Embed Size (px)
Citation preview
Making Data Real in
Outcomes-Based Work
Jeff Capobianco, PhD, LLP
Presentation Objectives
> Summary: A major component of the Affordable Care Act and
integrated health approaches is the ability to leverage data to better
understand how to improve the health of consumer populations. This
webinar will describe the components of population health management,
the important use of registries, and dashboards to inform care from the
frontline clinician to the CEO.
Objectives:
> Review of State of Data Collection in Healthcare
> Define Population Health Management;
> Review Ohio IH Health indicators;
> Define What a Data Registry is and How to Use It;
> Review How Data Collection Methods Can Drive QI:
> Review the Creation and use of data dashboards.
Why all the talk of
Big Data?
• In 2011 alone, 1.8 zettabytes of data were created globally.
• To put this into perspective, this volume of data equates to
200 billion, 2-hour long HD movies, which one person
would need 47 million years to watch in their entirety.
• What’s more, this volume of data is expected to double this
and every year going forward.
Source: http://ihealthtran.com/iHT2_BigData_2013.pdf
Why all the talk of
Big Data?
• U.S. health care data alone reached 150 exabytes in 2011.
• Five exabytes (1018 gigabytes) of data would contain all
the words ever spoken by human beings on earth.
• At this rate, big data for U.S. health care will soon reach
zettabyte (1021 gigabytes) scale and even yottabytes
(1024 gigabytes) not long after.
Source: TechAmerica Foundation, 2012
Why all the talk of
Big Data?
• Kaiser Permanente, the California-based health network
which has more than 9 million members, is estimated to
have between 26.5 petabytes and 44 petabytes of patient
data under management just from electronic health record
(EHR) data, including images and annotations.
• This amounts to the same amount of information contained
in 4,400 Libraries of Congress.
Source: Hartzband, 2011
Pt. Centered Medical
Home Domains
PCMH Domains
Enhance Access and Continuity
Identify and Manage Patient
Populations
Plan and Manage Care
Provide Self-Care
Support and Community Resources
Track and Coordinate
Care
Measure and Improve
Performance
>Population Management allows staff to look across
the entire population of people served to see if they
are improving and how much services cost.
>Measure preventative, chronic and acute care;
utilization affecting costs; patient experience and
report performance
>Use and monitor effectiveness of quality
improvement processes
Identify and Manage Patient
Populations
“The main reason seems to be a lack of integration
of health IT into clinical workflow in a way that
supports the cognitive work of the clinician and the
workflows among (partner) organizations, within a
clinic and within a visit.”
Source: Carayon & Karsh, (2010). AHRQ Publication No. 10-0098-EF
Why is Using Data
for Improving Health Outcomes so Difficult?
Collecting, Using, &
Sharing Health Data
BENEFITS
>More efficient workflow (e.g. less time spent
handling laboratory results)
>Improved access to clinical data
>Streamlined referral processes
>Improved quality of care--Better health outcomes
>Improved patient safety, including fewer
prescribing errors and fewer hospital readmissions
>Cost savings (e.g. eliminating costs of storing
paper records)
>Downsizing personnel
>Increased revenue (e.g. government incentives for
use of health IT)
>Pay-for-performance incentives
BARRIERS
>Lack of Leadership
>Lack of strategic plan for data use & health IT
>Costs of EHR implementation
>Cost of establishing and maintaining links
between EHRs and HIE networks
>Security and privacy issues
>Liability Provider’s concern to be held liable for
information from outside sources/labs
>Misaligned incentives (who pays and who
benefits)
>Provider reluctance to relinquish control of patient
information to competing systems
>Technical barriers (e.g. lack of interoperability
among EHRs)
>Lack of IT training and support
Source: Fontaine, Ross, et al. (2010). Systematic Review of HIE in Primary Care Practices, JABPM
Primary Drivers of Successful
Data Collection & Use
> Do you have a strategic plan for how data is shared and
leveraged to improve care? If you do, is it being
used/updated regularly?
> Is your leadership involved in the creation, articulation,
and monitoring of this plan?
> Are your IT, QI, finance, and clinical leads meeting
regularly to execute this plan?
> Requirements for analytics:
• Accessible High Quality Data
• Enterprise/Future Orientation
• Analytical Leadership
• Strategy Targets
• Analysts
Analytics at Work: Smarter
Decisions Better Results by Davenport, Harris & Morison
Analytics at Work: Smarter
Decisions Better Results by Davenport, Harris & Morison
Stage One: The Analytically Impaired
> The organization lacks one or several of the
prerequisites for serious analytical work,
such as data, analytical skills, or senior
management interest.
Stage Two: Localized Analytics
> There are pockets of analytic activity within
the organization, but they are not
coordinated or focused on strategic targets.
Analytics at Work: Smarter
Decisions Better Results by Davenport, Harris & Morison
Stage Three: Analytic Inspired
> The organization envisions a more analytic
future, has established analytic capabilities,
has a few strategic initiatives under way, but
progress is slow often because of lack of
leadership, future orientation, reliable data,
strategic targets or staffing/analysts.
Analytics at Work: Smarter
Decisions Better Results by Davenport, Harris & Morison
Stage Four: Analytic Agencies/Companies
> The organization has the needed human and
technological resources, applies analytics
regularly, and realizes the benefits across the
organization. But its strategic vision/focus is not
grounded in analytics, and it hasn’t turned
analytics to competitive advantage.
Analytics at Work: Smarter
Decisions Better Results by Davenport, Harris & Morison
Stage Five: Analytic Competitors
> The organization routinely uses analytics as a
distinctive business capability. It takes an
enterprise-wide approach, has committed and
involved leadership, and has achieved large-scale
results. It portrays itself both internally and
externally as an analytic competitor.
Analytics at Work: Smarter
Decisions Better Results by Davenport, Harris & Morison
Population Based
Care/Health Management
> Strategies for optimizing the health of an entire client
population by systematically assessing tracking, and
managing the group’s health conditions and treatment
response. It also entails approaches to engaging the entire
target group, rather than just responding to the clients that
actively seek care.
Approach to better managing all aspects of pt. health from wellness to complex care, by assessing health & health provision beyond a single episode of care. PHM helps clinicians assess their entire population and stratify it into various stages across the spectrum of health:
• Those who are well need to stay well by getting preventive tests completed
• Those who have health risks & need to change their health behaviors so they don’t develop the conditions they’re at risk for
• Those who have chronic conditions need to prevent further complications by closing care gaps and also working on health behaviors
source: adapted from phytel.com
Population Based
Care/Health Management
• PHM has traditionally been the work public health departments.
• Electronic Health Records & Registries will allow
clinicians/administrators to conduct PHM.
• The Affordable Care Act is making PMH a requirement.
What is Population
Health Management?
Components of PHM:
1. Knowing what to ask about your
population
2. Data registry describing your population
3. Use Dashboards for making data readable
4. Engage in QI Process to respond to the
findings
The Questions you want answers to
about your population are: 1. Who are you serving? Who are you not
serving but could/should be?
2. What are the costs for the average patient?
3. What kind of services are they getting, where,
& when?
4. What is the patient’s response to treatment?
5. What is the patient’s opinion of their care?
Expected Outcomes
from PCMH/ACA
> Decreased ED visits
> Decreased hospitalizations/Readmits
> Improved prevention, cardiac & diabetes care
> ↑ appointment access & proper med. use
> Better work environment: ↓ staff stress, ↑ staff
retention & recruitment
> ↓ disparities
> That’s great buy how do we track/eval this?
Meaningful Use is Driving HIT/Data Use Adoption
(source: Afia.com)
» Federal Incentive program to encourage use of HIT
» Required objectives to ensure physicians are “Meaningful Users” of HIT
Updated Ohio HH
Measures
Adult Child
1 Prenatal and Postpartum Care - Timeliness of Prenatal Care X X
2 Follow-Up After Hospitalization for Mental Illness - 7-day Follow-Up X X
3 Screening for Clinical Depression and Follow-up Plan X
4 Initiation and Engagement of Alcohol and Other Drug Dependence Treatment X X
5 Adolescent Well-Care Visits X
6 Controlling High Blood Pressure X
7 Smoking and Tobacco Use Cessation X X
8 Adults' Access to Preventive/Ambulatory Health Services X
9 Comprehensive Diabetes Care - LDL-C Screening X
10 Comprehensive Diabetes Care: HbA1c level Less Than 7.0% X
11 Cholesterol Management for Patients With Cardiovascular Conditions X
12 Adult Body Mass Index (BMI) Assessment X
13 Use of Appropriate Medications for People With Asthma - Total X
14 Appropriate Treatment for Children With Upper Respiratory Infection X
15 Ambulatory Care - Sensitive Condition Admission X X
16 Care Transition – Transition Record Transmitted to Health care Professional X X
17 Plan- All Cause Readmission X
18 Inpatient and Emergency Department (ED) utilization Rate X X
19 Medication Reconciliation Post-Discharge X x
IH Measures
Interpreting the
Specifications
The measure specifications will provide the following:
• Brief measure description.
• Definition of measure numerator.
• Definition of measure denominator.
• Exclusions to measure, if applicable.
• Description of report periods.
• Tables detailing the codes used to identify
denominator criteria, numerator criteria, and
exclusions (if applicable).
IH Measures
Interpreting the Specifications
Patient Registry
“…an organized system to collect uniform data
(clinical and other) to evaluate specified outcomes
for a population defined by a particular disease,
condition, or exposure, and that serves one or
more predetermined scientific, clinical, or policy
purposes.”
Source: Gliklich RE, Dreyer NA, eds. (2010).
Registries for Evaluating Patient Outcomes: A User’s Guide. 2nd ed.
Data, Information,
& Knowledge
> What is data?
Granular or unprocessed information. Data are symbols.
(e.g., 1 = smoker; 2=nonsmoker)
> What is information?
Information is data that have been
organized/contextualized and communicated in a
coherent and meaningful manner. Answers the who, what,
when, and where? (e.g., # of smokers on ABC team
offered stop smoking Evidence-based Treatment during
the past quarter)
> What is Knowledge?
Information evaluated and organized so that it can be
used purposefully. The capability to take information and
hear the story it is telling…to see the relationships
between and within information so you can act!
Knowledge answers the questions of how and why.
e.g. Team ABC achieved 60% of smokers being offered
a stop smoking EBP and this exceeded our target for
this quarter which was 50%. The QI process is working!
,
Data, Information,
& Knowledge
What is the ultimate purpose of
collecting & sharing data?
To turn it into action! (AKA QI)
Adapted from:
Kolhbacher, et al.
(2008) AHCMJ
Dashboards for Clinical & Admin.
Staff: Should allow the data to tell a story about the people being served &
the care provided
• Should be Simple to Start--target only a few key aspects of
population & their care
• Should be Colorful--use red, yellow, green to draw the eye
Complex Dashboard…
Simpler Dashboard…
That’s Nice…but…
• Our EMR doesn’t have a registry component to
help with PHM!
• We don’t have an EMR…yet!
• Our IT staff are hard to engage with helping with
this!
• Okay I’m sold but where do we start? How do we
design this?
Basic Dashboard
Pt Name/ID
Age Gender Race Axis I Axis II Axis III PCP PCP seen last
6 months? (Yes/No)
PHM & QI Process
• Once you have some data begin setting up benchmarks & risk cutoffs
for the data.
Examples include:
• % Patients seen PCP in last year (benchmark=100%)
• Risk cutoffs for BMI, A1c, etc.
• % Pts screening positive for depression
• Engage staff in Plan-Do-Study-Act Rapid Cycling to see if the data
changes as pts. respond.
Southeast , Inc.
HgbA1c by Cluster at Enrollment*
(5.7-6.4 at risk for diabetes)
2A 2B 3A 4A
n= 202 n = 80 n = 82 n = 43
*For enrollees with only a Baseline NOMs at time of analysis; **F (3,403) = 3.0, p < .05
Cluster**
At Risk
Ideas for Making Data
Meaningful to Frontline Staff
• Communicating/messaging/evangelizing the importance of data by
articulating how it is used for at the organization/strategic,
departmental, team, individual consumer, and staff levels.
• Incorporate discussion of data into all team interactions (e.g., make it
fun, include data in decision making about care, in team meetings, staff
training, etc.).
• Start with low-tech approaches (e.g., Data Jams! Where you provide a
fruit tray, data and a list of questions to get people talking).
• Train QI staff or high QI/UR/IT staff who can manipulate and help
convey the story your data is telling you.
Discussion
Questions?
What are you doing to drive data driven
decision making in your organization?
Some helpful links:
> http://www.integration.samhsa.gov/ (Great resource on everything
integration)
> http://www.integratedcareresourcecenter.com/ (Website detailing what
is happening with health reform in each state)
> http://www.chcs.org/ (Website focused on publicly funded healthcare
and the transformations underway)
> http://www.h2rminutes.com/main.html (Updates on the ACA for
professions—great site to sign up for email notices)
