#HASummit14 John Wadsworth Session #9 Getting The Most Out Of Your Data Analyst Vice President, Technical Operations, Health Catalyst.

#HASummit14

John Wadsworth

Session #9

Getting The Most Out Of Your Data Analyst

Vice President, Technical Operations, Health Catalyst 

#HASummit14

Today’s Agenda

Unlock the data

Analytic tools

Prove or analyze?

Analytic whiplash

Accepting the truth

#HASummit14

Poll Questions 1 - 2

Question #1 -

• How much time would you estimate your analysts spend gathering data (vs analyzing data)?

Question #2 –

• How much time would you estimate your analysts spend analyzing data?

3

#HASummit14 4

Unlock Your Data

#HASummit145

“I told you I wasn’t a hunter gather. I’m an analyst!”

#HASummit146

Conversation this week:

• Shared savings partnerships withholding money because of poor analytics

• Analysts spending 80% or more gathering data

• Data exists in multiple sources (EMR, costing, billing, patient satisfaction, etc.) that are not integrated

How hard can it be to gather the data for analytics?

Analysts - Hunting and Gathering

#HASummit147

Objective #1: Determine the total value of the money in your team bucket.

Rules• Work as a team. Everyone at your table needs to participate.

• Collect your bucket at corresponding colored locations around the room. Send 1 person (runner) from each table.

• Do not retrieve the bucket until you are given the “Go” signal. All teams start at the same time.

• With each task that you complete, ring the bell and you will be given an additional task.

• Complete as many tasks as possible as a team.

• Time limited, so work quickly!

Unlocking Your Data - Exercise

#HASummit14

Metadata: EDW Atlas Security and Auditing

Common, Linkable Vocabulary

FinancialSource Marts

AdministrativeSource Marts

DepartmentalSource Marts

PatientSource Marts

EMR Source Marts

HRSource Mart

Diabetes

…MANY more!

Pneumonia

Less TransformationMore Transformation

Catalyst Adaptive Data Warehouse

FINANCIAL SOURCES (e.g. EPSi, Peoplesoft,

Lawson)

ADMINISTRATIVE SOURCES

(e.g. API Time Tracking)

EMR SOURCE (e.g. Epic, Cerner)

DEPARTMENTAL SOURCES (e.g. Apollo)

PATIENT SATISFACTIONSOURCES

(e.g. NRC Picker, Press Ganey)

Human Resources(e.g. PeopleSoft)

Adaptive Data Warehouse Model

Surgery

#HASummit14

Poll Questions 3 - 4

•Question #3

•In your personal opinion, how important is the analyst role in your organization?

•Question #4

•How important is the role of analyst viewed by your organization?

9

#HASummit14 10

Analytic Tools

10

#HASummit1411

Objective #1: Group your coins by denomination AND stack them at least 5 coins high.

Rules• Work as a team. Everyone at your table needs to participate.

• Do not open the bucket until you are given the “Go” signal. All teams start at the same time.

• With each task that you complete, ring the bell and you will be given an additional task.

• Complete as many tasks as possible as a team.

• Your entire team MUST use the (hand) tools provided you for the complete exercise.

Analytic Tools - Exercise

#HASummit1412

Tools Support Transformation

• Structured Query Language (SQL or variant)

• Data analysis

• Visual representation of information

• Communicate meaningful story through the data

• Domain knowledge

From Hunter-Gather to Analyst

#HASummit14

Recommended Tools for Data-Driven Health System

• Source systems that support query (SQL)

• Let them get to the data

• Business intelligence development tools to build meaningful visualizations

• Cognos, Crystal Reports, Tableau, Qlikview, Excel

• An enterprise data warehouse (EDW)

• Start small and grow as needed

• Assumes data architects will extract, transform, load (ETL) and model data into warehouse

• Scalable platform to grow analytics

#HASummit14

Poll Questions 5 - 6

Question #5

•How often do you act on information provided to you by your analysts?

•Question #6

•Analysts – How often does management act on your analysis and/or recommendations?

14

#HASummit14 15

Prove This

15

#HASummit1416

Analyze the Decision to Build

• “We need to build an observation patient wing”‒ 3 year upward trend in observation patient volume through ED‒ Reimbursements dropping for obs patients get to inpatient or

ED acuity‒ Historically we had an observation wing

• Questions they wanted answered‒ How many beds do we need?

» Clinical data informed bed count estimates‒ What clinical staffing will be needed for the new wing?

» HR and clinical data justified staffing model‒ What will it cost to build the new wing?

» Costing data supported estimate of $.5M - $1M/bed for re-purposing existing beds $2.5M - $5M for 5 bed wing

• WAIT! Has the decision to build already been made?‒ If so, do you need an Analyst … or something else?

#HASummit1417

We asked, “What can the data tell us about the observation patients?”

~70% had chief complaint of chest pain ~90% existing patients in the hospital system ~80% with chest pain had former diagnosis of heart

failure from cardiology clinic/primary care ~75% arrived in ED between 5-10 PM Cardiology clinic closed at 5:00 PM

Analyst recommendation

Keep the cardiology clinic open until 10:00 PM Don’t spend the $2.5M - $5M for an observation wing

Analyze the Data to Inform a Decision

#HASummit14 18

Analytic Whiplash

18

#HASummit1419

“I could catch a trout on a dusty road.”

#HASummit1420

• Leadership discovers a problem

• Analyst assigned to provide insight

• Analyst & others study problem to define scope

• Data gathered then analyzed

• Patterns and correlations begin to emerge

• Leadership brings another problem for analysis or changes direction. Analyst told to “wrap it up and move to the next problem”.

Whiplash Cycle

#HASummit1421

• Analyst & others study problem to define scope

• Data gathered then analyzed

• Patterns and correlations begin to emerge

• Assumptions verified/refuted by knowledge experts closest to the work process being measured

• Adjust logic based on feedback.

• Iterate through process until all logic validated by process owners (in the trenches)

Give sufficient time for analysis, discovery and a recommendation.

Considerations for Improved Analytic Insight

#HASummit1422

• Insufficient time leads to half-baked analysis

• Incomplete analysis undermines credibility

• Lack of credibility creates further dissatisfaction with data and analytics

Risks of Under-resourced Analytics

#HASummit1423

• Perhaps… but before you hire more analysts, consider asking:

‒ Will more analysts get the needed time to do analysis?

‒ No? Increased capacity for incomplete analysis

• Analyst needs the time to work smarter, not harder.

Do You Need More Analysts?

#HASummit1424

• Remove prioritization burden from analysts

• Leadership become proficient with prioritization

• Leadership determine projects of highest priority

‒ Unified front – individual agendas undermine execution‒ Decide what projects will and will not be funded‒ Resist the lure of shiny, new objects‒ Commit resources for top projects to completion‒ Communicate results of prioritization to the masses

Minimize the whiplash of “urgent” projects

Leadership and Prioritization

#HASummit14 25

Accepting the Truth

25

#HASummit14

Poll Questions 7- 8

•Question #7

•On a scale of 1 to 5, how well do you trust information provided through your analysts?

•Question #8

On a scale of 1 to 5, how well does your culture support analysts delivering information that may be perceived as negative or undesirable?

26

#HASummit1427

• Honesty is the best policy for analytic credibility

• CLABSI reported or actuals?

• Confront the brutal facts

‒ “When you turn over rocks and look at all the squiggly things underneath, you can either put the rock down, or you can say, ‘My job is to turn over rocks and look at the squiggly things,’ even if what you see can scare the [heck] out of you.” – Jim Collins

Should I report the whole truth?

#HASummit1428

• Unlock the data for your analysts

• Get the right tools for your analysts and organization

• Leadership become proficient in prioritization

• Develop a culture of accepting the truth

Summary

#HASummit14

Analytic Insights

AQuestions &

Answers

#HASummit14 30

Session Feedback Survey

1. On a scale of 1-5, how satisfied were you overall with this session?1) Not at all satisfied2) Somewhat satisfied3) Moderately satisfied4) Very satisfied5) Extremely satisfied

3. On a scale of 1-5, what level of interest would you have for additional, continued learning on this topic (articles, webinars, collaboration, training)?

1) No interest2) Some interest3) Moderate interest4) Very interested5) Extremely interested

2. What feedback or suggestions do you have?

#HASummit14 31

Upcoming Keynote Sessions3:45 PM – 4:40 PM

13. Healthcare Reform 2.0: Anticipating What’s NextGovernor Mike Leavitt Founder and Chairman of Leavitt PartnersFormer Secretary of the Department of HHS

5:15PM – 6:00 PMReception

6:00PM – 7:00 PMDinner

7:00PM – 7:50 PM14. The Acceleration of Technology In The 21st Century:

Impacts on Healthcare and Ray KurzweilChairman, Kurzweil TechnologiesDirector of Engineering, Google

7:50PM – 8:30 PMEntertainment

Location

Main Ballroom