Presentation is loading. Please wait.

Presentation is loading. Please wait.

Corresponding author: Leandro Pecchia. 15 November 2013 AHP Algorithms and tools to support medical decision.

Similar presentations


Presentation on theme: "Corresponding author: Leandro Pecchia. 15 November 2013 AHP Algorithms and tools to support medical decision."— Presentation transcript:

1 Corresponding author: Leandro Pecchia. 15 November 2013 AHP Algorithms and tools to support medical decision making -Case studies- Leandro Pecchia

2 AHP Corresponding author: Leandro Pecchia. 2/32 RESEARCH BACKGROUND Me RF1 UNINA RF2 NOTT Ass. Prof. PHD 2007 Career path: : Assistant Professor, University of Warwick : Research Fellow (RF2), University of Nottingham : Research Fellow (RF1), UNINA* : PhD in Biomedical Engineering, UNINA* May 2005: BSc+MSc in Electronic Eng., UNINA* *UNINA= University Federico II of Naples, Italy Research interests: Biomedical signal processing and second level pattern recognition/data-mining Early stage Health Technology Assessment (HTA) and User Need Elicitation methods My main applications: active/healthy ageing: chronic cardiovascular diseases and falls in elderly Disease Management Programs, patient pervasive monitoring and Telemedicine f(stress, fat, salary, free-t,…)

3 AHP Corresponding author: Leandro Pecchia. 3/32 Contributions and outputs (1/2) Me Signal processing and pattern recognition for Cardiovascular disease (CVD) to identify Congestive Heart Failure (CHF) [early diagnosis] 2011, Long-term HRV & CHF detection, Med & Biol Engin & Computing, 49 (1): , Short-term HRV & CHF detection, IEEE T Inf Technolog in Biomed, 15 (1): to manage chronic CVD monitoring its damages and severity…[early detection of risks] 2012, HRV &Organ Damage in Hypertension, BMC Cardiovascular Disorders, 12: , Long-term HRV & CHF severity assessment, IEEE J. of Biom. and Health Informatics, 17(3): …also in remote monitoring applications:[telemedicine] 2011, Remote Health Monitoring of CHF, IEEE T Bio-Med Eng, 58 (3): , A feasibility study on telemedicine, Biomedical Engineering Online, 10: 49 Other signal processing and pattern recognition applications 2011, Nonlinear HRV for real-life stress detection, Biomedical Engineering Online 10: , Pupillometric analysis for assessment of gene therapy, Biomedical Engineering Online,11(1): , Infant cry analysis for early detection of Autism, ICHI2013 (+ submitting 2 journal papers)

4 AHP Corresponding author: Leandro Pecchia. 4/32 Contributions and outputs (2/2) Me Medical decision making is complex and multidisciplinary and needs: Quantitative knowledge: from the best available evidence (RCT, meat-analyses, network meta-analyses) Qualitative knowledge: to interpret the top of the EBM pyramid into everyday clinical practice Methods for the impact of BME researches: quantify qualitative knowledge User need elicitation using the Analytic Hierarchy Process (AHP) method 2013, User needs elicitation via AHP, BMC Medical Informatics and Decision Making, 3(1):2 2013, AHP & auto-injection of epinephrine, HIS2013, March 2013 in London. 2011, Factors affecting wellbeing in elderly, ISAHP 2011, Sorrento, Naples, Italy. Health Technology Assessment (HTA), especially for early stages of technology development 2013, HTA & AHP. In Studies in Fuzziness and Soft Computing, ed. Springer, Volume 305, 2013, HTA, Telemedicine, CHF. In Telehealthcare Computing and Engineering: Principles and Design. Science Publishers. ISBN (book chapter) Enhanced Remote Health Monitoring.. In Telehealthcare Computing and Engineering: Principles and Design, ed. Science Publishers. ISBN (book chapter) 2012, Network meta-analysis & mini-invasive surgery, Surgical Endoscopy, 2012 Jun 16, [Epub ahead of print] 2012, RCT for innovative biological drug, Hernia, 20 November 2012 Nov, [Epub ahead of print] 2011, Meta-analysis & minimally invasive surgery, Minimally Invasive Therapy & Allied Technologies, 21(3): June 2012, Treasurer of the HTA Division of International Federation of BME, IFMBE Risk factors for falls in elderly home dwelling Many intrinsic risk factors are related to physiological condition that can be detected 2011, Risk factors for falls, Methods of Information in Medicine, 50 (5): , Risk factors for falls, International Journal of the Analytic Hierarchy Process, 2 (2)

5 Corresponding author: Leandro Pecchia. 15 November 2013 AHP -Case study 1- SHARE Project*: home monitoring for patients suffering from Congestive Heart Failure (CHF) Leandro Pecchia *Smart Health and Artificial intelligence for Risk Estimation grant PON04a3_00139 to PM; Italian National Operational Programme for Research and Competitiveness. PI: Dr Paolo Melillo

6 AHP Corresponding author: Leandro Pecchia. 6/32 INTRO Congestive Heart Failure (CHF) is leading cause of hospitalization among the elderly in developed countries up to 50% of patients are rehospitalized within 3 months. Mortality ranges from 10% in patients with mild HF to 40% in severe cases COSTS: direct treatment costs of HF represent 2–3% of the total healthcare budget In literature there are three main models of care: Usual care (UC): outpatient follow-up (GP guided) as recommended by guideline Disease Man. Programs (DMP): UC + specialized doctors/nurses proactively at home Home Monitoring (HM): DMP + Information and Communication Technologies (ICT) The goal of HM, DMP and UC for CHF is to: QoL (or at least maintain stable) mortality (for CHF and for all causes), NHS costs by: bed days (CHF/all causes); readmissions (CHF/all causes) INTRO CS1

7 AHP Corresponding author: Leandro Pecchia. 7/32 GOALS Not all the HM is equally important and independent contribution of ICT is unclear Thus, the goals of the SHARE project are: To identify those elements that make HM more effective than DMP and UC (WP1) To design the most effective HM program To develop the ICT system to support such a HM To test its cost-effectiveness with 2 clinical trials GOALS WP1: HTA of HM WP2: HM and ICT platform R&D WP3: Algorithms R&D WP4: Observational Study WP5: Prospective study WP6: PM/Dissemination CS1

8 AHP Corresponding author: Leandro Pecchia. 8/32 METHODS 1/2 1.Meta-analyses were performed comparing DMPvsUC & HMvsUC Only well designed RCT were included Outcome considered: o QoL (or at least maintain stable) o mortality (for CHF and for all causes), o bed days (CHF/all causes); o readmissions (CHF/all causes) 2.Classified all the HM RCT according to: patients severity and HM complexity 3.We studied the correlations between the last 6 outcomes and patients severity HM complexity 4.According to these, the clinical trial and the ICT platform were designed 5.Algorithms for early detection of CHF worsening were Developed using public DB available Adapted using the information collected during the WP4 [now] Integrated into the ICT platform [2014] 6.Assessment of cost-effectiveness in WP5 [2016] METHODS CS1

9 AHP Corresponding author: Leandro Pecchia. 9/32 METHODS 2/2 HM RCT were classified according to patients severity and HM complexity Correlations effectiveness-severity and effectiveness-complexity were computed METHODS PATIENT SEVERITY NYHA CLASSES MEAN PATIENTS AGE EJECTION FRACTION NYHA II -- FREQUECY OF THE MONITORING Parameters & symptoms HM COMPLEXITY Only Symptoms - - Signal & Parameters & symptoms DATA MONITORED Weekly Daily CNYHA III -- CS1

10 AHP Corresponding author: Leandro Pecchia. 10/32 RESULTS Study selection: RESULTS 314 papers 23 DMP vs UC 8 HM vs UC 1DMP & HM vs UC 32 papers Full paper selection: 1 Not heart failure 7 Editorial or review 16 not an RCT 40 Other heart failure intervention or research 10 Study design 4 Other languages 5Short Follow-up 115 papers Title and abstract seletion: 8 Not heart failure 18 Invasive hemodynamic monitoring 43 Editorial or review 70 not an RCT 41 Other heart failure intervention or research 19 Study design CS1

11 AHP Corresponding author: Leandro Pecchia. 11/32 RESULTS RESULTS There is evidence that DMP are more effective that UC Reducing All-causes mortality Reducing readmission (All-causes and HF-related) It seems, but there is not evidence that DMPs reduce bed-days CS1

12 AHP Corresponding author: Leandro Pecchia. 12/32 RESULTS Mortality: HM vs UC All causesHF mortality RESULTS The HM RCT seems (no statistically significant result) more effective than UC CS1

13 AHP Corresponding author: Leandro Pecchia. 13/32 RESULTS Readmission: HM vs UC All causesHF mortality RESULTS The HM RCT seems (no statistically significant result) more effective than UC CS1

14 AHP Corresponding author: Leandro Pecchia. 14/32 RESULTS Bad days: HM vs UC All causesHF mortality RESULTS The HM RCT seems (no statistically significant result) more effective than UC CS1

15 AHP Corresponding author: Leandro Pecchia. 15/32 RESULTS HM RCT classification: Patients severity RESULTS NYHAEjection FractionMean Age Patient Severity Sherr 20083< Dendale, 20113< Dar, 20093> Koehler,20112&3< Antonicelli, 20102&3< Soran, 20082&3< Kulshreshtha, 20102&3< Giordano, 20102< Mortara, 20092< CS1

16 AHP Corresponding author: Leandro Pecchia. 16/32 RESULTS HM RCT classification: HM protocol complexity RESULTS SimptompsParametersSignalsFrequency Complexity KOEHLER, 2011 xECGDaily 6 KULSHRESHTHA, 2010 X Daily 5 DAR, 2009XX Daily 4 SORAN, 2008XX 4 GELLIS, 2012 X Daily 4 SHERR, 2008 X Daily 3 DENDALE, 2011 X Daily 3 ANTONICELLI, 2010XXECGWeekly 2 MORTARA, 2009XXECGWeekly 2 GIORDANO, 2010 XECGEach 15 days 1 CS1

17 AHP Corresponding author: Leandro Pecchia. 17/32 RESULTS HM outcomes vs Pz complexity and HM complexity: RESULTS PATIENT SEVERITY HM COMPLEXITY SurvivalSaved Bed daysReospedalization No significant correlation Evidence: HM is more effective for more severe patients (SHARE now focus on NYHA >2) More complex HM are more effective (SARE is designed accordingly) 17/32 CS1 significant correlation significant correlation significant correlation

18 AHP Corresponding author: Leandro Pecchia. 18/32 RESULTS How this informed the SHARE the clinical protocol? o Enrolling a proper number of severe cases [about 300 subjects in 12 months] o Acquiring daily useful symptoms, parameters and signals o These info will be daily reviewed to early detect patients worsening …and how the ICT platform? The DSS (WP3) will support clinician in modulating patient therapy o What these algorithms does and how they look like? RESULTS 18/32 NEXUS10(4), MindMedia -sensors: - EXG: ECG, EMG, EOG, EEG - SpO2, BVP, Body Temperature, Breathing acts, GSR -Communication: Bluetooth -Memory: Up to 7 days recording memory -Costs: from £5k to £12k, according to the sensors BioHarmess3, Zephyr -sensors: ECG, 3axial accellerations, Breathing -up to 3 days recording memory -Communication: Bluetooth/ZigBee -Memory: Up to 7 days recording memory -Cost: £350 (10) or £200 (>10) Healthcare Professionals (App) BIOMEDICAL SIGNAL PROCESSING (wearable devices) REMOTE PROCESSING (DDS) WARNING APPROPRIATE INTERVENTION CS1

19 AHP Corresponding author: Leandro Pecchia. 19/32 RESULTS RESULTS 19/32 Autonomous Nervous System (ANS) controls human equilibrium (homeostasis) Normal subjects show a good degree of variability in body functionalities, reflecting a continuous state of unstable equilibrium Unstable equilibrium is complex to control, but allows faster state changes This allow humans to react promptly to: – internal changes (i.e. emotions, stress) – external treads (i.e. the lion…) Monitoring these changes we can estimate the status of a subject and how stable it is… CS1

20 AHP Corresponding author: Leandro Pecchia. 20/32 RESULTS RESULTS 20/32 Signal processing: features extraction in time-, frequency-, nonlinear- domain Pattern recognition: signal/patient classifications normal vs CHF, mild vs severe, damage vs sane Signal pre-processing: filtering, beat recognition (normal vs abnormal),… Peculiarities of these signals: Bandwidth (0- few tens of Hz) Low S/N ratio in band No stationary (FFT cannot be used!) Strong non linearity and high dependence from parameters (chaos?) Problems for pattern recognition Limited cases Natural patterns (no human-generated) Last but not least… …our methods/results are needed by clinicians that cold be not skilled in mathematical methods! CS1

21 AHP Corresponding author: Leandro Pecchia. 21/32 RESULTS RESULTS 21/32 ECG PREPROCESSING HRV DETECTION HRV FEATURES EXTRAC. CART TRAIN/TEST FEATURES COMBINATION PERFORMANCE ASSESSMENT Detection: long & short HRV (Guidelines says that 12-leads ECG is not enough to diagnoses HF) Severity (NYHA) assessment HRV to identify Congestive Heart Failure (CHF) A) 2011, Med & Biol Engin & Computing 49 (1): B) 2011, IEEE T Inf Technolog in Biomed 15 (1): HRV to manage CHF monitoring its severity… C) IEEE J. of Biom. and Health Informatics, 17(3): D) BMC Cardiovascular Disorders, 12:105 [organ damages] A) B) C) CS1

22 AHP Corresponding author: Leandro Pecchia. 22/32 CONCLUSIONS CONCLUSIONS Not all the HM strategies are equally effective: HM is more effective on more severe patients more complex HM interventions seems more effective than less complex ones However, the increased quantity of information requires: Reliable technological solution Smart algorithms to extract the useful information Integrated management strategies The preliminary results of the algorithms developed are promising on public DB These SHARE trials will generate reliable databases for the adaptation of these algorithms.

23 Corresponding author: Leandro Pecchia. 15 November 2013 AHP -Case Study 2- A software tool to support the Health Technology Assessment (HTA) and the user need elicitation of medical devices via the Analytic Hierarchy Process (AHP) L. Pecchia 1, F. Crispino 2, S. Morgna 3 1 University of Warwick, The United Kingdom 2 Business Engineering, Avellino, Italy 3 University of Nottingham, The United Kingdom

24 AHP Corresponding author: Leandro Pecchia. 24/32 AHP for HTA & User Need Elic. HTA/UNE MULTIDIMENSIONAL EVALUATION IDENTIFY EXISTING TECHNOLOGIES CLINICAL EPIDEMIOLOGICAL ECONOMICAL … ETHIC/ SOCIAL DATA ANALYSIS RELATIVE ASSESSMENT NEED ANALYSIS PRIORITIZATIONINDIVIDUATIONCLASSIFICATION PERFORMANCEEFFICACYEFFICIENCY How to prioritize the needs? How to measure the fitting between MD performance and needs?? How to measure the MD performance in non-clinical domains? INTRODUCTIONMETHODRESUTLSCOMCLUSIONS CS2

25 AHP Corresponding author: Leandro Pecchia. Developing (or selecting) the health technology for a clinical problem (i.e. congestive heart failure) Developing (or selecting) the health technology for a clinical problem (i.e. congestive heart failure) 25/32 mortality worsening … … usability education service … … Initial cost ReadmissionC. qaly … … Technological domain (services/spare parts/ Human F) [Medical Eng.] Technological domain (services/spare parts/ Human F) [Medical Eng.] Economical domain (costs) [Hosp. Managers] Economical domain (costs) [Hosp. Managers] AHP for HTA Hierarchy via an exempla Clinical domain (effectiveness/utility) [clinicians/cardiologists/ger.] Clinical domain (effectiveness/utility) [clinicians/cardiologists/ger.] ALTERNATIVE 1 Disease Management Program ALTERNATIVE 1 Disease Management Program ALTERNATIVE 3 Active Implantable Device ALTERNATIVE 3 Active Implantable Device ALTERNATIVE 2 Telemedicine Telemedicine AHPAHP How important is each need for the assessment? [needs prioritization] How each alternative satisfy each factor? [MD performance] How each alternative fit with the goal?[MD/Goal fitting] INTRODUCTIONMETHODRESUTLSCOMCLUSIONS CS2

26 AHP Corresponding author: Leandro Pecchia. 26/32 AHP method pairwise comparisons Process Much less important Less important Equally important More important Much more Numerical values N1>N2 & N2>N3 => N1 >> N3 N1 > N3 N1 < N3 (5) (3) (1) (1/3) (1/5) NEEDS INDIVIDUATION JUDGEMENTS MATRIX (J) TREE OF NEEDS DATA POOLING QUESTIONNAIRES RELATIVE IMPORTANCE OF NEEDS ALTERNATIVES PERFORMANCE ASSESSMENT ALTERNATIVES PRIORITIZATION CONSISTENCY RATIO(CR) IF CR >0.1 Eigenvector (priorities) Eigen value (coherence) AHP EXPERTS INTRODUCTIONMETHODRESUTLSCOMCLUSIONS RELATIVE IMPORTANCE OF NEEDSCATEGORIES CS2

27 AHP Corresponding author: Leandro Pecchia. 27/32 AHP method Analytic needs prioritization Method worsening mortality usability education service Initial cost ReadmissionC. qaly Developing (or selecting) the health technology for a clinical problem (i.e. congestive heart failure) Developing (or selecting) the health technology for a clinical problem (i.e. congestive heart failure) Technological domain (services/spare parts/ Human F) [Medical Eng.] Technological domain (services/spare parts/ Human F) [Medical Eng.] Economical domain (costs) [Hosp. Managers] Economical domain (costs) [Hosp. Managers] Clinical domain (effectiveness/utility) [clinicians/cardiologists/ger.] Clinical domain (effectiveness/utility) [clinicians/cardiologists/ger.] INTRODUCTIONMETHODRESUTLSCOMCLUSIONS ALTERNATIVE 1 Disease Management Program ALTERNATIVE 1 Disease Management Program ALTERNATIVE 3 Active Implantable Device ALTERNATIVE 3 Active Implantable Device ALTERNATIVE 2 Telemedicine ALTERNATIVE 2 Telemedicine CS2

28 AHP Corresponding author: Leandro Pecchia. AHP for HTA AHP 28/32 ALTERNATIVE 1 DMP ALTERNATIVE 1 DMP ALTERNATIVE 2 Telemedicine ALTERNATIVE 2 Telemedicine worsening mortality usability education service Initial cost ReadmissionC. qaly ALTERNATIVE 3 Active Implantable D. ALTERNATIVE 3 Active Implantable D. Global importance Global importance INTRODUCTIONMETHODRESUTLSCOMCLUSIONS CS2

29 AHP Corresponding author: Leandro Pecchia. Method: AHP for HTA/User need elicitation Applications: Publication in Healthcare whit the App Models: to be downloaded and adapted in your study Community: experts willing be involved The system a web tool with App AHP 29/32 INTRODUCTIONMETHODRESUTLSCOMCLUSIONS CS2

30 AHP Corresponding author: Leandro Pecchia. 30/32 AHP for HTA Hierarchy via an exempla AHPAHP INTRODUCTIONMETHODRESUTLSCOMCLUSIONS Users: The elicitor: design/pilot the hierarchy and the questionnaires, invites domain experts and final responders, pool the results; generate the report; publish the results on the web portal. The domain expert: review the hierarchy/questionnaires, suggest final responders or other domain experts; The final responder: under invitation, download the hierarchy answer the questions. CS2

31 AHP Corresponding author: Leandro Pecchia. 31/32 AHP for HTA Hierarchy via an exempla AHPAHP Two possible scenarios: S1: Local elicitor, domain experts and the final responders in the same place Using the APP to speed-up the process and find consensus S2: Remote elicitor, domain experts and the final responders NOT in the same place, Using the APP and the portal to cooperate to the study via the web. Functionalities: Create the Hierarchy: problem definition/hierarchy draft Download an existing hierarchy: to be used as starting model (only S2) Invite domain experts: study piloting (only S2) Amend the hierarchy (only S2) Invite responders (only S2) Participate to the study Analyse and Pool results Generate a report Publish: upload on the portal hierarchy ¦¦ results ¦¦ reports¦¦papers INTRODUCTIONMETHODRESUTLSCOMCLUSIONS CS2

32 AHP Corresponding author: Leandro Pecchia. 32/32 CONCLUSIONS Concluding INTRODUCTIONMETHODRESUTLSCOMCLUSIONS This is the first tool specifically designed to: perform shared decision making in healthcare involve lay-users into the decisional process (paramount important for HTA) applying the AHP to the HTA/the user need elicitation in healthcare Medical Decision making is complex (…not necessary difficult!) Methods have to be: Reliable Well tested according to clinical practices Intelligible (no black boxes) Easy to use/understand for people not skilled in maths Traceable (you may have to prove that you did the best you could after years)

33 AHP Corresponding author: Leandro Pecchia. Thank you! Leandro

34 AHP Corresponding author: Leandro Pecchia. 34/33 HTA standard methods HTA MULTIDIMENSIONAL EVALUATION IDENTIFY EXISTING TECHNOLOGIES CLINICALEPIDEMIOLOGICALECONOMICAL … ETHIC/SOCIAL DATA ANALYSIS DISSEMINATION OF INFORMATION MONITORING RELATIVE ASSESSMENT (VERSUS BENCHMARK) NEED ANALYSIS SCORINGINDIVIDUATIONCLASSIFICATION PERFORMANCEEFFICACYEFFICIENCY Example 3 – not cost effectiveExample 5 – highlighting data requirements Example 1 – cost effective Example 4 – price optimisation EFFECT DIFFERENCE (QALY?) COST DIFFERENCE reduce price Reduce data uncertainty * NHS NICE willingness-to-pay threshold range of £20,000-£30,000 per QALY. *

35 AHP Corresponding author: Leandro Pecchia. 35/33 HTA Limits of standard methods HTA limits VS L Pecchia, MP Craven, Early stage Health Technology Assessment (HTA) of biomedical devices. The MATCH experience. World Congress on Medical Physics and Biomedical 2012, May 2012, Beijing, China.

36 AHP Corresponding author: Leandro Pecchia. 36/33 Headroom Analysis Method 1 p1p1 30k£/QALY QALY p3p3 DEVICE Production Costs OTHER NHS Costs HEADROOM p2p2 U C p2p2

37 AHP Corresponding author: Leandro Pecchia. 4 states Markov Models disease/worsening/exacerbation/dead MM HAVE DISEASE (C1,U1) DEAD (C2,U2) 1-p d -p w -p e WORSE DISEASE (C3,U3) pwpw p wb worsening (pw < p w ) INNOVATE DEVICE p ed 1-p eb -p ew -p ed EXACERBATION (C4,U4) p we pepe pdpd p wd p eb HAVE DISEASE (C1,U1 ) DEAD (C2,U2 ) WORSE DISEASE (C3,U3) pwpw p wb p ed EXACERBATION (C4,U4) p we P we pepe pdpd p wd p eb exacerbation (p e < p e & p we < p we ) U C C/ U =30K/QALY 1-p wb -p wd -p we 1-p d -p w -p e 1-p eb -p ew -p ed 37/33

38 AHP Corresponding author: Leandro Pecchia. Markov Models & AHP MM&AHP HAVE DISEASE (C1,U1) DEAD (C2,U2) 1-p d -p w -p e WORSE DISEASE (C3,U3) pwpw p wb INNOVATE DEVICE p ed 1-p eb -p ew -p ed EXACERBATION (C4,U4) p we pepe pdpd p wd p eb HAVE DISEASE (C1,U1 ) DEAD (C2,U2 ) WORSE DISEASE (C3,U3) pwpw p wb p ed EXACERBATION (C4,U4) p we pepe p d ± p p wd p eb 1-p wb -p wd -p we 1-p d -p w -p e 1-p eb -p ew -p ed What if some information are missing (eHTA)? Missing data can be estimated using AHP... …and using sensitivity analysisto estimate worst/best cases. 38/33


Download ppt "Corresponding author: Leandro Pecchia. 15 November 2013 AHP Algorithms and tools to support medical decision."

Similar presentations


Ads by Google