clinical and health economic relevance of the swedish...

SICOT 2008PHESICOT/SIROT XXIV SICOT/SIROT XXIV TriennialTriennial World Congress 2008World Congress 2008

PlenaryPlenaryspeaker,27 August 2008, Hong Kongspeaker,27 August 2008, Hong Kong

Clinical and Health Economic Relevance of The Swedish Hip Arthroplasty Register

P.HerbertsMD, PhD.Professor

Department of OrthopaedicsSahlgrenska University Hospital,

Göteborg, Sweden

SICOT 2008PHE

Acknowledgment to coAcknowledgment to co--workersworkers

Johan KJohan Käärrholm rrholm GGööran Garellickran Garellick

& & The entire Register StaffThe entire Register Staff

SICOT 2008PHE

Serious consideration should be given to establishing a central register to keep a finger

on the pulse of total implant surgery on a nation-wide basis

Sir John CharnleyInternal Publication No 39, 1972

The idea of a national Register

SICOT 2008PHE

The Swedish The Swedish HipHip Arthroplasty RegisterArthroplasty Register--A A ProspectiveProspectiveObservationalObservationalStudyStudy

• Started 1979 and has had a profound impact on the results of all THR surgery in Sweden.

• Owned by the Swedish Orthopaedic Society

• Supported by the National Board of Health and Welfare.

SICOT 2008PHE

The Swedish RegisterThe Swedish RegisterInternet addressInternet address

• http://www.jru.orthop.gu.se/

• All data collection and feedback through this site since Jan. 1st 1999.

SICOT 2008PHEThe MissionThe Mission

• To improve the general outcome of total hip replacement by outcome assessment

• To establish a continuous learning process

• To control quality with focus on the procedure – i.e. enables cost-utility analysis

• To give public information of results

SICOT 2008PHE

Feedback of results is the most essential feature for compliance and will make

continued clinical responsibilityand accountability feasible

Confidential and public feedbackon-line and Annual Reports

www.jru.orthop.gu.se

SICOT 2008PHE

The Swedish The Swedish THATHA RegisterRegister1979 1979 -- 20072007

• 284,630 primary THR • 27,690 revision THR

SICOT 2008PHE

Two levels of Definition for FailureTwo levels of Definition for Failure

• Traditional: Revision of the implant (since 1979)

• The patient not satisfied or low HRQoLat follow-up (2002)

SICOT 2008PHE

Results after 30 yearsResults after 30 years

The result of this continuous outcome assessment and disseminationof results back to the professionis a constant improvement of the 10-year survivorship on a nation-wide basis

SICOT 2008PHE

10-years survivorship

time interval – index operation

% survivors

10-year: 94%

SICOT 2008PHE

• Survival statistics based on patient-and implantimplantrelated factors.

• Kaplan-Meier and regression analysis.• In 2007 more than 50% of the THRs were done

with 3 cemented implant combinations.

Restriction of Implant ChoiceOne reason for improvement

SICOT 2008PHE

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

11000

12000

13000

14000

15000

79 81 83 85 87 89 91 93 95 97 99 01 03 05 07

Primary THR in Sweden 1979 - 2007

Cemented

Hybrid

Uncemented

Reversed hybrid

Resurfacing

SICOT 2008PHEAll cemented THRs

256 689 primary THRs, 22 641 revisions, 1979-2007

01000

20003000

40005000

60007000

80009000

1000011000

1200013000

1400015000

16000

79 81 83 85 87 89 91 93 95 97 99 01 03 05 07

RB 1979 – 2007: 8.1%RB 1992 – 2007: 9.8%

SICOT 2008PHE

Uncemented THRs12 289 primary THRs, 2 569 revisions,

1979-2007

0

200

400

600

800

1000

1200

1400

1600

1800

2000

79 81 83 85 87 89 91 93 95 97 99 01 03 05 07

RB 1979 – 2007: 17.3%RB 1992 – 2007: 20.3%

SICOT 2008PHE

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Aseptic loosening/osteolysisInfection, dislocation, fractureMisc.

Reasons for revision1998 - 2007

SICOT 2008PHE

Improvement of surgical techniqueMost important reason for improvementSignificant factors in regression analysis

Pulsative lavageDistal femoral plugProximal femoral sealVacuum mixing

SICOT 2008PHE

Cementing techniqueCementing techniqueOA and aseptic looseningOA and aseptic loosening

Risk ratio

95% confidence

limits

Pulsatile lavage 0.72 0.66-0.79

Proximal femur seal 0.79 0.72-0.87

Distal femoral plug 0.87 0.80-0.94

SICOT 2008PHE

Introduction of new technologyIntroduction of new technologyi.e. uncemented fixationi.e. uncemented fixation

The problem is that surgeons by their nature are attracted to new ideas and concepts.New designs are used with very little supporting evidence

SICOT 2008PHE

70

75

80

85

90

95

100

0 1 2 3 4 5 6 7 8 9 1011 121314 1516

follow -up year

perc

enta

ge n

ot reo

pera

ted

Survival cemented fixation in blue

Survival uncemented fixation in red

(n=170.413 1992-2007)

N=170.413

nn

SICOT 2008PHE

90

92

94

96

98

100

0 1 2 3 4 5 6 7 8 9 10

follow -up year

perc

enta

ge n

ot reo

pera

ted

bnvSurvival cemented fixation in blue

Survival uncemented fixation in red (n=115.959, 1998-2007)

SICOT 2008PHE

98

99

100

0 0,25 0,5 0,75 1 1,25 1,5 1,75 2

follow -up year

perc

enta

ge n

ot r

evis

ed

Survival cemented fixation in blue

Survival uncemented fixation in red

Revised within 2 years (1998-2007)

SICOT 2008PHE

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

uncemented cemented

total revision

liner exchange

SICOT 2008PHE

Secur-Fit/OmnifitSecur-Fit/OmnifitAll observations, 1979-2000All observations, 1979-2000

0 2 4 6 8 10 12 14 16 18 20 22

years postoperatively

70

75

80

85

90

95

100percent not revised

Revision Cup: 8 y = 79,1%, n = 457Revision Stem: 8 y = 98,5%, n = 457

SICOT 2008PHE

CLS Spotornoall diagnoses and all reasons

80

85

90

95

100

0 2 4 6 8 10 12

years postoperatively

pe

rce

nt

no

t re

vise

d

1992-2004, 12y = 97.6% (95.4-99.7), n = 551

1992-2004, 12y = 98.8% (97.6-100), n = 551

SICOT 2008PHE

95

96

97

98

99

100

0 0,5 1 1,5 2 2,5 3

follow -up year

perc

enta

ge n

ot rev

ised

All other fixation methods

Resurfacing

SICOT 2008PHE

NARANordic Arthroplasty

Register Association

SICOT 2008PHE

NARA history:NARA history:

• Danish Hip Arthroplasty Register 1995

• Norwegian Arthroplasty Register 1987

• Swedish Hip Arthroplasty Register 1979

SICOT 2008PHE

Material:Material:

• A total of 280 201 THR:s

• Denmark 69 242

• Sweden 140 821

• Norway 70 138

SICOT 2008PHE

ResultsResults 1010--year year survivalsurvival::

• Denmark 91.9% (91.5-92.3)

• Sweden 93.9% (93.6-94.1)

• Norway 92.6% (92.3-93.0)

9 596/280 201 revised

SICOT 2008PHE

SICOT 2008PHE

Fig. 4 Survivorship curves (with 95% confidence int ervals) for total hip arthroplasty implants in the United States, Sweden, and Norway.

Kurtz S. M. et.al. J Bone Joint Surg 2007:89:144-15 1

SICOT 2008PHE

Effect of Arthroplasty Registries

• The argument that information from a Registry is not as good as a well designed research study does not exist. They are different things.

• The fact that Registries are more effective at improving clinical outcomes than research studies is clearly established.

SICOT 2008PHE

Register improvement since 2002 Register improvement since 2002 Increase the sensitivity for failure definitionIncrease the sensitivity for failure definition

• Since 6 years we capture PROM by use of EQ-5D – HRQoL, pain and satisfaction

• Web based registration and feed-back.• It will enable a large scale of cost-utility

studies.

SICOT 2008PHE

The PROM instrument (patient The PROM instrument (patient related outcome measurement) related outcome measurement)

• Charnley classification (A, B, C)

• Pain – VAS (0-100)

• EQ-5D – HRQoL

• Satisfaction – VAS (0-100)

SICOT 2008PHE

EQEQ--5D5D

• Self-reported health related QoL

• Five dimensions

• EQ-5D index from 0-1

SICOT 2008PHE

• Preop: questionnaire (10)

• 1 year: questionnaire (11)

• 6 years: questionnaire + X-ray

• 10 years: questionnaire + X-ray

The project implies Follow up for all primary THRs in Sweden

SICOT 2008PHE

SICOT 2008PHE

paperless – time saving

methodologically attractive

the patient is “forced” to answer all

questions in order to go further

no missing values!

SICOT 2008PHEThe outcome resultsThe outcome results--Specific Specific

gains for the clinicgains for the clinic

• Every clinic can log in on-line

• Own results versus the country

• Pre and postop values for EQ-5D

• Pre and postop values for pain, satisfaction and Charnley category

SICOT 2008PHE

EQ-5D index comparable with an aged and

gender matched population (0.76 – 16 000

inhabitants)

SICOT 2008PHE

Register gainswith awith astandardized followstandardized follow--up routineup routine

• Patient reported outcome is present

• Increased sensitivity with a complementary failure definition.

• Decreased number of “unrecorded”failures.

SICOT 2008PHE

• Since 2 years the National Board of Health and Welfare in Sweden desires inclusion of patient reported outcome in all Quality Registers

•…and “faster” performance indicators

Furthermore

SICOT 2008PHE

• Patient related parameters: pain satisfaction QoL

• Reoperation @ 2 years

• Revisions @ 5 and 10 years

• Cost-effectiveness analysis

Four outcome dimensions in the Register

SICOT 2008PHE

ReRe--op @ 2 year: dislocation/deep infection op @ 2 year: dislocation/deep infection

• high patient morbidity

• technically demanding

• very expensive

• high failure rates

• often bad patient related outcome

SICOT 2008PHEOpen variables from the Register per hospital Open variables from the Register per hospital

on the on the homehomepage:page:• 5-year implant survival

• 10-year survival

• reoperation @ 2 year

• satisfaction

• pain relief

• EQ-5D gain @ 1, (6 and 10 year)

• 90-days mortality

• cost

1999

2005

2006

SICOT 2008PHE

The Clinical Value Compass

ClinicalOutcome

Cost and Utility

Patient Satisfaction

Functional HealthQoL

Batalden and Nelson,

Dartmouth Medical School.

SICOT 2008PHE

Clinical Value CompassTHA surgery in Sweden - range of mean values

(+/-1SD)Patient Satisfaction

PainRelief

EQ-5D indexGained

90 DaysMortality

Costs

Reoperationsw ithin 2 years

ImplantSurvival 5 years


SICOT 2008PHE

Clinical Value CompassTHA surgery in Sweden


PainRelief

EQ-5D indexGained

90 DaysMortality

Costs


ImplantSurv. 5 years


Clinical Value CompassHospital A


PainRelief

EQ-5D indexGained

90 DaysMortality

Costs


ImplantSurv. 5 years


In this example, this hospital has worse outcome in five different dimensions (cardinals).

SICOT 2008PHE

Sundsvall

mainly revisions due to

recurrent dislocations

Sundsvall

Case mix

High volume central hospital

SICOT 2008PHE

Reoperation @ 2 years varies: 0 – 4,8% •

SICOT 2008PHE

• local analysis

• improvement programme

• no dislocation last year

• saving: 5 X 20.000 €

• this example shows the true mission of the Register“I wasn’t aware

of our highcomplication

rate.”

SICOT 2008PHE

The goal with open disclosure of clinical results is to initiate a local learning and

improvement process at each department

SICOT 2008PHE

• It will improve the entire process

• THR is not an operation - it is a procedure

Clinical Value Compass Thinking

SICOT 2008PHE

The case-mix factor is the largest individual factor that leads to misinterpretations of register results.

The Case-mix problem

Comparisons are difficult

SICOT 2008PHE

”Case-mix”-variablesnation-wide mean values (percentage)

Charnley CategoryA/B

OA

60 years or older

females

•gender•diagnosis•age•Charnley Category

The case-mix isdefined by specific

criteria:

SICOT 2008PHE

SICOT 2008PHE

The The CaseCase--MixMix ProblemProblem

The Charnley classification is a highlysignificant predictor concerning patient related outcome – both for disease-specificand generic instruments

SICOT 2008PHE

Patient reported resultsPatient reported results

Charnley category C patients varies

Preop frequency: 31 - 55%

SICOT 2008PHEPatient related outcome per clinic @ 1year

2002 - 2005

SICOT 2008PHE

90-day mortalityprimary THR 2003 - 2007

0‰

5‰

10‰

15‰

20‰

25‰

30‰

35‰

40‰

45‰

50‰

one tick - one unit

SICOT 2008PHEThe Clinical Value Compass

ClinicalOutcome

Cost and Utility


Functional HealthQoL

Batalden and Nelson,

Dartmouth Medical School.

SICOT 2008PHE

"... It´s a waste of time to concentrate on disease and costs.

Cost-effectiveness or utility of intervention should be measured!!...”

Alan Williams:

Health economical evaluation:

SICOT 2008PHE

• new scientific field

• decision makers often sceptical

– no incentive for long-term results

– not interested of total societal costs

– ”our budget is in balance!”

Health economy is controversial

SICOT 2008PHE

Health economical evaluation

The most important ”income” or profit in health care – is patient utility – quality of life improvement

SICOT 2008PHE

• linking with county databases concerning costs, resources, waiting lists …

Since 2004 cost is derived by

SICOT 2008PHE

• most exact reimbursement system in Sweden

• 40 of 79 units

• nation-wide (as a standard) introduced 2009 -2010

Cost Per Patient = CPP data base

SICOT 2008PHE

• mean cost = 78 000 SEK (12 800 $)

• range = 56 000 – 147 000 SEK

Nation wide mean costs:

SICOT 2008PHE


Costs ( A – B)

gained HRQoL x duration

Costs/QALY gainedQuality Adjusted Life Years

SICOT 2008PHE

Cost-utility: 78 000/(0,36 x 10)

QALY: 22 000 SEK (ca 3 540 $)Not adjusted for inflation, aging, death and reoperations!

SICOT 2008PHE

SICOT 2008PHE

• even better if all societal costs were known

Costs/QALY gained

SICOT 2008PHE

• 73 000 SEK /1 year (12 000 $)• 14 000 X 73 000 = 1 bil SEK!

Waiting time mean costs2 700 patients:

SICOT 2008PHE

THR is not only one of the best operations ever introduced but also one of the most

cost-effective


SICOT 2008PHE

We can and need to monitor and describe our current and changing

practice continuously, and then provide this information public to all parties

The Swedish Hip Register has shown

SICOT 2008PHE

Advantages of a National RegistryAdvantages of a National Registry

• Independent

• Prospective data

• Comparative outcomes

• Simultaneously compares all treatments

• Very large numbers

• Provides data that is not available from any other source

• Describes and Monitors current/changing practice

• Able to identify outcome outliers

SICOT 2008PHE

Advantages of a National RegistryAdvantages of a National Registry

• Includes all centers, no performance bias

• Wide applicability and relevance

• Can be used to answer multiple questions

• Answers questions not possible to do in any other way

• Hospital, Regional and International comparisons

• The information they provide improves performance

• Result in considerable savings /very cost effective

• Hypothesis generation

SICOT 2008PHE

Improvement of the ArthroplastyRegistries

• Establish a mechanism with the purpose of ensuring and maintaining continual quality improvement

• Provide hospital and community based comparative outcomes data

• Establish a mechanism for continuous public reporting of outcome measurements

SICOT 2008PHE

Thank You for Your Thank You for Your AttentionAttention

Department of OrthopaedicsSahlgrenska University Hospital,

Göteborg, Sweden

SICOT 2008PHE

ConclusionConclusion

• Outcome assessment through the Register has had a profound impact on total hip replacement care in Sweden.

• These efforts must be a continuous process – a steady state situation will never develop.

• Enormous potential for clinical research

SICOT 2008PHE

ConclusionConclusion

• For the healthcare providers– large economic savings and public information of results

• For the patient optimal treatment modalities identified and described on the web

• For the orthopaedic community outcome facts are present - confidential and public

SICOT 2008PHE

clinical and health economic relevance of the swedish...

Documents