Health Affairs, 24, no. 2 (2005): 516-526 doi: 10.1377/hlthaff.24.2.516 © 2005 by Project HOPE	New Online   House Health Reform Bill   Paying for Reform   Vetting AHIP's Report   HIV/AIDS Costs   Brief: Insurance Reform   HA Blog Top 10

This Article Abstract Figures Only Reprint (PDF) Online Appendix Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (8) Citing Articles via Google Scholar Google Scholar Articles by Fremont, A. M. Articles by Rector, T. Search for Related Content PubMed PubMed Citation Articles by Fremont, A. M. Articles by Rector, T. Related Collections Access To Care Insurance Coverage Business Of Health Minority Health Research And Technology Safety-Net Systems Consumer Issues Demography

DataWatch

Use Of Geocoding In Managed Care Settings To Identify Quality Disparities

Abstract

 
Tracking quality-of-care measures is essential for improving care, particularly for vulnerable populations. Although managed care plans routinely track quality measures, few examine whether their performance differs by enrollee race/ethnicity or socioeconomic status (SES), in part because plans do not collect that information. We show that plans can begin examining and targeting potential disparities using indirect measures of enrollee race/ethnicity and SES based on geocoding. Using such measures, we demonstrate disparities within both Medicare+Choice and commercial plans on Health Plan Employer Data and Information Set (HEDIS) measures of diabetes and cardiovascular care, including instances in which race/ethnicity and SES have distinct effects.

Although managed care plans routinely collect and monitor performance on quality measures, such as the National Committee for Quality Assurance (NCQA) Health Employer Data and Information Set (HEDIS), few plans examine these data for possible disparities based on enrollees’ race/ethnicity or SES. A key barrier is the fact that few plans routinely collect race/ethnicity and SES information, which could then be linked to performance on quality measures.² The reasons plans do not collect this information reflect concerns such as uncertainty about the legality of collecting racial/ethnic data and fear that some consumers or advocacy groups would assume that plans were using this information inappropriately, perhaps to avoid enrolling or reenrolling minority patients.

Despite these concerns, plans’ interest in collecting these data to identify and target potential disparities is increasing.³ At least one large managed care organization (MCO) has already embarked on a well-publicized and -received effort to collect race/ethnicity information to identify and address disparities. Although specific results of this effort are not yet available, the plan’s initial experiences suggest that it takes several years for even a committed plan to obtain this information from a sufficient sample of enrollees to adequately assess disparities.

Plans’ interest in examining disparities in care is being facilitated by the Centers for Medicare and Medicaid Services (CMS), which decided to provide Medicare+Choice (M+C) plans with race/ethnicity information on their enrollees. This decision was supported by several studies that linked the CMS race/ethnicity information to HEDIS performance data on selected measures from a national sample of M+C plans.⁴ These studies showed substantial disparities on a variety of measures and helped demonstrate the feasibility of using CMS race/ethnicity data and HEDIS measures to identify disparities. Nevertheless, since these studies involved only M+C enrollees, it remains unclear whether disparities also exist among commercial enrollees. Further, although CMS race/ethnicity data may offer a reasonable alternative to collecting data directly from enrollees, most managed care enrollees are in commercial plans, which lack access to CMS data. These data do not include information about enrollees’ SES other than Medicaid eligibility. Finally, these studies used data pooled across a national sample of several hundred plans, so it is unclear whether disparities are limited to a minority of plans or are consistent across plans.

For this study we estimated enrollees’ race/ethnicity and SES indirectly based on geocoding and then used that information to examine disparities on HEDIS measures of quality of care for cardiovascular disease and diabetes.

Study Data And Methods

Top Study Data And Methods Study Results Discussion NOTES

 
Geocoding is a method in which certain characteristics of a person are estimated based on characteristics of the person’s area or neighborhood. Plans can perform this straightforward process quickly and inexpensively. Although geocoding does not provide precise estimates of a given person’s characteristics, it can provide reasonably accurate estimates when applied to a group or population.⁵

We used geocoded measures to examine racial/ethnic and SES disparities for M+C and commercial plan enrollees. We analyzed HEDIS process-of-care measures for diabetes and cardiovascular care and also conducted secondary analyses of disparities on HEDIS intermediate outcome measures. To assess the validity of using geocoded measures in estimating enrollees’ race/ethnicity and associated disparities in care, we compared results on a subset of measures for M+C enrollees when race is based on geocoding versus individual-level race information from the CMS. To assess how disparities are distributed across plans, we also examined disparities on selected measures separately within individual plans.

Data sources. For our primary analyses (disparities on selected HEDIS process measures), we obtained data from nine M+C and ten commercial plans. When data were obtained in 2000, these plans served approximately 195,116 Medicare and 2,151,050 commercial enrollees across four regions. Data on enrollment, patient address, medical care encounters, and claims were contained in a research database maintained by the MCO of which the nineteen plans were part.

The research database lacked chart data needed to calculate HEDIS intermediate outcome measures. For these analyses, we obtained enrollee-level performance data directly from plans. Unfortunately, only one in the research database used for our primary analyses was able to provide intermediate outcome data during the data collection phase. However, we obtained intermediate outcome data from nine additional plans in the same MCO. Together these ten plans included three M+C and seven commercial plans, representing 92,483 M+C and 1,763,292 commercial enrollees, respectively.

Measures of race/ethnicity and SES. Our measures of race/ethnicity and SES were derived from geocoded data. This approach takes advantage of the correlation between neighborhoods’ sociodemographic characteristics and certain characteristics of their residents. We geocoded to the census block group level, which corresponds to a small neighborhood with approximately 1,000 residents and tends to provide more precise estimates than larger areas such as ZIP codes.

Some characteristics are well suited to estimating from geocoding; others are not. For example, geocoding can be used to identify people likely to be black, since racially segregated neighborhoods persist in many U.S. regions and, thus, people living in predominantly black neighborhoods are likely to be black themselves.⁶ Geocoding is less well suited for identifying likely Asians and Hispanics, since these groups tend to live in less segregated neighborhoods. Other techniques such as surname analyses have proved to be more reliable methods of determining the race/ethnicity of Hispanics and Asians when this information is not otherwise available.⁷ Thus, in this study we focus on a single geocoded measure of race/ethnicity: whether or not an enrollee lives in a predominantly black block group, defined as a neighborhood where more than 66 percent of residents are black.

Geocoding also can provide reasonable estimates of SES. For example, enrollees living in poor neighborhoods tend to be far less affluent than those living in neighborhoods without poverty. We computed several different SES measures, but since results were similar for each measure, we report only one measure, living in a poor neighborhood, which we defined as a block group where more than 20 percent of the residents have an income below the federal poverty level.

Geocoding process. We used a vendor, Mapping Analytics, to convert enrollees’ addresses to the corresponding census block group number. All but 8 percent of addresses were matched to the block group level, which is an excellent rate. Enrollees with matched and unmatched addresses were similar with respect to age, sex, insurance type, plan, medical condition, and race/ethnicity (for M+C enrollees). We then linked each enrollee’s block group number to selected information about that block group from 1990 census data (2000 census data were not yet available) and calculated the measure of race/ethnicity and SES described above.

Although our geocoded measures of SES were previously validated, the geocoded measure of race/ethnicity was not.⁸ To help validate this measure, we also obtained race/ethnicity information on individual M+C enrollees from the CMS. Individual-level measures of race/ethnicity (such as self-reported race/ethnicity) for commercial enrollees were unavailable.

Quality measures. Our analyses focused on six HEDIS 2000 measures of whether specific processes of care were performed for eligible enrollees: in diabetics, an annual check of glycosolated hemoglobin (HbA1c), low density lipoprotein (LDL), and urine protein levels, as well as a dilated eye exam; beta-blocker prescription for myocardial infarction (MI) patients; and LDL check in patients after a cardiac event.⁹ We focused on the process measures because the NCQA permits plans to calculate these from administrative data alone. Thus, plans can easily calculate these measures for all eligible enrollees, increasing the likelihood of detecting disparities. We used this approach, calculating performance (that is, percentage of eligible enrollees receiving an indicated service) for all eligible enrollees.

We also assessed disparities for four HEDIS intermediate outcome measures: adequately controlled LDL cholesterol after a cardiac event, blood pressure in hypertensives, and HbA1c and LDL in diabetics. For this paper we treated assessments with these measures as a secondary analysis, since we had data from only a few plans and enrollees for them. In contrast to the process measures, for which we could assess care for all eligible enrollees directly from administrative data, we were limited to samples of approximately 411 enrollees (the NCQA’s suggested sample size) or fewer from each of the ten plans that provided us with these data. Consequently, our capacity to detect racial/ethnic and SES disparities was generally more limited than for the process measures.

Analysis. In our primary analyses, we used chi-square tests to compare the percentage of eligible patients in each racial/ethnic or SES subgroup who received the service specified by each process measure. We conducted all comparisons separately for M+C and commercial enrollees. Since specifications were designed so that all eligible patients should receive the indicated care regardless of other factors, we focused on unadjusted comparisons as recommended by the NCQA.

We also conducted several secondary analyses. First, we compared results of the racial disparities analyses above using the geocoded measure of race/ethnicity with those obtained when a more direct measure of race/ethnicity from CMS data was used. Second, we used multivariable logistic regression to compare the adjusted probability of M+C and commercial plan enrollees in each subgroup receiving specified care. To address the possibility that any disparities observed reflect differential enrollment of minority or low-SES enrollees into plans with lower average performance, we initially adjusted for the specific plan from which the enrollee obtained care, and then we added sociodemographic characteristics including enrollees’ age, sex, and, depending on the model, race or SES (or both). Third, to assess the extent to which disparities varied within plans, we repeated the primary analyses at the individual plan level. We limited these analyses to the four diabetes process measures, which had the largest denominators (the number of eligible enrollees). In our final analyses, we examined unadjusted and adjusted racial/ethnic and SES disparities on the intermediate outcomes measures.

Study Results

Top Study Data And Methods Study Results Discussion NOTES

 
The number of enrollees eligible for each HEDIS process measure and their characteristics are shown in Exhibit 1. Far more enrollees met NCQA eligibility requirements for the diabetes measures than for the two cardiac measures. M+C enrollees were older than commercial enrollees, with a higher percentage living in predominantly black or poor neighborhoods. A higher percentage of diabetic-than cardiac-eligible enrollees also lived in these neighborhoods. Exhibit 2 shows similar information for the HEDIS intermediate outcome measures.

When we compared estimated performance and disparities in M+C plans based on CMS measures of race/ethnicity for individual enrollees with results using our geocoded measure, the results were essentially the same (Exhibit 4). Indeed, disparities differed by less than one percentage point for all six measures.

Measures of race and SES each remained significant on several measures despite statistical adjustments for both. For example, although the patterns of racial and SES disparities were similar, the adjusted analyses for LDL check in diabetic M+C enrollees not only showed a disparity of eight percentage points associated with race (p <.01), but also revealed a distinct effect of SES that was associated with an additional disparity of eight percentage points (p <.01).

Exhibits 5 and 6 illustrate the results of our disparities analysis within individual plans. When we observed a disparity on a given measure (such as LDL check for diabetics) in the analyses of data pooled across the plans (that is, the primary analysis), we found that about half of the individual plans had a significant disparity as well. Several other plans showed similar but insignificant differences in care. Interestingly, disparity size was not consistently associated with a plan’s overall performance on a given measure.

View larger version (23K): [in this window] [in a new window]   EXHIBIT 5 Racial Disparities In LDL Checks Among Diabetic Enrollees In Medicare+Choice Plans In The Managed Care Organization Studied, 2000

View larger version (25K): [in this window] [in a new window]   EXHIBIT 6 Racial Disparities In LDL Checks Among Diabetic Enrollees In Commercial Plans In The Managed Care Organization Studied, 2000

Top Study Data And Methods Study Results Discussion NOTES

Both race and SES effects. Our study also demonstrates the importance of examining the effects of both race and SES. Although race/ethnicity and SES are correlated and disparities associated with these two characteristics linked, our study shows that on several quality measures, race and SES exert independent effects. That is, although the patterns of racial and SES disparities were similar, both characteristics remained significant on a number of measures, even though the effect of the other characteristics was taken into account. Although some of the causes of these disparities may be the same for low-income and minority enrollees, in other instances different factors may be responsible. Plans can use this information to identify and target these contributing factors.

Medicare and commercial plans. Our study also highlights potentially important differences in the pattern of disparities for M+C and commercial plans. Although the overall pattern of disparities for M+C and commercial enrollees was similar, there were distinct differences on some measures, and disparities were generally smaller for enrollees in commercial plans.

Commercial enrollees were younger than M+C enrollees, so it is possible that physicians’ practice styles differed for these two groups. However, HEDIS measures are designed to take clinical factors into account so that all eligible enrollees should have received appropriate care as specified by these measures. Commercial and M+C enrollees also may differ in terms of their SES and the resources available to them, since commercial enrollees either are employed by or have a family member with an employer that provides relatively high levels of health insurance, or they have the resources to buy such insurance themselves. These and related factors could have contributed to some of the differences in results. More research is needed to clarify these issues.

Study limitations. Our study had several limitations. We used a geocoded (indirect) measure of race/ethnicity. However, we found relatively little misclassification or bias when we used a geocoded versus a direct measure of race/ethnicity from CMS data. Less than 11 percent of our sample was misclassified; 9.5 percent of those were blacks living outside predominantly black neighborhoods. Moreover, disparity results were the same regardless of which race/ethnicity measure was used. Other studies suggest higher rates of misclassification among enrollees in commercial plans, particularly in regions where more blacks live outside of predominantly black neighborhoods or tend to be more affluent.¹⁰ Nonetheless, these studies also showed reasonably accurate estimates of enrollees’ racial/ethnic composition and care disparities when the geocoded data were analyzed at the group level.

Because we compared "blacks" with "other," rather than exclusively with whites, our analyses may have underestimated racial disparities, since the minorities included in the "other" category may have received worse care than whites. We may have further underestimated disparities because of our limited power to detect disparities on several measures. Finally, since all study plans were from one MCO, results may not be generalizable to other MCOs.

We recommend that plans use indirect methods such as geocoding to obtain race/ethnicity and SES data on plan enrollees and use these data to begin identifying care disparities and targeting interventions to eliminate them. Plans should simultaneously move toward developing direct data collection methods such as enrollees’ self-reports.

Editor's Notes

 
Allen Fremont (fremont{at}rand.org), Chloe Bird, and José Escarce are with RAND Health in Santa Monica, California. Arlene Bierman is with the University of Toronto (Ontario). Steve Wickstrom, Mona Shah, Thomas Horstman, and Thomas Rector are with the Center for Health Care Policy and Evaluation, Minneapolis, Minnesota. Fremont is also affiliated with the Department of Medicine, West Los Angeles Veterans Affairs Medical Center and with the Division of General Internal Medicine, University of California, Los Angeles (UCLA) in Westwood. Escarce shares this latter affiliation as well.

This research was supported by Agency for Healthcare Research and Quality Contract no. 290-00-0012. The authors thank Nancy Krieger for her sage advice during the initial stages of this project and Nicole Lurie for constructive comments on an earlier draft. They also thank Shelley Wiseman and Judy Bearer for helpful editing and assistance preparing the manuscript. Portions of this work were presented at the Society of General Internal Medicine Annual Meeting, Atlanta, May 2002, and the AcademyHealth Annual Meeting, Washington, June 2002.

NOTES

Top Study Data And Methods Study Results Discussion NOTES

 

1. A. Fremont and N. Lurie, "The Role of Racial and Ethnic Data Collection in Eliminating Disparities in Health Care," in Eliminating Health Disparities: Measurement and Data Needs, ed. M. Ver Ploeg and E. Perrin (Washington: National Academies Press, 2004), 202–231.
2. A.S. Bierman et al., "Addressing Racial and Ethnic Barriers to Effective Health Care: The Need for Better Data," Health Affairs 21, no. 3 (2002): 91–102[Abstract/Free Full Text]; and K. Fiscella et al., "Inequality in Quality: Addressing Socioeconomic, Racial, and Ethnic Disparities in Health Care," Journal of the American Medical Association 283, no. 19 (2000): 2579–2584.[Abstract/Free Full Text]
3. Agency for Healthcare Research and Quality, "Major Health Plans and Organizations Join AHRQ to Reduce Racial and Ethnic Disparities in Health Care," Press Release, 14 December 2004, www.ahrq.gov/news/press/pr2004/dispcolpr.htm (4 January 2005).
4. E.C. Schneider, A.M. Zaslavsky, and A.M. Epstein, "Racial Disparities in the Quality of Care for Enrollees in Medicare Managed Care," Journal of the American Medical Association 287, no. 10 (2002): 1288–1294[Abstract/Free Full Text]; B.A. Virnig et al., "Racial Variation in Quality of Care among Medicare+Choice Enrollees," Health Affairs 21, no. 6 (2002): 224–230[Abstract/Free Full Text]; and A.M. McBean et al., "Racial Variation in the Control of Diabetes among Elderly Medicare Managed Care Beneficiaries," Diabetes Care 26, no. 12 (2003): 3250–3256.[Abstract/Free Full Text]
5. N. Krieger, D.R. Williams, and N.E. Moss, "Measuring Social Class in U.S. Public Health Research: Concepts, Methodologies, and Guidelines," Annual Review of Public Health 18 (1997): 341–378.[CrossRef][Web of Science][Medline]
6. D.S. Massey and N.A. Denton, "Hypersegregation in U.S. Metropolitan Areas: Black and Hispanic Segregation along Five Dimensions," Demography 26, no. 3 (1989): 373–391.[Web of Science][Medline]
7. D.R. Nerenz et al., "Eliminating Racial/Ethnic Disparities in Health Care: Can Health Plans Generate Reports?" Health Affairs 21, no. 3 (2002): 259–263[Free Full Text]; and A. Abrahamse, P. Morrison, and N. Bolton, "Surname Analysis for Estimating Local Concentration of Hispanic and Asians," Population Research and Policy Review 13, no. 4 (1994): 383–398.[CrossRef]
8. N. Krieger et al., "On the Wrong Side of the Tracts? Evaluating the Accuracy of Geocoding in Public Health Research," American Journal of Public Health 91, no. 7 (2001): 1114–1116.[Abstract]
9. A complete listing of process measures is available in Online Appendix 1, content.healthaffairs.org/cgi/content/full/hlthaff/24/2/516/DC1.
10. W. Chen, D.B. Petitti, and S. Enger, "Limitations and Potential Uses of Census-based Data on Ethnicity in a Diverse Community," Annals of Epidemiology 14, no. 5 (2004): 339–345[CrossRef][Web of Science][Medline]; and R. Kwok and B. Yankaskas, "The Use of Census Data for Determining Race and Education as SES Indicators: A Validation Study," Annals of Epidemiology 11, no. 3 (2001): 171–177.[Medline]

                      What's this?

This article has been cited by other articles:

				J. X. Zhang, E. S. Huang, M. L. Drum, A. C. Kirchhoff, J. A. Schlichting, C. T. Schaefer, L. J. Heuer, and M. H. Chin Insurance Status and Quality of Diabetes Care in Community Health Centers Am J Public Health, April 1, 2009; 99(4): 742 - 747. [Abstract] [Full Text] [PDF]

				M. Maney, C.-L. Tseng, M. M. Safford, D. R. Miller, and L. M. Pogach Impact of Self-Reported Patient Characteristics Upon Assessment of Glycemic Control in the Veterans Health Administration Diabetes Care, February 1, 2007; 30(2): 245 - 251. [Abstract] [Full Text] [PDF]