Health Care Data and Analytics In Alberta Alberta

Health Care Data and Analytics In Alberta – Alberta Health Services Perspective Stafford Dean – Senior Program Officer Analytics , Alberta Health Services

Outline • AHS background and current situation • What we have learned about analytics from other high performing health systems • Measurement frameworks in health care • Lessons learned • Challenges • Provincial Perspectives 2

3

Analytics - Background • Merged 12 organizations with varying cultures and levels of maturity in data, measurement and reporting. • Initial merger work for AHS was focused on standardization of data, measurement and reporting. • Foundations have been developed over the past five years – data repositories, data governance, automated reporting, and standardized performance reporting. • Today AHS’ CEO, CMIO, CFO and other leaders have identified that as an organization we have to make better use of our vast data holdings and analytics capability to help the organization transform health care. 4

Current Situation – Data Warehouse • Oracle/Informatica data warehouse – traditional relational data system • Secondary Use - data stores that are created from administering and delivering health services is leveraged, stored, and made accessible so that it can be used to support quality improvement, performance reporting, and to support learning • Benefits of the EDW: • • • Increases analytical efficiency - analysts spend more time analyzing, less time managing data Benefits - Increases the value of information generated from AHS’ data assets by linking data across systems of care Enables report automation 5

Alberta Health Services Health Information For Secondary Use Health & Health Related Data Sources AHS Clinical and Corporate data fed from AHS Provincial Clinical and Corporate Applications Secondary Use Data Analytics Repository Adult Patient Experience Survey (HCAHPS) Ambulatory Admit Discharge Transfer Alberta Population Cost Utilization Child Patient Experience Survey (HCIES) Clinical Risk Group (CRG) Clinical Utilization Continuing Care Diagnostic Imaging Emergency Department Health Workforce Human Resources Workforce Laboratory Perinatal Hospital Based Pharmacy Surgical Inpatient and Ambulatory Abstract Other AHS Data Repositories Artissan Cancer EMS Human Resources Mental Health MIS Master Data (Patient, Provider , Location, etc. ) External Data Sources AH CIHI Stats Canada Pharmacy Information Alberta Population Cost Network Utilization Alberta Wait Time Reporting Population Health Provincial Registry Census Postal Code Claims Vital Stats Delivery Site National Ambulatory Care Reporting System Delivery Mechanisms Internal AHS Analytics Repository Reporting via Tableau Cancer Repository Reporting via Micro strategy Corporate Repository via Oracle Financials Various other tools used for analytics including SAS, SPSS, SQL, Oracle, Informatica, etc. External AHS Analytics Reporting via Tableau Portal

Current Situation – Reporting • • Report automation is a key initiative in DIMR to provide timely and easy accessible relevant information and to reduce resources spent on ADHOC requests. Enables various drill down capability and variable groupings (allows for various levels of details from a high level down to very specific geographies or patient types). 3, 000+ dashboards have been created so far In the spirit of transparency, fostering a workforce that is fact based, and comfortable using data to manage business and create insights, the dashboards are available to all AHS staff unless the data is patient or provider identifiable. Initial dashboard work is largely organized around an indicator and developed for AHS analysts. This is a major contributor to a single source of truth. If individuals go to the tableau reporting environment for facts, they will retrieve the same information, i. e. standardized reporting. Reporting Governance is now the challenge AH Tableau Portal 7

8

Current Situation – AHS Analytics • AHS is largely an analytically immature organization with analytical and data teams scattered and unconnected across the organization. Some pockets of excellence. • Need for an overarching analytic strategy that links clinical quality improvement across clinical, operations, and corporate analytic functions. • Significant data gaps to drive quality management and assess value for money… limited clinical, cost, experience and outcome data. • Important health system functions are not properly supported with analytics (e. g. , integrated health system planning, formal economic evaluations of major interventions). 9

AHS is largely analytically immature Stage 5 Analytical Competitors Stage 4 Analytical Companies AHS Stage 3 Analytical Aspirations Stage 2 Localized Analytics Stage 1 Analytically Impaired Taken from Analytics at Work: smarter decisions, better results (Davenport, Harris & Morison, 2010) 10

Industry Findings – Quality First • • • AHS has reached out to leading health systems that use analytics to drive clinical improvement. Strategy is about outcome and quality management, not about analytics. Analytics is one piece to help reach high performance. High performing organizations combine clinical, financial data, and patient experience data and other data to capture a holistic picture of quality and identify improvement opportunities. They focus their efforts on a small number of high-priority, high-impact projects and align all business units around these priorities. Teams focus on quality not costs. Costs will take care of themselves if we improve quality and outcomes. 11

Alberta’s Health System Measurement Framework

“Triple Aim” • Proposed by Berwick and Nolan in 2007 to re-vision healthcare around 3 core values • What would it look like if health care were aligned to: • The Triple Aim requires the simultaneous pursuit of: – Improved health – Enhanced experience of care – Reduced cost per capita

Dale Sanders – SPARC Launch - 2015 14 14

Triple Aim – Measurement Fronts • Outcomes • Experience • Cost • Workforce Health (Quadruple AIM? ) 15

Triple Aim - Outcomes • Clinical Outcomes – minimal • Patient Reported Outcomes – just started 16

Triple Aim - Experience • AHS - HCHAPs (Hospital Consumer Assessment of Healthcare Providers and Systems) - Inpatients • HQCA – administer LTC, SL, Home Care, ED • Need an patient experience strategy 17

Inpatient Experience Data https: //tableau. albertahealthservices. ca/#/workbooks/12798/views 18

Patient Experience – Self Service Tableau Reports

Triple Aim - Cost • Largely top down costing – pushing $$ down to the event level • Case costing – better but middle ground • Moving towards bottom up activity based costing leveraging EMRs/EHRs 20

Health Spending Per Capita Direct Spending - Primary and Specialty, ED/UC, Hospital Clinics, Inpatients Comparison of Actual and CRG –Age Weighted Spending Per Capita (CRG = Clinical Risk Group) 22

Inpatient Total Cost Versus Variation in Cost https: //tableau. albertahealthservices. ca/#/workbooks/6378/views 23

Total Medical Costs Versus Variation in Cost https: //tableau. albertahealthservices. ca/#/views/Total. Costvs. Cost. Varation by. CRG/CRGCost. Variation 2? : iid=1 25

High Performing Health Systems – Quality First! • • • AHS has reached out to leading health systems that use analytics to drive clinical improvement. High performing organizations combine clinical, financial data, and patient experience data and other data to capture a holistic picture of quality and identify improvement opportunities. They focus their efforts on a small number of high-priority, high-impact projects and align all business units around these priorities. Teams focus on quality not costs. Costs will take care of themselves if we improve quality and outcomes. More than just a data and analytics system– Content/Best Practice, implementation/adoption 27

What does Success look like? • AHS has clearly articulated and understood analytic functions that complement each other (clinical, Zone (operational), and corporate) • Bottom up clinical and operational analytics is the primary focus for health system transformation • Analytic capacity is embedded into the clinical and operational business and formally connected and supported through a hub and spoke model 28

Hub and Spoke Model Seniors Health Primary Care/CDM Surgery Closely connected core functions: The goal is to establish a single source of truth, scale, and the development of best practices to answer the key strategic questions for top executives. – develop the most important standing analytic functions. Create / Strengthen governance for analytic services: (1) Prioritization and transparency related to projects and resource allocation; (2) Stronger governance for data, data sources, measures, and methods employed prior to an analysis being chartered and/or published by AHS; (3) Balancing of analytic resources across functions. Distributed clinical, zone analytics: Rebalance resources to have a net increase of embedded analytics within the clinical and Zone/operational teams. This analytics function needs to have detailed knowledge of both the clinical workflows as well as high-level interpretive skills to support clinician-driven improvements. Public Emergency Care Health Corporate Functions Respiratory Population Health System Planning Resource Allocation Economic Evaluation Performance Management Strategic Clinical Support Services Bone and Joint Coordination with external organizations / academic centers to drive research: Establish a single point of coordination related to data and resources that supports the research agenda with academic institutions and other external organizations. Zones Addictions & Mental Health Cardiac Cancer Critical Care 29

What does Success look like? FOR CLINICIANS: • Clinical program managers have timely reporting of key clinical, process, outcome and cost measures balanced across the Quality Matrix for Health. • Measurement and reporting is available to the front line and meaningful for those that provide care. • Clinicians are able to see clinician-specific reporting that shows how they compare to best practice and how they compare to their peers. • Clinical program areas have the data and capacity to address their most important clinical questions. 30

Key Concepts • More flexible, agile data warehouse philosophy – staged levels of data access and late binding – bring all the data in with limited context or transformation (raw data materials) then make the data available in raw form up to very structured forms • Analytics is different than traditional BI – more fluid new, use cases all the time • Trust the analysts. Allow access, let them be creative and do not use restriction as a means to minimize risk. Manage what is produced not just what is accessed. Over governance is as bad as too little governance. 31

Key Concepts • Clinical data from secondary use of EMR/EHR data is the key to truly be able to mange clinical service delivery • Trust the analysts. Allow access, let them be creative and do not use restriction as a means to minimize risk. Manage what is produced not just what is accessed. Over governance is as bad as too little governance. • In system versus in data warehouse reporting • Real time Versus Lagged reporting • Next Generation EMR/EHRs? 32

Dale Sanders – Health Analytics Summit - 2014

“ Aspirational Healthcare Imagine if your physician could say this to you: “I can make a health optimization recommendation to you, informed not only by the latest clinical trials, but also by our local and regional data about patients like you; the real-world health outcomes over time of every patient like you who has had your illness; and the level of your interest and ability to engage in your own care -- and in turn I can tell you within a specified range of confidence, which treatment has the greatest chance of success for a patient specifically like you and how much that treatment will cost. "

Closed Loop Analytics Lifecycle New insights and knowledge created Transaction Systems Questions asked analysis undertaken Real-time clinical decision support Data extract, transform and load (ETL) Data Warehouse

What we have learned so far– data sharing • • Developed a shared data model program that develops relationships and facilitates data sharing among distributed analytic teams in AHS. Start with an analytic or reporting need that important to the particular area The data that is needed to address the need is not currently accessible to the distributed analytic/data team Work together in a formal way to develop the analytic product Develops relationships between DIMR analysts and the distributed analysts Distributed teams learn about data holdings in the repository and we learn about data holdings in the distributed teams Win - win – the potential value by working together is greater than working in isolation. 36

The Shared Data Model Cancer Care Repository Perinatal Population Health Operating Room Critical Care Personally identifiable data Cardiac Care Alberta Data Repository /DIMR Zone data Trauma Seniors Emergency Department Mental Health The Open Data Model has… § § § A centrally managed core Controlled access to personally identifiable data Distributed departmental / subject-area databases that can link with the core Robust master data and identifier keys that link central and departmental / subject-area data Open Data Model Privacy and Anonymization are foundational issues that must resolved early 37 in the Roadmap 37

Shared Data Model - DI https: //tableau. albertahealthservices. ca/#/w orkbooks/15189/views 38

Challenges • Currently have limited clinical data accessible for secondary use • Difficulty with the complexity EMR/EHR clinical data • Variation with how providers interact with clinical systems – adds further data variation • Lack of data scientists to extract analytic value from various sorts of data • Limited numbers of individuals that have direct access to clinical data – choke points • Reporting Governance • Primary data collection functionality – Major Pain Point 41

Challenges • Big Data – not there yet, but it is coming. Contact data to clinical data, to genomic data, bio monitoring data, social media data, … will pose all sorts of technical and analytical challenges • How do we get there – what is the roadmap? 42

44

Provincial Perspectives • AHS is building an analytics strategy that can and should plug and play into broader provincial data initiatives, such as PHAN or SPARC • Alberta has a huge opportunity if we work together • Lots of capacity and talent across many organizations, AHS, HQCA, Universities, … • Data on over 4. 2 M individuals, Alberta could create a true data lab and analytics system to answer important clinical and social questions 45

Discussion and Questions? ? ? 46