The Czech National Health Care Information System (NH-IS) and its strategy in building population-based reporting Evropská unie Evropský sociální fond Operační program Zaměstnanost Regional reporting of NH-IS
I. Introducing NH-IS Evropská unie Evropský sociální fond Operační program Zaměstnanost Regional reporting of NH-IS
The Czech National Health Care Information System NH - IS NH-IS can contribute to the optimization of health care system and to the monitoring of population of health. NH-IS = very complex system Statistical surveys (clinical reports on care, economy of providers) Health care registries (real world observations, based on representative reporting) Reference information systems (ehealth key components) International collaboration and reporting (Health Data OECD, Eurostat, EHIS, EHLEIS, )
Key mission of the Czech NH-IS I. Evidence-based medicine NH-IS! Clinically based evidence 14% 12% 10% 8% 6% 4% 2% 0% UL: 0,7 TP: 0,9 MO: 3,6 DC: 6,4 LI: 37,8 CL: 10,8 JN: 30,3 Počet (na 10 000 osob) CH: 28,9 SO: 24,7 TC: 5,4 DO: 4,0 KV: 8,1 CV: 0,9 LN: 3,5 LT: 4,1 ME: 3,8 MB: 7,6 KD: 5,4 RA: 5,3 PH: 11,1 NB: 3,3 AX: 12,6 PS: 12,5 BE: 1,0 PZ: 9,4 KO: 13,8 RO: 8,8 PM: 22,9 KH: 2,1 PJ: 12,9 PB: 4,2 BN: 2,5 SM: 24,2 JC: 12,9 HB: 2,8 TU: 12,8 HK: 52,3 PA: 21,7 CR: 11,2 NA: 8,4 RK: 13,0 UO: 9,7 SY: 8,3 SU: 1,1 JE: 1,9 BR: 0,8 OP: 6,2 OT: 12,8KA: 7,1 KT: 9,4 ST: 3,0 PT: 43,4 PI: 7,6 CB: 29,3 TA: 3,5 PE: 3,2 JH: 32,0 JI: 22,9 TR: 3,6 ZR: 2,1 BO: 3,7 BK: 1,0 BM: 3,9 PV: 3,2 VY: 0,9 OL: 0,8 KM: 2,4 PR: 2,8 ZL: 2,7 NJ: 25,7 VS: 14,8 FM: 36,3 CK: 13,1 ZN: 6,1 UH: 1,5 HO: 0,5 BV: 0,7 Everybody knows what to do! Does really everybody do it?
Key mission of the Czech NH-IS II. Basic elements of effective e-health information system Czech approach: 1 2 3 4 - e-identification/certification of health care professionals - e-identification/certification of health care providers - monitoring of patient s trajectory in the health care system - quantification of consumed health care services: classification, extent and cost National registry of health care professionals National registry of health care providers Health care records in centralized data store of reimbursement companies
II. Building NH-IS: main challenges Evropská unie Evropský sociální fond Operační program Zaměstnanost Regional reporting of NH-IS
Effective system must answer very complex questions How to facilitate and control implementation of clearly given evidence-based guidelines? and how to update them? Interventions How to control population based interventions in a way which supports equity of care? and how to interlink consequent steps? How to control population and its health status (mortality, morbidity, survival,.)?
Effective system must cover multiple and heterogeneous data sources DATA INTEGRATION and reporting? HOW TO MANAGE IT? SEPARATED KEY INFORMATION SOURCES Diagnostics NON-STANDARDIZED INFORMATION SYSTEMS Clinical monitoring Health care payers
Effective system cannot increase administrative burden of health care professionals? The Czech strategy is to develop a system utilizing already generated / collected administrative data and health care records
III. Three views defining modern NH-IS Evropská unie Evropský sociální fond Operační program Zaměstnanost Regional reporting of NH-IS
I. Management view Integration of very complex data Cost Optimization of networks; reasonable distribution of care Outcomes Processes Therapy Health technology assessment; cost effectiveness Inpus Diagnostics Risks QA/QC: Health care quality assessment
II. System view Comprehensive coverage of all segments of care Diagnostics & therapy Prevention screening Primary care Follow-up Supportive care Palliative care Really comprehensive national health care data collection cannot be based on hundreds of separated registries for individual diagnoses
III. Legal background including data integration and protection policy Ensuring representative data is a significant step to guarantee the system functionality; this also includes the legislation.! This applies particularly to the Amendment to Act No. 372/2011 Coll., on Health Services, which would make it possible to analyse data from health care payers.!!
Conclusion: NH-IS architecture National registry of health care professionals Targeted surveys National registry of health care providers Specialized registries Health care records in centralized data store of reimbursement companies
Example of outcomes: The Czech National Cancer Care Information System Czech National Cancer Control Program Information Platform
IT infrastructure for monitoring of cancer care Cancer centres network as a regional managing system Epidemiology Population-based registries Population and treatment burden National Cancer Registry Hospitals Specialized registries Hospital information systems Local and national registries Monitoring of health care EHR Primary care (GPs, gynaecologists) Hospital care Specialized care and cancer centres Equity of health care Structure of health care Results of health care Quality of health care REPORTS Distribution of health care Volume of health care Data validation
Examples of IS functionality: Hospital-based monitoring Self benchmarking against centralized repository Dissemination of reports among centers CENTRALIZED DWS Representative base for clinically relevant analyses of consecutively treated patients
Set of indicators: health care quality & population-based health indicators
EXAMPLE OF RESULTS - I. Modern comprehensive data-based reporting www.svod.cz Czech National Cancer Control Program Information Platform
National on-line data-based reporting www.svod.cz www.kolorektum.cz www.rektum.cz www.mefanet.cz
Treated patients / year Counts per 100 000 inhabitants Counts per 100 000 inhabitants Examples of reporting: I. Population level & II. Hospital level Model diagnosis: colorectal carcinoma Main trends: incidence & mortality 600 Main trends: prevalence +49.1% Clinical stages: primary diagnosis 100% TNM 2. TNM 3. TNM 4. TNM 5. TNM 6. TNM 7. 500 80% 400 300 60% 200 40% incidence mortality Incidence Mortality Absolute counts in 2013 8 066 3 716 Counts per 100 000 in 2013 76.7 35.4 100 0 1990 1992 1994 %: trend change 2003 2013 1996 1998 2000 2002 2004 2006 2008 2010 2012 Prevalence (31/12/2013) Absolute counts 54 982 Counts per 100 000 523,1 20% 0% 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 Stage of the disease 2001 2003 2005 2007 2009 2011 2013 1 2 3 4 unknown not recorded Distribution of care among regions/centers 600 500 400 300 Volume of cancer care: capacity mapping 2006-2014 N = 54 360 Benchmarking of outcome measures: Survival after given medication vs. EBM trials 1,0 0,8 0,6 Medián OS Registr CORECT medián OS - 28,4 měsíce Studie AVF2107g* medián OS - 20,3 měsíce Studie NO16966 * medián OS - 21,2 měsíce 200 0,4 100 0,2 0 1 6 11 16 21 26 31 36 41 46 51 Rank of hospitals 0,0 0 12 24 36 48 60 72 84 96 Čas (měsíce)
Examples of reporting: III. Predictive mapping of cancer burden Model diagnosis: colorectal carcinoma CRC (C18 C20) Stage I Stage II Stage III Stage IV TOTAL INCIDENCE PREVALENCE (95 % CI) (95 % CI) ---------- Prediction: 2016 ---------- 2050 21 376 (1903; 2197) (21 136; 21 616) 1951 (1844; 2057) 2117 (2010; 2226) 1631 (1359; 1903) 8037 (7298; 8777) 19 104 (18 877; 19 331) 15 114 (14 912; 15 316) 7083 (6945; 7221) 65 331 (64 911; 65 751)
Predictive mapping of cancer burden model concept Population-based predictive models Expert inputs and corrections Adjusted for clinical stage Stagespecific estimate of prevalence of treated patients Surviving cancer patients with relapse/progression of primary disease Primarily treated patients Relapse/ progression Scenario 1 (%) Scenario (%) Scenario (%) Scenario (%) Cost calculation
Example of clinical outcome assessment Model diagnosis: colorectal carcinoma Population-based monitoring CRC: 5-yr relative survival Cohort 1990 1994 Cohort 1995 1999 Period 2005 2009 Period 2010 2014 Hospital-based benchmarking Stage 1 64,9 % 76,2 % 87,6 % 91,8 % Stage 2 48,4 % 62,9 % 73,7 % 79,4 % Stage 3 40,0 % 41,8 % 54,5 % 62,2 % Stage 4 12,0 % 10,7 % 13,9 % 16,2 % Total 47,9 % 51,7 % 59,4 % 65,4 %
EXAMPLE OF RESULTS - II. Data-driven cancer screening - population based design - Czech National Cancer Control Program Information Platform
Examples of IS functionality: Management of population-based screening Selection of people to be addressed Prospective mode Backward monitoring Czech PB screening: system of birthday invitation driven by health care payers. Retrospective mode
Example of reports on the Czech PB colorectal cancer screening program CRC screening: FOBT regional coverage CRC screening: age-specific coverage FOBT positivity: time trend and regional profile Men and women aged over 50 Coverage in percents Coverage by screening Men and women aged over 50 Positivity 10% SO: 31,8 KV: 32,2 CH: 24,0 TC: 25,0 DO: 31,1 PJ: 33,0 KT: 26,6 UL: 28,5 LI: 34,7 TP: 32,9 MO: 35,0 CL: 32,4 JN: 27,4 CV: 38,6 LN: 32,6 PT: 22,0 PB: 31,1 PI: 35,7 ST: 21,8 CK: 30,3 LT: 37,4 KD: 29,5 RA: 22,7 CB: 30,2 DC: 26,6 AX: 23,9 PS: 30,0 PZ: 23,4 BE: 27,7 RO: 36,0 PM: 25,5 MB: 36,7 ME: 26,9 PH: 26,6 BN: 23,4 TA: 31,6 PE: 33,4 JH: 30,1 NB: 33,1 KO: 25,6 KH: 24,4 SM: 23,4 JC: 25,8 HB: 30,4 JI: 33,6 TU: 29,8 HK: 26,9 PA: 30,5 CR: 33,7 TR: 40,1 ZR: 29,6 ZN: 34,3 NA: 36,9 RK: 36,5 UO: 37,5 SY: 33,1 BO: 29,5 BK: 36,0 BM: 30,0 BV: 31,9 SU: 43,4 PV: 37,9 VY: 29,7 HO: 35,0 JE: 35,1 OL: 39,8 BR: 33,8 PR: 40,5 KM: 37,4 ZL: 37,8 UH: 39,1 OP: 34,6 OT: 31,5KA: 31,3 NJ: 27,2 VS: 33,8 FM: 25,5 < 25,0 25,0-30,0 30,0-35,0 35,0-40,0 > 40,0 50% 40% 30% 20% 10% 0% One-year interval Two-year interval Muži Ženy 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ 8% 6% 4% 2% 0% 2006 2007 2008 2009 2010 2011 2012 2013 2014 14% Positivity 12% 10% 8% 6% CR 7,2% Range in regions 5,9-8,6% Year 4% 30,8% 21,8% 43,4% Age group 2% 0% Region Coverage in time trend: 2013 vs. 2014 Share of primary care specialists Primary screening Colonoscopy regional coverage Men and women aged over 50 DC: 4,3 UL: 4,7 LI: 5,4 TP: 2,6 MO: 2,7 CL: 4,1 JN: 4,6 Difference (%) < 3,0 3,0-4,0 Women aged over 50 Year 2011, N = 310 393 examinations 151 (0,1%) other/unknown expertise Year 2012, N = 325 631 examinations 77 (0,02%) other/unknown expertise Men and women aged over 50 UL: 9,0 TP: 1,7 MO: 7,8 DC: 7,5 CL: 21,3 LI: 69,9 JN: 33,8 Number per 10,000 < 10,0 10,0-20,0 CV: 2,0 LT: 4,5 SM: 5,8 TU: 4,2 SO: 2,2 KV: 3,7 LN: 4,6 ME: 5,2 MB: 5,3 JC: 6,3 NA: 2,5 CH: 3,9 KD: 5,8 RA: 5,2 PH: 4,5 NB: 4,9 HK: 4,3 AX: 4,5 RK: 4,4 PS: 4,6 TC: 5,2 BE: 3,7 PZ: 4,8 KO: 4,0 PA: 3,5 RO: 5,0 PM: 2,2 UO: 4,2 KH: 4,5 CR: 3,8 PJ: 3,4 PB: 6,0 BN: 6,3 DO: 2,5 HB: 6,5 SY: 7,2 JE: 6,3 SU: 3,9 BR: 3,7 OP: 4,7 OT: 3,9KA: 4,2 4,0-5,0 5,0-6,0 > 6,0 86,6% 13,3% 86,4% 13,6% SO: 35,8 KV: 16,4 CH: 46,4 TC: 12,0 DO: 13,4 CV: 0,9 PS: 13,8 PJ: 22,7 LN: 20,6 RA: 8,5 RO: 8,6 PM: 25,5 PB: 21,0 LT: 22,4 BE: 2,5 PZ: 26,2 ME: 16,0 KD: 12,3 PH: 27,7 AX: 33,4 BN: 8,4 MB: 27,9 NB: 13,1 KO: 41,5 KH: 11,0 SM: 28,8 JC: 14,2 HB: 8,1 TU: 18,6 HK: 86,9 PA: 49,4 CR: 26,5 NA: 13,2 RK: 40,8 UO: 14,6 SY: 15,2 SU: 6,5 JE: 43,8 BR: 1,9 OP: 10,5 OT: 24,3KA: 19,5 20,0-30,0 30,0-40,0 > 40,0 PI: 2,3 OL: 4,2 KT: 3,4 TA: 4,1 PE: 4,2 ZR: 5,8 NJ: 3,2 FM: 4,2 ST: 4,5 BK: 3,9 PV: 7,3 PR: 4,9 JI: 3,4 BO: 4,5 JH: 3,0 VS: 4,2 PT: 2,3 TR: 5,0 BM: 4,1 VY: 4,8 KM: 4,9 CB: 4,7 ZL: 3,8 Year 2013, N = 339 424 examinations 88 (0,03%) other/unknown expertise Year 2014, N = 423 660 examinations 74 (0,02%) other/unknown expertise KT: 37,5 ST: 37,3 PT: 103,3 PI: 9,2 CB: 26,0 TA: 9,3 PE: 26,1 JH: 55,7 JI: 36,0 TR: 37,8 ZR: 26,3 BO: 10,1 BK: 2,2 BM: 10,1 PV: 4,8 VY: 1,5 OL: 3,3 KM: 3,1 PR: 9,4 ZL: 17,4 NJ: 35,2 VS: 20,6 FM: 65,7 CK: 4,2 ZN: 3,3 BV: 5,8 HO: 6,4 UH: 6,6 CK: 12,7 ZN: 21,5 BV: 2,8 HO: 1,7 UH: 15,1 +4,4% +2,0% +7,3% 86,7% 13,3% Practitioner 88,8% 11,2% Gynecologist 23,2 0,9 103,3
PB CRC screening: improvement 2010 -> Colorectal cancer epidemiology Colorectal cancer screening
EXAMPLE OF RESULTS - III. Data-driven changes in cancer care management www.onconet.cz Czech National Cancer Control Program Information Platform
www.onconet.cz Diagrams of cancer care available for each region PDF download Regional models of cancer care, presenting professionals and navigating patients Interactive maps Access points www.onconet.cz
Regional models of cancer care Comprehensive cancer care Map of facilities involved in comprehensive cancer care Types and numbers of facilities Diagram of cancer care Link to a regional Cancer Centre Regional news www.onconet.cz Detail of a health care facility
Cancer Centres On-line Comprehensive cancer care Equipment characteristics Clinical research Performance www.onconet.cz and more
Target area of CCCN Together, South Moravian Region and the Vysočina Region (target area for CCCN) account for 18% of the total area of the Czech Republic. Although these two regions are geographically next to each other, their remote parts are very different and provide thus representative sample for piloting of CCCN. South Moravian Region Vysočina Region Both regions Population (as of 31/12/2015) 1 175 025 509 475 1 684 500 Area (km2) 7 195 6 796 13 991 Population density (per km2) 163 75 120 Number of districts 7 5 12 Number of municipalities 673 704 1 377 Total length of roads and motorways (km, estimation) 4 500 5 000 9 500 Total length of railway network (km, estimation) 800 650 1 450
Pilot CCCN: South Moravian Region and Vysocina Region cancer care infrastructure Type of health care facility www.onconet.cz CCCN Comprehensive Cancer Centre [n=1] Cancer Centres [n=3] Children s cancer centres [n=1] Haemato-oncology centres [n=1] Constituent parts of CCCN [n=4] HCF cooperating with CCs [n=20] Mammography screening centres [n=13] Colonoscopy screening centres [n=29] LCTHs and hospices [n=20] All [n=92]
Conclusion and Current Challenges Evropská unie Evropský sociální fond Operační program Zaměstnanost Regional reporting of NH-IS
Personalized medicine needs personalized data collection and reporting Hospital IS (EHR) Individualized tracking of patient flow = future of cancer care control. Thank you very much for your attention