AMPLIFYING INTELLIGENCE IN HEALTHCARE PATIENT FLOW EXECUTION GOALS - PowerPoint PPT Presentation

AMPLIFYING INTELLIGENCE IN HEALTHCARE PATIENT FLOW EXECUTION

GOALS Showcase how applied Share lessons learned intelligence can Present the solution and sucess factors for increase hospital bed with a strong business applying AI in health availability and care oriented vision industry quality

WELCOME! AMPLIFYING INTELLIGENCE IN HEALTHCARE PATIENT FLOW EXECUTION Cláudia Laselva Fábio Ferraretto Chief Nursing & Operations Officer Chief Data Scientist Albert Einstein Jewish Hospital Accenture Latin America

HOSPITAL ALBERT EINSTEIN

Purpose, Mission, Vision, Precepts and Values Overview Purpose 1 Deliver healthier lives by handing a drop of Einstein to every citizen. To offer excellence in the field of healthcare, education, and social Mission 2 responsibility, as a way of highlighting the Jewish community’s contribution to Brazilian society. To be a leader and an innovator in medical and hospital care, a reference in Vision 3 managing knowledge, and recognized for its commitment to social responsibility. Mitzvah (Good Deeds) Refuah (Health) 4 Chinuch (Education) Precepts Tsedakah (Social Justice) • Honesty • Diligence 5 • Truthfulness • Competence Organizational Values • Integrity • Fairness 5

Einstein Hospital Key Figures 81,5% Operations Overview occupation rate 86,2% occupation 339,3 Private System rate² appointments 675,5k 640 beds 340,5k appoitments 3,28 l ength of stay ED cases 730,3k ED cases¹ 4.237 births (HMMD + UPA Campo Limpo) 8.498 births 32.884 surgeries Unified Health System (Public) 11.174 surgeries 5,1mi 240 beds Hospital Municipal Dr exams 5,51 lengh of stay 2,9mi exams Moysés Deutsch 55,8 k 174 beds Hospital 28,2k hospital leaves Municipal Vila 5,68 length of stay hospital leaves Santa Catarina ¹ HMMD + UPA Campo Limpo ² HMVSC e HMMD 6

Business Context Operational Efficiency Burning Plataform Hospital experienced a margin deterioration despite of significant revenue growth captured. Call for Action Call for Action + 10.6mi Increase operational efficiency to Increase operational efficiency to reduce investiments on capacity reduce investiments on capacity new private customers expansion expansion Health Market 614 No signicant improvement of 2007-10 average length of stay indicator +31% 577 523 499 493 Period revenue increase Volume Growth -6.9p.p. EBTIDA reduction 2007 2008 2009 2010 2011 Margin Loss Operational Beds

PATIENT FLOW MANAGEMENT PROGRAM

PATIENT FLOW MANAGEMENT PROGRAM The Central Concept "Managing patient flow is one way to improve health services. Adapting the relationship between capacity and demand increases patient safety and it is essential to ensure that patients receive the right care, at the right place, at the right time, all the time . "

PATIENT FLOW MANAGEMENT PROGRAM Program Drivers and Principals Program Principals Increase capacity Break Systemic availability Silos Vision Deliver high care Scientific Data quality standards Evidence Reliability Maximize patient Process KPIs experience Review Monitor

PATIENT FLOW MANAGEMENT PROGRAM Program Organization The program aimed to elliminate waste of time and resources among the patient flow through a process optimization oriented methodology. Phase 1 Phase 2 Phase 3 • Process mapping • Multidisciplinary team • Action plan building formalization • Process decoupling • KPIs and indicators among emergency, • Process gaps confirmation elective and outpacient confirmation • Detailed workplan and • Future state design • Prioritization plan follow up 27 +70 548 70 operational areas team members actions KPIs and indicators involved accountable implemented monitored daily

PATIENT FLOW MANAGEMENT PROGRAM How we measure success The reduction of average length of stay increase bed availability, increase patient safety and postpone investments on capacity expansion. Relation between Length of Stay and Virtual Capacity Relação entre redução do TMP e Ganho Incremental de leitos 4.10 3.96 3.87 97 3.86 3.81 3.75 3.64 3.51 3.40 virtual beds added 2009 2010 2011 2012 2013 2014 2015 2016 2017 20 34 36 44 to serve patients 54 74 97 117 Length of Stay (days) Virtual Capacity (beds) TMP Incluíndo Day Clinic (dias) Capacidade Virtual (leitos)

HOW TO CONTINUE CREATING VIRTUAL CAPACITY AND ENSURING HIGH CARE QUALITY STANDARDS?

APPLIED INTELLIGENCE FOR PATIENT FLOW Rethinking Patient Flow Experience The New Patient Flow Experience

APPLIED INTELLIGENCE FOR PATIENT FLOW Solution’s Components Prioritized 25% of bed First bed specialty demand comes allocation reduced from ER length of stay business reason PROBABILITY OPTIMIZED OF ER PATIENT ADMISSION ALLOCATION component objective Antecipate Maximize patient visibility of ER allocation in first demands for bed specialty capacity planning

APPLIED INTELLIGENCE FOR PATIENT FLOW Probability of ER Admission Variables from every patient stage in ER are ingested and used... EXAM/TEST SPECIALIST 1 ST DOCTOR CHECK SCREENING HOSPITALIZATION RESULTS DOCTOR CHECK 1 2 3 4 5 VARIABLES VARIABLES VARIABLES VARIABLES • Reference tables • Patient record • Medicines • Specialty and standards • Specialty • Blood Exams • Doctor • Test results • Vital signs • Image exams • Image reports • Severity index • Patient in Observation PREDICT ACCURATELY AS EARLY AS POSSIBLE

APPLIED INTELLIGENCE FOR PATIENT FLOW Probability of ER Admission ...but modelling approach is based on 10’ time windows. Business Hyphotesis: treatment approach evolves depending on how patient’s clinical condition respond to previews prescriptions, such as medication and exam results. + Probability for Admission Increase of 21.p.p Prescrition Package 3 in F1 score (production more info and Prescrition Package 2 approach review database) more info and Prescrition Package 1 approach review - - + time

APPLIED INTELLIGENCE FOR PATIENT FLOW Probability of ER Admission Dataset used from new CERNER EMR: 66K 6.5k 10k ER cases Admissions Types of medicines prescribed. 1.1k active JANUARY TO principles JULY OF 2017 1.8k 2.5k 87 Types of blood tests and Different medical Different form fields image exams prescriptions in EMR

APPLIED INTELLIGENCE FOR PATIENT FLOW Probability of ER Admission The variety of features were a powerful tool against lack of historic data set. Patient Specialist evaluation Nutrition Services Pharmacy • Medical record • Evaluation • Fasting, General, Pasty, .. • Medication • Age • Type of Drug • Gender Physiotherapy Admnistration Screening • Respiratory Physiotherapy • Specialty • Non-invasive ventilation Labs Patient assistance • Protocols (Heart Attack, Stroke, • Analgesic Physiotherapy • Exams • Evaluation of pain level Sepsis) • Respiratory and Motor Physiotherapy • Emergency room flow • Screening complaint • Observation room flow • ESI Respiratory Therapy • O 2 saturation • SPO2 • Oxygen therapy • Oncologic patient? Radiology • Intervention after fall? • X-Ray Neurodiagnosis • Level of consciousness • Tomography • Electroencephalogram • Insulin dependent • Ultrasonography • Level of pain • Magnetic Resonance Procedures • Temperature • Endoscopy • Respiratory frequency Cardiovascular • US Obstetric Transvaginal • Blood Pressure • Electrocardiogram • Angiography • Admission Date Cardiology • Doppler echocardiogram

APPLIED INTELLIGENCE FOR PATIENT FLOW Probability of ER Admission Accuracy vs. Explainability: METRIC TRAIN TEST MODEL 80% split 20% split 92% 87% R.N.N. 80% 78% Logistic Regression ACCURACY 78% 75% Random Forest 0,75 0,69 R.N.N. 0,61 0,56 Logistic Regression F1 SCORE 0,55 0,49 Random Forest R.N.N. = Recurrent neural network