Shaping healthcare quality through data

Transcription

1 Shaping healthcare quality through data The role of public health insurers in Germany Christoph J. Rupprecht Head of the department for Health Policy and Health Economics AOK Rheinland/Hamburg- Die Gesundheitskasse

2 Evolution... Information, communication and knowledge... Communicational shortcomings 20-40% of the services data recording In the health care data processing system: communication Actual situation in the health care system Sectoral care as barrier Duplication of the medical knowledge every 5 years Globalisation and european integration The diagnostic and therapeutic spectrum of care is getting more and more complex 2

3 Contracts: Identity, fields of action, data and perspective Establishment of innovative and insuree-friendly models Improvement of low-threshold access for vulnerable patient groups such as socially disadvantaged, migrants, people with disabilities, people with dementia (Billstedt/Horn) Structured treatment programs in DMP succession Further development of care in a meaningful way ("life air"), enable transitions (psychiatry) Shaping the transition from science to healthcare (gene sequencing) Interlinking new and proven interaction possibilities (apps and personal contact options) Developing alliances (self-help, interest groups, consumer protection)

4 The scope of routine claims data Health services data Outpatient sector ICD-10 diagnosis data (outpatient and inpatient) ATC Code for medication prescriptions Invoicing data from auxiliary service providers (wheelchairs, clutches, physiotherapy ) Emergency care transportation services Rehabilitation services Supplemented by Patient satisfaction surveys QoL surveys (SF 36 etc.) Inpatient sector DRG Codes Length of stay Admission diagnosis Discharge diagnosis Patient demographics Age Gender Address Social security status Disability status Income Provider information Lifelong physician/pharmacist number Lifelong practice number Hospital ID 4

5 Advantages and drawbacks of routine claims data Advantages of routine billing data: 1) Full view on patient trajectory across providers and sectors 2) No additional data collection effort 3) High spatial granularity allowing for a zoom on healthcare delivery at city level 4) Highly regulated area with decades of practice Drawbacks of routine data: 1) Generally no clinical data (exception: e.g. Disease Management Program Diabetes) 2) Availability time-lag of approximately 6 to 9 months in outpatient care 3) Difficult to operationalise for outcome quality, limited number of use cases 5

6 Results of a patient satisfaction survey: rates of recommendation 95% 90% 85% Ø Federal average: 82 % 80% 75% 70% 65% 60% 149 hospitals in Rhineland region with at least 75 responses I each blue line represents one hospital Source: results from questionnaires collected in the framework of a joint survey by AOK Rheinland/Hamburg, Barmer GEK und hkk 6

7 Elective hip surgery: results from the QSR analysis 1,0 = risk adjusted event rate corresponds to federal average Below 1,0 = risk adjusted event rate is better than federal average Over 1,0 = risk adjusted event rate is worse than federal average I each blue line represents one hospital 7

8 Pflegeberaterinnen der AOK

9 Examples of innovative projects that run on routine healthcare data 9

10 AOK participation in a select number of projects INVEST Billstedt-Horn: integrated care network at city district level ACD: holding physician networks accountable for quality Predictive modelling Arena: combating antibiotic resistance through data-based quality feedback in physician networks AOK claims data evaluation Physician scorecards TeLiPro: establishing a telemedical lifestyle intervention for Diabetes Typ 2 patients Patient-tailored interventions 10

11 Project INVEST: Establishing an integrated care system at city district level A deprived neighbourhood in Hamburg: Billstedt-Horn 11

12 Project INVEST background A deprived city district: Hamburg district Billstedt-Horn characterised by inhabitants with mostly low socio-economic status Above average per capita healthcare expenditure Citizens with mixed ethnic origins Poor access to healthcare services Average age at time of death: 71 12

13 Actors and processes in project INVEST Health insurers' administrative data Data delivery Data protection Consent verification Declaration of participation of the service providers and insureds Datastorage Datacheck Data evaluation Participation declarations, program participation, Treatments, and surveys Feedback for providers Therapy Prevention Program participation Services Participating providers Health care controlling Diagnostics Vulnerable insurants Diabetes Multi-morbility Migration background Interventional objectives: Improve medication safety/prescription Reduce unnecessary hospital admissions Activate patients, improve access ( Gesundheitskiosk ) 14

14 INVEST Billstedt/Horn project consortium insurers providers research organisations Gesundheit für Billstedt/Horn UG Ärztenetz Billstedt-Horn e.v. OptiMedis AG Stadtteilklinik Hamburg NAV-Virchow-Bund e.v. Cooperating partners 15

15 The concept of ambulatory care sensitive conditions In 2012: 5,04 mio. inpatient cases classified as sensitive to ambulatory care Of which 3,52 mio. considered avoidable 7,2 bn EURO avoidable expenditure Most likely causes: Insufficient outpatient coordination Insufficient cross-sectoral coordination 2015 report by WHO Regional Office for Europe 17

16 ACD Methods 1 1. Step: Quantitative analysis of claims data by AOK RH, AOK NW and TK as well as routine data by regional physician associations a) Identify provider networks b) Analyse quality of care c) Observe care pathways Quelle: Projektposter ACD,

17 ACD Methods 2 2. Step: Identified networks are randomly assigned to an intervention and a control group; intervention group receives quality feedback based on data analysis; healtheconomic effects are analysed Cluster randomisation of physican networks Intervention group: Quality feedback Control group Endpoint: hospital admissions Endpoint: hospital admissions Quelle: Projektposter ACD,

18 TeliPro: a life-style intervention for diabetic patients 21

19 Diabetes mellitus share of patients, 2015 Kreis Kleve Kreis Heinsberg Aachen Kreis Aachen Kreis Viersen Kreis Düren Kreis Wesel Krefeld Rhein-Kreis Neuss Duisburg Mülheim a.d.r. Rhein-Erft-Kreis Kreis Euskirchen Düsseldorf Essen Kreis Mettmann Wuppertal Remscheid Solingen Oberhausen Mönchengladbach Leverkusen Rheinisch- Bergischer Kreis Köln Bonn Klassengrenzen Rhein-Sieg-Kreis Hamburg 10,1% - < 10,9% (6) 10,9% - < 11,2% (5) 11,2% - < 11,7% (6) 11,7% - < 12,0% (5) 12,0% - < 12,8% (6) Oberbergischer Kreis Kreis Kleve Kreis Euskirchen Aachen Oberbergischer Kreis Rheinisch-Berg. Kreis Kreis Wesel Rhein-Sieg-Kreis Düsseldorf Solingen Kreis Mettmann Rhein-Kreis Neuss Kreis Heinsberg Kreis Viersen Mülheim a. d. Ruhr Hamburg Bonn Rhein-Erft-Kreis Köln Leverkusen Kreis Düren Wuppertal Kreis Aachen Remscheid Oberhausen Mönchengladbach Krefeld Duisburg Essen 10,1% 10,1% 10,2% 10,4% 10,7% 10,8% 11,0% 11,0% 11,1% 11,1% 11,2% 11,2% 11,3% 11,5% 11,5% 11,6% 11,7% 11,8% 11,8% 11,9% 11,9% 11,9% 12,0% 12,1% 12,1% 12,2% 12,2% 12,8% 0% 5% 10% 15% ø Rheinland/Hamburg 11,4% ICD10-Code: E10-E14 = 2010 ø 9,9% Source: AOK Rheinland/Hamburg, standardised according to federal census 22

20 Prevalence of select chronic conditions comparison of insureers with and without diabetes, % 0,9% 30% 25% 20% 0,8% 0,7% 0,6% 0,5% 15% 0,4% 10% 0,3% 0,2% 5% 0,1% 0% Diabetisches Fußsyndrom Ulcus cruris Dekubitus Retinopathie Herzinsuffizienz Koronare Herzkrankheiten Nierenleiden 0,0% Herzinfarkt kein Diabetes Diabetes Source: AOK Rheinland/Hamburg, comparison based on matching process 23

21 Cases per Insuree years Cases per Insuree years Inpatient admissions as a consequence of Diabetes mellitus Soc. Security recipients vs. regular employed, Heart attack cases (diabetics) social security recipients vs. regular Beschäftigte ALG-II-Bezieher 0 Beschäftigte ALG-II-Bezieher Quelle: AOK Rheinland/Hamburg, indirekt standardisiert 24

22 TeLiPro Intervention Individual coaching based on bloodsugardata, body weight and physical acitivty as input factors 25

23 Telemedical life-style intervention for Diabetes patients (TeLiPro) Background: insulin dependence can be reduced, if diet is changed Objective: improve health and quality of life of Typ 2 diabetics through a long-term and sustainable change of life-style using telephone coaching in addition to routine care Avoid permanent dependence on insulin-medication Provide 1:1 data-driven telephone coaching 26

24 Relevance of TeliPro Achieving an insulin-free, high quality of life for diabetic patients Address individual needs of patients, especially for people with low health literacy Identify and address patient groups with specific needs Scaleability of telemedical solution 27

25 Improving care for patients with diabetic foot syndrome Background: Diabetic Foot Syndrome (DFS) is a severe complication of Diabetes; in coordinated care networks, it is largely preventable Objective: reduce major amputation rates AOK s DFS care contract: Establishment of a formal network of physicians responsible for DFS patients with clear structural requirements on the kind of expertise that needs to be available Definition of wound assistant qualifications Appointment within 24 hrs Hospital referral only to specialised units Quality management system Clear financial incentives for surgeons to treat DFS conservatively 28

26 Anteil an Versicherten mit Diabetischem Fußsyndrom Improvement of major amputation rate, DMP vs. non-dmp participation 5% 4% 3% 2% 1% 0% DMP-Teilnahme keine DMP- Teilnahme 29

27 Conclusion Routine claims data is one vital tool to drive innovation in health service delivery Insurers can use their data to: Control quality of care Analyse weaknesses in the care delivery system (over- and undersupply) Identify patient groups with particular needs (low health literacy, access problems) Build new care contracts based on insights gained from data analysis AOK Rheinland/Hamburg is actively engaged in a high number of projects that make use of routine data to improve healthcare 30

28 Innovative Treatment Molecular diagnostics and personalized therapy for lung cancer (University of Cologne): An integrated care contract as a motor for innovation Patient from Cologne with ROS1 translocation under therapy with ROS inhibitor 31

29 Multitude of potential starting points target orientated therapy in oncology EGFR- Inhibitors Cyclin dependent Kinaseininhibitors Inhibitors of the aerobic glycolysis Preservation of proliferative signal cascade Circumvention of growth inhibition Immune-activating Anti-CTLA4 mab Deregulation of cellular energy balance Prevention of destruction by the immune system proapoptotic BH3-Mimetika Apoptosisresistance Tumor growth Unlimited Replication Telomeraseinhibitors PARP-Inhibitors Instability of the genome and mutation Stimulation of the angiogenesis Activation of invasive growth and metastasis Tumor stimulating inflammation Selective antiinflammatory agents Lung-Ca e.g. EGFR, ALK, MET, ROS-1, KRAS, RET Inhibitors of the VEGF-Signalling Inhibitors HGF/c-Met Source: Adapted from Steward J, Naeymi-Rad N, et al. 32

30 Implementation of molecular diagnostics and personalized therapy in lung cancer Network Genomic Medicine (NGM) lung cancer of the University of Cologne Many years of experience with mutation analysis and high method competence (high number of NGS-based diagnostics per year) High level of consulting competence Continuous accompanying evaluation Participation in relevant clinical studies Knowledge and guarantees of the latest therapeutic approaches Second opinion for patients Patient-oriented study participation Contract on the basis of integrated health care ( 140a-d SGB V [old]). Start with the AOK Rheinland/Hamburg in April 2014 Accession of other German health insurance companies: KKH, BKK Novitas, BKK VBU, BIG direkt gesund, BKK Viactiv, Barmer, SBK, and TK. Genome profiling Source: aacc.org 33