Developing a Strategic Analytics Framework that Drives Healthcare Transformation

Developing a Strategic Analytics Framework that Drives Healthcare

Transformation

Developing a Strategic Analytics Framework that Drives Healthcare

Transformation

Trevor Strome, MSc, PMP

Blog: http://HealthcareAnalytics.info

Twitter: @tstrome

Presentation Content Based on Chapter 3 of:

Healthcare Analytics for Quality and Performance Improvement

http://healthcareanalytics.info/

https://twitter.com/tstrome

https://twitter.com/tstrome

http://www.amazon.com/gp/product/1118519698/

Developing an Analytics Strategy Framework that Drives Healthcare Transformation Copyright © 2014 Trevor Strome

Healthcare Analytics and the Information Value Chain

2

Performance Objectives

Quality Goals

Improvement Approach

Data

Business Processes

Analytics

What DID Happen

What IS Happening

What Will Happen

Decisions & Actions

Outcomes Evaluation

Healthcare analytics is the system of tools, techniques, and people required to consistently and reliably generate the accurate, validated, and trustworthy business and clinical insight needed to take appropriate actions and achieve measurable, desired outcomes.

System Insight


Problem Domain

• Healthcare organizations (HCOs) are facing increasing quality, financial, and regulatory pressures, and must transform to achieve sustainability.

• The three fundamental information needs of healthcare improvement are to identify: – What quality/performance/safety aspects need to improve?– What processes must change to result in improvement?– What change (if any) has occurred?

3


Going beyond reporting…

• Healthcare organizations require better insight into their operations and accountability for their performance.

• Healthcare organizations must allow for creative use of available data and analytic tools to foster decision making – in real time and near the point of care.

• To keep up with pace of change, analytics development needs to adopt an agile approach which values innovation and experimentation.

4


Analytics - Attaining the right insight at the right time

What Happened?(Reports)

What’s Happening Now?

(Alerts)

What Will Happen?(Extrapolation)

How and Why Did It Happen?(Modeling)

What’s the next best action?

(Recommendation)

What’s the best/worst that can

happen?(Prediction, Simulation)

Past Present Future

Information

Insight

Notes:

Adapted from: Davenport TH, Harris JG, & Morison R. Analytics at Work. Boston: Harvard Business School Publishing Corporation, 2010.

5

Developing an Analytics Strategy Framework that Drives Healthcare Transformation Copyright © 2014 Trevor Strome 6

What is an Analytics Strategy?

• A strategy that ensures analytics development and capabilities are in alignment with enterprise quality and performance goals– avoids the “all dashboard, no improvement” syndrome

• Helps to achieve optimal use of analytics– can mean the difference between a “collection of reports” versus

a high-value information resource

• Analytics Strategy should align with other relevant strategies including: – Business Intelligence (BI) strategy– Information Technology (IT) strategy– Quality Improvement (QI) strategy


Building and Executing a Successful Framework

• Understand requirements– Review strategy components with stakeholders– Identify how analytics are currently used– Determine what capabilities will be needed (short & long term)

• Identify gaps and mitigate risks– List known/potential gaps and their mitigation approaches– Prioritize gap mitigation based on impact, effort, & cost

• Execute plan– Assign task owners and target implementation deadlines– Monitor progress and apply mid-course corrections

7

Strategic Analytics System Framework


Analytics System

Business & Quality Context

Stakeholders & Users

Processes & Data

Tools & Techniques

Team & Training

Technology & Infrastructure

Analytics System Framework

9

An effective analytics system is more than simply a reporting/BI tool layered on top of a data source.


Adding SWOT to Strategy

• Traditional “SWOT” analysis can be layered onto the components (and sub-components) of analytics strategy.

10

Strengths Weaknesses Opportunities Threats



Data & Processes

Tools & Techniques

Team & Training



Analytics Strategy



Processes & Data

Tools & Techniques

Team & Training



Business Context: Enterprise Goals, Objectives, and Strategy

• Goals:– Are what the organization is aiming to achieve. – Define the performance and quality targets of the organization– Answer “why” the organization is (or should be) engaging in

certain activities

• Strategy– Outlines how the organization expects to achieve its goals

• Analytics must provide insight into past, current, and anticipated future progress towards meeting the enterprise goals.


Aligning Strategic and Tactical Quality Objectives

• Analytics is the “glue” which ties strategic objectives and tactical activities together.

• Objectives of unit- or department-based improvement initiatives should, where possible, align with the quality objectives of the organization as a whole.– Prevents misdirected/wasted activity– Enables the HCO to monitor progress and evaluate outcomes

Strategic Level Strategic Objectives

Analytics Metrics Indicators Targets

Tactical Level Tactical Objectives

A reminder that the customer (“the patient”) is the ultimate reason for the work we’re doing.

13

Voice of the customer


Quality Strategy / Improvement Approach

• Quality Strategy outlines the steps and approach the organization is going to be taking to achieve quality goals/objectives.

• Which QI approaches are utilized (i.e., Lean, Six Sigma) will impact what data is required, how it is analyzed, and how it is communicated.

• Analytics development teams and quality improvement teams must work closely together – to ensure that information requirements of users and the delivery

by via analytics are in sync.

• When executing the analytics strategy, always ask “are we taking appropriate and necessary steps towards achieving the organization’s quality and performance goals?”


Analytics Strategy



Processes & Data

Tools & Techniques

Team & Training



Stakeholder Analysis

• A stakeholder is a person (or group of persons) that are:– impacted by, users of, or otherwise have a concern (or interest

in) the development and deployment of analytical solutions throughout the healthcare organization.

• When developing an analytics strategy, it is important to understand what each of the likely analytics stakeholders will require, and develop approaches to ensure they are getting what they need.

16


HCO Stakeholder Types

17

Stakeholder Description

Patient The person whose health an healthcare experience we’re trying to improve with the use of analytics

Sponsor The person who supports and provides financial resources for the development and implementation of the analytics infrastructure

Influencer A person who may not be directly involved in the development or use of analytics, but who holders considerable influence over support of analytics initiatives.

Customer / User A person in the HCO who accesses analytical tools, or uses the output of analytical tools, to support decision making and to drive action.


Analytics Use Cases

• A use case is a brief description of how analytics will be used by a stakeholder. Analytics use cases can help to:– identify any gaps in analytics capabilities, and – reduce the likelihood that critical analytics needs will be missed.

• Analytics use cases help identify:– what data elements are most important and what indicators will

be necessary to calculate, and – what types of usability and presentation factors (such as

dashboards, alerts, and mobile access) need to be considered.

• TIP: Develop high-level use cases when outlining the analytics strategy, and drill down in more detail as new analytical applications are designed and built.

18


Example Analytics Use Cases

19

Customer / user

Sample use case(s)

Physician Uses personalized performance report to adjust care practices.

Unit manager Determine which patients are likely to exceed length of stay targets.

QI team leader Identify bottlenecks in patient flow.Evaluate outcomes of QI initiatives.

Healthcare executive

Evaluate and monitor overall performance of the organization.

Processes & Data

Analytics Strategy



Processes & Data

Tools & Techniques

Team & Training



Data considerations

• Data is the “raw material” of analytics.

• Modern computerized clinical systems (such as electronic medical records) contain dozens if not hundreds of individual data elements. – The potential exists for thousands of possible data items from

which to choose for analytics.

• An analytics strategy must consider:– how to determine which data is necessary for quality and

performance improvement– how the data is managed and its quality assured– how data links back to business processes for necessary

context.

21


Data Issue Example

Data Sources • What are the sources of data?• What data is necessary to address key

business issues?

Data Quality • How good is the quality of available data?• Is the data “good enough” for analytics?• What gaps in data exist?• Does metadata exist?

Data governance • Who is responsible for data management, governance, and stewardship?

• What policies and procedures exist?

Business Processes • What business processes and procedures align with important quality issues?

• What data is available for measuring processes? Are proxy measures available?

Data Considerations for Analytics Strategy

22


Business Processes

• Business processes provide essential context to the data.

• Most quality improvement methodologies monitor progress and evaluate performance and outcomes using indicators based on process data. – This requires a strong alignment between key business

processes and the data that measures those processes.

• As part of the analytics strategy, you should consider:– if and how current business processes are documented, and – how data items are mapped to these documented business

processes.

• TIP: stacks of Visio charts becomes unmanageable very quickly!

23


Big Data and Analytics

• Big data “describes large volumes of high velocity, complex, and variable data that require advanced techniques and technologies to enable the capture, storage, distribution, management, and analysis of the information.”1

• Big Data represents big opportunity– U.S. health care data alone reached 150 exabytes in 2011. – Big data for U.S. health care will soon reach zettabyte (1021

gigabytes) scale and even yottabytes (1024 gigabytes) not long

after.

24

1. Hartzband, D. D. (2011). Using Ultra-Large Data Sets in Health Care. 2011 Sessions (p. 3). e-healthpolicy.org.


Where Does Big Data Come From?

• Web and social media data: Clickstream and interaction data from social media such as Facebook, Twitter, Linkedin, and blogs.

• Machine-to-machine data: Readings from sensors, meters, and other devices.

• Transaction data: Health care claims and other billing records.

• Biometric data: Fingerprints, genetics, handwriting, blood pressure, medical images, retinal scans, and similar types of data.

• Human-generated data: Unstructured and semi-structured data such as electronic medical records (EMRs), physicians’ notes, email, and paper documents.

25

SOURCE: Institute for Health Technology Transformation.


• Using appropriate indicators that align between tactical and strategic levels are necessary. – Tactical-level sub-indicators should align with strategic indicators– Some tactical-level-specific indicators might be necessary for

initiatives that are important at a program, department, or unit level, but don’t directly align with strategic goals.

Indicator

Sub-Indicator 1

Sub-Indicator 2

Sub-Indicator 3

Strategic Level

Tactical Level

Tactical Indicator 1

Using Appropriate Indicators

26


Example Strategic and Tactical Indicator Alignment

27

95% of patients admitted from ED achieve EDLOS < 8hrs

Time to physician

assessment

Time to consult

answered

Time to consult decision

Strategic Level

Tactical Level

Time to inpatient bed

assigned

Time to patient left ED

Analytics Tools and Techniques

Analytics Strategy



Processes & Data

Tools & Techniques

Team & Training



Common Analytical Applications

29

Analytical Application Description

Statistical • Used for deeper statistical analysis not available in “standard” business intelligence or reporting packages

Visualization • Used for developing interactive, dynamic data visualizations that aid with analysis

Data Profiling • Helps to understand and improve the quality of an HCO’s data.

Data Mining • Analysis of large data sets to uncover unknown or unsuspected relationships.

Text Mining • Analysis of unstructured, text-based data to extract high-quality information.

Online Analytical Processing

• Allows analysts to interactively explore data by drilling-down, rolling up, or “slicing and dicing” data.


Inventory of Existing Analytical Tools

• Analytical tools must meet the requirements of analysts building analytics solutions/applications, and the end-users who will rely on the resultant information and insight.

• Conduct an inventory of existing analytics tools to determine if:

– Capability is missing that will be required– Existing capability exists that may not be widely known

• Identify viable best-of-breed vendor solutions that meet requirements; custom-build from scratch if necessary or if participating in research.

30

Team and Training

Analytics Strategy



Processes & Data

Tools & Techniques

Team & Training



Team Development Considerations

• PEOPLE are a critical consideration when developing or expanding an analytics capability within a healthcare organization

• Although having the best tools are nice, having the best (and right) people is critical to achieving the goals and objectives of the HCO

• An analytics strategy must consider:– What kinds of people (and the skills they bring) are necessary– The optimal size and composition of the team– Roles and degree of specialization– What gaps in skills exist, and what training is required– How to attract the best analytical talent– How to retain the analytic talent within your HCO– Optimal organizational structure

32


Organizational Considerations

• Different resource management models exist for analytics teams:– “centralized” analytics office– “distributed” analytics resources– “virtual” center of excellence / competency center (combines

best aspects of centralized and distributed models)

33

Virtual Business Intelligence / Analytics Competency Centre

Senior Management

Decision Support Services

(Analytics)

Central (“Core”) Analytics Analysts

Surgery Program

Program Analytics Resource

Medicine Program


Emergency Program


Technology and Infrastructure

Analytics Strategy



Processes & Data

Tools & Techniques

Team & Training



Healthcare BI and Analytics Technology and Infrastructure

35

Source:

Evelson, B. It's Time to Reinvent your BI Strategy. Forrester Research, Inc.

Reporting and analytics are the “tip of the iceberg” regarding the business intelligence technology stack.


Analytics Stack

Presentation

Visualization Dashboards Reports

Alerts Mobile Geospatial

Quality & Performance Management

Processes Indicators Targets

Improvement strategy Evaluation strategy

Analytics

Tools Techniques Team

Stakeholders Requirements

Deployment Management

Data

Quality Management Integration

Infrastructure Storage

Business Context

Objectives Goals Voice of patient

Focus on the Business

• An abstracted BI stack helps maintain focus on key components of analytics required to address business and clinical goals.

36



• Analytics and reporting are the tip of the iceberg in the business intelligence stack.

• The current, near-term, and long-term analytics needs of the HCO must drive how analytics-related technological capabilities are acquired. The exact complement of tools will depend on the overall needs of the HCO.

• The analytics strategy is an important input to IT hardware and infrastructure strategies and planning as hardware and other system upgrades are considered.

37

Strategy Execution


Strategy Execution Summary

• It is important to implement and adhere to the analytics strategy

• Plan for and schedule activities to address identified gaps

– Establish a selection criteria to determine what projects will get emphasis in light of needs of the business and analytics strategy

– Prioritize activities and desired capabilities to balance resources as new (possibly conflicting) work arises

• Monitor progress towards achieving goals of the analytics strategy

• Ensure that the strategy is a living document that serves as a roadmap for guiding action and doesn’t become “shelfware”


Gap Analysis

• Identify important gaps between current and future state, what the corrective action(s) will be, who owns the actions, and what the due date for corrective actions is.

40

http://www.mindtools.com/pages/article/gap-analysis.htm

Category Current State Target State Corrective Action Priority Owner Due Date



Data & Processes

Tools & Techniques

Team & Training


http://www.mindtools.com/pages/article/gap-analysis.htm


Prioritizing Gap Corrective Actions

• Use the Impact / Effort matrix to help quantitatively determine priority for addressing analytics gaps.

41

Q1

Imp

ac

t (i

nc

rea

sin

g)

Effort/Resources Required (increasing)

Q4

Q2 Q3

Low impact, Low effort“Consider”

High impact, Low effort“Immediate”

High impact, High effort“Evaluate”

Low impact, High effort“Avoid”


For More Information

Author: Trevor Strome

Email: [email protected]

Twitter: @tstrome

Blog: http://HealthcareAnalytics.info

Book: Healthcare Analytics for Quality and Performance Improvement

http://HealthcareAnalyticsBook.com(Published by John Wiley & Sons, Inc, and available on Amazon.com)

mailto:[email protected]

http://twitter.com/tstrome

http://twitter.com/tstrome

http://healthcareanalytics.info/

http://healthcareanalyticsbook.com/



http://www.amazon.com/gp/product/1118519698

Developing a Strategic Analytics Framework that Drives Healthcare Transformation

Health & Medicine

techniques

techniques

large data

infrastructure

analytics

users processes

processes

users data