/2019 Hospital Guide 2010
15/38
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Hospital Standardised Mortality Ratios (HSMRs) Thisisabroadmeasureacrossthemajorityofactivityinahospitalwhereriskofdeathis
signicant.Becauseitcoverssomuchactivity,itisanexcellentscreeningtoolforidentifyingwheretheremaybeproblemswithavoidablemortality.
ower than expected mortality Ratio Higher than expected mortality Ratio
AintreeUniversityHospitalsNHSFoundationTrust 85 Barking,HaveringandRedbridgeUniversityHospitalsNHSTrust 116
AshfordandStPetersHospitalsNHSTrust 90 BuckinghamshireHospitalsNHSTrust 118
BarnetandChaseFarmHospitalsNHSTrust 88 CityHospitalsSunderlandNHSFoundationTrust* 1 14
BartsandTheLondonNHSTrust 89 DerbyHospitalsNHSFoundationTrust 112
BradfordTeachingHospitalsNHSFoundationTrust 81 EastSussexHospitalsNHSTrust* 110
CambridgeUniversityHospitalsNHSFoundationTrust 81 GeorgeEliotHospitalNHSTrust* 113
EastKentHospitalsUniversityNHSFoundationTrust 79 HullandEastYorkshireHospitalsNHSTrust 117
EpsomandStHelierUniversityHospitalsNHSTrust 90 IsleofWightNHSPCT* 115
FrimleyParkHospitalNHSFoundationTrust 85 MidCheshireHospitalsNHSFoundationTrust 1 14
ImperialCollegeHealthcareNHSTrust 80 NorthamptonGeneralHospitalNHSTrust* 112
LeedsTeachingHospitalsNHSTrust 91 PennineAcuteHospitalsNHSTrust 110
MaidstoneandTunbridgeWellsNHSTrust 92 RoyalBoltonHospitalNHSFoundationTrust 116
MidStaffordshireNHSFoundationTrust 87 ShrewsburyandTelfordHospitalNHSTrust* 117
NorthBristolNHSTrust 90 SouthLondonHealthcareNHSTrust* 109
NorthWestLondonHospitalsNHSTrust 87 SouthportandOrmskirkHospitalNHSTrust* 113
PlymouthHospitalsNHSTrust 86 TheDudleyGroupofHospitalsNHSFoundationTrust 115
RoyalFreeHampsteadNHSTrust 72 TheRoyalWolverhamptonHospitalsNHSTrust* 116
SalfordRoyalNHSFoundationTrust 84 UniversityHospitalsBirminghamNHSFoundationTrust 109
ShefeldTeachingHospitalsNHSFoundationTrust 92 WesternSussexHospitalsNHSTrust* 107
StGeorgesHealthcareNHSTrust 84
TauntonandSomersetNHSFoundationTrust 89
TheNewcastleuponTyneHospitalsNHSFoundationTrust 90
TheWhittingtonHospitalNHSTrust 84
UniversityCollegeLondonHospitalsNHSFoundationTrust 72
UniversityHospitalsBristolNHSFoundationTrust 86
WestMiddlesexUniversityHospitalNHSTrust 86
Results for the three mortality indicators
*DenotestrustswhichdidnothavehighHSMRslastyear
DenotestrustswithhighHSMRsforthepastsixyears
MORTAIT
Hospital Guide 2010 16
8/8/2019 Hospital Guide 2010
16/38
Hospital Guide 2010 17
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Standardised mortality ratiosThisbasketcontainsveofthe56conditionsthatcomprisetheHSMR:heartattacks,stroke,pneumonia,
congestiveheartfailureandbrokenhips.
ower than expected mortality Ratio Higher than expected mortality Ratio
AshfordandStPetersHospitalsNHSTrust 83 CountyDurhamandDarlingtonNHSFoundationTrust 113
BradfordTeachingHospitalsNHSFoundationTrust 87 DerbyHospitalsNHSFoundationTrust 115
EastKentHospitalsUniversityNHSFoundationTrust 81 EastandNorthHertfordshireNHSTrust 118
FrimleyParkHospitalNHSFoundationTrust 84 GreatWesternHospitalsNHSFoundationTrust 117
ImperialCollegeHealthcareNHSTrust 83 HullandEastYorkshireHospitalsNHSTrust 115
MidStaffordshireNHSFoundationTrust 74 RoyalBoltonHospitalNHSFoundationTrust 118
NorthWestLondonHospitalsNHSTrust 88 ShrewsburyandTelfordHospitalNHSTrust 117
PlymouthHospitalsNHSTrust 87 SouthLondonHealthcareNHSTrust 112
RoyalFreeHampsteadNHSTrust 79 SurreyandSussexHealthcareNHSTrust 121
StGeorgesHealthcareNHSTrust 87 TheRoyalWolverhamptonHospitalsNHSTrust 121
UniversityCollegeLondonHospitalsNHSFoundationTrust 73
UniversityHospitalsBristolNHSFoundationTrust 84
WestMiddlesexUniversityHospitalNHSTrust 78
Deaths after surgery Thisindicatorlooksatunexpecteddeathsamongsurgicalpatients.
ower than expected mortality Ratio Higher than expected mortality Ratio
ChelseaandWestminsterHospitalNHSFoundationTrust 26 HullandEastYorkshireHospitalsNHSTrust 166
WinchesterandEastleighHealthcareNHSTrust 46 TheNewcastleuponTyneHospitalsNHSFoundationTrust 137
UniversityHospitalsBirminghamNHSFoundationTrust 157
UniversityHospitalofNorthStaffordshireNHSTrust 153
Hospital Guide 2010 17
Look back at pages 11-13
Why did we choosethese indicators?
MORTAIT
Results for all trusts are available at www.drfosterhealth.co.uk/hospital-guide
Source:SecondaryUsesService(SUS)data2009/10
8/8/2019 Hospital Guide 2010
17/38
Hospital Guide 2010 18
Strokeisahighnationalprioritybutthereisa
markedvariationinstandardsofcare.Thiswas
highlightedbytheNationalAuditOfce(NAO)
in2005andrecentlyin2010.1
AstheUKsthirdbiggestkiller,stroketakes
thelivesofmorewomenthanbreastcancer.
There are 110,000 strokes in England each
year,andalmostathirdofeveryonewhohas
onediesfromit.2Thosewhosurviveoftenhave
permanentdisabilities.TheNAOestimatedthat
thedirectcostofcaringforpeoplewhohavea
strokeis3bnayearandthewidereconomic
costsare8bn.3
Under pressure to improve
TheDepartmentofHealths2007StrokeStrategy
recognised thescaleof theproblemand set
outtomakeimprovementsapriority. 4Butwhat
impacthasthestrategyhadintermsofhelping
hospitalstoadheretobestpracticeandimprove
patientoutcomes?
Inthis years HospitalGuidethe DrFoster
teamhasfocusedonkeyindicatorsofquality
and outcomes that stretch across the stroke
carepathway,inotherwordsacrossthemany
differentstagesoftreatmentandcareforthis
particularcondition.
Strokeexcellence across a care pathwayTo understand quality, you need to measure the aspects of care that
patients are most concerned about. This often boils down to looking
at the detail around individual conditions. Here we focus on the care
that patients receive when they have a stroke.
The key to providing high-quality stroke care is making sure that everyone who
has a stroke is admitted directly to a stroke unit and spends all of their time there.
The evidence is strong that these units, staffed with a multi-disciplinary team of stroke
specialists, improve outcomes and reduce stroke mortality.
Recent years have seen a dramatic improvement in the number of stroke units, the
number of patients treated there and the length of time they stay. Stroke patients
and their families will want to assess these different aspects of care, as well as the
quality of specic units.
The Sentinel Stroke Audit, which is carried out every two years by the Royal College
of Physicians, provides a wealth of data about stroke unit provision. New, real-time
measurement of the hyper-acute phase of stroke care is also coming on stream with
the introduction of the Stroke Improvement National Audit Programme. Both of
these are important measures for clinicians and commissioners, but they are also
vital tools for helping patients to assess their care.
Just as vital is data about post-hospital stroke provision, and this is much thinner
on the ground. The Care Quality Commission (CQC) is carrying out a one-off review
of post-hospital stroke provision to be published later this year, but in future we
will need to see more systematic measurement of the quality and quantity of stroke
services in the community, building on the baseline provided by the CQC.
Joe Korner is director of communications at The Stroke Association
The information we needby The Stroke Association
8/8/2019 Hospital Guide 2010
18/38
Hospital Guide 2010 19
Full results are available at www.drfosterhealth.co.uk/hospital-guideSTROKE
Taking action early
Ofthesixperformanceindicators,thersttwo
measuredinterventionsthatshouldtakeplacein
thecriticalperiodofcarestraightafterastroke:
Theproportionofpatientsreceivingabrain
scanonthesameornextday.
Theproportionofpatientsgiventhrombolytic
orclotbustingdrugswithin24hours.
Receiving a brain scan promptly is thebest
way tocorrectly diagnosethe nature ofthe
stroke.Withoutthis information itmaynot
bepossibletostartappropriatetreatmentas
soonasrequired.In2009/10thehighestrateofpatientshavingabrainscanbythenextday
was87percentatNorthMiddlesexUniversity
HospitalNHSTrust.Elsewheretherateswere
aslowas42percent.
Thrombolysiscanmakea bigdifferenceto the
patientsrecoverybutthisrequiresskilledteams
onsite.Notalltrustsaresetuptoprovidethis
care.Insomeareas,networksarebeingsetup
sothat,ifahospitalcannotprovidetreatment,
patientsaretransferredquicklytonearbyunits
whichcan.Thrombolysisratesvariedfrom0. 2
to17percent.Youcanseeallthescanningand
thrombolysisresultsonourwebsite.
Quality care from start to nish
The following four indicators were chosen to
helpdemonstratethequalityofoutcomes:
Theproportionofstrokeadmissionsthatlead
topneumoniaduetoswallowingproblems,
whichshouldnothappenifcareteamshavecarriedoutastandardcheck.Ratesvariedfrom
2to12percent.
Theproportionofpatientsreturningtotheir
usual place of residence following hospital
treatmentwithinaperiodof56days,which
impliessuccessfulrehabilitation.Ratesvaried
from55to85percent.
Therateofemergencyreadmissionstohospital
aftertreatmentforastroke,whichhighlights
returnvisitsthatcouldpossiblybeavoided.
Ratesvariedfrom44percentbelowaverage
to58percentaboveaverage.
Thestandardisedmortalityratio,whichcan
highlightpreventabledeaths.Ratesvariedfrom
34belowaverageto66aboveaverage.
No room for complacency
Tobuildapictureofoverallperformance,wehave identied trusts that have performed
signicantly better or worse than expected
acrossallsixindicators.
Tobeinourbestperformersbasket,trusts
hadtodoverywellintwoormoreofoursix
indicators, and not be below average on any
oftheothers.Likewiseourworstperformers
are belowaverage onat leasttwo indicators,
without doing particularly well on any ofthe
rest.Youcanseetheresultsforthesetrustson
pages20-21,whilethefulllistingsforallother
trustsareavailableonourwebsite.
It is clear that there have been measurable
improvementsinthewaytheNHSdealswith
strokesandthattheStrokeStrategyismaking
a difference tothe number ofdeaths.In fact
theNAO estimated that, since 2006, stroke
patientschancesofdyingwithin10yearshave
fallenfrom71to67percent.5Butthereisstilla
worryinglevelofvariationincare.Yourchance
ofsurvivaloryourqualityoflifeifyoudo
survivestillvariesaccordingtowhichhospital
youareadmittedto.
DID OU KNOW? 97% of trusts have a specialist stroke unit.
The stroke data in the 2010 Hospital Guide provides an invaluable means
of comparing performances across all English acute services for stroke.
Two relevant outcomes for stroke are readmissions and standardised
mortality and it is very encouraging that most hospitals perform well on these
two measures. A move towards measuring in- and out-of-hospital mortality
would be a further advance given that many stroke patients now benet from
early supported discharge, sometimes within 72 hours of admission.
The landscape of acute stroke care is changing dramatically in the UK,
and the data presented here is unlikely to reect this. For instance, this
year a new model of care in London has already helped to increase rates
of intravenous thrombolysis to 12 per cent for February to July 2010,
compared with 3.5 per cent for the same period in 2009.
Of course thrombolysis rates are only a small part of the story. There is also
a need to demonstrate improved outcomes of functional recovery following
treatment, improvement in the percentage of patients returning to their
previous life roles, and patient satisfaction with the care provided. It is to behoped that the 2011 Hospital Guide looks across the whole stroke pathway,
rather than just acute care.
Dr Charles Davie is consultant neurologist at the Royal Free Hampstead NHS Trust,
stroke lead for University College London Partners and clinical stroke lead for the North
Central London Cardiac and Stroke Network
What happens
to patients after theyleave hospital?
Dr Charles Davie
E
PE
R
T
O
P
I
N
I
O
N
8/8/2019 Hospital Guide 2010
19/38
Hospital Guide 2010 20
The care pathway:best and worst performing trusts
across six indicators
DIAGNOSIS
AND PREENTION
URGENT
TREATENT
Performing
brain scans the same
or next day
Providing
thrombolytic drugs
within 24 hours
EastKentHospitalsUniversityNHSFoundationTrust
DerbyHospitalsNHSFoundationTrust
NorthumbriaHealthcareNHSFoundationTrust
SouthendUniversityHospitalNHSFoundationTrust
SouthTeesHospitalsNHSFoundationTrust
TheQueenElizabethHospitalKingsLynnNHSTrust
Barking,HaveringandRedbridgeUniversityHospitalsNHSTrust
BasildonandThurrockUniversityHospitalsNHSFoundationTrust
Blackpool,FyldeandWyreHospitalsNHSFoundationTrust
GeorgeEliotHospitalNHSTrust
IsleofWightNHSPCT
LeedsTeachingHospitalsNHSTrust
NottinghamUniversityHospitalsNHSTrust
WestMiddlesexUniversityHospitalNHSTrust
Full results are available at www.drfosterhealth.co.uk/hospital-guideSTROKE
Key
Exceedsexpectation Meetsexpectation BelowexpectationNotapplicable
Bestperformers
Worstperformers
Continuedo
ver
8/8/2019 Hospital Guide 2010
20/38
Hospital Guide 2010 21
Full results are available at www.drfosterhealth.co.uk/hospital-guideSTROKE
ACUTE TREATENT HOSPITAL DISCHARGE SECONDAR
PREENTION
Pneumonia due
to swallowing problems
Standardised
mortality ratio for stroke
Discharge home
within 56 days
Emergency
readmissions for stroke
Source:SUSdata2009/10
8/8/2019 Hospital Guide 2010
21/38
Hospital Guide 2010 22
With an ageing population, the demand fororthopaedic services has grown steadily over
thepast20years.Thisisparticularlytruefor
hip and knee replacements: 125,000 took
placeinEnglandin2009/10.Theannualcost
formedicalandsocialcareintheUKforhip
fracturesaloneisabout2bn,settoriseto
2.2bnby2020.1Forthesereasonswehave
focusedonorthopaedicsasakeyspecialty.
Avoiding readmissions
First we identied all patients who were
readmittedwithin28daysafterahiporknee
replacement. For both these outcomes, the
majority oftrusts performed to theexpected
standard in 2009/10. However, for hips,two
trusts had high readmission rates Leeds
Teaching Hospitals NHS Trust (75 per cent
aboveaverage)andTheNewcastleuponTyne
HospitalsNHSFoundationTrust(63percent).
Incontrast,twotrustshadlowrates Northern
Devon Healthcare NHS Trust (67 per cent
below average) and Royal Devon and ExeterNHSFoundationTrust(35percent).
Reducing the need for revisions
In2009/10morethan2,000patientshadto
havetheirhiporkneereplacementrevisedor
manipulated. Wear and tear does mean that
replacementswillnotlastforever.Butformost
patientstheydolastfor15to20years.
Hip revision rates varied from 0 to 3.5 per
cent,andkneerevisionsfrom0to2.1percent.
Twotrusts had high rates for hip revisions
Frimley Park Hospital NHS Foundation Trust
andNorthumbriaHealthcareNHSFoundation
Excellence in
orthopaedicsa team approachWhen measuring clinical effectiveness, it is essential to look at
the overall performance of care teams, not just hospitals, so that
the results are meaningful to patients. This year we have assessed
a basket of six indicators for the quality of orthopaedic care.
8/8/2019 Hospital Guide 2010
22/38
Hospital Guide 2010 23
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Trustandonewashighforkneerevisions
GuysandStThomasNHSFoundationTrust.
Sixteen trusts have done particularly well on
this indicator. Visit www.drfosterhealth.co.uk
toseethefulllistofresults.
Death following a hip fracture
Hipfractures(orfracturedneckoffemur)also
representamajorexpensefortheNHSand
arethemostcommonreasonforadmissionto
anorthopaedicward.Morethan70,000hip
fractureshappeneachyearintheUK,whichis
likelytoincreaseto101,000by2020.2
Moreover, patients who suffer a hip fracture
haveahighmortalityratio:about10percentof
peoplewithahipfracturediewithinonemonth,
andaboutathirdwithin12months.In2009/10
nearly6,000peoplediedinhospital.However,
when examining standardised mortality ratios,
all trusts performed as well as expected, and
CambridgeUniversityHospitalsNHSFoundation
Trust had an especially lowratio(46 per cent
belowaverage).
Operating straightaway
Patientswho falland break theirhips should
havethemoperatedonwithintwodays.Thisis
acrucialtimeframe,notonlyasitisacceptedbest practice from the National Institute for
HealthandClinicalExcellence(NICE),butalso
asthereisaprovenlinkbetweenadelayin
theoperationandanincreasedriskofdeath.3
Worryingly, in our analysis, 21 per cent of
trustshadratesthatweresignicantlylow.The
percentageoperatedonwithintwodaysvaried
from34to94percent.Againthefullresults
areshownonourwebsite.
Forapictureoftrustsoverallperformance,we
have identied those performing signicantly
wellorpoorlyacrossthesixindicators(using
the same criteria asfor stroke, see page 19).
Thesetrustsareshownintheboxtotheright.
ORTHOPAEDIS
Joining up
the dataTom WainwrightRobert Middleton
Best performers AiredaleNHSTrust
NorthernDevonHealthcareNHSTrust
RoyalDevonandExeterNHSFoundationTrust
RoyalSurreyCountyNHSFoundationTrust
UniversityHospitalsCoventryandWarwickshireNHSTrust
WestSuffolkHospitalNHSTrust
Worst performer LeedsTeachingHospitalsNHSTrust
Combining the indicators
Truststhathaveperformedsignicantlywellorpoorlyacrossthesixindicators:
E
PE
R
T
O
P
I
N
I
O
N
Understanding more about the effect of our orthopaedic interventions is
imperative, so there is great value in tracking the data we see on these pages.
There have also been recent developments in the way that we can assess the
quality of hip and knee replacements. In September 2010, the rst data on
Patient Reported Outcome Measures (PROMs) was published by The NHS
Information Centre for health and social care. PROMs use ve different
questionnaires to evaluate patients health. The initial data is termed
experimental but shows us that 96 per cent of hip replacement patients
and 91 per cent of knee replacement patients recorded a joint-related
improvement after their operation.
Benchmarking is a hugely powerful tool in helping hospitals to identify
areas where improvements are needed, as well as increasing transparency
for the general public so they can make choices about where to be treated.
It should be noted that the variation in outcomes across providers and
therefore the inferred difference in quality is often greater than we might
imagine. For example, case-mix-adjusted average length of stay varies by
over seven days across hospitals for knee replacement. We would like to
see more sensitive and discriminative data; at present the data we have
does not provide patients with enough detail to choose between different
hospitals by making meaningful judgements about quality.
A step towards more meaningful data could happen by bringing together
the various databases that we already have. While the Dr Foster data is very
useful in isolation, it would be strengthened further by aggregation with the
other major databases such as PROMs, the National Joint Registry and the
national hip fracture audit.
Robert Middleton is consultant orthopaedic surgeon and Tom Wainwright
is clinical researcher in orthopaedics, both at The Royal Bournemouth
and Christchurch Hospitals NHS Foundation Trust
8/8/2019 Hospital Guide 2010
23/38
Hospital Guide 2010 24
Urologyexcellence in operations
Guidelines recommend that
surgery for urological cancer
be carried out in specialist
centres to improve quality.1
Simon Carter, consultant
urologist at Imperial College
Healthcare NHS Trust,
shows how the guidelines
are being implemented
and identies leading trusts
for these types of operation.
Urologicalcancers,suchasthosefoundinthebladder,prostateorkidney,affectmorepeople
thanbreastcancereachyear. 2Itisrecommended
practice to carry out operations for these
conditionsinlargerorganisationswherethe
oftencomplexproceduresareperformedmore
frequently.Inthissectionwelistsomeofthe
leadingtrustsinthetreatmentoftheseillnesses.
Operationstotreatbenignprostatediseaseare
performedinawidernumberofunits;wehave
comparedhowoftentheprocedureneedstobe
redonewithinthreeyears.
Surgery for pelvic cancer
Therearegoodreasonstobelievethatcentres
whichcarry outlarge numbers ofsurgery for
prostateand bladdercancerhave consistently
highstandards.Wehave thereforeidentiedthe19truststhatperformedhighnumbersof
operationsonpeoplewithprostateorbladder
cancerfrom2007/08to2009/10(seepage25).
Inaddition,wehaveidentiedwhichofthose
trusts alsoperformed a signicant numberof
laparoscopicprostatectomyoperationsduring
2009/10.Theseoperationsoffer considerable
benetstothepatientintermsofthespeedof
operatingandthespeedofrecovery.However,
notalltrustshavesurgeonswhoareexpertin
thesetechniques.
The2002guidelinesfromNICEdemandedthat
pelvic urological cancer surgeryshouldonly
beundertaken in units where morethan 50
proceduresareperformedeveryyear.
%l
aparoscopic
2006/07
10%
20%
30%
40%
50%
60%
2007/08 2008/09 2009/10
Financial year
Large Medium Small All
0%
ore prostatectomies are now being performed by laparoscopic techniqueSource: SUS data 2006/07 to 2009/10.
8/8/2019 Hospital Guide 2010
24/38
Hospital Guide 2010 25
Full results are available at www.drfosterhealth.co.uk/hospital-guideUROO
Toagreatextent,theguidelinesarenowbeing
achieved.Over3,500prostatectomyoperations
are being performed in the large trusts each
year (dened as trusts doing more than 50
operationsayear).Fewerthan1,000arebeing
performedinsmallormedium-sizedtrusts.The
useoflaparoscopicorkeyholetechniqueshas
also increased dramatically, predominantly in
thelargercentres.
Prostatecancersurgeryisontheincrease.The
totalnumberofoperations(bothlaparoscopic
and open)has risenby 33per cent between
2005/06and2009/10.Largetrustshaveseen
ariseof166percentduringthattime,while
mediumtrustsactivityhasfallenby61percent.
In2006/07only15percentofprostatectomies
werecodedasbeingalaparoscopicprocedure;
this rate has increased each year and by
2009/10 stood at44per cent.The upward
trendistrueofbothlargeandmediumtrusts
(showninthegraphonpage24).
Removalofthebladder,knownasacystectomy,
isanothermajorsurgicalprocedurewithmany
potential complications, and it is probably
best performed by surgical teams with great
experience.In2005/06largetrusts(thosedoing
more than 25 operations a year) performed
only21percentofcystectomies(304intotal).
Bycontrast,by2009/10largetrustsaccounted
for63percentofcystectomies(1,005).
Isitpossibletosaythatthequalityofoperative
urologicalcarehasimproved?Whencomparing
truststhatperformmanyprocedureswiththose
thatonlydoamodestnumber,simplemeasures
haveshownthattheirratesofreadmissionsand
lengthofstayareverysimilar.
However,theincreasinguseofsophisticated
surgicaltechniquesisconcentratedinasmall
numberoftrusts,andpatientsaremorelikelyto
getthefullrangeofoptionsinthelargercentres.
Surgery for benign conditions
Conventionalurologicalproceduresforlower
urinary tract symptoms and acute retention
of urine both benign conditions continue
tobeundertakeninawiderangeofhospitals,
with varying quality. One such procedure is
transurethralresectionoftheprostate(TURP).
Measuringthe need torepeatthis operation
withinthreeyearscouldbe a novelway of
lookingatquality,especiallyasthedatacanbeusedtoseewhenoperationsareperformedin
anytrustinEngland.
TrustswithlowratesforrepeatTURPoperations
withinathree-yearperiod:
AintreeUniversityHospitals
NHSFoundationTrust
Barking,HaveringandRedbridgeUniversity
HospitalsNHSTrust
DerbyHospitalsNHSFoundationTrust
GeorgeEliotHospitalNHSTrust
IpswichHospitalNHSTrust
LutonandDunstableHospital
NHSFoundationTrust
NottinghamUniversityHospitalsNHSTrust
RoyalBoltonHospitalNHSFoundationTrustShrewsburyandTelfordHospitalNHSTrust
SouthportandOrmskirkHospitalNHSTrust
UniversityHospitalsofLeicesterNHSTrust
WarringtonandHaltonHospitals
NHSFoundationTrust
YeovilDistrictHospitalNHSFoundationTrust
TrustswithhighratesforrepeatTURPoperations
withinathree-yearperiod:
MidEssexHospitalServicesNHSTrust
NorthernLincolnshireandGooleHospitals
NHSFoundationTrust
TheQueenElizabethHospitalKingsLynn
NHSTrust
Trusts performing high numbers of urological operations
on pelvic cancer patients.Source: SUS data 2007/08 to 2009/10.
Number of cases
NorthBristolNHSTrust* 391
GuysandStThomasNHSFoundationTrust* 376
CambridgeUniversityHospitalsNHSFoundationTrust* 330
TheNewcastleuponTyneHospitalsNHSFoundationTrust* 279
ShefeldTeachingHospitalsNHSFoundationTrust* 268
NorfolkandNorwichUniversityHospitalsNHSFoundationTrust 230
ImperialCollegeHealthcareNHSTrust* 225
LeedsTeachingHospitalsNHSTrust 201
HullandEastYorkshireHospitalsNHSTrust 193
CityHospitalsSunderlandNHSFoundationTrust 189
RoyalBerkshireNHSFoundationTrust* 182
NottinghamUniversityHospitalsNHSTrust* 179
SouthTeesHospitalsNHSFoundationTrust 168
RoyalDevonandExeterNHSFoundationTrust 150
UniversityCollegeLondonHospitalsNHSFoundationTrust 140
PlymouthHospitalsNHSTrust 132
AshfordandStPetersHospitalsNHSTrust 128
MedwayNHSFoundationTrust 125
CentralManchesterUniversityHospitalsNHSFoundationTrust 106
*Trustswithmorethan40laparoscopicprostatectomiesin2009/10
8/8/2019 Hospital Guide 2010
25/38
Hospital Guide 2010 26
Wherever medicine is practised in the world, unsafe treatment results in
errors and harm to patients. A key part of efforts to improve safety is to
accurately measure and monitor the way in which it is being addressed.
Saetythe foundation for quality
See pages 7-9
Who have we named ourtrusts of the year?
8/8/2019 Hospital Guide 2010
26/38
Hospital Guide 2010 27
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Ayearagoweratedhospitaltrustsonarangeofaspectsofpatientsafety.Thisyearweseeif
thesituationhasimproved.Forallthemeasures
below,fullresultsareavailableonourwebsite
atwww.drfosterhealth.co.uk.
Complying with safety alerts
Inour2009surveyweaskedallNHShospital
trustswhethertheyweremeetingbasicsafety
requirements.Thesearealertsissuedbythe
NationalPatientSafetyAgency(NPSA)which
warnhospitalsofpotentiallydangerouspractice
andadvisethemwhattodotoavoidharmto
patients.Seventruststoldusthat,foroneor
moreofthesealerts,itwouldtakethematleast
sixmonthstobecomecompliant.
Thisyearweaskedaboutalertsthatwereissued
in2009/10.Notrustsaiditwouldtakelonger
thansixmonthstobecomecompliantwiththese
alerts.Thisisaclearimprovementonlastyear.
However, threetrusts saidthey werestill notcompliantwithatleastoneofthealertsand
requiredafurtherthreemonthstodothis.The
alertsincluded:reducingtheriskofoverdosing
thedrugmidazolam,andinappropriateuseof
oralbowel-cleansingsolutionspriortosurgery.
Thethreetrustswere:
SouthendUniversityHospital
NHSFoundationTrust
StGeorgesHealthcareNHSTrust
WesternSussexHospitalsNHSTrust
ThesedelayscomedespitetheNPSAclearly
stating that its guidance should be acted on
immediately.Adateforcompletionisalsogiven.
Track and trigger systems
InlastyearsHospitalGuidewereportedabout
trackandtriggersystems.Theseareregularobservationsmadebynurses,designedtopick
updeteriorationinapatientscondition.
Lastyear64percentoftrustssaidtheyhadthis
systeminplaceforallacutepatients,andthis
yearwearepleasedtoseethishasrisento79
percent.Butthismeansthatoneinvetrusts
stilldonothaveatrackandtriggersystemin
place.Tondouttheresultsforyourlocaltrust
gotowww.drfosterhealth.co.uk.
Commitment by hospital boards
Weaskedalltrustswhethertheyhaveaboard
representativeresponsible for patientsafety,
whether they discuss patients safety at all
board meetings, and whether theyhave clear
denitionsthatenableittobemonitored.This
year,aslastyear,100percentoftrustsconrmed
thattheydid.
Infection control
Againweaskedalltrustswhethertheyhavean
antibioticpharmacist(whohasakeyrolefor
managing infection risks), whether they have
pre-assessmentclinics to screenall patients
forinfectionspriortoadmission,andwhether
theytreatthose patientscarrying aninfection
beforeadmittingthem(bytreatmentthrough
adecolonisationroutine).
Lastyear86percentoftrustssaidthattheydid
allthesethings.Thisyearithasrisento97percent.However, Walsall HospitalsNHS Trust
and UniversityHospitalsof MorecambeBay
NHSTrusttoldusthattheydonotemployan
antibioticpharmacist.
Reporting incidents when they happen
Disclosingpatientsafetyincidentsthroughthe
NationalReportingandLearningServiceisanimportantelement ofmanagingsafety. After
last years guide, this voluntary system now
includessomemandatoryreportingamove
wewelcome.In2008/09,trustsonaverage
reportedveincidentsper100admissions.This
hasrisento5.7in 2009/10.Ahigherrateis
generallyregardedasapositivesignbecauseit
showsawarenessoferrorsandnear-missesand
acultureoffreedomtoreport.Thetrustswith
thelowestratesofreportingare:1
MidYorkshireHospitalsNHSTrust
(2.1incidentsper100admissions)
JamesPagetUniversityHospitals
NHSFoundationTrust(2.7)
WinchesterandEastleighHealthcare
NHSTrust(2.7)
We have looked at how many patient safetyincidentswererecordedateachhospitaltrust
in2009/10usingroutinedata.Weknowthat
this underestimatesthe scaleof theproblem
becausetherecordingofdataisstillnotaccurate
enoughtogiveatruepicture.However,wecan
saythefollowing:
Pressure sores Approximately6,000patients
wererecordedashavingpressuresoreswhile
inhospital.Unfortunatelywedonotknowhow
manypatientsdevelopedthese after arriving,ratherthanbeforehand.Butwecansaythat,
inthetrustswiththehighestrates,morethan
3.5percentofpatientswererecordedashaving
pressuresores.
PATIENT SAFET
1 Is patient safetyimproving?
2 Measuring patient safety how big is the problem?
99% of all trusts responded to the Hospital Guide survey.
Only George Eliot Hospital NHS Trust and University HospitalsBirmingham NHS Foundation Trust failed to submit a response.
8/8/2019 Hospital Guide 2010
27/38
Hospital Guide 2010 28
Full results are available at www.drfosterhealth.co.uk/hospital-guidePATIENT SAFET
Pulmonary embolisms (PEs)Morethan30,500
admissionswererecordedashavingPEswhile
inhospitalin2009/10.Thislife-threatening
conditionisapotentialcomplicationfollowingastayinhospital,andthereismuchthathospitals
candotoreducetheriskofitoccurring.We
found that the rate of recorded PEs varied
widelybetweentrusts(seethediagramonpage
29),withthehighestratesover3.5timesgreater
thanthelowest.
Obstetric tears Tearingduringchildbirthcan
resultinincontinenceandtheneedforfurther
treatment.Risksoftearing,however,canbe
reducedthroughsafemanagementofpatients.
More than 13,000 women were recorded as
havingexperiencedanobstetrictearin2009/10
(with a delivery that was not assisted with
forceps)andthehighestratesweremorethan
sixtimesgreaterthanthelowest.Pleasevisit
www.drfosterhealth.co.uktondoutifyour
localtrusthasahighorlowrate.
Accidental punctures or lacerations Almost
10,000 hospital patients were recorded as
having sufferedfrom an accidental puncture
orlacerationin2009/10.Thisgureisalmostcertainlyanunder-recording,andeachoneof
theseeventscouldhavebeenavoided.
Post-operative haemorrhagesMorethan2,000
patientswererecordedashavingsufferedfrom
post-operative intestinal bleeding. This often
requiresfurthersurgerytotreatandcanbea
life-threateningcomplication.Again,thelevels
recordedarelikelytobeanunderestimate.
Post-operative sepsisThisisanotherpotentially
life-threatening complication. Around 1,300
patientsundergoingsurgerywererecordedas
alsohavingsepsis.Itmustbeassumedthatin
mostcasesthesepsiswastheresultofsurgery.
The information we needby Lifeblood: The Thrombosis Charity
Deep vein thrombosis is hard to spot. It does not always cause any physical swelling
or redness as the textbooks say, so it is often clinically silent. We know from the
many calls and stories we receive in the Lifeblood ofce that many people have their
symptoms ignored by health professionals as they do not t the textbook description.
Knowing how many PEs are actually occurring allows us to monitor how well the
condition is being prevented in hospital. We are grateful to Dr Foster for its work in
trying to establish the numbers admitted with a diagnosis of PE, and we are just as
disappointed that the data is so poor due to the coding system around DVT and PE.
2010 has been a watershed, with the Department of Health setting nancial incentives
for hospitals in England to assess the VTE risk of all adult admissions. Now, thanks to
the Dr Foster team, there is added proof that coding in this area must be improved.
Here is just one of many stories about the serious harm from hospital-acquired clots:
The risk of a blood clot wasnt even mentioned
Amy was diagnosed with juvenile arthritis aged six, had her rst hip replacement at
15 and has now had seven new hips. At her last operation in 2005, aged 32, she did
not receive routine thromboprophylaxis pre or post-operatively. Afterwards she was on
crutches and not allowed to bear any weight, increasing her risk of developing a blood
clot. But this wasnt mentioned and she was sent home without any information.
A month later her leg swelled up. At rst she put it down to her arthritis and the strain
of hopping about, but it got worse and one day her leg had trebled in size. She called
her GP, who lifted her legs, asking if they hurt; she said they did not and he left saying
nothing was wrong. But the pain became so severe that her friend took her to hospital.
She was diagnosed with a DVT (a large clot in her thigh), given drugs and kept in bed for
two weeks. After being sent home with compression stockings, her leg returned to its
normal size in a fortnight. Then she was given three months warfarin and discharged.
After one DVT, the risk of a second increases. Amy developed a clot two years laterafter a long-haul ight and is now on life-long warfarin. She has frequent pain and
swelling, but her GP only gave her stockings when she showed him a Lifeblood leaet.
Professor Beverley Hunt is medical director at Lifeblood: The Thrombosis Charity
8/8/2019 Hospital Guide 2010
28/38
Hospital Guide 2010 29
Patients assessed for VTE No. of trusts
100% 6
91-99% 7
61-90% 59
31-60% 36
1-30% 24
0% 1
Didnotanswerthequestion 14
3 Preventing blood clotsin hospital
Recognition of the burden of hospital-acquired VTE in England has been
consigned to the too difcult box for too many years. In 2005, when
we rst became involved in the VTE prevention journey in the NHS in
England, there was little appetite for exploring routine data in any detail or
for considering the data in new ways.
However, under the leadership of Sir Liam Donaldson and now Sir Bruce
Keogh, it has become clear that improving coding can provide insight into
the number of people each year who develop a hospital-acquired thrombosis.
We have been working with the Dr Foster team to create an evidential
basis on which to dene hospital-acquired thrombosis. This approach was
endorsed recently by the Academy of Medical Royal Colleges through the
work of the National Quality Boards VTE sub-group. We are also working
with trusts in the south-west to try to improve consistency in coding.
We are still some way from understanding the true incidence of hospital-
acquired VTE, and use of discharge coding remains difcult. Post-mortem
studies suggest that, in a third of patients where death is caused by a PE,
the correct diagnosis is not even suspected beforehand.
Highlighting PEs in this years Hospital Guide is a welcome contribution to
our understanding of the size of the VTE issue. Just as important, though,
is the impetus that this published data may provide in stimulating the NHS
to locally discuss, publish, use and improve local data on VTE.
Dr Anita Thomas OBE is national clinical director for VTE and Tim Brown
is national VTE prevention programme lead, both for the Department of Health
Measuringblood clots moreaccurately
Tim BrownDr Anita Thomas
E
PE
R
T
O
P
I
N
I
O
N
Avenousthromboembolism(VTE)isablood
clot which develops in a part of the body,
usuallythe leg.Deep veinthrombosis (DVT)
isacommontypeofthiscondition.Partof
aclotmaybreakoffandlodgeinthearteries
thatsupplythelungs,resultinginapulmonary
embolism(PE).Thiscanoftenbefatal.
TheDepartmentofHealthhasmadesurethat
thepreventionofVTEisamajorpriority,and
it isa key component of the CQUIN scheme
(CommissioningforQualityandInnovation).
All adult patients admitted to hospital mustnow be risk-assessed for VTE,and trusts will
berequiredtodoaroot-causeanalysisofall
conrmedcasesofhospital-acquiredVTE.2
Theriskofpatientsdevelopingabloodclotis
increasedby mostsurgicaland somemedical
treatmentsandconditions.In2001,forexample,
JohnHeitfromtheMayoClinicintheUSreported
thatincidence of VTEis morethan 100-fold
higheramong hospitalised patients compared
withcommunityresidents.3Inaddition,in2005
theHealthSelectCommitteereportedthatVTEs
couldaccountfor 25,000 preventabledeaths
eachyear.4
Weaskedtrustsinourquestionnaire,What
percentageofpatientsarerisk-assessedforVTE
onadmission?Theytoldusthefollowing:
Themajorityoftrustswereabletoreporthow
manypatientswererisk-assessed.However,itis
aconcernthat15trustseithertoldustheywere
notassessing any patientsor were unable to
providetheinformation.Also,mosttrustsneed
toincreasetheirassessmentratessignicantly
inordertoprotectpatientsfromrisk.Seeour
websiteforfulllistings.
Variation in the rates of pulmonary embolismSource:SUSdata2009/10,99.8%controllimits.
ofall147trusts
22exceedexpectation 98meetexpectation 27arebelowexpectation
Full results are available at www.drfosterhealth.co.uk/hospital-guidePATIENT SAFET
8/8/2019 Hospital Guide 2010
29/38
Hospital Guide 2010 30
Full results are available at www.drfosterhealth.co.uk/hospital-guidePATIENT SAFET
Why we cannot tell you which trusts
are best at preventing blood clots
Fromthehospitaldatarecords,weidentiedallpatientswhowererecordedashavingsuffered
fromaPE.Wethentookthosepatientsforwhom
itwasrecordedasasecondarydiagnosisafter
admissionforadifferentconditionorprocedure.
Wealsotookthosepatientswhowereadmitted
tohospitalwithaprimarydiagnosisofPEwho
had been treated in hospital for a different
conditionwithinthepreviousthreemonths.This
isbecausetheriskofdevelopinganembolism
following hospital treatment continues for up
to90days.5
Our analysis revealed approximately 30,500
PEsduring2009/10.Weknowthatthisisan
undercount because many patients with PEs
arenotcodedassuch,andinsteadarerecorded
ashavinghadunspeciedcomplicationscaused
bytheirtreatment.
Asthediagramonpage29shows,thereiswide
variation in the number of PEs recorded at
eachtrust.Thisrangesfrom47percentbelow
averageto174percentaboveaverage.However,wedonotknowthetruerateofPEsforeach
trust. We are also uncertain about the exact
numberofPEscausedbyinappropriatecare
or inadequatepre-treatment riskassessment.
Nevertheless,allhospitaltrustsmustchange
their coding systems where necessary and
ensurethattheycomplywiththemandateto
systematicallyinvestigateallunexpectedPEs.
We are approaching The NHS Information
Centreforhealthand social care,askingit toissueexplicitguidancearounduniformcoding
practices.DrFosterhopestorevisitthistopic
earlynextyearandbeabletoidentifytrustsand
theirrates.
Formorethanadecade,successivereportshaveattemptedtoquantifythenumberofmedical
mistakes(oradverseevents)thattakeplacein
ourhospitals.In2000theDepartmentofHealth
estimated that harm is caused to patients in
around10percentofadmissions,orataratein
excessof850,000ayear.6
Tenyearslaterwearestillquotingguresbased
onresearchestimates.TheNHScannotstate
categorically how many medical errors take
place in its hospitals.Not all hospital chief
executivesknowexactlyhowmanypatientsare
harmedintheirunitseachyear,thereforethey
cannotknowexactlyhowtheirorganisationwill
preventharmtoallfuturepatients.
DrFosterhasexaminedroutinedatatotryto
gain a picture of the medical mistakes being
recorded.IntheUS,theAgencyforHealthcare
ResearchQuality(AHRQ)hasbeendeveloping
waysofmeasuringtheseforseveralyears,and
inparalleltheDrFosterUnitatImperialCollege
Londonhasbeentranslatingthemethods.Thesehave been featured inpastHospital Guides.
The AHRQ has also introduced a composite
indicator,aggregatingthiscollectionofmeasures
intoan overall hospital score.7 TheUnit has
reproducedthiscompositeindex,usingsixofthe
20indicators:
Pressuresores(decubitusulcer).
Deathsaftersurgery(seepages12-13for
adenition).
Bleedsorbruisesaftersurgery(post-operative
haemorrhageorhaematoma).
Post-operativerespiratoryfailure.
Post-operativesepsis.
Accidentalpunctureorlaceration.
OtherindicatorsrecommendedbytheAHRQ
wereexcludedduringtheresearch,eitherasthe
reportednumbersweretoosmallortherewas
notenoughcondenceinthecoding(iehowwelltheinformationwasrecorded).
In2009/10,acrossthesesixindicators,more
than27,000 potentialadverse events were
reported.Again,thisdatadoesnotgiveusa
completepicturebecausesometrustsarebetter
atrecordingtheinformationthanothers.Infact
we cansee that, in general,trustswith high
ratesofincidentsarethosethattendtohave
more complete records about their patients.
Itisnotthattheyhavehigherratesofmedical
errors,theyaresimplybetteratrecordingwhat
happens.Whatismore,truststhatarebetterat
recordinginformationarelikelytobebetterat
managingtheproblems.
On page 31 we list the trusts that are
good at recording data, as well as those
that are relatively poor at it.
4 Preventing adverseevents in hospitals
DID OU KNOW? Trusts reported 56 incidents of wrong site
surgery, as well as 150 foreign objects that were left inside patients
after an operation.
Its clear that all trust
boards should have thesafety of their patients as
the number one priority.
No matter how dedicated
and professional the
nursing staff are, things
do go wrong and we need
to work even harder at
tackling avoidable mortality
and adverse events.
DrPeterCarter,ChiefExecutiveandGeneralSecretary,RoyalCollegeofNursing
8/8/2019 Hospital Guide 2010
30/38
Hospital Guide 2010 31
Full results are available at www.drfosterhealth.co.uk/hospital-guide
Good data-recording and high adverse events
What this may mean:Thecodingismoreaccuratethanmany
othertrusts.Thoselistedbelowhaveahighrateofadverseevents
whencomparedwiththerestoftheNHSandthis,coupledwith
theaccuratecoding,meansthattheremaybepotentialproblems
here.Thedatashouldbeinvestigatedtorulethisout.
CentralManchesterUniversityHospitalsNHSFoundationTrust
DoncasterandBassetlawHospitalsNHSFoundationTrust
LancashireTeachingHospitalsNHSFoundationTrust
LutonandDunstableHospitalNHSFoundationTrust
NorthBristolNHSTrust
NottinghamUniversityHospitalsNHSTrust
PlymouthHospitalsNHSTrust
RoyalDevonandExeterNHSFoundationTrust
SalfordRoyalNHSFoundationTrust
SherwoodForestHospitalsNHSFoundationTrust
SouthendUniversityHospitalNHSFoundationTrust
StGeorgesHealthcareNHSTrust
TheNewcastleuponTyneHospitalsNHSFoundationTrust
UniversityHospitalsofLeicesterNHSTrust
WarringtonandHaltonHospitalsNHSFoundationTrust
WestHertfordshireHospitalsNHSTrust
WesternSussexHospitalsNHSTrust
Good data-recording and low adverse events
What this may mean: Thecodingismoreaccuratethanmany
othertrusts.The low rates are promising because theysuggest
thatfewermedicalerrorsareoccurringinthesetrusts.However,
ofcourse,allerrorsshouldbeinvestigated.
BedfordHospitalNHSTrust
RoyalCornwallHospitalsNHSTrust
RoyalLiverpoolandBroadgreenUniversityHospitalsNHSTrust
SandwellandWestBirminghamHospitalsNHSTrust
SouthDevonHealthcareNHSFoundationTrust
UniversityHospitalofSouthManchester
NHSFoundationTrust
Wrightington,WiganandLeighNHSFoundationTrust
Poorer data-recording and high adverse events
What this may mean:Codingratesarelowcomparedwithother
trusts,yetincidentsarehigh.Thissuggeststhatthetruerateof
incidentsmaybeevenhigher,asnotallarebeingrecorded.The
recordedincidentsshouldbeinvestigated,andthereshouldalso
beanassessmenttoseeifmoreshouldhavebeenreported.
CambridgeUniversityHospitalsNHSFoundationTrust
LeedsTeachingHospitalsNHSTrust
MidEssexHospitalServicesNHSTrust
NorthCumbriaUniversityHospitalsNHSTrust
OxfordRadcliffeHospitalsNHSTrust
Poorer data-recording and low adverse events
What this may mean:Codingratesarelowcomparedwithother
trusts. This maybe thereasonfor the lowratesof incidents.
Areviewmayneedtotakeplacetoensurethatadverseeventsare
beingrecorded.
AiredaleNHSTrust
Ba