Appendix A: Analysis slides of each case study a. Type 2 Diabetes b. - - PDF document

FP7-241741 – MANAGED OUTCOMES 31.12.2012 Deliverable 4: Report on inventory and analysis of European practises in selected countries

Appendix A: Analysis slides of each case study

• a. Type 2 Diabetes
• b. Stroke
• c. Hip Osteoarthritis
• d. Dementia

30-12-2012 1

MANAGED OUTCOMES

MANAGED OUTCOMES

An operations management and demand based approach to regional health service delivery systems for Diabetes type 2 patients

Meeting European health care system challenges by learning from differences between management practices in six EU countries

MANAGED OUTCOMES

Contents

• 1. Background information MO project
• 2. Results
• 3. Interpretation of results
• 4. Key findings

30-12-2012 2

MANAGED OUTCOMES

MANAGED OUTCOMES

1 Background MO project

MANAGED OUTCOMES

The Managed Outcomes Project

 All EU countries are experiencing the same problems in healthcare: the population is aging, causing an increasing demand for healthcare services: availability of trained personnel and funding is limited, while new medical treatments are more effective but more expensive. In order to cope with these constraints, the European healthcare systems need to improve to better consider the cost-beneficial provision of health outcomes, rather than just health outputs.  The MANAGED OUTCOMES project is based on the notion that healthcare

• utcomes and cost-benefits are affected by the efficiency of service production,

the regional structure of healthcare delivery and the degree to which people are empowered to participate in the co-production of their care. These relationships are insufficiently understood and need to be studied to meet the objectives of the European health strategy.  The main objective of Managed Outcomes is to develop and disseminate rich but practical conceptual models and a toolkit for improving the health service production system.

30-12-2012 3

MANAGED OUTCOMES

The Managed Outcomes project

 The project is performed by a consortium of universities and consultancy

• rganisations:
• Aalto University (AALTO) - Finland
• Erasmus University Rotterdam (EUR)- Netherlands
• Otto-Friedrich-Universität Bamberg (Universität Bamberg) - Germany
• Universidad Politechnica de Valencia (UPVLC) - Spain
• European Hospital and Healthcare Federation (HOPE)- Belgium
• Riel Miller - Xperidox Futures Consulting - France
• Ethniki Scholi Dimosias Ygeias Eidikos Llogariasmos Erevnon (NSPH) - Greece
• Balance of Care Group - UK
• Innovation in Leraning Institute (ILI) - Germany
• Forum Virium - Finland

 In six EU countries (FI, FG, GR, NL, SP, UK) cases studies are performed for four costly health care demands that are challenging EU healthcare systems:

• Diabetes type 2
• diabetes type 2
• Hip-osteoarthritis
• Dementia

MANAGED OUTCOMES

Diabetes type 2 services as EU healthcare challenge

• Diabetes type 2 is a disease with growing prevalence in the ageing

population in Europe.

• The cost of insulin dependent patients place a large burden on

healthcare expenditures.

• Though healthcare systems differ much between countries in the

EU, the treatment protocols used by medical, nursing and paramedical professionals are the same, so therefore there is a lot to learn from how we organize and manage the services for diabetes type 2 patients at operational level

• The differences in outcomes of diabetes type 2 health services in

different countries might be explained to a considerable extent in the way we organize the processes for delivering the services

30-12-2012 4

MANAGED OUTCOMES

Diabetes type 2 as a health systems challenge

 96% of diabetes mellitus cost is spent for type 2  A high proportion of patients with risk factors for diabetes-related complications are not adequately

• controlled. Improvements in disease management and

monitoring are therefore required to ensure that guideline targets are met, thus reducing the long-term complications of Type II diabetes.

Reference:

• A. Liebl, M. Mata, E. Eschwège (2002). Evaluation of risk factors for development of

complications in Type II diabetes in Europe. Diabetologia, 45:S23–S28

Reference: to be added

MANAGED OUTCOMES

Literature on diabetes type 2 (services)

• European Diabetes Policy Group (1999) Guidelines for diabetes

care: a desktop guide to type II diabetes mellitus. International Diabetes Federation (European Region). Diabet Med 16: 716–730

• Jönsson B (2000) Revealing the cost of Type II diabetes in Europe.

Diabetologia supplement (Code 2 study)

• A. Liebl, M. Mata, E. Eschwège (2002). Evaluation of risk factors for

development of complications in Type II diabetes in Europe. Diabetologia, 45:S23–S28

• additional literature also in WP3-report

30-12-2012 5

MANAGED OUTCOMES

Methodology

Demand Services User journey Resources Outcomes Operational model Operations Management Practice Population User Experiences Regional Setting Costs & Reimbursement

MANAGED OUTCOMES

Methodology

• Operational model: formal description of the demand, services,

user journey, resources and outcomes, and their quantitative relationships

• Operations management practice: the planning, management and

innovation of services; the collaboration between partners providing services

• Outcomes: the impact of services on health status, measured by

providers (quality indicators) and experienced by users (satisfaction)

• User experiences: the view of users on services and their

performance, measured in a survey

• Costs & reimbursement: the costing of resources for services and

the financing of services

30-12-2012 6

MANAGED OUTCOMES

Demarcation

MANAGED OUTCOMES

Region Keski-Suomi (Finland) Herakleion (Greece) NWN & DWO (Netherlands) Tower Hamlets (UK) Valencia, La Fé (Spain) Bamberg (Germany)

Region Keski-Suomi regiona Herakleion Regional Unit Nieuwe Waterweg… Tower Hamlets (London) La Fé (Valencia) Bamberg Population 247246 (2009) 304270 (2011) 443281 (2008) 233329 (2010) 202621 (2009) 214269 (2009) %population >40 years 53% 47% 49% 28% 49% 55% Area (km2) 14474 2641 273 21 133 1223 Population density (hab./km2) 17 115 1623 11272 8201 175

Case instance regions

30-12-2012 7

MANAGED OUTCOMES

Age groups

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

Keski-Suomi (FI) Bamberg (DE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK)

40-49 50-59 60-69 70-79 80+

MANAGED OUTCOMES

MANAGED OUTCOMES

2 Results

30-12-2012 8

MANAGED OUTCOMES

Demand segments

Number Name Description DS1 Prevention care Population that is targeted for prevention care. DS2 Diabetes care stage 1 Patients with diabetes type II needing lifestyle advice. DS3 Diabetes care stage 2 Patients with diabetes type II needing lifestyle advice and oral medication. DS4 Diabetes care stage 3 Patients with diabetes type II needing lifestyle advice, oral medication and insulin injections. DS5 Diabetes care stage 4 Patients with complicated diabetes type II needing specialized care.

MANAGED OUTCOMES

Patient journey - model

DS2 Exit DS3 DS4 DS5 Exit Exit

3 yrs 10 yrs 10 yrs 10 yrs 1.2% 3.2% 6.0% 0.5% 1.0% 3.0%

30-12-2012 9

MANAGED OUTCOMES

Patients Keski- Suomi (Finland) Herakleion (Greece) NWN & DWO (Netherlands) Tower Hamlets (UK) Valencia, La Fé (Spain) Bamberg (Germany) DS1 DS2 4758 2331 2683 5156 121 DS3 10086 13040 8084 4052 2090 DS4 2723 5991 1451 1516 1732 DS5 63 Total 17630 21362 12218 11203 10724 3943 Population 272784 299689 443281 330464* 233329* 202621 214269

Patients per DS

* Population registered at GP’s, not population in area

MANAGED OUTCOMES

Keski- Suomi (Finland) Herakleion (Greece) NWN & DWO (Netherlands) Tower Hamlets (UK) Valencia, La Fé (Spain) Bamberg (Germany) DS1 DS2 1.72% 0.78% 0.81% 3.18% 0.06% DS3 3.70% 4.35% 2.45% 2.00% 0.98% DS4 1.00% 2.00% 0.44% 0.75% 0.81% DS5 0.02% N/A N/A N/A N/A Total/ Area Total 6.4% 7.1% 3.7% 4,8% 5.9% 1.8% Prevalence DM (I+II) OECD data 2011 Age 20-79 5.7% 6.0% 5.3% 3,5% 6.6% 8.9%

Prevalence of Diabetes

30-12-2012 10

MANAGED OUTCOMES

Prevalence of diabetes

0% 1% 2% 3% 4% 5% 6% 7% 8%

Keski-Suomi (Finland) Herakleion (Greece) NWN & DWO (Netherlands) Tower Hamlets (UK) Valencia, La Fé (Spain)

Total DS4 DS3 DS2

MANAGED OUTCOMES

Length of stay in demand segments

10 20 30 40 Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) LoS in DS2 LoS in DS3 LoS in DS4

30-12-2012 11

MANAGED OUTCOMES

Relative distribution of DS2-DS4

0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 % Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) DS4 DS3 DS2

MANAGED OUTCOMES

Services

S1 - Screening S2 - Diagnosis S3 - Treatment with lifestyle advice S4 - Treatment with oral medication and lifestyle advice S5 - Insulin stabilization S6 - Treatment with insulin,

• ral

medication and lifestyle advice

DS2 DS4 DS3

30-12-2012 12

MANAGED OUTCOMES

Number of resources in diabetes care

2 4 6 8 10 12 14 16 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK)

MANAGED OUTCOMES

Resources: main care provider

Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) GP resource 60 € / hour 70 € / hour 42 € / hour 66 € / hour 31 € / hour 120 € / hour Nurse resource 29 € / hour 20 € / hour 10 € / hour 35 € / hour 23 € / hour 35 € / hour

30-12-2012 13

MANAGED OUTCOMES

Hours of care per service per year

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) S3 Lifestyle S4 Oral + lifestyle S6 Insuline+ Oral + Lifestyle

MANAGED OUTCOMES

Hours of care service S4 (oral medication)

1 2 3 4 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) Hours per year Follow–up visit & prescription of medicine Diet consultation Education Eye care Foot care

• ther

30-12-2012 14

MANAGED OUTCOMES

Hours of care service S6 (insulin care)

0,0 1,0 2,0 3,0 4,0 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) Hours per year Follow–up visit & prescription of medicine Diet consultation Education Eye care Foot care

• ther

Insulin injection by professional Delivering medication by professional

MANAGED OUTCOMES Hours of care over patient lifetime and LoS DS2-DS4

5 10 15 20 25 30 35 40 45 50 10 20 30 40 50 60 70 80 90 100 Years Hours

S5-S6 (DS4) S4 (DS3) S2-S3 (DS2) Total LoS in DS2-DS3-DS4

30-12-2012 15

MANAGED OUTCOMES

Average hours of care per year during patient life time DS2-DS4

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 Hours

MANAGED OUTCOMES

Costs of chronic diabetes care per year

200 400 600 800 1000 1200 1400 1600 1800 2000 S3 Lifestyle S4 Oral + lifestyle S6 Insuline+ Oral + Lifestyle Euros/yr Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK)

30-12-2012 16

MANAGED OUTCOMES

Costs of S4 per patient per year

100 200 300 400 500 600 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) Cost per year Lab test & analysis Follow–up visit&prescription medicine Eye care Foot care Self–test glucose monitoring Oral medication

• ther

Data missing

MANAGED OUTCOMES

Costs of S6 per patient per year

200 400 600 800 1000 1200 1400 1600 1800 2000 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) Cost per year (euros)

Lab test & analysis Follow–up visit&prescription medicine Diet consultation Eye care Foot care Self–test glucose monitoring Insulin medication Oral medication

Data missing

30-12-2012 17

MANAGED OUTCOMES Costs of care over patient lifetime and LoS in DS2-DS4

5 10 15 20 25 30 35 40 45 50 5000 10000 15000 20000 25000 Keski-Suomi (FI) Bamberg (GE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Years Euros

S5-S6 (DS4) S4 (DS3) S2-S3 (DS2) Total LoS in DS2-DS3-DS4 MANAGED OUTCOMES

Average costs of care per year per patient (lifetime)

€ - € 100 € 200 € 300 € 400 € 500 € 600 € 700 € 800 € 900 Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP)

30-12-2012 18

MANAGED OUTCOMES

Costs per balanced patient

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

Percentage of patients with HbA1c < 53 mmol/mol € - € 500 € 1 000 € 1 500 € 2 000 € 2 500

Cost per balanced patient per year

MANAGED OUTCOMES

Average costs per year

€ - € 100 € 200 € 300 € 400 € 500 € 600 € 700 € 800 € 900

avg cost per patient per year

€ - € 10 € 20 € 30 € 40 € 50 € 60 € 70

Cost per inhabitant

30-12-2012 19

MANAGED OUTCOMES

Operations managment and planning

Components Type Description Planning models Strategic This refers to the planning of services at the aggregation level of the unit or the

• rganization with a relatively long time horizon (in years).

Tactical This refers to the planning of services at an aggregate level (e.g. a group of patients) and at a medium time horizon (e.g. months or years). Operational This refers to the day to day planning of services for individual patients in which resources are allocated to an individual patient. Executional This addresses the real time management and planning of service operations. Operations management responsibilities and structure Organizational structure and hierarchy This refers to organizational structure and hierarchy (division into e.g. medical disciplines, service provision points, segments, et cetera). Improvement and

• rganizational

development Quality certificates European Foundation for Quality Management (EFQM) model, or EFQM based models and certificates. Quality strategy Quality management approaches, such as Total Quality Management, Lean Management, Six Sigma, Process optimization Process optimization methods, such as Business Process Reengineering, Theory of Constraints, et cetera. Information management IT support systems

• Tracking and tracing
• Electronic Patient Records
• Service Planning Software
• Demand management
• Workforce planning
• Financial information systems

MANAGED OUTCOMES

Operations management and planning

Components Type Finland Germany Greece Netherlands Spain United Kingdom Planning models Strategic Plan for access time Tactical Operational Executional

• Task division
• Detailed work

procedures in GP

• ffices
• Referral

procedure Operations management responsibilities and structure Organizational structure and hierarchy

• Organization to

coordinate all practices in the region Improvement and

• rganizational development

Quality certificates Quality improvement

• Visitatie (for GPs and

Physiotherapist)

• Guideline-based

practices

• Podoscreeners
• Fundusscreeners

Process

• ptimization

Information management IT support systems

• IPCI
• Zorgdomein
• Labsystem

30-12-2012 20

MANAGED OUTCOMES

Operations Management

Keski-Suomi (FI) Bamberg (DE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) Screening program Yes Yes Patient out-of-pocket for medication 0 - 10% Patient segmentation method Type of care Type of care Type of care Type of care Type of care Care balance National/regional diabetes care standard Yes Yes Yes Patient freedom of choice Low High High Collaboration 1 organized form of collaboration Some Yes 2 partners in diabetes service Some 3 steering group 4 working groups Some 5 development programme Yes Yes

MANAGED OUTCOMES

Operations Management

Keski-Suomi (FI) Bamberg (DE) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) Information management 1 Diabetes patient IT system partial Yes No 2 electronic patient record system fragmented Yes Yes 3 patient tracking system no Yes 4 performance measurement system no Yes Yes Process management 1 case manager Nurse GP GP 2 performance monitoring No Yes Yes (GP funding principle) Innovation 1 Quality improvement Yes Yes Yes 2 Process improvement Yes Yes Yes 3 Information systems improvement Yes 4 Regional diabetes plan Yes Yes Yes

30-12-2012 21

MANAGED OUTCOMES

Questionnaires

Country Investigated institutions Distributed Returned Response rate Included England 7 3343 475 14.2% 313 Finland 9 436 183 42.0% 183 Germany 5 462 286 61.9% 282 Greece 4 600 179 29.8% 179 The Netherlands 5 779 400 51.3% 387 Spain 1 625 115 18.4% 115 Total 31 6245 1638 26.2% 1459

MANAGED OUTCOMES

Health status (Survey)

10 20 30 40 50 60 70 80 90

England Finland Germany Greece The Netherlands Spain

EQ-5D utility index (Dolan 1997 EQ-5D visual analogue scale Satisfaction with health

30-12-2012 22

MANAGED OUTCOMES

Service (Survey)1

10 20 30 40 50 60 70 80 90 100

England Finland Germany Greece The Netherlands Spain

Service

ServQual comparison with best and worst service Satisfaction with service

MANAGED OUTCOMES

MANAGED OUTCOMES

3 Interpretation of results

30-12-2012 23

MANAGED OUTCOMES

Hours of care & balanced patients

Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 % balanced patients Hours of care (average per patient per year)

MANAGED OUTCOMES

Cost of care & balanced patients

Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 200 400 600 800 1000 % balanced patients Cost of care (average per patient per year)

30-12-2012 24

MANAGED OUTCOMES

Comparative analyses

• Method: Fixed Proportion Technology

Comparisons by calculating indices:

1. Calculate indices: inputi/ outcomej 2. Normalize and compare each index with the minimum (best practice) 3. Calculate average over all indices

Darold T. Barnum and John M. Gleason, Measuring efficiency under fixed proportion technologies, Journal of Productivity Analysis, Volume 35, Number 3 (2011), 243-262.

• Adaptations for Managed Outcomes Project

1. Use one country as reference point 2. Calculate the difference with reference country for each case instance

• Remarks:

– We are studying further how we can apply this technique for presenting our results. At the moment we have sometimes interpreted input and output in a less strict sense which is not impeccable; also when we use survey results as output and calculate ratios, this is not fully appropriate, as the underlying scale is an interval scale and the FPT method assumes linearity. – Therefore the results in slides 46-56 will be confirmed only after peer-review in scientific journal publication process. All the results are aimed to be published by the consortium and the partners.

MANAGED OUTCOMES Hours of care compared with patient outcomes

Analysis Input Output Hours of care to perceived health

• Average yearly hours of

care S2-S6

• EQ-5D utility index
• EQ-5D visual analogue scale
• Satisfaction with health

Hours of care to health status • Average yearly hours of care S2-S6

• HbA1c % < 53 mmol/mol
• Problems with lower extremities
• Problems with sight

30-12-2012 25

MANAGED OUTCOMES

Average yearly hours of care to health outcomes

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

Average yearly hours of care to patient perceived health

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

Average yearly hours of care to patient clinical outcomes

MANAGED OUTCOMES

Average yearly hours of care, perceived health and clinical

• utcomes

Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 0,2 0,4 0,6 0,8 1 Clinical outcomes Perceived health

30-12-2012 26

MANAGED OUTCOMES

Average yearly cost of care compared with patient outcomes

Analysis Input Output Costs of care to perceived health

• Average yearly cost
• EQ-5D utility index
• EQ-5D visual analogue scale
• Satisfaction with health

Costs of care to health status

• Average yearly cost
• HbA1c % < 53 mmol/mol
• Problems with lower extremities
• Problems with sight

MANAGED OUTCOMES

Costs of care, perceived health and clinical outcomes

Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 0,2 0,4 0,6 0,8 1 Clinical outcomes Perceived health

30-12-2012 27

MANAGED OUTCOMES

Secondary prevention

Analysis Input Output

prevention for complications

• Service volume S3
• Service Volume S4
• No problems with lower

extremities

• No problems with sight

prevention for insulin dependancy

• Service volume S3
• Service Volume S4
• Avg time before insulin

treatment

MANAGED OUTCOMES

Secondary prevention (service hours against outcomes

Keski-Suomi (FI) Herakleion (GR) NWN & DWO (NL) Valencia (SP) Tower Hamlets (UK) 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 Complications (extremities/sight) Movement to insulin dependancy

30-12-2012 28

MANAGED OUTCOMES

MANAGED OUTCOMES

4 Key findings

MANAGED OUTCOMES Key findings: main differences in key demand

and operations parameters

• Relative distribution of patients over demand segments:

– DS2: 11-48% – DS4: 12-28%

• The length of stay in demand segments:

– DS2: 2-5 yrs – DS3: 9-13 yrs – DS4: 8-19 yrs

• The number of hours of direct care per patient lifetime:

– 70-90 hrs

• The main care-giver:

– GP or nurse

• Costs per patient year during patient lifetime:

– 609 – 1081 Euro’s

30-12-2012 29

MANAGED OUTCOMES Key findings: main differences in key outcome

parameters

• Health status (EQ5D, survey)

– 0.67-0.81 (Dolan index)

• Quality of services

– 68-90 (SERVQUAL-score)

• Percentage of patients with HbA1c < 53 mmol/mol

– 42-77%

• Complications

– Lower extremities: 6-17% – Eyesight: 9-29%

MANAGED OUTCOMES

Key findings: regional systems

• The number of care-givers and impact on
• utcomes
• The main care-giver and outcomes
• The primary-secondary roles in diabetes care

30-12-2012 30

MANAGED OUTCOMES

Key findings: good practices

• The amount of care delivered does not necessarily corresponds

with the best health outcomes

• There are huge differences in the costs per demand segment.

Regions that are successful in keeping patients in the lower demand segments (Keski Suomi and Valencia) are more cost-

• effective. However, Keski-Suomi is spending much more money.
• NWO-DWD and Valencia are doing well in finding a balance

between efficiency and outcomes

– Valencia puts less efforts in it and has relative good outcomes – NWN DWO puts more efforts in it and has better outcomes, as well as perceived as clinical

• NWO-DWD and Valencia are doing well preventing complications

and insulin dependency

• Care frequency and adherence to diet and medication

MANAGED OUTCOMES

Key findings: recommendations

• The Managed Outcomes approach provides a

multiperspective view on the design and the performance of the regional diabetes care delivery system; This facilitates a systematic approach to improvement of diabetes care.

• There is a lot of difference between health care practices

in provision of services; further study to relate the

• perational description to evidence based clinical

guidelines

• Improving the operational modelling (now aggregate

description) by collecting operations data to individual patients

30.12.2012 1

MANAGED OUTCOMES

MANAGED OUTCOMES

An operations management and demand based approach to regional health service delivery systems for stroke patients

Meeting European health care system challenges by learning from differences between management practices in six EU countries

MANAGED OUTCOMES

Contents

• 1. Background information MO project
• 2. Results
• 3. Relating inputs to outcomes
• 4. Key findings

30.12.2012 2

MANAGED OUTCOMES

MANAGED OUTCOMES

1 Background MO project

MANAGED OUTCOMES

The Managed Outcomes Project

 All EU countries are experiencing the same problems in healthcare: the population is aging, causing an increasing demand for healthcare services: availability of trained personnel and funding is limited, while new medical treatments are more effective but more expensive. In order to cope with these constraints, the European healthcare systems need to improve to better consider the cost-beneficial provision of health outcomes, rather than just health outputs.  The MANAGED OUTCOMES project is based on the notion that healthcare

• utcomes and cost-benefits are affected by the efficiency of service production,

the regional structure of healthcare delivery and the degree to which people are empowered to participate in the co-production of their care. These relationships are insufficiently understood and need to be studied to meet the objectives of the European health strategy.  The main objective of Managed Outcomes is to develop and disseminate rich but practical conceptual models and a toolkit for improving the health service production system.

30.12.2012 3

MANAGED OUTCOMES

The Managed Outcomes project

 The project is performed by a consortium of universities and consultancy

• rganisations:
• Aalto University (AALTO) - Finland
• Erasmus University Rotterdam (EUR)- Netherlands
• Otto-Friedrich-Universität Bamberg (Universität Bamberg) - Germany
• Universidad Politechnica de Valencia (UPVLC) - Spain
• European Hospital and Healthcare Federation (HOPE)- Belgium
• Riel Miller - Xperidox Futures Consulting - France
• Ethniki Scholi Dimosias Ygeias Eidikos Llogariasmos Erevnon (NSPH) - Greece
• Balance of Care Group - UK
• Innovation in Learning Institute (ILI) - Germany
• Forum Virium - Finland

 In six EU countries (FI, FG, GR, NL, SP, UK) cases studies are performed for four costly health care demands that are challenging EU healthcare systems:

• Diabetes type 2
• Stroke
• Hip-osteoarthritis
• Dementia

MANAGED OUTCOMES

Stroke networks as EU healthcare challenge

• Stroke is one of the most occurring causes of death in the EU
• The services required for treating stroke and care for stroke

patients are very costly and place a large burden on health care systems, patients and families

• Though healthcare systems differ much between countries in the

EU, the treatment protocols used by medical, nursing and paramedical professionals are the same, so therefore there is a lot to learn from how we organize and manage the services for stroke patients at operational level

• The huge differences in outcomes of stroke services in different

countries might be explained to a considerable extent in the way we organize the processes for delivering the services

30.12.2012 4

MANAGED OUTCOMES

Stroke disease

Rapidly developing clinical signs of focal

• r global disturbance of cerebral

function leading to death or lasting more than 24 hours with no apparent cause other than a vascular one.

1WHO Stroke control programme, 1972

MANAGED OUTCOMES

Helsingborg declaration on stroke

• Pan European Consensus meeting on stroke management in 1995 in Helsinborg

Sweden, organised by the WHO Aboderin I, Venables G, Asplund K. Stroke management in Europe. Journal of Internal Medicine, 1996, 240:173-180

• Second Helsingborg meeting in 2006. Targets for stroke to be realised in 2015
• n:

– Organization of stroke services: all patients with stroke will have access to a continuum of care from organized stroke units in the acute phase to appropriate rehabilitation and secondary prevention measures – Management of acute stroke: more than 85% of stroke patients survive the first month after stroke – Prevention: stroke mortality is reduced with at least 20% from the level of 2005 – Rehabilitation: more than 70% of surviving patients are within three months independent in their activities of daily living – Evaluation: established system for routine collecting of data required for evaluation

• f services
• K. Kjellström, B. Norving, A. Shatchkute. Helsingborg declaration 2006 on

European stroke strategies. Cerebrovascular diseases 2007;23:229-241

30.12.2012 5

MANAGED OUTCOMES

The six EU case study settings

Region Population Population density Age 70+ (%) Case – National Incidence ischemic stroke Case - National Incidence haemorrhagic stroke Case - National FI: Keski-Suomi 273 000 14 13% - 12% 197 - 182 40 – 48 FG: Erlangen 236.264 368 12% - 15% 110 - 23 - GR: Athens 3 191 329* 1076 10% - 10% 107 - 19 - NL: Tilburg 341.313 492 10% - 10% 198 – 239 26 – 42 SP: Valencia 266 320 2002 11% - 12% 179 - 188 107 - UK: Brighton 365 000 3044 12% - 12% 123 - 185 36 - 33*

*National incidence data based on NICE/England

MANAGED OUTCOMES

The six stroke case studies

Region Hospitals involved Size (# beds) Total cases Ischemic strokes Hemorrhagic Strokes FI: Keski-Suomi Keski-Suomi Central Hospital (KSCH) 400 756 538 238 FG: Erlangen University Hospital Erlangen 1300 613 508 105 GR: Athens Alexandra University General Hospital of Athens (AUGHA) 482 181 150 26 NL: Tilburg

• St. Elisabeth Hospital (EH)

Tweesteden Hospital (TH) 673 576 773 657 88 SP: Valencia Hospital La Fe (FH) 1297* 763 478 285 UK: Brighton Royal Sussex County Hospital (RSCH) 600 579 449 130

*Number of beds reduced in new built hospital 2010

30.12.2012 6

MANAGED OUTCOMES

The six stroke networks

Region Partners in stroke network FI: Keski-Suomi (KSCH) Ambulance, Emergency department (24/7), stroke unit / neurology ward, neurosurgery / neurosurgical ward (Kuopio University hospital), rehabilitation centres in primary care, nursing homes FG: Erlangen Ambulance, Emergency department (24/7), stroke unit / neurology ward, neurosurgery / neurosurgical ward, homecare / rehabilitation centre GR: Athens (AUGHA) Ambulance, Emergency department (24/every 4th day, shared service with 13

• ther hospitals), stroke unit (5 beds) / internal medicine ward, neurosurgery /

neurosurgical ward (other hospital) NL: Tilburg (EH, TH) Ambulance, Emergency department (24/7), stroke unit / neurology ward, neurosurgery / neurosurgical ward (EH), homecare, rehabilitation centre , nursing homes SP: Valencia (FH) Ambulance, Emergency department (24/7), stroke unit / neurology ward, neurosurgery / neurosurgical ward, rehabilitation unit, hospital at home, nursing homes UK: Brighton (RSCH) Ambulance, Accident & Emergency department (24/7), stroke unit / neurology ward, neurosurgery / neurosurgical ward (other hospital), community rehabilitation team / step down facility

MANAGED OUTCOMES

Methodology

Demand Services User journey Resources Outcomes Operational model Operations Management Practice Population User Experiences Regional Setting Costs & Reimbursement

30.12.2012 7

MANAGED OUTCOMES

Methodology

• Operational model: formal description of the demand, services,

user journey, resources and outcomes, and their quantitative relationships

• Operations management practice: the planning, management and

innovation of services; the collaboration between partners providing services

• Outcomes: health status, measured by providers (quality

indicators) and experienced by users, satisfaction

• User experiences: the view of users on services and their

performance, measured in a survey

• Costs & reimbursement: the costing of resources for services and

the financing of services

MANAGED OUTCOMES

Literature on stroke (services)

• Leonid Churilov, Geoffrey A. Donnan. Operations Research for

stroke care systems: an opportunity for The Science of Better to do much better. Operations Research for Health Care, Volume 1, Issue 1, March 2012, pp. 6-15.

• Soojin Park and Lee. H. Schwamm. Organizing regional stroke

systems of care. Current Opinion in Neurology, 2008,21, pp.43-55.

30.12.2012 8

MANAGED OUTCOMES

Demarcation

Focusing on acute hospital care involving specialist stroke services. Excluding prevention and long term rehabilitation

MANAGED OUTCOMES

MANAGED OUTCOMES

2 Results

30.12.2012 9

MANAGED OUTCOMES

Data

• Operational model: incidence, number of stroke cases, services,

resources, costs: 2009/2010 data

• Operational performance: access time, percentage trombolysis,

door-to-needle time, length of stay, mortality : 2009/2010 data

• Patient survey: questionnaire in 2011

Keski- Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Questionnaires distributed 600 366 126 625 306 346 Response rate 31.7% 34.4% 51.6% 35.8% 33.0% 34.7% Questionnaires included 160 110 52 210 72 94 percentage males (s.) 42,0% 61,2% 46,2% 60,2% 62,9% 67,0% average age (s.) 69,8 67,0 73,9 70,1 66,0 73,9 percentage ischemic strokes NA 88,2% 80,8% 91,4% NA 100,0%

MANAGED OUTCOMES

Analysis stroke process and performance

Demand Services Resources Outcomes User journey

Population structure Stroke incidence Design of services Service mix Hyper acute service delivery Use of services Length of stay Use of resources Clinical outcomes Content of services Patient experiences Unit costs of resources Patient reported

• utcomes

Costs & outcomes Costs of stroke services

30.12.2012 10

MANAGED OUTCOMES

Results Contents

Demand Services Resources Outcomes User journey

Population structure Stroke incidence Design of services Service mix Hyper acute service delivery Use of services Length of stay Use of resources Clinical outcomes Content of services Patient experiences Unit costs of resources Patient reported

• utcomes

Costs & outcomes Costs of stroke services

MANAGED OUTCOMES

Population structure (operational model)

GR national data = 1st Attica region; UK national data = Sussex region

0% 2% 4% 6% 8% 10% 12% 14% 16% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

70+ years, % of population in case study area 70+ years, % of population in country

30.12.2012 11

MANAGED OUTCOMES

Incidence ischemic & hemorrhagic strokes

(operational model)

GR: incidence of 319,4 per 100.000 inhabitants 45-84 years based on paper from 1999, translated to overall incidence figure 50 100 150 200 250 300 350 Case National Case National Case National Case National Case National Case National Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

Ischemic Hemorrhagic

MANAGED OUTCOMES

Demand Services Resources Outcomes User journey

Population structure Stroke incidence Design of services Service mix Hyper acute service delivery Use of services Length of stay Use of resources Clinical outcomes Content of services Patient experiences Unit costs of resources Patient reported

• utcomes

Costs & outcomes Costs of stroke services

Results Contents

30.12.2012 12

MANAGED OUTCOMES

Design of services (operational model)

0% 20% 40% 60% 80% 100% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Percentage percentage ischemic strokes via stroke unit percentage ischemic strokes direct to ward percentage neurosurgical hemorrhagic strokes operated elsewhere Criteria for stroke unit GR/SP/UK: ischemic stroke less than 24 hours evolution or Ischemic stroke in progression or TIA repetition. Excluding: patients with Barthel < 85, Rankin > 2, disabilities, short life expectancy, dementia For case instances with a split of flows, we will use DS4a as the flow via the stroke unit and DS4b as the flow directly to a medical ward

MANAGED OUTCOMES

Service mix (operational model)

5 10 15 20 25 30

Diagnosis (S1) Treatment ischemic stroke (S2) Rehabilitation (S4) Rehabilitation (S4) 6 4 5 5 7 5 5 7 7 Treatment ischemic stroke (S2) 9 3 7 6 10 9 7 6 3 Diagnosis (S1) 10 10 14 14 13 16 16 12 12 DS4a DS4b DS4a DS4b DS4a DS4b Keski- Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

Erlangen data incomplete DS4a flow via stroke unit, DS4b: flow directly to medical ward

30.12.2012 13

MANAGED OUTCOMES

Hyper acute phase: service delivery

Onset to specialist care (h) (operational performance) 1 2 3 4 5 6 7 8 9 10 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Door-to-needle time in ED (m) (operational performance) 30 40 50 60 70 80 90 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Hours until diagnosis (survey) 2 4 6 8 10 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Time from ED arrival to CT-scan (h) (operational performance) 0,5 1 1,5 2 2,5 3 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Door-to-door time in ED (h) (operational performance) 1 2 3 4 5 6 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Minutes until medical help to arrive (survey) 10 20 30 40 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

MANAGED OUTCOMES

Thrombolysis

Percentage trombolysis 0% 2% 4% 6% 8% 10% 12% 14% 16% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

30.12.2012 14

MANAGED OUTCOMES

Results Contents

Demand Services Resources Outcomes User journey

Population structure Stroke incidence Design of services Service mix Hyper acute service delivery Use of services Length of stay Use of resources Clinical outcomes Content of services Patient experiences Unit costs of resources Patient reported

• utcomes

Costs & outcomes Costs of stroke services

MANAGED OUTCOMES

Consultation and advice (survey)

0% 20% 40% 60% 80% 100% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

Advice impact of stroke on life (s.) Discussion impact of stroke on life Advice concerning risk factors (s.) Discussion advice concerning risk factors

30.12.2012 15

MANAGED OUTCOMES

Rehabilitation practices (survey)

0% 20% 40% 60% 80% 100% 120% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) 0,0 5,0 10,0 15,0 20,0 25,0 30,0 Rehabilitation prescribed (s.) Rehabilitation received Days until rehabilitation (average) (s.) MANAGED OUTCOMES

Therapeutic actions (survey)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Special medication prescribed (s.) Special diet recommended (s.) Special activity recommended (s.) Special therapy recommended (s.)

30.12.2012 16

MANAGED OUTCOMES

Length of stay (operational performance & survey)

Length of stay (operational performance)

2 4 6 8 10 12 14 16 DS4a DS4b DS4a DS4b DS4a DS4b Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Length of stay in stroke unit or ward (d) Length of stay elsewhere in hospital (d) Nights at hospital (survey) 4 8 12 16 20 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

SP:

RHB unit in hospital MANAGED OUTCOMES

5,0 5,2 5,4 5,6 5,8 6,0 6,2 6,4 6,6 6,8 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Up-to-date equipment (s.) Service on time React promptly Polite Personal attention (s.) Careful communication (s.)

Quality of services (survey)

30.12.2012 17

MANAGED OUTCOMES

Quality of services (survey)

72,0 74,0 76,0 78,0 80,0 82,0 84,0 86,0 88,0 90,0 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) SERVQUAL short form Evaluation in comparison with best and worst imaginable service Satisfaction with service

MANAGED OUTCOMES

Results Contents

Demand Services Resources Outcomes User journey

Population structure Stroke incidence Design of services Service mix Hyper acute service delivery Use of services Length of stay Use of resources Clinical outcomes Content of services Patient experiences Unit costs of resources Patient reported

• utcomes

Costs & outcomes Costs of stroke services

30.12.2012 18

MANAGED OUTCOMES Use of rehabilitation resources (operational model)

(physio, speech, occupational, social worker, diet)

Total minutes of rehabilitation services

50 100 150 200 250 300 Keski-Suomi (FI) Erlangen (FG) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

Total minutes of rehabilitation services

50 100 150 200 250 300 DS4a DS4b DS4a DS4b DS4a DS4b Keski- Suomi Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Total minutes of rehabilitation services

MANAGED OUTCOMES

Unit costs of resources (selection operational model)

Keski- Suomi (FI) Erlangen (FG) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Ambulance 500 37,5 310 310 310 ED care 302 11 180 124 124 Stroke unit 1300 661 412 584 450 Neurology ward 500 299 320 Medical ward 560 400 Physiotherapy 76 6 33 20 20 Discharge ward 150 150 Shortstay ward 224

30.12.2012 19

MANAGED OUTCOMES

Results Contents

Demand Services Resources Outcomes User journey

Population structure Stroke incidence Design of services Service mix Hyper acute service delivery Use of services Length of stay Use of resources Clinical outcomes Content of services Patient experiences Unit costs of resources Patient reported

• utcomes

Costs & outcomes Costs of stroke services

MANAGED OUTCOMES

Clinical Outcomes (operational performance)

2 4 6 8 10 12 14 DS4a DS4b DS4a DS4b DS4a DS4b Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Mortality ischemic stroke during hospital stay Mortality ischemic stroke 1 month Mortality ischemic stroke 3 month Mortality ischemic stroke 12 month Mortality Ischemic stroke OECD 1 month

30.12.2012 20

MANAGED OUTCOMES

Patient reported Outcomes (survey)

Health status at admission (self-reported Modified Rankin Scale)

0% 20% 40% 60% 80% 100% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

No symptoms at all Symptoms, but able to carry out usual duties and activities Unable to perform all usual activities, but able to look after own affairs Some help required, able to walk Unable to walk without assistance Bedridden

Health status at survey (self-reported Modified Rankin Scale)

0% 20% 40% 60% 80% 100% Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

No symptoms at all Symptoms, but able to carry out usual duties and activities Unable to perform all usual activities, but able to look after ow n affairs Some help required, able to w alk Unable to w alk w ithout assistance Bedridden

MANAGED OUTCOMES

Patient reported Outcomes (survey)

Health status (self-reported Modified Rankin Scale, states 3-6, survey)

• 20,0

0,0 20,0 40,0 60,0 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

Status immediately after stroke (s.) Status at time of survey (s.) Change in status before-after

30.12.2012 21

MANAGED OUTCOMES

Patient reported Outcomes (survey)

Health status at survey

0,54 0,56 0,58 0,6 0,62 0,64 0,66 0,68 0,7 0,72 0,74 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) 58 60 62 64 66 68 70 72 Dolan utility index for EQ-5D Visual analoque scale of EQ-5D

MANAGED OUTCOMES

Patient reported Outcomes (survey)

44 46 48 50 52 54 56 58 60 62 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) 10 20 30 40 50 60 70 80 90 Satisfaction with health Adaptation of life to impairments caused by stroke

30.12.2012 22

MANAGED OUTCOMES

Costs of stroke services

Costs of stroke services

1000 2000 3000 4000 5000 6000 7000 8000 9000 Keski-Suomi (FI) Erlangen (FG) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) Diagnosis (S1) Treatment Ischemic stroke (S3) Rehabilitation (S4) MANAGED OUTCOMES

MANAGED OUTCOMES

Relating inputs to outcomes 3 Comparative analysis

30.12.2012 23

MANAGED OUTCOMES

Comparative analyses input-outcomes

• Method: Fixed Proportion Technology (FPT)

Comparisons by calculating indices:

1. Calculate indices: inputi/ outputj 2. Standardize and compare index with minimum (best practice) 3. Calculate average over all indices

Darold T. Barnum and John M. Gleason, Measuring efficiency under fixed proportion technologies, Journal of Productivity Analysis, Volume 35, Number 3 (2011), 243-262.

• Remarks:

– We are studying further how we can apply this technique for presenting our

• results. At the moment we have sometimes interpreted input and output in a

less strict sense which is not impeccable; also when we use survey results as

• utput and calculate ratios, this is not fully appropriate, as the underlying scale is

an interval scale and the FPT method assumes linearity. – Therefore the results in slides 46-56 will be confirmed only after peer-review in scientific journal publication process. All the results are aimed to be published by the consortium and the partners.

MANAGED OUTCOMES 1 rehabilitation efforts and health outcomes

Analysis Input Output Rehabilitation efforts versus perceived health

• Minutes of rehabilitation

services provided

• EQ-5D Dolan utility index
• EQ-5D visual analogue scale
• Satisfaction with health
• Change in health status

30.12.2012 24

MANAGED OUTCOMES Rehabilitation efforts versus perceived health

Efficiency rehabilitation efforts versus perceived health 0,2 0,4 0,6 0,8 1 Keski-Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK)

MANAGED OUTCOMES

Analysis Input Output Cost of stroke service versus perceived health

• Costs of ischemic stroke
• EQ-5D Dolan utility index
• EQ-5D visual analogue scale
• Satisfaction with health
• Change in health status

Costs of stroke service versus clinical outcomes

• Costs of ischemic stroke
• Mortality ischemic stroke, during

hospital stay

• Mortality ischemic stroke, one

month

2 Costs of stroke service and health outcomes

30.12.2012 25

MANAGED OUTCOMES

Costs of stroke service and health outcomes

Costs versus health outcomes

Keski-Suomi (FI) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9

Costs versus clinical outcomes Costs versus perceived health MANAGED OUTCOMES

Analysis Input Output Speed of delivery in hyper-acute phase versus perceived health

• Onset to specialist care
• Arrival ED to CT-scan
• Door-to-needle time
• Minutes until medical help
• Hours until diagnosis
• EQ-5D Dolan utility index
• EQ-5D visual analogue scale
• Satisfaction with health
• Change in health status

Speed of delivery in hyper-acute phase versus clinical outcomes

• Onset to specialist care
• Arrival ED to CT-scan
• Door-to-needle time
• Minutes until medical help
• Hours until diagnosis
• Mortality ischemic stroke,

during hospital stay

• Mortality ischemic stroke, one

month

3 Speed of delivery and health outcomes

30.12.2012 26

MANAGED OUTCOMES

Speed of delivery and health outcomes

MANAGED OUTCOMES

4 Consultation and perceived health

Analysis Input Output Risk advice versus perceived health

• Advice on impact on life
• Discussion on impact
• Advice on risk factors
• Discussion on rik factors
• EQ-5D utility index
• EQ-5D visual analogue scale
• Satisfaction with health
• Change in health status

Therapy advice versus perceived health

• Special medication
• Special diet
• Special activity
• Special therapy
• Mortality ischemic stroke during

hospital stay

• Mortality ischemic stroke, one

month

30.12.2012 27

MANAGED OUTCOMES

Consultation versus perceived health

MANAGED OUTCOMES 5 Service performance and service outcomes

Analysis Input Output Speed versus service outcomes

• Onset to specialist care
• Arrival ED to CT-scan
• Door-to-needle time
• Minutes until medical

help

• Hours until diagnosis
• ServQUAL
• Evalation in comparison with best

and worst scenario

• Satisfaction with services

Costs versus service outcomes

• Costs of ischemic stroke
• ServQUAL
• Evalation in comparison with best

and worst scenario

• Satisfaction with services

30.12.2012 28

MANAGED OUTCOMES Service performance versus service outcomes MANAGED OUTCOMES

Overview of analyses on efficiency

Analysis Keski- Suomi (FI) Erlangen (G) Athens (GR) Tilburg (NL) Valencia (SP) Brighton (UK) 1 Rehabilitation efforts versus health outcomes Service time versus perceived health 0,22 0,33 0,42 0,33 1,00 0,46 2 Costs versus health outcomes Costs versus perceived health 0,48 0,48 0,44 0,80 0,76 0,68 Costs versus clinical outcomes 0,22 0,30 0,49 1,00 0,79 3 Speed versus health outcomes Speed versus perceived health 0,54 0,60 0,69 0,77 0,26 0,47 Speed versus clinical outcome 0,45 0,79 0,80 0,60 0,75 4 Consult and recommendation versus perceived health Risk factor advice vs. perceived health 0,86 0,75 0,61 0,87 0,62 0,88 Therapy recommendation vs. perceived health 0,50 0,58 0,35 0,74 0,53 0,70 5 Rehabilitation versus perceived health Rehabilitation vs. perceived health 0,64 0,64 0,60 0,75 0,60 0,66 6 Service performance and service outcomes Costs versus service outcomes 0,48 0,48 0,52 0,65 1,00 0,62 Speed versus service outcomes 0,56 0,64 0,81 0,67 0,36 0,46

30.12.2012 29

MANAGED OUTCOMES

4 Key findings

MANAGED OUTCOMES

Key findings: range in inputs &performance

• There are large variations in practices

Average Minimum Maximum 1 Incidence 152 107 198 2 Thrombolysis 8% 3% 14.3% 3 Minutes of rehabilitation 231 172 319 4 Length of stay 10.6 8.0 13.9 5 Costs 6989 4193 8504 Average Minimum Maximum 1 Time ED-CT 0.95 0.25 2.6 2 Mortality during stay 6.3% 3% 10.1% 3 Mortality 1 month 7.3% 3.7% 13.1% 4 Mortality 1 month OECD 9.9% 5.8% 12.9% 5 Satisfaction with service 83.0 77.8 86.9 6 Satisfaction with health 0.67 0.60 0.72

30.12.2012 30

MANAGED OUTCOMES

Key findings: range OM practice

• Differences in production system:

– All stroke patients via stroke unit and then to neurology ward

• r triage at ED for split between stroke unit flow medical ward flow

– Hemorrhagic strokes operated at own hospital or in another more specialized hospital – Role of the stroke unit: central focus point of the stroke service or partner in the chain

MANAGED OUTCOMES

Key findings: regional systems

• Differences in regional embedding:

– Some stroke services are consisting of ambulance, stroke unit and medical ward – Others are embedded in a regional structure in which all partners in stroke (general practitioners, ambulance, hospital, rehabilitation centre, nursing home, homes for elderly, home care) participate – The regional collaboration around stroke services ranges between a very light structure with a meeting of partners once a year to a platform for evaluation and development of the stroke services

30.12.2012 31

MANAGED OUTCOMES

Key findings: good practices

• The best performance for costs related to health outcomes is found in the

cluster Valencia-Tilburg-Brighton (see slide 51):

– Valencia due to the low costs despite lower health outcomes – Tilburg and Brighton due to the high health outcomes despite high costs

• As there are large differences in unit costs per service, it is better to look at
• ther measures that generate costs such as the amount of efforts put in

rehabilitation or the length of stay. The best performance in rehabilitation efforts related to perceived health is realized in Valencia, due to the low number of minutes available for rehabilitation during hospital stay.

• The best performance of speed of service delivery in the hyper-acute phase

related to health outcomes is realized in Tilburg-Athens-Brighton:

– Tilburg due to the combined performance on speed and health outcomes – Athens and Brighton more due to the high performance on speed

• The best performance of advice efforts on risks and therapies related to

perceived health is realized in Tilburg and Brighton

• The best performance of service performance related to service outcomes is

realized in Athens-Tilburg-Valencia:

– Valencia due to the low costs – Athens and Tilburg due to the high speed of delivery in the hyper-acute phase

MANAGED OUTCOMES

Key findings: recommendations

• The Managed Outcomes approach provides a

multiperspective view on the design and the performance of the regional stroke services delivery

• system. Adoption of this approach in stroke services by

management supports a systematic approach to improvement of stroke services.

• There are large differences between health care practices

in provision of services; further research is required to relate the operational description to evidence based clinical guidelines

• Improving the operational modelling (now based on

aggregate description by experts) by collecting data on

• perational performance on individual patients.

30.12.2012 1

MANAGED OUTCOMES

MANAGED OUTCOMES

An operations management and demand based approach to regional health service delivery systems for hip osteoarthritis patients

Meeting European health care system challenges by learning from differences between management practices in six EU countries

MANAGED OUTCOMES

Contents

• 1. Background information MO project
• 2. Results
• 3. Interpretation of results
• 4. Key findings

30.12.2012 2

MANAGED OUTCOMES

MANAGED OUTCOMES

1 Background MO project

MANAGED OUTCOMES

The Managed Outcomes Project

 All EU countries are experiencing the same problems in healthcare: the population is aging, causing an increasing demand for healthcare services: availability of trained personnel and funding is limited, while new medical treatments are more effective but more expensive. In order to cope with these constraints, the European healthcare systems need to improve to better consider the cost-beneficial provision of health outcomes, rather than just health outputs.  The MANAGED OUTCOMES project is based on the notion that healthcare

• utcomes and cost-benefits are affected by the efficiency of service production,

the regional structure of healthcare delivery and the degree to which people are empowered to participate in the co-production of their care. These relationships are insufficiently understood and need to be studied to meet the objectives of the European health strategy.  The main objective of Managed Outcomes is to develop and disseminate rich but practical conceptual models and a toolkit for improving the health service production system.

30.12.2012 3

MANAGED OUTCOMES

The Managed Outcomes project

 The project is performed by a consortium of universities and consultancy

• rganisations:
• Aalto University (AALTO) - Finland
• Erasmus University Rotterdam (EUR)- Netherlands
• Otto-Friedrich-Universität Bamberg (Universität Bamberg) - Germany
• Universidad Politechnica de Valencia (UPVLC) - Spain
• European Hospital and Healthcare Federation (HOPE)- Belgium
• Riel Miller - Xperidox Futures Consulting - France
• Ethniki Scholi Dimosias Ygeias Eidikos Llogariasmos Erevnon (NSPH) - Greece
• Balance of Care Group - UK
• Innovation in Leraning Institute (ILI) - Germany
• Forum Virium - Finland

 In six EU countries (FI, FG, GR, NL, SP, UK) cases studies are performed for four costly health care demands that are challenging EU healthcare systems:

• Diabetes type 2
• Stroke
• Hip-osteoarthritis
• Dementia

MANAGED OUTCOMES

Hip osteoarthritis

• Osteoarthritis is the most common type of arthritis, especially

among older people

• Osteoarthritis of the hip can result from several different patterns
• f joint failure
• Osteoarthritis in the hip can cause pain, stiffness, and severe

disability

• The main curative treatment is surgery:
• Total Hip Replacement (THR) or
• Hip Resurfacing

30.12.2012 4

MANAGED OUTCOMES

Total Hip Replacements (THR) – provider perspective

• The incidence ratio of THR has been shown to vary from 1.25 to

4.7 between regions of the same country (Keskimäki et al. 1994, Birkmeyer et

• al. 1998, Pedersen et al. 2005, Dixon et al. 2006)
• 75 % of the costs of THR are related to the surgery process:
• peration and in-hospital stay (Peltokorpi & Kujala 2006)
• Lower provider volume has been associated with longer hospital

stay after THR surgery (Doro et al. 2006, Judge et al. 2006, Mäkelä 2010), with higher costs (Kreder et al. 1997, Martineau et al. 2005, Mitsuyasu et al. 2006) and also by increased mortality and complications (Katz et al. 2011, Kreder et al. 1997,

Lavernia & Guzman 1995, Solomon et al 2002)

• High provider volume of arthroplasty operations has been

associated with increased productivity (Torkki 2011)

MANAGED OUTCOMES

Region – HIP OA

Region Case hospitals Population Hip OA cases Age 60+ (%)

Finland (Keski-Suomi) Keski-Suomi Central Hospital 273 000 444 24 % Germany (Erlangen) Greece (Larisa) University General Hospital of Larisa 730 115 103 24 % Netherlands (Tilburg) St Elisabeth Ziekenhuis Tweesteden Ziekenhuis 430 955 340 104 22 % Spain (Valencia) Hospital La Fe 266 320 97 22 % UK (Brighton) SWLondon Elective Orthopaedic Centre 2 484 500 1273 24 %

30.12.2012 5

MANAGED OUTCOMES

Methodology

Demand Services User journey Resources Outcomes Operational model Operations Management Practice Population User Experiences Regional Setting Costs & Reimbursement

MANAGED OUTCOMES

Methodology

• Operational model: formal description of the demand, services,

user journey, resources and outcomes, and their quantitative relationships

• Operations management practice: the planning, management and

innovation of services; the collaboration between partners providing services

• Outcomes: the impact of services on health status, measured by

providers (quality indicators) and experienced by users (satisfaction)

• User experiences: the view of users on services and their

performance, measured in a survey

• Costs & reimbursement: the costing of resources for services and

the financing of services

30.12.2012 6

MANAGED OUTCOMES

Demarcation

This study focuses on operative phase of the care: hip replacement operations

MANAGED OUTCOMES

MANAGED OUTCOMES

2 Results

30.12.2012 7

MANAGED OUTCOMES

Incidence of Total Hip Replacements (THR)

1 2 3 4 5 6 7 8

<60 60-69 70-79 >79

Operations per 1000 inhabitants Age

Incidence per 1000

Keski-Suomi (Finland) Larisa (Greece) Tilburg (Netherlands) Valencia (Spain) Brighton (UK)

MANAGED OUTCOMES

Supply analysis – HIP OA

• Annual TJA (Total joint arthroplasty) volume of 376 – 2709 (UK)
• Annual TJA volume per surgeon 47-226
• Nurse intensity per OR and per bed is highest in UK and lowest in Greece

557 376 705 660 2709 88 47 77,5 226

500 1000 1500 2000 2500 3000

Arthroplasty operations per hospital Arthroplasty operations per surgeon

4,5 2,7 3,8 5,8

0,7 0,3 0,7 0,9

1 2 3 4 5 6 7

OR nurses /OR Ward nurses / bed

30.12.2012 8

MANAGED OUTCOMES

Patient Journey analysis – HIP OA

• The waiting times are shortest in Netherlands (~60 days)
• The waiting time of hip arthroplasty operation is approximately 250 days in Spain,

Greece and Finland

90 90 21 44 61 170 150 42 194 44

100 200 300 Keski-Suomi Larisa Tilburg Valencia SW London

Waiting time to specialist (d) Waiting time to operation (d)

5,8 8,5 5,0 6,6 4,5 0,0 5,0 10,0 Keski-Suomi Larisa Tilburg Valencia SW London

Length of stay in hospital (d) Length of stay in rehab (d)

60 % home 74 % home 93 % discharged home 70 % home 84 % home

MANAGED OUTCOMES

Cost analysis – HIP OA

• The implant costs are lowest in UK (high volume?) and greatest in Greece
• The total costs per operation varies from 5500 eur (Greece) to 7500 eur

(Finland)

5822 3216 6122 5159 4926 1350 2200 1400 1500 1174

1000 2000 3000 4000 5000 6000 7000 8000

Other costs Implant costs

0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 % 100 %

Implant Ward Surgery Outpatient

Total costs Distribution of costs

30.12.2012 9

MANAGED OUTCOMES

Outcomes – clinical indicators

• The number of complication is lowest in UK, especially in Reoperations
• In infections, classifications may be different?

4,1 % 2,9 % 2,0 % 0,80% 0,9 % 1,8 % 1,0 % 3,0 % 0,50% 0,8 % 1,4 % 1,9 % 4,0 % 0,1 %

0% 1% 1% 2% 2% 3% 3% 4% 4% 5% Keski-Suomi Larisa Tilburg Valencia SW London

Reoperations % Repositions % Infections %

MANAGED OUTCOMES

MANAGED OUTCOMES

3 Interpretation of results

30.12.2012 10

MANAGED OUTCOMES

Differences in production system

• The majority of the hospitals were quite similarly
• rganized in terms of volume and specialization

– Exceptionally, the SW London Hospital was focused

• nly joint replacements and the annual volume was

multiple compared to the other units

• The differences in process practices are difficult to

be found

– In upper level “the hospitals have similar processes” – the differences in results may become from daily management and detailed process prescriptions

MANAGED OUTCOMES

Productivity vs unit size

1 2 3 4 5 6 7 8 9 1000 2000 3000 Days Annual no of TJAs

Length of stay

Length of stay

1000 2000 3000 4000 5000 6000 7000 8000 1000 2000 3000 € Annual no of TJAs

Costs

Costs

30.12.2012 11

MANAGED OUTCOMES

Differences in user experiences I

91,57 84,74 88,98 83,72 92,22 88,3

10 20 30 40 50 60 70 80 90 100 SW London Keski-Suomi Larisa Tilburg Valencia Average

Satisfaction with service

MANAGED OUTCOMES

Differences in user experiences II

0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 Time at decision for operation

SW London Keski-Suomi Larisa Tilburg Valencia Average

0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 Time directly before operation

SW London Keski-Suomi Larisa Tilburg Valencia Average : Baseline health statusa

aThe scales are scored from 1 for ‘no pain at all’ or ‘no impairments at all’ respectively to 5 ‘extreme pain’ or ‘extreme impairments’ respectively.

30.12.2012 12

MANAGED OUTCOMES

Differences in user experiences III

0,5 1 1,5 2 2,5

SW London Keski-Suomi Larisa Tilburg Valencia Average

0,7 0,72 0,74 0,76 0,78 0,8 0,82 0,84 EQ-5D utility index (Dolan 1997) (B)

SW London Keski-Suomi Larisa Tilburg Valencia Average : Baseline health statusa

aThe scales are scored from 1 for ‘no pain at all’ or ‘no impairments at all’ respectively to 5 ‘extreme pain’ or ‘extreme impairments’ respectively.

MANAGED OUTCOMES

Differences in outcomes

• The number of complication is lowest in UK and Spain, especially in Reoperations
• In infections, classifications may be different?

4,1 % 2,9 % 2,0 % 0,80% 0,9 % 1,8 % 1,0 % 3,0 % 0,50% 0,8 % 1,4 % 1,9 % 4,0 % 0,1 %

0% 1% 1% 2% 2% 3% 3% 4% 4% 5% Keski-Suomi Larisa Tilburg Valencia SW London

Reoperations % Repositions % Infections %

30.12.2012 13

MANAGED OUTCOMES

Clinical Quality vs unit size

0,0 % 0,5 % 1,0 % 1,5 % 2,0 % 2,5 % 3,0 % 3,5 % 4,0 % 4,5 % 500 1000 1500 2000 2500 3000

Reoperations % Repositions % Infections %

This study supports the results of earlier studies: the clinical quality of high-volume units is better than low-volume units (Katz et al. 2011, Kreder et al. 1997,

Lavernia & Guzman 1995, Solomon et al 2002)

MANAGED OUTCOMES

Differences in regional costs

5000 10000 15000 20000 25000 30000 35000 40000

Cost per >60 year people

Cost per >60 year people

• 15000
• 10000
• 5000

5000 10000 15000

Cost difference per >60 year people

Incidence related difference Unit cost related difference

The regional cost is calculated by unit cost x regional incidence of operations

• The right figure indicates that the regional costs are more influenced by inciden

than the unit costs (process efficiency)

30.12.2012 14

MANAGED OUTCOMES

Cost-effectiveness compared to the incidence ratio of THR

2000 4000 6000 8000 10000 12000

• 2
• 1

1 2 3 4

Cots per EQ5D Incidence Cost per EQ5D The regions with higher incidence have higher cost per EQ5D and higher cost per less pain

Higher cost per EQ5D Higher Incidence Higher cost per EQ5D Lower Incidence Lower cost per EQ5D Higher Incidence Lower cost per EQ5D Lower Incidence

500 1000 1500 2000 2500 3000 3500

• 3
• 2
• 1

1 2 3 4 5

Cost per less pain Incidence Cost per less pain

Higher cost per Less pain Higher Incidence Higher cost per Less pain Lower Incidence Lower cost per Less pain Higher Incidence Lower cost per Less pain Lower Incidence Larisa Valencia Keski-Suomi Tilburg Tilburg Keski-Suomi Valencia Larisa

MANAGED OUTCOMES

Summary of results based on MO data

• In terms of regional cost-effectiveness the results indicate that:

– The differences in outcomes are minor – The differences in regional costs are mostly derived from incidence of operations: the unit costs (process efficiency) have minor role

• Our data shows that there are multiple differences in incidences (operations per

population) between regions

• The differences in unit costs are less than 40 % between hospitals

 The key question in improving the regional cost-effectiveness is to understand the reasons for differences in incidences

• In care processes the biggest differences are in ward care: the length of stay and

patients discharged to home

– These may become from differences in rehabilitation practices, attitude and intensity, which are hard to measure

• In terms of processes, the SW London has lowest complication level, shortest

length of stay and short waiting times

– The major observed differences in service producer perspective are the annual volume of arthroplasty operations and focus on those operations – Could the volume be an explanation?

30.12.2012 15

MANAGED OUTCOMES

MANAGED OUTCOMES

4 Discussion with the literature and other data sources based on MO results

MANAGED OUTCOMES

The OECD data has similar results in terms of incidences compared to MO data

30.12.2012 16

MANAGED OUTCOMES

There seems to be correlation between total Healthcare costs per capita and incidence of Total Hip Arthroplasty operations per capita (OECD)

• Based on interviews, the decision to operate is subjective: can the availability
• f resource increase the incidence?

Datapoints European Countries belonging to OECD Source: OECD Health Data 2011

R² = 0,6784 50 100 150 200 250 300 350 1 2 3 4 5 6 Incidence of Total Hip Arthroplasty operations (per 100 000) Total Healthcare costs per capita (PPP 1000 $)

MANAGED OUTCOMES Literature on association between volume and process

quality

• Solomon et al 2002 suggested, that surgeon volume and hospital volume are the

best indicators of orthopaedic adverse events in patients undergoing THR surgery

• Lower provider volume has been associated with longer hospital stay after THR

surgery (Doro et al. 2006, Judge et al. 2006)

• Katz et al. (2001): patients treated with THR at hospitals and by surgeons with

higher annual caseloads had lower rates of dislocation.

• Battaglia et al. (2006) and Shervin et al. (2007): a positive association between

higher hospital and surgeon volumes and lower rates of hip dislocation was

• found. (Literature reviews)
• Kreder et al. (1997): association between high surgeon volume and low rate of

revisions within three months and within one year.

• Lavernia and Guzman (1995): Surgeons with a low volume of THAs were

associated with a higher mortality rate than high volume surgeons.

• Doro et al. (2006) Mortality increased with decreasing hospital case volume, and

the lowest volume hospitals had an adjusted odds ratio of 1.9.

30.12.2012 17

MANAGED OUTCOMES

The complete Finnish data of public hospitals performing THAs (34 hospitals)

2,3 % 2,2 % 1,8 %

0,0 % 0,5 % 1,0 % 1,5 % 2,0 % 2,5 %

Hospitals < 250 THAs per year Hospitals 250-500 THAs per year Hospitals > 500 THAs per year

Re-operation within 12 months

Re-operation within 12 months

56% 57% 78%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Hospitals < 250 THAs per year Hospitals 250-500 THAs per year Hospitals > 500 THAs per year

Patients at home 1 week after operation Patients at home 1 week after operation

The share of complications is lowest and inpatient episode shortest In hospitals having greater annual volume of THAs

MANAGED OUTCOMES

Patient Journey analysis – HIP OA

• The waiting time of operation is a common problem in the EU
• Since the healthcare costs per capita (resources) have correlation with

incidence of operations:

• Does the increase of resources lead to increased number of operations or shortened waiting

times?

90 90 21 44 61 170 150 42 194 44

50 100 150 200 250 300 Keski-Suomi Larisa Tilburg Valencia SW London

Waiting time to specialist (d) Waiting time to operation (d)

30.12.2012 18

MANAGED OUTCOMES

Summary

• The other data sources and earlier literature

supports our findings

– The reasons for differences in incidences is a key question to understand in publicly-funded or insurance-based regional systems – From service provision point of view, the volume of specific operations of the service provider seems to be a critical factor in terms of process quality – The differences in outcomes or in process practices are minor from regional cost-effectiveness point of view

MANAGED OUTCOMES

MANAGED OUTCOMES

4 Key findings

30.12.2012 19

MANAGED OUTCOMES

Key findings: general notions

• The regional costs and cost-effectiveness are mostly

related to the incidence ratio

– There are multiple differences in incidences of operations between regions – The technical efficiency has less significant role

• The waiting time to operation is a common problem in

European regions

– Further research concerning balance between waiting times, service level and cost-efficiency is proposed

• Higher volume of Joint replacements is associated with

better productivity and lower level of complications

MANAGED OUTCOMES

Key findings: regional systems

• The hip replacements are typically performed in

multi-specialty units: central or university hospitals

• The typical volume of hospital is few hundred

replacement operations per year

• In the literature, many studies propose focusing

the operations to the focused units having > 1000

• perations annually

30.12.2012 20

MANAGED OUTCOMES

Key findings: good practices

• Fast track surgery combined to the optimum level
• f nurses in the ward

– The need of special care in hip replacement operations is 3-5 days

• Volume of specific operations per service

production unit is essential not the volume of hospital – In addition, the literature suggests also specialization in resource level

MANAGED OUTCOMES

Key findings: recommendations

• From regional perspective, the key question in terms of costs and cost-

effectiveness is defining the incidence-level of elective operations

– The decision to operate is subjective: “how mild/severe symptoms lead to

• peration?”

– In publicly-funded (or insurance) system the marginal utility of the money do not limit the demand: prioritization (public) vs. out-of-pocket (private) -system to be considered in elective operations

• From service provision or production perspective, the results and earlier

literature indicate that the productivity and clinical quality are increased in specialized high-volume centers

• From customer perspective, the focus should be in shortening the waiting time

to operation

– This problem is strongly linked to the first problem (incidence) – The problem can be solved by hierarchy (defined incidence-level  number of resources needed) or markets (competition)

30.12.2012 1

MANAGED OUTCOMES

MANAGED OUTCOMES An operations management and demand based approach to regional health service delivery systems for dementia patients

Meeting European health care system challenges by learning from differences between management practices in six EU countries

MANAGED OUTCOMES

The Managed Outcomes Project

• The main objective of the MANAGED OUTCOMES project is to develop and

disseminate practical conceptual models and a toolkit for improving the health service production system

• It is based on the premise that healthcare outcomes and cost-benefits are

affected by:

• the efficiency of service production
• the regional structure of healthcare delivery
• the degree to which people are empowered to participate in the co-production of

their care

• Case studies have been undertaken in six EU countries (FI, FG, GR, NL, ES, UK) for

four costly and challenging demands for all EU healthcare systems:

• Type 2 Diabetes
• Stroke
• Hip-osteoarthritis
• Dementia

30.12.2012 2

MANAGED OUTCOMES

Project partners

• These include a mix of university and consultancy organisations:
• Aalto University - Finland
• Erasmus University Rotterdam - Netherlands
• Balance of Care Group - UK
• Otto-Friedrich-Universität Bamberg - Germany
• Ethniki Scholi Dimosias Ygeias Eidikos Llogariasmos Erevnon - Greece
• Universidad Politechnica de Valencia – Spain
• Xperidox Futures Consulting - France
• Innovation in Learning Institute - Germany
• Forum Virium - Finland
• European Hospital and Healthcare Federation - Belgium

MANAGED OUTCOMES

The challenge of dementia

• Dementia is an increasing issue for both care management and

resource usage for older populations across the EU

• The services required are costly and can be complex to organise
• ver a long time period. There is a large burden on health care

systems, patients and families

• Integrated care is assumed to be an important process

characteristic

• Little research has been done in this field to explore the link

between care processes and quality, and this case study, in particular, represents an innovative experiment

30.12.2012 3

MANAGED OUTCOMES

References

• National Audit Office (2007), Improving services and support for

people with dementia. (London: NAO) ISBN: 9780102945614

• Burns A, Iliffe S, 2009, Dementia. BMJ 2009;338:b75

doi:10.1136/bmj.b75

• Burns A, Iliffe S, 2009, Alzheimer’s disease. BMJ 2009;338:b158

doi:10.1136/bmj.b158

• Kümpers S, Mur I, Hardy B, Van Raak A, Maarse H, 2006, Integrating

dementia care in England and The Netherlands: four comparative local case studies. Health & Place 12 (2006) 404–420

MANAGED OUTCOMES

Managed Outcomes project methodology - I

Demand Services User journey Resources Outcomes Operational model Operations Management Practice Population User Experiences Regional Setting Costs & Reimbursement

30.12.2012 4

MANAGED OUTCOMES

Managed Outcomes project methodology - II

• Operational model: formal description of the demand, services,

user journey, resources and outcomes, and their quantitative relationships

• Operations management practice: planning, management and

innovation in services, and the collaboration between partners providing services

• Outcomes: the impact of services on health status, measured by

user survey (EQ-5D), and by providers (quality indicators)

• User views and satisfaction on services and their performance
• Economic modelling of resources

MANAGED OUTCOMES

Dementia case study demarcation

The starting point for this case study is the notion that the ‘degree of integration’ of services has an important impact on health outcomes and that the best way of understanding this is to look at the impact which the regional framework of community services impacts on the lengths of stay of dementia patients in acute hospital settings.

Continuing care

Acute medical care Post acute care

Home Care Home

30.12.2012 5

MANAGED OUTCOMES

KEY

• SE1. Primary care

consultation (GP or nurse) yes Triggering symptoms Memory illness?

• SE2. Memory

assessment service Diagnosis (mild dementia) Progressing memory illness? Referral elsewhere Diagnosis Comorbidities?

• SE4. Comprehensive

geriatric assessment Multidisciplinary Care Plan

• SE13. MH nurse
• SE12. Community nurse
• SE11. Home care
• SE16. Social worker
• SE14. Specialist adviser
• SE15. Physiotherapist

Advance Care Plan (ACP)

• SE17. Telecare service

Exacerbation/ delirium/ trauma

• SE6. A&E/ Medical

Assessment

• SE7. Acute inpatient

admission

• SE8. Mental health

assessment Diagnosis

• SE9. Medical

treatment Can patient live at home?

• SE10. Step down

care SE5. Multidisciplinary team review

• SE18. Care home
• SE19. Palliative care

yes

Figure 2.1 Dementia Case: Process Flow

Can patient live at home? no not sure no Event Action Action Patient event Information integration Other process step S4 Acute hospital care S1 Primary care S2 Memory assessment & treatment S3 Care planning S5 Long term care packages

• SE3. OPMH beds

yes no yes

Dementia pathways representation

This simplified process flow illustrates the complexity of the potential pathways crossing between primary, secondary and long-term care settings.

MANAGED OUTCOMES

Dementia case study settings

Area (sq km) Total population % 65+ years Keski-Suomi (FI) 19,950 272,784 17.7 % Syros (GR) 84 19,793 19.5 % Lincolnshire (UK) 5,921 745,575 20.2 % Valencia (SP) 134 204,569 16.4 % Rotterdam (NL) 319 60,000 12.7 % Nuremberg (DE) 187 505,664 20.7 %

30.12.2012 6

MANAGED OUTCOMES

Dementia prevalence

Prevalence estimates can be contentious. For example, in the UK, the national estimate is based on age-based extrapolation of research data from early 1990s which may not take full account of improvements in dependency levels generally of older people (due to improvements in their health). Estimated dementia prevalence Estimated prevalence %

• f 65+

Diagnosed cases Diagnosed as % of prevalence Keski-Suomi (FI) 4,915 10.2 % 2,622 53.3 % Syros (GR) 228 5.9 % 166 72.8 % Lincolnshire (UK) 10,833 7.2 % 4,005 37.0 % Valencia (SP) 3,491 10.4 % 2,208 63.2 % Rotterdam (NL) 480 6.3 %

• Nuremberg (DE)

6,699 6.4 %

• NB: Local data systems do not enable some figures to be defined

MANAGED OUTCOMES

Patient and carer survey

• Complex to initiate given the nature of the condition and ability to access

patients

• Carer inclusion important:
• as proxy for patient where necessitated by their condition
• in their own right (health status, burden of caring)
• Variable response rates:

Case Study Total returned Used in main analyses Lincolnshire (UK) 62 39 Keski-Suomi (FI) 70 20 Nurembourg (DE) 59 23 Syros (GR) 84 40 Rotterdam (NL) 7 4 Valencia (SP) 39 29

Notes: Not all respondents answered all questions, so many analyses based on smaller numbers. NL excluded from most survey-related analyses due to very small numbers of respondents.

30.12.2012 7

MANAGED OUTCOMES

Characteristics of people with dementia in the survey

aEstimated by subtracting the mean number of months since survey divided by 12 from mean age at the time of survey. bbased on 6 respondents only.

% female Mean age in years at the time of survey Mean age in years at the time of admissiona Mean year of admission Mean number of months between admission and survey % patients permanently in care home at the time of survey Lincolnshire (UK) 56 84 83 2011/11 5 32 Keski-Suomi (FI) 53 78 77 2010/09 14.8 29 Nuremberg (DE) 62 83 82 2011/11 8.3 59 Syros (GR) 71 84 82 2009/10 23 18 Rotterdam (NL)b 40 84 83 2011/07 7.7 17 Valencia (SP) 73 79 78 2011/02 9.2 22 All countries 62 82 80 2011/02 12.1 32

MANAGED OUTCOMES

Characteristics of carers in survey

NB: Rotterdam excluded due to small numbers reporting

% women Mean age at time of survey % carers with higher education % who are spouses

• f patients

% carers with

• ther paid work

Lincolnshire (UK) 57 67 38 60 17 Keski-Suomi (FI) 46 61 93 50 35 Nuremberg (DE) 44 66 83 50 32 Syros (GR) 70 60 54 32 30 Rotterdam (NL)

• Valencia (SP)

72 66 32 43 14

30.12.2012 8

MANAGED OUTCOMES

Dementia and acute hospital admissions

16.5 % 24.1 % 21.4 % 22.7 % 35.0 % 31.0 % 33.1 % 57.8 % 36.0 %

0% 10% 20% 30% 40% 50% 60% 70%

Keski-Suomi (FI) Syros (GR) Lincolnshire (UK) Valencia (SP) Rotterdam (NL) Admissions / est. dementia popn. Admissions / diagnosed dementia pts.

MANAGED OUTCOMES

Average lengths of stay in hospital settings

Comparisons of average length of stay in hospital for patients with and without dementia

3.46 7.69 14.5 7.38 10.0 9.9 13.9 3.58 5.50 11.0 9.37 6.4 6.5

2 4 6 8 10 12 14 16 Including dementia patients Health Centre (FI only) Excluding dementia patients

30.12.2012 9

MANAGED OUTCOMES

Discharge destinations following acute hospital episode

Non-acute care: eg Spain (hospital at home); UK (intermediate care home placement)

83% 0% 3% 7% 5% 8% 3% 4% 9% 17% 41% 39% 11% 96% 73% 63% 43% 46% 3% 14% 12% 11% 7%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Keski-Suomi (FI) Syros (GR) Lincolnshire (UK) Valencia (SP) Rotterdam (NL) Nuremberg (DE) Deaths Home Non-acute care Other hospital/short-term care

MANAGED OUTCOMES

Length of stay and % discharged home

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 2 4 6 8 10 12 14 16 % Discharged home Average LOS (days)

Syros GR Keski-Suomi FI Valencia SP Lincolnshire UK Rotterdam NL Nuremburg DE

30.12.2012 10

MANAGED OUTCOMES

Carer opinions of hospital-based dementia services

10 20 30 40 50 60 70 80 90 100 Percentage of respondents Awareness of the confusion Appropriate treatment considering confusion Consultation about support at home Consultation about treatment Longer stay than necessary

NB: Rotterdam excluded due to small numbers reporting

MANAGED OUTCOMES

Levels of confusion in patients living at home

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Percentage of people with dementia % severe % moderate % mild % none

NB: Rotterdam excluded due to small numbers reporting

This shows the carer’s estimation of the patient’s level of dementia at home at the time of the

• survey. Note FI values in particular implying many more patients with moderate to severe

dementia are in care homes.

30.12.2012 11

MANAGED OUTCOMES

5 10 15 20 Percentage carer responses

Move to care home directly after hospital stay

NB: Syros value based on 36 responses Rotterdam excluded due to small numbers reporting NB: Rotterdam excluded due to small numbers reporting

Long-term care home placements: actual and expected

0.5 1 1.5 2 2.5 3 3.5 4 4.5 Lincolnshire (UK) Keski-Suomi (FI) Nuremberg (DE) Syros (GR) Rotterdam (NL) Valencia (SP) Certainty of admission (1-7)

Expectation of move to a care home within the next year

MANAGED OUTCOMES

Burden to informal carer

NB: Rotterdam excluded due to small numbers reporting

10 20 30 40 50 60 70 Lincolnshire (UK) Keski-Suomi (FI) Nuremberg (DE) Syros (GR) Rotterdam (NL) Valencia (SP) Standardised BSFC score

Standardised ‘Burden Scale for Family Caregivers’. 100 = no burden; 0 = maximum burden

30.12.2012 12

MANAGED OUTCOMES

Carer hours of care per year by BSFC score

23

Keski-Suomi FI Syros GR Lincolnshire UK Valencia SP Nuremburg DE 1000 2000 3000 4000 5000 6000 20 25 30 35 40 45 50 55 Carer's hour of care per year BSFC score

NB: Rotterdam excluded due to small numbers reporting

MANAGED OUTCOMES

Mean overall measures of perceived service quality

Satisfaction scale: 0 = worst, 100 = best possible Survey form completed by informal care giver as proxy for the patient. SERVQUAL = A multiple-item scale for measuring consumer perceptions of service quality.

10 20 30 40 50 60 70 80 Satisfaction SERVQUAL Satisfaction with service

NB: Rotterdam excluded due to small numbers reporting

30.12.2012 13

MANAGED OUTCOMES

Health of patient (EQ-5D) - as reported by carer

This shown standardised across the five EQ-5D dimensions (Dolan index): 0 for death and 0.91 for best imaginable health state. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Lincolnshire (UK) Keski-Suomi (FI) Nuremberg (DE) Syros (GR) Rotterdam (NL) Valencia (SP) Health index score

NB: Rotterdam excluded due to small numbers reporting

MANAGED OUTCOMES

Comparison between quality of life of people with dementia and their carers

People with dementia Carers

10 20 30 40 50 60 70 80 90 Lincolnshire (UK) Keski-Suomi (FI) Nuremberg (DE) Syros (GR) Rotterdam (NL) Valencia (SP) Index score

NB: Rotterdam excluded due to small numbers reporting 10 20 30 40 50 60 70 Index Score Utility index according to Dolan Stimulus anchored evaluation of health via Visual Analogue Scale

30.12.2012 14

MANAGED OUTCOMES FI GR UK S D 1000 2000 3000 4000 5000 6000 7000 8000 0,1 0,2 0,3 0,4 0,5 0,6 0,7 Hours fo total care per year per patient Patient's EQ-5D

Hours of total care (health professionals + carer) per year per patient by patient's EQ-5D

27

Higher resources, lower QALY

MANAGED OUTCOMES

Key characteristics of dementia services

This analysis summarises a checklist of groups of processes undertaken for each case study area. This may involve more than one organisation. The ‘Integration index’ represents the proportion of respondents giving positive answers for each group of processes (max = 1.0) Notes:

• Keski-Suomi: high use of shared information systems and access to information held by other agencies
• Syros: lower intensity of skilled assessment/memory clinic; no training program for professionals; good early dementia

screening; no guidelines

• Lincolnshire: use training programs in hospital; involvement of patient and carers in discharge planning
• Valencia: emphasis on community-based multidisciplinary teams, and guidelines for overall care of dementia

0.63 0.10 0.08 0.00 0.68 0.35 0.65 0.40 0.65 0.29 0.63 0.50 0.60 0.20 0.55 0.60 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Keski-Suomi (FI) Syros (GR) Lincolnshire (UK) Valencia (SP)

Integration index

Information Systems Pre-Admission Hospital Post-discharge

30.12.2012 15

MANAGED OUTCOMES

Key findings: general notions

• Complex care network needs integration and coordination
• Development of community based services is tied to the

willingness of carers to support people with dementia, and the extent of support given to them to do so

• Finnish patients at home had much lower levels of confusion

recorded than elsewhere; also earlier diagnosis of dementia than elsewhere, which appears to lead to earlier permanent admission to care home

• Uncertainty over value of memory clinics, given lack of possible

processes that could be initiated by early diagnosis.

MANAGED OUTCOMES

Key findings: scenarios

• Full integration

– Integrated community-based care – Identification of patients – Early and comprehensive response to problems – Carer involvement

• Hospital coordination

– Assessment on admission to acute hospital – Liaison and outreach nurses – Data sharing with PHC and social care

• PHC coordination

– Opportunistic assessment by GPs – Community liaison staff (nurses or social care) – Data sharing with other providers

30.12.2012 16

MANAGED OUTCOMES

Key findings: regional systems

• Developing role of PHC
• Good information systems central to development of

processes

• Viability dependent on good support to carers

MANAGED OUTCOMES