The Use of Indirect Race and Ethnicity Data in Reducing Health Disparities: A Healthplan Model Diversity Rx October 20, 2010 Peggy Payne, MA, CDE CIGNA.

The Use of Indirect Race and Ethnicity Data in Reducing Health Disparities:

A Healthplan Model

Diversity RxOctober 20, 2010

Peggy Payne, MA, CDE

CIGNA

2

© 2010 CIGNA

Mission: To help the people we serve improve their health, well-being and sense of security

Customers: 46 million customers worldwide

History: 1792 Insurance Company of North America1865 Connecticut General Life Insurance Company1982 Connecticut General and INA form CIGNA

Team: 25,000 Employees WorldwideHeadquarters: Philadelphia, PA

Medical Medical

Dental Dental

Vision Vision

Behavioral Behavioral

Voluntary Voluntary

Disability Disability

Accident Accident

Life Life

InternationalInternational

About CIGNA

3

© 2010 CIGNA

What are health disparities?

Meaningful differences in the health status (risk, prevalence, outcomes, etc) of one sub-population compared to another

No single operational definition, many factors to consider and many ways to look at the data

We focus on percent differences among racial/ethnic groups using the group with the most favorable rate/outcomes as the point of comparison1

1http://www.cdc.gov/nchs/data/series/sr_02/sr02_141.pdf

4

© 2010 CIGNA

Why are health plans interested in demographic data?

To know our customers to provider superior customer service

To identify need for customized interventions

To develop targeted communications

To add meaning to our outcomes

5

© 2010 CIGNA

Spectrum of Difficulty in Defining, Collecting and Understanding Demographic Data

Gender, Age

Most universal descriptors

Collect at enrollment

Location

Useful for examining regional differences

Mailing address collected at enrollment, not 100% accurate

P.O. boxes limit the precision

Race/Ethnicity

Most common descriptor -health disparities

Limited collection; not collected at enrollment

Can be indirectly estimated based on home address and last name

Social/Cultural

Income, education, urban/rural, etc.

significant implications on underlying causes and opportunities for intervention, not collected

Can be estimated based on home address, difficult to assess reliability with no ‘true’ data available for validity checks

Easy Challenging

6

© 2010 CIGNA

The RAND Model2

Uses address and last name by combining the race/ethnicity distribution of where a person lives (census block or tract data) with the race/ethnicity distribution of people with the same last name to determine the probability of a person belonging to each OMB category

The estimates only work aggregated for a population and have limited use at individual level.

How is it possible to estimate race/ethnicity?

1 Office of Management and Budget (OMB) http://minorityhealth.hhs.gov/templates/browse.aspx?lvl=2&lvlID=172

2 http://www.rand.org/pubs/external_publications/EP20060804/

OMB Race/Ethnicity Categories1: American Indian or Alaskan Native Asian Black or African American Native Hawaiian or Other Pacific Islander White Hispanic or Latino

7

© 2010 CIGNA

So what is CIGNA doing with the data?

Analysis– Examining our utilization and HEDIS measures

to identify disparities by race/ethnicity.

– Using Hispanic/Latino ethnicity as a proxy for Spanish language to estimate volume of Spanish-speaking.

Action– Creating pilot initiatives to address disparities identified.

– Training staff to be culturally sensitive.

– Mobilizing staff with knowledge of a specific racial/ethnic culture to serve a subject matter experts to guide the development of culturally-sensitive materials and interventions.

8

© 2010 CIGNA

Lessons Learned

Understand the people behind the numbers.

Race/ethnicity estimates are just that – estimates.

Measure twice, cut once.

Looking at the data from multiple angles will lead to a more informed conclusion and the most effective allocation of resources.

Don’t outsmart your common sense.

A large percent difference observed among various groups doesn’t necessarily mean that action is required.

Questions?Peggy Payne

[email protected] 949.709.1119

“CIGNA” and the “Tree of Life” logo are registered service marks of CIGNA Intellectual Property, Inc., licensed for use by CIGNA Corporation and its operating subsidiaries. All products and services are provided exclusively by such operating subsidiaries, including Connecticut General Life Insurance Company and CIGNA

Health and Life Insurance Company and not by CIGNA Corporation.