disparities in quality are documented, we lack the tools to reliably estimate what it takes to reduce them. Finally, no convincing evidence exists that improving quality, regardless of disparities considerations, actually generates a positive ROI. Although a recent meta-analysis found evidence of a positive ROI for interventions to improve quality for some diseases, such as congestive heart failure, the evidence for a positive ROI for improving the quality of care for many other conditions is mixed.11
Quantifying the social case related to reducing disparities is even more challenging. The social case may be measured, for example, in terms of increased wages or productivity from better health and reductions in absenteeism, or averted costs for disability payments or to the end-stage renal disease (ESRD) program from reductions in diabetes-related complications. A 2002 study conducted for the American Diabetes Association found that diabetes care in the United States accounted for $132 billion in direct and indirect costs.12 Minority populations, who disproportionately suffer from diabetes and related conditions, tend to have worse outcomes and to generate a significant portion of these costs even when insured. Although Sandeep Vijan and colleagues estimated a per person, per year cost of nearly $3,200 in lost productivity as a result of diabetes, robust methodologies for estimating racial/ethnic or socioeconomic disparities in lost productivity are not well developed.13 And as previously mentioned, with regard to health care interventions, it is not clear how to quantify the level of societal investments that must be made, in either the short or the long run, to eliminate or even reduce these disparities and realize the social benefits.
Several NHPC plans have explored aspects of the business and social cases associated with interventions to improve quality and reduce disparities. Their experiences corroborate the challenges outlined above.14 Some found that these efforts require substantial access to financial data, including the program costs required to develop, implement, and operate initiatives, and to health care claims to assess changes in utilization patterns over time—data that are often challenging to obtain. Further, they report that such analyses are mainly suited to interventions and outcomes that are measurable in a reasonably short time frame, since the business case becomes less meaningful to any given organization over time, because of membership churning, near-term financial priories and pressures, and ability to accurately forecast impacts. They note that the business case would be easier to assess if interventions were implemented with strong evaluation designs that could isolate intervention effects associated with the business case. But such studies are complex and resource-intensive. In the absence of valid design and comparison groups, though, it will be difficult to isolate true financial savings from artifacts caused by factors such as selection and regression to the mean.
A final complication is that one health care entity’s short-term ROI may be another one’s loss. For example, if a health plan can save money by reducing emergency department (ED) and inpatient care for congestive heart failure, the local hospital may well suffer a loss of revenue. Such financing misalignments may serve as disincentives for addressing disparities. Stakeholders need to understand the financial implications for the multiple organizations involved (purchaser, plan, hospital, and physician) and may need to realign financing so that there can be cost sharing—and gain sharing—of any savings. However, methods for efficiently doing the necessary financial analysis or the actual risk sharing have not yet been fully developed.
Steps Can Be Taken To Reduce Disparities
Despite these challenges, a combination of business and quality improvement principles may still be able to guide health care organizations seeking to reduce disparities. For example, using Pareto charts and the 80-20 rule—that 80 percent of the problem, be it costs or disparities, arises from 20 percent of patients—they can begin to focus their efforts and, ideally, target scarce resources more effectively. Even absent gold-standard data on patients’ race/ethnicity, they can focus on health care settings, whether a hospital or a provider’s office, that simultaneously serve large numbers of minority patients and provide poorer-than-average quality of care. Support for this approach can be found in work by Peter Bach and colleagues, which found that approximately 80 percent of African Americans were cared for by 20 percent of physicians, often in under-resourced settings in which providing high-quality care was challenging.15 Targeting intervention opportunities can also be improved by using geographic information system (GIS) tools to map and highlight concentrated areas of poor quality as small as the census-tract level. For instance, several NHPC plans are using these approaches to highlight local “hot spots” and salient characteristics of those areas to help develop and target interventions.16
In their paper, Acevedo-Garcia and colleagues argue that such pockets of poor care for minority populations have their roots in racial segregation, whose impacts are felt broadly, including in low-opportunity neighborhoods that also likely generate many of the nonmedical determinants of health.17 In these and other respects, instead of going it alone in these neighborhoods, a health care organization may reap greater efficiencies—and results—by partnering with other community stakeholders (such as employers, community groups, public health agencies, and others) on broader initiatives to improve quality and reduce disparities, both for its own ROI and for the broader social good. Policymakers increasingly accept that interventions within the personal health care delivery system, while important, may do little to address the root causes of those disparities, and to be effective, these efforts must be combined with broader community and policy efforts.18 Partnerships among multiple health care organizations, with significant market share in low-opportunity neighborhoods, could lower the intervention costs for any one health care organization. Additional partnering with other community stakeholders can also increase the likelihood of effectiveness and sustainability of interventions. The Centers for Disease Control and Prevention (CDC) has found through its Racial and Ethnic Approaches to Community Health (REACH) and Steps programs that interventions that involve multiple partners may also be more likely to close the gap in intermediate outcomes (such as diabetes or lipid control).19
The experiences of organizations in working toward fulfillment of the IOM’s equity aim call into question whether all interventions require a strictly defined business case and force us to reconsider, also, the social case for doing so. They highlight the reality that even if making the business case for addressing disparities is ultimately necessary, accurately determining the business case in the current environment will continue to be challenging. Further complicating the calculation is the likelihood that the ratio of returns to costs will improve, as routine collection and use of race/ethnicity data become more common and widespread in quality monitoring and improvement efforts and more effective interventions and supporting infrastructure are developed. What is more clear is that the business case for any one plan or other health care entity to further engage in addressing disparities could also be helped by changes within and external to the larger health care system—at the individual, health system, employer, and societal levels—that better align incentives, regulations, and interests of various stakeholders in ways that promote elimination of disparities and improve equity along with the other key aims laid out by the IOM.
Nicole Lurie (..firstname.lastname@example.org) is senior natural scientist and the Paul O’Neill Alcoa Professor at RAND in Arlington, Virginia. Stephen Somers is president and chief executive officer of the Center for Health Care Strategies (CHCS) in Hamilton, New Jersey. Allen Fremont is a natural scientist and sociologist at RAND in Santa Monica, California. January Angeles is a program officer at the CHCS. Erin Murphy is a research assistant at RAND in Arlington. Allison Hamblin is a program officer at the CHCS. RAND and the CHCS are two of the coordinating and managing partners of the National Health Plan Collaborative, whose activities are recounted in this Perspective.
The authors acknowledge funding from the Agency for Healthcare Research and Quality and the Robert Wood Johnson Foundation, which are coordinating and managing partners of the National Health Plan Collaborative (NHPC). The views expressed here are solely those of the authors and do not necessarily reflect those of the NHPC or its funders.
↵ D. Acevedo-Garcia et al., “Toward a Policy-Relevant Analysis of Geographic and Racial/Ethnic Disparities in Child Health,” Health Affairs 27, no. 2 (2008): 321–333. Abstract/FREE Full Text
↵ Institute of Medicine, Crossing the Quality Chasm: A New Health System for the Twenty-first Century (Washington: National Academies Press, 2001).
↵ S. Leatherman et al., “The Business Case for Quality: Case Studies and an Analysis,” Health Affairs 22, no. 2 (2003): 17–30 Abstract/FREE Full Text ; R.Z. Goetzel et al., “Return on Investment in Disease Management: A Review,” Health Care Financing Review 26, no. 4 (2005): 1–19 Medline ; and Center for Health Care Strategies, The Return on Investment Evidence Base: Identifying Quality Improvement Strategies with Cost-Saving Potential, November 2007, http://www.chcs.org/publications3960/publications_show.htm?doc_id=576275 (accessed 10 December 2007).
↵ N. Lurie, M. Jung, and R. Lavizzo-Mourey, “Disparities and Quality Improvement: Federal Policy Levers,” Health Affairs 24, no. 2 (2005): 354–364. Abstract/FREE Full Text
↵ K. Fiscella and A.M. Fremont, “Use of Geocoding and Surname Analysis to Estimate Race and Ethnicity,” Health Services Research 41, no. 4, Part 1 (2006): 1482–1500 Medline ; and M. Elliott et al., “A New Method for Estimating Racial/Ethnic Disparities where Administrative Records Lack Self Reported Race/Ethnicity,” Health Services Research (forthcoming).
↵ For example, Aetna, which has been at the forefront of primary data collection for more than five years, reports that it has data on just over 35 percent of enrollees.
↵ National Health Plan Collaborative, Reducing Racial and Ethnic Disparities and Improving Quality of Health Care, November 2006, http://www.chcs.org/NationalHealthPlanCollaborative/images/641_104_NHCP_summary_V3.pdf (accessed 13 December 2007) ; and George Washington University School of Public Health and Health Services, “Collecting Data on Patient Race, Ethnicity, and Primary Language Is Helping Hospitals Improve the Quality of Care: Robert Wood Johnson Foundation Program Helps Hospitals Decrease Disparities, Increase Patient-Centeredness in Heart Care,” September 2007, http://rwjf.org/files/research/issuebrief092007.pdf (accessed 10 December 2007).
↵ NHPC, Reducing Racial and Ethnic Disparities.
↵ Leatherman et al., “The Business Case.”
↵ M.H. Chin et al., “Interventions to Reduce Racial and Ethnic Disparities in Health Care,” Medical Care Research and Review 64, no. 5 Supp. (2007): 7S–28S. Abstract/FREE Full Text
↵ Goetzel et al., “Return on Investment” ; and Leatherman et al., “The Business Case.”
↵ P. Hogan et al., “Economic Costs of Diabetes in the U.S. in 2002,” Diabetes Care 26, no. 3 (2003): 917–932. Abstract/FREE Full Text
↵ S. Vijan, R.A. Hayward, and K.M. Langa, “The Impact of Diabetes on Workforce Participation: Results from a National Household Sample,” Health Services Research 39, no. 6, Part 1 (2004): 1653–1669 CrossRefMedline ; and M. Berger et al., “Alternative Valuations of Work Loss and Productivity,” Journal of Occupational and Environmental Medicine 43, no. 1 (2001): 18–24. Medline
↵ Conference call with the health plans, 14 August 2007.
↵ P.B. Bach et al., “Primary Care Physicians Who Treat Blacks and Whites,” New England Journal of Medicine 351, no. 6 (2004): 575–584. CrossRefMedline
↵ NHPC, Reducing Racial and Ethnic Disparities.
↵ Williams et al., “Toward a Policy-Relevant Analysis.”
↵ T. Miller, “Making a Difference in Differences for the Health Inequalities of Individuals,” Health Affairs 26, no. 5 (2007): 1235–1237. Abstract/FREE Full Text
↵ See the REACH program home page, http://www.cdc.gov/reach; and the Steps program home page, http://www.cdc.gov/steps/index.htm.
Hi there! Click one of our representatives below and we will get back to you as soon as possible.