Predictive Analytics for hospital management Hans Levenbach, Delphus, Inc. and Paul Savage, HCI-LLC Email: hlevenbach@delphus.com ISF 2010 San Diego, CA June 21, 2010 reserved 1
Introduction Overview Predictive Analytics something new? Approaches and methods model complexity vs data volume Supporting the Hospital Value Chain Geography and product lines - Patient care activity Multiple competitor and product-mix forecasting Competing for new hospital locations Simulations Hospital closings - Berger Commission type simulations Data architecture and reporting Multi-State & Current Perspective On Demand (Saas( Saas) ) Virtual Forecasting reserved 2
We have to bring the science of management back into Healthcare Donald Berwick, MD Institute of Healthcare Improvement reserved 3
Predictive Analytics Something New? According to Wikipedia: Predictive analytics encompasses a variety of techniques from statistics, data mining and game theory that analyze current and historical facts to make predictions about future events.
Predictive Analytics - Methods The use of current and past data, in conjunction with statistical, structural or other analytical models and methods, to determine the likelihood of certain future events CLUSTERING DECISION TREE As you move down the spectrum, the complexity of the approaches and their implementation increase, while the volume of data decrease FORECASTING Predictive methods cover a range from relatively simple classification and forecasting to more advanced techniques such as dynamic modeling and simulation MONITORING& ADVISING SIMULATION& SCENARIO PLANNING reserved 5
Framework for SaaS Planning SPARCS Data Resource Multi-year Multihospital Information Mining & Analytics Product Line Trends Spatial Analysis Market Shares Budget PEER Planner Dashboard Forecasting & Collaborative Planning Competitive Simulation Model Interactive reserved 6
All Patients Decision Tree: Descend tree using any available differentiating attributes; natural, derived or inferred; separately or in combination Primary Tumor Site Surgery Psychiatry Colon: Classification by Stage Surgery New Born C1 Cardiology Maternity Cardiac > 100+ product line/service combinations Oncology Pulmonary > 220 Hospitals > Payor Class > Region reserved 7
Intelligent Dashboard Environment Manage strategic opportunities Monitor competitive environment Enhance physician relations Report writing capability reserved 8
Dashboard History, Adjustments and Forecasts Hospital Client reserved 9
Report Writing and Additional Capabilities Physician Activity Reports Market Level Assessment Simulation Modeling Product Level Demography Benchmarking: Quality, Safety & Performance reserved 10
Physician Level Reporting Attending Physician Activity Surgeons Activity Measures of Loyalty Ranking Trends and Life Cycle Analysis reserved 11
Physician & Group Practice Analysis
3 0 0 2 8 0 Total Group Practice Inpatient Activity ( 8 0 % p re d i c t i o n i n t e r v a l s ) 19 % S e a s o n a l R a n g e 2 6 0 2 4 0 2 2 0 A v g. 2 16 / Mo 2 0 0 1 8 0 1 6 0 1 4 0 1 2 0 1 0 0 Low er P-L (10%) Upper P-L (90%) History History T-C Forecast T-C J a n- 0 4 Ma y - 0 5 Oc t - 0 6 Fe b- 0 8 J ul - 0 9 No v - 1 0 reserved 13
1 0 0 % Percent of Group Practice by Hospital 8 0 %p re d ic t io n int e rvals 9 0 % 8 3 % 8 0 % 7 0 % 6 0 % 5 0 % 4 0 % 3 0 % P rim a ry Ho s pita l S ha re o f P ra c tic e S e c o nda ry Ho s pita l S ha re o f P ra c tic e G-S_History G-S_History T-C G-S_Forecast T-C N-I_History N-I_History T-C N-I_Forecast T-C 2 0 % 1 0 % 17 % 0 % J a n- 0 4 Ma y - 0 5 Oc t - 0 6 Fe b- 0 8 J u l - 0 9 No v - 1 0 reserved 14
1 2 0 New Physician Growth ( 8 0 % p re d ic t io n int e rva ls ) 1 0 0 8 0 6 0 4 0 2 0 Lower P-C (10%) Upper P-C (90%) History History L-C Forecast L-C 0 J a n- 0 4 Ma y - 0 5 Oc t - 0 6 Fe b - 0 8 J u l - 0 9 No v - 1 0 reserved 15
Transition to Mature In-Patient Activity 7 0 ( 8 0 % pr e di c t i o n i nt e r v a l s ) 6 0 5 0 4 0 3 0 2 0 1 0 Forecast T-C History History T-C 0 J a n - 0 4 M a y - 0 5 O c t - 0 6 F e b - 0 8 J u l - 0 9 N o v - 1 0 reserved 16
Senior Partner In-Patient Activity 7 0 6 0 5 0 ( 8 0 % p re d ic t io n int e rvals ) Low er P-C (10%) Upper P-C (90%) History History T-C Forecast T-C 4 0 3 0 2 0 Avg 26/ Mo 1 0 0 J a n- 0 4 Ma y - 0 5 Oc t - 0 6 Fe b - 0 8 J ul - 0 9 No v - 1 0 reserved 17
8 0 Mature Partner (Stable) Activity (80% prediction intervals) 7 0 6 0 Avg 50/Mo 5 0 4 0 3 0 2 0 10 Low er P-L (10%) Upper P-L (90%) History History T-C Forecast T-C 0 J a n - 0 4 Ma y - 0 5 Oc t - 0 6 Fe b - 0 8 J u l - 0 9 No v - 10 reserved 18
Hospital & Product Line Analysis
Community Hospital Forecast Patient Activity Patients/Mo. 2000 1900 1800 1700 1600 1500 22,000 Pts/Yr 20,000 Pts/Yr 1400 reserved 20 Jan-06 Jan-07 Jan-08
Simulation Modeling : Competing For New Hospital Locations
Simulation Geographical Map reserved 22
Simulated Map of Hospitals reserved 23
Extending Visibility Into The Enterprise Consistency Traceability Highly Aggregated Graphical Display, Dashboards, Interactive Aggregated Data, Model Execution Limited Drill Down Executive User Functional User Power User More Detail Standard & Ad hoc Reporting - Parameter-Driven by Users at Run-Time - Sorting, Selection, Filtering, Drill-Down Utilizing Standard Functions and Models Complete Raw Data Direct Access to Detailed Raw Data - Just give me the data in Excel Model Development & Deployment reserved 24
Questions? Contact: Hans Levenbach Email: hlevenbach@delphus.com URL: www.delphus.com reserved 25
SaaS Implementation: System & Data Architecture DATA PRESENTATION/ CONSUMPITON LAYER Predictive Models Report Management Test Management Model/Version Management APPLICATION/ MODEL MANAGEMENT LAYER Data Warehouse Multi- Dimensional Data Store Data Files Data Sets DATA STORAGE LAYER Integration Cleansing Data Quality Formatting Aggregation DATA INTEGRATION LAYER Hospital Data Operations Data Financial Data External Data DATA SOURCE LAYER reserved 26