Racial/Ethnic Disparities in Quality of Ambulatory Care for Chronic Physical Health Conditions: The effects of physician and care setting characteristics

1
Racial/Ethnic Disparities in Quality of
Ambulatory Care for Chronic Physical Health
Conditions The effects of physician and care
setting characteristics

• Sponsored by The Robert Wood Johnson Foundation,
  New Connections Program
• Academy Health Conference June 2008
• Rhonda BeLue PhD
• The Pennsylvania State University

2
Overview

• Background
• Rationale
• Study Objectives
• Methods
• Data
• Measures
• Analysis
• Results
• Conclusions and Implications

3
Background
4
Background Disparities in Health Care Quality

• Racial/Ethnic inequities in exist in multiple
  domains of quantity and quality of care safety,
  timeliness, effectiveness, efficiency, equity,
  and patient-centeredness
• (Aaron 2003, Aaron 2003, Kirby 2006, IOM,
  Mayberry 2006, Ma 2005, Weisfeld 2005).

5
Background

• Disparities exist across a wide variety of
  treatments for multiple conditions including
  treatment for CVD, Heart failure and diabetes
  (IOM, Unequal Treatment).
• It is believed that poor quality of care for
  ethnic minorities is linked to poor health care
  outcomes. (Lavizzo-Mourey 2005)

6
Background

• In fact, it has been shown that improvements in
  quality of care for all US consumers are
  necessary (Asch 2006).
• Wide variation also exists in racial disparities
  across geographic lines and care settings
• (Baiker 2005, Baiker 2004, Wennberg 2006) Bach
  2004, 2005).

7
Background

• Despite the documented existence of inequities in
  healthcare quality, more work is needed to
  understand and test strategies to improve the
  quality of healthcare for ethnic minority
  populations (Beach 2006).

8
Background

• African Americans are more likely to see health
  care providers in facilities with inadequate
  recourses and by providers with lesser
  credentials than facilities where whites receive
  care (Epstein 2004, Bach 2004).
• African Americans are also likely to have poorer
  continuity of care largely due to lack of regular
  site of care.

9

• African Americans are more likely to be seen in
  hospital clinics and community health centers
  where the chances of seeing the same provider
  across visits are low (Doescher 2001).

10
Background

• Peter Bach et al (2004) found that elderly Blacks
  and Whites are treated at racially homogeneous
  facilities that are either largely White or
  African American.

11
Background

• Elderly blacks receive care at facilities which
• 1) provided more charity care
• 2) had higher percentage of revenue from
  Medicaid
• 3) were more likely to practice in a low-income
  neighborhood and
• 4) were less likely to be board certified in
  their primary specialty.
• 5) Physicians treating mostly white patients
  were more likely to indicate that they could
  confidently provide quality care and access to
  referrals, specialty care, and ancillary services

12
Rationale

• Understanding the characteristics of health care
  facilities can inform interventions and policy
  making related to consumer access to care and
  choice of health care setting, resource
  management and allocation in settings that treat
  racial/ethnic minorities receive care

13
Contribution

• This study adds to the literature by
  investigating the relationship between
  healthcare setting context and quality of care
  received for chronic conditions in adults ages 18
  and older.
• Diabetes will be used as an illustration for this
  presentation

14
Objectives

• Assess
• 1) racial/ethnic differences in the
  characteristics of the facilities where racial
  and ethnic minorities receive care
• 2) The relationship between quality of care for
  diabetes and characteristics of the care setting

15
Conceptual Framework The Chronic Care Model

• Summarizes the basic elements for improving care
  in health systems at the community, organization,
  practice and patient levels.
• Community characteristics resources and policies
• Health system characteristics clinical
  information systems, design and delivery system
• Provider characteristics prepared and proactive
• Patient characteristics activated patient

16
Methods
17
Data and Sample

• The 2005 National Ambulatory Medical Care Survey
  (NAMCS) were used for this investigation.
• NAMCS uses a multistage stratified probability
  sample of patient visits in ambulatory care
  settings to enable nationally
• NAMCS is designed obtain objective information
  about ambulatory medical care services in the
  United States
• Whites, Blacks, and Hispanics aged 40 and older
  with Diabetes
• Several other ACS conditions were explored

18
Measurement
19

• Patients with conditions (diabetes) of interest
  were identified via ICD9 code (as indicated by
  the NAMCS diagnosis variable) and confirmed by
  physician report.
• First, second and third diagnoses were included
• Checked against physician report

20
Measures Facility and Physician Characteristics

• Facility
• Solo or group practice
• Ownership
• Lab Testing available
• Difficulty with referrals
• EMR
• of revenue
• Claims submitted electronically

• Physician
• Employment status
• Does email consults
• Telephone consults
• Hospital visits
• Time spent with patient

21
Measures Patient Characteristics

• Demographics age, gender, insurance status
• Comorbid illnesses bmi, total number of chronic
  conditions, number of medications
• Number of visits in the past 12 months
• Number of Medications

22
Measures Dependent Variables

• Quality measure(s) were derived based on measures
  from
• The National Quality Measures Clearing House
• Selected measures relevant to ambulatory care
  settings

23

• Quality indicators were calculated as the
  percentage of visits in which the patient
  received appropriate quality of care divided by
  the total number of visits.

24
Diabetes Quality Outcome

• Ambulatory care management of diabetes measure
• Diabetes
• Process
• of patients who received a HA1c test
• Should be taken every 3 months, especially in
  those with poor glycemic control

25
Analytic Strategy
26
Analytic Strategy Aim 1

• Chi-square tests using Stata survey procedures
  were employed to examine the relationship between
  race/ethnicity and care setting characteristics

27
Analytic Strategy Aim 2

• Logistic regression analyses were employed to
  examine the relationship between race, provider
  and facility characteristics and quality
  indicator controlling for patient demographic and
  health status indicators
• Sample/design weights were incorporated
• GEE for parameter estimation
• Assessed moderation-within race/ethnicity models
• Bonferroni adjustment for multiple comparisons
• Modeled the probability of receiving HGBA

28
Results
29
Sample Characteristics

• Sample represents a total of 15858 patient
  encounters among those over 40,
• of patient encounters among those with diabetes
  (weighted)
• N3078 diabetics
• White 11.5
• Black 18.9
• Latino/a 16.8

30
Race and Outcome Measures
31
Diabetes

• Among those who have diabetes, approximately at
  any encounter
• 13.9 of whites receive an HA1c screening
• 7.4 of blacks receive an HA1c screening
• 8.4 of Latino/as receive an HA1c screening
• Significant at Plt0.001

32
Results Aim 1
33
Results Aim 2-Whites

• More that 50 or revenue from Medicare-
• OR 0.24 ( 0.1, 0.5)
• Difficulty Referring to Medicaid
• OR0.35 ( 0.2, 0.7)

34
Results Aim 2-Blacks

• Seen in a solo practice
• OR 1.8(1.7,2.1)
• More that 50 or revenue from Medicare-
• OR 0.4(0.1,0.7)
• On Site Lab
• OR 5.7(1.5,7.2)

35
Results Aim 2-Latinos

• Seen in a solo practice
• OR 2.1(1.6,2.7)
• On Site Lab
• OR 1.5(1.5,7.2)

36
Limitations

• Cross-sectional data
• Lack of financial data to accurately asses level
  of resources
• Need of composite score or better interpretation
  of what facility characteristics mean
• Patient preferences for care setting
• Combine several years to increase N for minority
  groups and to allow for more comparisons

37
Conclusions

• Future health services and quality initiatives
  may benefit from focusing on improving resources
  in care settings in order to improve quality and
  treatment of chronic conditions in racial and
  ethnic minorities

38
Acknowledgements

• The Robert Wood Johnson Foundation, New
  Connections Program
• Dr. Debra J. Perez
• Dr. Margarita Alegria
• Junior Investigator Forum Colleagues