Best practices in collecting and processing data in CRC screening and after it

Transcription

1 Best practices in collecting and processing data in CRC screening and after it The potential of harmonized information policy in effective national implementation of CRC screening Ladislav Dušek, Czech Republic Institute of Biostatistics and Analyses Masaryk University, Brno, Czech Republic

2 I. Best practice requires robust methodical framework Institute of Biostatistics and Analyses Masaryk University, Brno, Czech Republic

3 HOW to optimize and manage CRC screening? European Council Recommendation (2003/878/EC) The European Parliament Declaration (2010) European Guidelines (2010) Population-based CRC screening - public and democratic - personalized - controlled Comprehensive guidelines must be effectively implemented in real world clinical practice Addressed invitation Coverage monitoring Participation rate Interval cancers Follow-up controls Compliance rate Detection rate Population impact Here, ICT plays very important role!

4 WHICH DATA items should be monitored? Europe against Cancer: Optimisation of the Use of Registries for Scientific Excellence in research WP5: Interface of cancer registries with cancer screening programmes A. Anttila, A. Ponti, G. Ronco, S. Lönnberg, N. Malila, A. Chil, J. Fracheboud, S. Törnberg, M. Zakelj, L. Karsa Performance indicators Individual-level data

5 Reality in the CRC screening implementation: 27 EU countries = 27 approaches? Zavoral M., Suchanek S., Zavada F., Dusek L., et al., WJG 15(47), 2009? The practical implementation of the CRC screening in Europe is evidently heterogeneous and not well reported, although methodical standards are clearly given. WHAT TO DO?

6 OECD Health at a Glance, 2011 The same reality also in the other programmes

7 The same reality also in the other programmes OECD Health at a Glance, 2013 OECD Health Policy Studies, 2013 Cancer Care ASSURING QUALITY TO IMPROVE SURVIVAL

8 Can data-based communication help? Czech experience as practical example Ferlay J, et al. GLOBOCAN 2008, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 10 [Internet]. Slovakia Hungary Czech Republic The Netherlands Norway Denmark Italy Belgium Ireland Germany Slovenia Luxembourg Croatia Portugal United Kingdom Spain Bulgaria France (metropolitan) Sweden Serbia Iceland Austria Malta Switzerland Estonia Republic of Moldova Lithuania Poland FYR Macedonia Finland Russian Federation Belarus Ukraine Latvia Romania Montenegro Cyprus Bosnia Herzegovena Greece Albania CRC: ASR(W) Other countries Neighbouring countries Czech Republic

9 II. Best practice in data collection requires comprehensive information system Institute of Biostatistics and Analyses Masaryk University, Brno, Czech Republic

10 Comprehensive system must cover multiple and heterogeneous data sources SCREENING PROGRAMME(S)? SEPARATED KEY INFORMATION SOURCES Diagnostics NON-STANDARDIZED INFORMATION SYSTEMS Clinical monitoring Health care payers

11 Solution? Respect the health care system 14 regional coordination offices Epidemiology Cancer care 189 health care facilities Equity Colonoscopy Diagnostics 160 colonoscopy centers QA / QC REPORTS Primary care FOBT GPs gynaecologists Follow-up Compliance

12 Examples of IS functionality: I. Management of population-based screening Selection of people to be invited Prospective mode Backward monitoring Retrospective mode

13 Examples of IS functionality: II. Clinical (hospital-based) monitoring Self benchmarking against centralized repository CENTRALIZED DWS Representative base for clinically relevant analyses of consecutively treated patients Dissemination of reports among centers

14 Examples of IS functionality: III. National on-line data-based reporting

15 Examples of reporting generated by the Czech National Cancer Control System: I. Population level Model diagnosis: colorectal carcinoma Counts per inhabitants Main trends: incidence & mortality incidence mortality ,7 % - 9,4 % %: trend change Incidence Mortality Absolute counts in Counts per in ,2 36,2 Counts per inhabitants Main trends: prevalence ,1 % %: trend change Prevalence ( ) Absolute counts Counts per ,7 Clinical stages: primary diagnosis 100% 80% 60% 40% 20% 0% TNM 2. TNM 3. TNM 4. í TNM 5. TNM Stage of the disease unknown not recorded Survival of patients in time trends Stochastic predictions of incidence and prevalence Stochastic predictions of therapeutic burden Colorectal carcinoma (C18-C20) 5yr relative survival (95% IC) All patients 44.7 ( ) 50.1 ( ) stage ( ) 73.3 ( ) stage ( ) 61.2 ( ) stage ( ) 46.8 ( ) stage ( ) 12.6 ( ) * Period analysis Colorectal carcinoma (C18-C20) Predictions for 2014 Incidence Prevalence Stage I (1918; 2267) (18 27; ) Stage II (1796; 2070) (16895; 17477) Stage III (2106; 2415) (12697; 13213) Stage IV (1956; 2305) (7 404; 7 800) Stage unknown 256 (165; 349) objective reasons (2 300; 2 526) Stage unknown 71 (46; 98) not recorded TOTAL (7987; 9504) (58223; 60579) Colorectal carcinoma (C18-C20) Newly treated patients in 2014 Stage I (1695; 2004) Stage II (1679; 1934) Stage III (1976; 2265) Stage IV incidence (1314; 1549) Disseminated relapses / progressions (1693; 2014) TOTAL (8 357; 9766)

16 Examples of reporting generated by the Czech National Cancer Control System: II. Screening program Model diagnosis: colorectal carcinoma CRC screening: regional coverage Pokrytí populace v procentech 40% 35% 30% 25% 20% 15% 10% 5% 0% CRC screening: age-specific coverage jednoletý interval dvouletý interval Muži Ženy Věk pojištěnce 10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0% QA/QC indicators in CRC screening % 12% ČR 6,9 % Rozsah v krajích 5,2-8,6 % 35% 10% 30% 8% 25,5% 16,0% 38,9% 25% 20% 15% 10% 5% 0% 5,4% 10,5% 11,5% 12,4% 13,4% 14,3% 15,9% 17,9% 18,5% 22,6% 24,8% 25,5% 6% 4% 2% 0% Coverage in time trend: Share of primary care specialists Primary screening colonoscopy Vývoj pokrytí 2012 vs (procentní body) Rok 2009, N = vyšetření vč. 324 (0,1%) jiná/neznámá odbornost 93,7% 6,2% Rok 2010, N = vyšetření vč. 186 (0,1%) jiná/neznámá odbornost 85,9% 14,1% Počet (na osob) Rok 2011, N = vyšetření vč. 151 (0,1%) jiná/neznámá odbornost Rok 2012, N = vyšetření vč. 77 (0,02%) jiná/neznámá odbornost 86,6% 13,3% 86,4% 13,6% +6,9% +1,1% +12,2% Praktický lékař (odbornost 001) Gynekolog (odbornost 603) 10,9 0,5 52,3

17 Examples of reporting generated by the Czech National Cancer Control System: II. Clinical centers Model diagnosis: colorectal carcinoma Distribution of care among regions/centers Treated patients / year Volume of primary care: capacity of CCCN N = Benchmarking of outcome measures: Survival after given medication vs. EBM trials 1,0 0,8 0,6 0,4 0,2 Medián OS Registr CORECT medián OS - 28,4 měsíce Studie AVF2107g* medián OS - 20,3 měsíce Studie NO16966 * medián OS - 21,2 měsíce Rank of hospitals 0, Čas (měsíce) Migration of patients /Example of one CCC/ Incidence and prevalence of treated CRC patients: benchmarking Benchmarking of outcome measures: 5yr survival population comparisons N = Care only in one facility Migration in primary therapy Migration after primary therapy Migration in follow-up Unknown Cumul. counts Counts / / /11 Incidence Prevalence Average CCC Range 2008/ / / / / / / / /11 Rok

18 III. Current challenge for all of us: harmonized implementation of CRC screening in clinical practice Institute of Biostatistics and Analyses Masaryk University, Brno, Czech Republic

19 Challenges and questions Excellent example: European cancer observatory - We need to standardize (harmonize): 1. E-data capture systems 2. Integration tools for heterogeneous data 3. Data mining tools 4. Standard national and European reporting 5. Employment of population cancer registries 6. IT guidelines for addressed invitation to screening 7. Communication guidelines 8. (E) - learning approaches 9. Legislative support for merging of different data sources

20 Legislative regulation of personal data handling? What is the acceptable extent? Individualized tracking of patient flow will it be possible in EU? Thank you very much for your attention