Objective: The Robert Wood Johnson Foundation’s (RWJF’s) Aligning Forces for Quality (AF4Q) initiative aimed to advance healthcare quality and equity in 16 communities across the United States through multi-stakeholder alliances of healthcare payers, providers, and consumers. Our objectives are (1) to summarize the major approaches and activities undertaken by the AF4Q alliances that were most successful in tracking and implementing programs that aimed to reduce local healthcare disparities by race, ethnicity, and primary language spoken (REL), and socioeconomic status (SES); and (2) to identify the major lessons learned from the successes and failures of the AF4Q alliances to inform other equity-focused initiatives.
Methods: We analyzed data from 6 rounds of key informant interviews conducted between 2010 and 2015, and triannual progress reports submitted by the alliances to RWJF between 2008 and 2015.
Results: Of the 16 AF4Q alliances, 2 succeeded in developing communitywide systems to track local healthcare disparities, 5 alliances implemented substantive programs that aimed to reduce local disparities, and 3 alliances were successful in disparity measurement and program implementation. The alliances that were most active in addressing disparities tended to have long-established relationships with relevant community organizations, focused on improving the quality of care provided by safety-net providers, and shifted quickly toward working to address disparities even if their initial efforts to stratify performance measures by REL failed.
Conclusion: Few alliances were able to develop communitywide systems to track local healthcare disparities or implement large-scale initiatives to reduce disparities during the 7 years that these objectives were advanced by the AF4Q initiative. Establishing robust local disparity-tracking systems and establishing productive relationships with key community stakeholders took substantial time. The AF4Q experience suggests that efforts to reduce disparities should not be held up by disparity measurement challenges.
Am J Manag Care. 2016;22:S413-S422
Eliminating health disparities has long been promoted as a key public health priority, yet healthcare systems and communities throughout the United States continue to struggle to achieve this aim.1 Aligning Forces for Quality (AF4Q), one of the Robert Wood Johnson Foundation’s (RWJF’s) foremost initiatives to improve the health and healthcare of communities across the United States, embraced the promotion of healthcare equity as a cornerstone of the program. Through the AF4Q initiative, which ran from 2006 to 2015, multi-stakeholder partnerships (hereafter referred to as alliances) of healthcare providers (hospitals and primary care centers), purchasers (health plans and large employers), and consumers (patients and community members), from each of the 16 geographically and demographically diverse communities, worked to advance the quality of care and provide models for national healthcare reform. This article provides an overview of the major activities undertaken by the 16 AF4Q multi-stakeholder alliances as they worked to reduce healthcare disparities, examines the alliance and community factors that were linked with success in advancing community capacity to address local disparities, and discusses the major lessons learned from the alliances’ experiences as they strove to simultaneously advance healthcare quality and equity.
Evolution of the Health Equity Agenda in the AF4Q Initiative
Reducing healthcare disparities was not an explicit objective of the AF4Q initiative when it launched in 2006. In the initial phase of the AF4Q program, alliances were asked to focus on advancing the overall quality of healthcare in their communities through attention to 3 main areas: healthcare system performance measurement and public reporting, implementation and dissemination of quality improvement strategies, and promotion of greater consumer engagement in healthcare.2 Reducing healthcare disparities by race, ethnicity, and primary language spoken (REL) was added as an explicit objective of the AF4Q initiative in 2008.
The alliances were expected to include stakeholders most relevant to achieving success in the original program aims of advancing public reporting, quality improvement, and community engagement, and were selected for participation in the AF4Q initiative based, in large part, on their anticipated potential to achieve success in these areas. However, because the alliances were not selected based on their anticipated potential to reduce disparities, they started the AF4Q program with variable levels of engagement with community organizations and other stakeholders that represent or serve racial and ethnic minority populations, varying levels of commitment to prioritizing efforts to reduce racial and ethnic disparities, and generally low levels of expertise combatting disparities. Few participating alliances had significant prior experience in addressing healthcare disparities, and only 1 of the AF4Q grant-holding organizations included disparities reduction as part of their mission prior to the AF4Q initiative.
A logic model outlining the assumptions and expectations of how the multi-stakeholder alliances would advance healthcare equity was developed by the evaluation team during the early years of the AF4Q initiative and is depicted in Figure 1. Briefly, to be effective, alliances were expected to engage relevant stakeholders to drive efforts aimed at reducing disparities. A focus on equity was to be integrated throughout the core AF4Q programmatic areas of performance measurement and reporting, quality improvement, and consumer engagement. Healthcare quality and outcome data stratified by REL would direct and refine the alliances’ efforts to achieve equitable health outcomes. Given the complexity of factors that contribute to health disparities, it was expected that factors internal and external to the AF4Q initiative would influence the strategies pursued by the alliance leadership and partnering stakeholders in each community.
The 16 AF4Q alliances were allowed some flexibility in addressing healthcare disparities, but they were guided by an evolving set of equity-focused directives laid out by the AF4Q National Program Office (NPO) as depicted in Figure 2. Initially, the alliances were charged with advancing the routine and standardized collection of data regarding patients’ REL by hospitals, primary care practices, and healthcare plans. These data were linked with healthcare quality and patient outcome measures to examine healthcare system performance stratified by REL. Healthcare performance data could be stratified at the hospital or primary care practice level to guide institution-level equity initiatives, or aggregated across the community to identify community-level trends and equity priorities. In 2011, the alliances were directed to focus on stratifying quality measures by REL and incorporating the reduction of healthcare disparities into their quality improvement and consumer engagement strategies. Beginning in 2013, the equity targets were expanded to include the reduction of disparities by markers of socioeconomic status (SES) (ie, privately insured vs uninsured or publicly insured) and geography (ie, rural vs urban) in addition to REL. During this final phase of the AF4Q initiative, alliances were encouraged to focus on achieving their self-determined equity objectives.
Methods
We used a qualitative approach to examine the equity-focused component of the AF4Q initiative across the 16 alliances. Our data sources included interviews with alliance directors and staff, triannual progress reports submitted by the alliances to RWJF, and alliance disparity verification reports. Six rounds of semi-structured interviews with alliance directors and disparities staff leads were conducted from 2010 to 2015. The interviews were conducted in person or by telephone, were 60 to 90 minutes in length, audio recorded, transcribed, and entered into Atlas.ti (Version 2013; Cologne, Germany), a qualitative software program. The interview transcripts were coded using a deductive and inductive approach.3 Data were initially coded using pre-identified theoretical constructs underlying disparity measurement and reduction. New codes were added during coding to capture additional themes, and this final set of disparities-related codes was applied to all transcripts. Each transcript was coded by one team member and reviewed by a different team member for quality assurance, and coding discrepancies were resolved through discussion. Qualitative data with the following codes were used to develop the verification summaries described below: activities, REL data collection, REL data use, and underserved populations. Qualitative data coded as barriers, facilitators, external forces, and advice were used to identify lessons learned. Triannual progress reports were submitted by the alliances to RWJF from 2008 through 2015. These provided information regarding activities, accomplishments, and challenges in each AF4Q program area. Based on data from the interviews and triannual reports, we drafted verification reports for each alliance that were submitted to the alliance directors and disparity staff leads from June 2015 through April 2016 for review and correction.
Our analysis consisted of a multi-step process. First, we created summary documents for each alliance detailing their activities in REL data collection, data reporting regarding local healthcare disparities, and disparities reduction interventions. Two authors independently ranked individual alliances’ efforts in each of these framework categories as high or low, and consensus was reached during team discussion. Alliances were considered to be high in disparity measurement if they produced quality reports stratified by REL or markers of SES on a repeated basis that included data from multiple providers and multiple payer types, whether or not these reports were publicly reported. Alliances were considered to be high in disparity-focused activities if they had several low-reach but high-intensity activities, or at least one high-reach activity. Low-reach activities were defined as those that involved less than 25% of hospitals, primary care practices, or healthcare consumers in the community; high-reach activities were those that involved at least 25%. Low-intensity activities were defined as those that involved a single or very small number of interactions (eg, a quarterly health fair at a church), targeted only one member of the healthcare team (eg, cultural competency training for physicians), or did not provide ongoing patient or provider support (eg, provision of language-concordant educational materials to patients with low English proficiency). High-intensity activities were defined as those that involved healthcare delivery redesign (eg, patient-centered medical home implementation), consisted of multiple members of the healthcare team (eg, care transitions program involving social work, nursing, and linkages to community resources), or included recurrent interactions with patients or care teams (eg, chronic disease management programs with patient self-management support). Activities that did not fall into one of these categories were ranked as high or low intensity based on group discussion and consensus.
Results
Overall, from 2008 to 2015, the AF4Q alliances made low to modest gains toward routinely tracking local healthcare disparities, addressing disparities through their consumer engagement activities, and integrating attention to reducing disparities with their quality improvement strategies. This section summarizes the experiences of the 16 alliances, overall, as they worked to measure and reduce local healthcare disparities, describes the strategies and characteristics of the most successful alliances, and highlights the lessons learned from the alliances’ achievements and failures.
REL Data Collection
Most of the 16 alliances were not able to achieve high rates of standardized REL data by area hospitals and primary care practices by the end of the AF4Q initiative (Figure 3). None of the alliances succeeded in advancing the collection of REL data by healthcare plans.
Stratifying Healthcare Performance Measures by REL
Alliances struggled to advance the routine stratification of healthcare system performance measures by REL. Only 5 alliances reported that hospitals or primary care practices in their community reviewed their own quality measures stratified by REL on a regular basis. Although 7 alliances succeeded in aggregating data from multiple hospitals or primary care practices within their service area to examine community-level disparities by REL at least once, only 4 alliances produced and released these stratified quality reports, privately (to the alliance leadership and participating healthcare providers) or publicly, on a regular basis.
Examples of Success in Tracking Local Healthcare Disparities
Overall, 5 alliances met the threshold for high achievement in monitoring local disparities by producing quality reports on a repeated basis that stratified data by REL or markers of SES from multiple providers and multiple payer types (Table 1). Four of these 5 alliances had expertise in performance measurement prior to the AF4Q initiative, and 3 had data reporting systems that were based on electronic health record (EHR) data rather than health insurance claims data.
The alliances in Cleveland and Minnesota were the most successful overall in establishing communitywide systems for tracking healthcare disparities. The quality reporting systems for both of these alliances used EHR data submitted via secure data portals. Medical groups in both communities can review their quality measures stratified by REL through the data portals. In Cleveland, they can also review their performance stratified by educational attainment and insurance type. Cleveland’s alliance has publicly reported stratified quality measures since 2009; measures are stratified by REL at the regional level and by SES at the regional and medical group levels. Minnesota’s alliance has produced annual publicly available health equity reports which stratify quality measures by race and ethnicity for patients insured through Medicaid. Their equity reports have also included quality measures stratified by health insurance type (eg, Medicaid vs private insurance) at the state level. In 2015, Minnesota’s health equity reports started including quality measures stratified by REL for patients with all types of health insurance at the statewide, regional, and medical group levels.
Challenges to Tracking Local Healthcare Disparities
The other 10 alliances stumbled at different points along the path in establishing communitywide systems for tracking healthcare disparities. Alliances that relied on claims data were only able to stratify quality measures by race and ethnicity for the Medicaid population because Medicaid was the only payer that routinely collected this data from enrollees. Alliances that were not able to achieve high rates of standardized REL data collection by area hospitals or primary care practices, and relied on EHR data for their quality reports, struggled to move forward with the development of public equity reports due to concerns about the quality of the available REL data in the EHRs. Even when they achieved high rates of REL data collection, some alliances were not able to aggregate data from different healthcare systems due to poor interoperability and other technical challenges. Some of the alliances that passed all these hurdles were still unable to produce equity reports on a regular basis due to resource constraints and competing priorities.
Lessons Learned From the AF4Q Experience in Promoting Local REL Data Collection and Disparities Monitoring
In general, having an established, high-quality, and reliable healthcare performance measurement system based on EHR, rather than claims-based data, facilitated the development of robust systems for tracking disparities at the individual healthcare institution or communitywide level. Alliances that had to devote a lot of resources to building their quality measurement or public reporting systems had little remaining social and financial capital to apply toward creating and maintaining disparity-tracking systems. However, a robust healthcare quality measurement system was not sufficient for establishing a communitywide disparity-monitoring system. For many of the AF4Q alliances that were otherwise strong in quality measurement, the development of disparity-tracking systems were hampered by: insufficient resources to promote the standardized collection of REL data by hospital or health center staff, lack of standardized REL data fields to enable data aggregation across institutions, and a lack of buy-in regarding the importance of tracking performance by REL due to low perceived REL diversity within the service area.
Nonetheless, some alliances were able to overcome the reliance on claims data or low levels of reliable REL data in local EHR systems by focusing on tracking disparities by health insurance type or other markers of SES. The alliances in Washington and Oregon, for example, produced publicly available quality reports stratified by payer type at the state and regional levels. They disseminated these reports to their respective Medicaid agencies and providers serving large numbers of Medicaid recipients.
Addressing Disparities Through Consumer Engagement
As noted by Greene et al in this supplement, consumer engagement activities in the AF4Q initiative addressed multiple dimensions of engagement. These included the promotion of greater self-care (eg, through the adoption of healthy behaviors and chronic disease self-management), the promotion of consumeristic healthcare-seeking behaviors (eg, more effective communication with healthcare providers and selecting providers based on quality or value), and the engagement of consumers in the governance of healthcare organizations (eg, through participation in alliance governance or healthcare quality improvement teams).4 Examples of how the alliances strove to address disparities through consumer engagement initiatives are listed in Table 2. Most of the alliances aimed to improve the adoption of healthy behaviors and greater self-management by racial and ethnic minorities by targeting and tailoring their health education campaigns and self-management support initiatives to minority populations; however, these efforts were on a small scale and were not sustained due to resource limitations. Promoting consumeristic healthcare-seeking behaviors among minority populations was a less common approach to aligning consumer engagement and equity-focused goals, as many alliance leaders did not buy into the notion that “shopping” for healthcare providers based on quality or value was a realistic option in their communities, particularly in areas with shortages of primary care providers.5 A few alliances partnered with community members, community leaders, and community-based institutions to foster the development of greater community capacity to address the social determinants of health, but with the exception of the Healthier Roxbury Coalition sponsored by Boston’s alliance, these efforts were small in scale. Incorporating equity champions into the alliance leadership was very rare. Only 2 alliances included equity champions in the alliance leadership, but this appeared to be due to serendipity in both cases, as the equity champions were a part of the leadership team before disparities reduction became an objective of the AF4Q initiative.
Addressing Disparities Through Quality Improvement
Overall, compared with consumer engagement initiatives, alliances made greater strides toward incorporating efforts to reduce healthcare disparities into their improvement activities. However, only half of the alliances met the threshold for a high level of activity focused on reducing disparities (defined as several low-reach but high-intensity activities or at least one high-reach activity) (Table 1).
Hospitals and primary care practices in the AF4Q communities were invited to participate in several 18-month-long equity-focused learning collaboratives sponsored by RWJF and organized by the AF4Q NPO during the early years of the program. These included the Equity Quality Improvement Collaborative, which included 8 hospitals from 5 communities and focused on reducing disparities in cardiac care; the Improving Language Services initiative, which included 32 hospitals from 12 communities, and aimed to ensure access to language services by qualified medical interpreters for patients with limited English proficiency; and the Equity Improvement Initiative, which included 9 outpatient practices from 4 communities and helped these sites to target disparities identified through their REL-stratified performance data. The results and lessons learned from these initiatives have been published elsewhere.6-8
Separate from these NPO-led learning collaboratives, the alliances’ approaches to advancing healthcare equity through quality improvement tended to fall into 3 general categories: implementation of programs focused on eliminating an observed disparity for a racial or ethnic minority group; implementation of programs to raise the quality of care provided by federally qualified health centers (FQHCs) or other healthcare providers serving large numbers of minority, low-income, or uninsured community members; or reliance on improving quality generally, a “rising tide lifts all boats” approach.
Targeting Observed Disparities
Six alliances partnered with local FQHCs to implement programs targeting observed local disparities. For example, in Cleveland and Detroit, local health centers implemented multimodal programs to improve the rates of blood pressure control among African American patients.9 These programs were generally small-scale and often included only 1 FQHC, but were also high intensity, including some combination of patient self-management support, links to social services, cultural competency training, performance feedback, and practice coaching.
Four alliances partnered with public health agencies and other social service providers to address local disparities, and these efforts were generally larger in scale. Kansas City’s alliance, for example, partnered with the University of Missouri and the Missouri Department of Health and Human Services, in support of a program to improve the asthma care of children with Medicaid across several major school districts, covering over 300 schools, through educational programs for the children, parents, school nurses, and physicians.10
Raising the Quality of Care for Vulnerable Populations
Seven alliances’ equity-related efforts centered on raising the quality of care provided by FQHCs and other providers serving large numbers of racial and ethnic minorities, low-income, or uninsured individuals. Smaller-scale efforts included quality improvement learning collaboratives and care coordination programs including up to 10 health centers. The largest-scale initiatives were implemented in the statewide alliances and included over 100 health centers. The alliances in Maine and Minnesota, for example, partnered with their respective departments of health to support their statewide Health Homes initiatives which aimed to promote enhanced primary care services and care coordination for Medicaid recipients. The alliances’ role in these partnerships ranged from providing healthcare quality data and statistical support to even greater efforts such as practice coaching, technical support, and support for learning collaboratives.
A Rising Tide Lifts All Boats
Three alliances did not implement specific programs targeting REL or SES disparities, but instead pursued the strategy that generalized quality improvement initiatives would raise the quality of care overall and lead to a reduction in disparities, in effect a “rising tide lifts all boats” approach. In these alliances, the major quality improvement initiatives generally included providers and payers serving historically underserved populations, but these populations were not the focus of the initiatives.
Linkages Between Disparity Measurement and the Implementation of Disparities Reduction Initiatives
In the AF4Q initiative, success in developing effective systems for tracking and reporting disparities in the community, whether publicly or privately, was not closely tied to the robust implementation of programs to reduce disparities (Table 1). Alliances that were most advanced in the measurement and reporting of disparities were not consistently the most active in the implementation of interventions to reduce disparities. Some alliances that were unsuccessful in measuring disparities were quite active in the implementation of interventions to reduce disparities. Six alliances made little inroads toward measuring or addressing disparities in their communities.
Although the link between measurement and action for disparities was not consistent, there were some powerful examples of how local data on disparities can inform disparities-targeted interventions. The leaders of Cleveland’s alliance initially articulated a strong belief that a focus on raising the quality of care among local primary care providers, overall, would lead to reduced disparities. However, they observed that while disparities in process of care measures improved, disparities in health outcomes widened. This motivated them to shift strategies and target quality improvement initiatives to racial minorities and safety-net providers. In Maine, the alliance leaders found that disparities in health outcomes in the state were even larger by SES than by race or ethnicity. This led them to partner closely with FQHCs and MaineHealth, Maine’s Medicaid program, to improve access to care for the poor and uninsured. Oregon’s alliance partnered with a coalition of more than 100 organizations dedicated to promoting healthy and sustainable communities in the Portland metropolitan area to contribute to the Regional Equity Atlas, which promotes changes in public policy, planning, and strategic investments to eliminate disparities.11 In these ways, local data on health disparities stimulated the development of new approaches and partnerships to eliminate disparities.
Lessons Learned From the AF4Q Experience in Addressing Disparities Through Consumer Engagement and Quality Improvement Initiatives
Although the alliances used a range of approaches in their efforts to implement initiatives aimed at reducing disparities, a common feature of the most successful alliances was the development of strong partnerships with organizations or institutions serving large numbers of minority, poor, or historically underserved populations. For example, the alliances in Cleveland and Memphis had strong equity champions from the beginning of the AF4Q initiative, and the alliance in Boston partnered closely with their public health department early in the program. The alliances of South Central Pennsylvania and Western New York did not have strong relationships with safety-net providers at baseline, but built these relationships over time and developed small-scale, but high-intensity, equity-focused programs. Three of the statewide alliances, Minnesota, Oregon, and Maine, succeeded in partnering with their state Medicaid departments and were able to support the implementation of large-scale, equity-focused initiatives. Factors external to the AF4Q initiative also contributed to the success of potential partnerships. For example, the alliances in states that did not participate in the Affordable Care Act’s (ACA’s) Medicaid expansion reached out to their state Medicaid agencies, but—with one exception—did not find them to be responsive.
Discussion and Lessons Learned
The AF4Q initiative was an ambitious program that aimed to simultaneously improve the quality and equity of healthcare in diverse communities across the United States by leveraging the power of multi-stakeholder collaboration. In particular, the 16 AF4Q alliances were expected to engage other stakeholders relevant for advancing health equity, improve their community’s capacity to track local disparities, and incorporate attention to advancing healthcare equity into other AF4Q program areas, including increased consumer engagement in healthcare and quality improvement. Based on the achievements and shortcomings of the AF4Q alliances in these efforts, several key lessons were identified and their implications for healthcare leaders, community leaders, and policy makers are discussed below.
First, engaging the “right” stakeholders to address local health disparities was an important early step for each of the alliances, but there were many different types of disparities-relevant stakeholders who could partner effectively with the alliances. Almost all the alliances were able to partner with FQHCs in their communities. While most of these partnerships resulted in small pilot programs, a few alliances established effective partnerships with networks of FQHCs, or were able to expand initiatives to a much larger scale by partnering with public health departments and state Medicaid agencies. The largest programs leveraged funding from Health Homes and other initiatives from the Centers for Medicare & Medicaid Services (CMS). Establishing productive relationships with government partners was generally successful in states with administrations that were supportive of publicly financed approaches to expanding healthcare access, including the ACA’s Medicaid expansion. Several alliances partnered effectively with institutions that are not traditionally considered to be a part of the healthcare system (eg, schools, social service agencies, and churches) but these partnerships generally took more time to establish, and the initiatives resulting from these partnerships were almost exclusively very small in scale.
Second, while the strategy to measure first then act has intuitive appeal as a logical approach to combatting healthcare disparities, the experience of the alliances indicates that local disparity measurement is not necessary or sufficient for tackling disparities—at least not with the measurement strategy promoted by the AF4Q initiative. In their attempt to build local healthcare disparities monitoring systems based on claims or EHR data, many of the alliances spent years addressing technical challenges and assuring data quality. Minnesota’s alliance, for example, spent 5 years building a data submission platform and ensuring the quality of their REL data before releasing their first communitywide disparities report. Despite these measurement delays, Minnesota and a few other alliances still charged ahead to implement programs that targeted disparities that were easier to measure (ie, by health insurance status) or disparities that were evident, based on analysis of data from other sources, including national databases such as the Behavioral Risk Factor Surveillance System or epidemiological data from state or city health departments. Many alliances, however, got stuck in the measurement phase and failed to make substantive efforts to address local disparities. Additionally, some alliances, even after succeeding in building strong disparity measurement systems, still failed to make substantive efforts to address disparities. The alliances often cited a lack of expertise in how to reduce disparities or insufficient resources to support disparity reduction programs.
It is important to remember the healthcare quality data landscape during the early years of the AF4Q initiative. The Health Information Technology for Economic and Clinical Health Act was enacted in early 2009, but stages 1 and 2 of the Meaningful Use program did not launch until late 2010 and 2014, respectively. Thus, for the vast majority of the AF4Q program, a lack of standardization of EHR fields (eg, for race and ethnicity) and poor interoperability between EHR platforms were common. In this environment, the resources needed to establish communitywide disparity-tracking systems were high. It is possible that as EHR standardization and interoperability improve, and as local healthcare quality measurement systems spread and mature, the costs of establishing clinically based disparity-tracking systems will decrease. Until then, communities interested in initiating these types of disparity-tracking efforts should consider the costs and value of building such local disparity-tracking systems, particularly relative to the costs or value of using established state and national healthcare quality or public health databases.
We noted several examples of how data from the alliances’ disparity measurement systems catalyzed efforts to address those disparities. However, there were also many cases when the alliances moved forward to implement evidence-based approaches to addressing known national health disparities even in the absence of local data. As suggested by others, efforts to measure and act upon local disparities can occur simultaneously rather than sequentially.12
Our study has several limitations. We do not have disparities outcome data for the AF4Q communities. Thus, while we grouped alliances based on the robustness of their disparity measurement systems and the reach and intensity of their disparity-focused activities, we were not able to group the alliances based on their relative success in reducing disparities. While we drew from data collected longitudinally over the course of the AF4Q initiative to examine factors and approaches that were common among the higher-achieving alliances, due to the qualitative nature of the data, we cannot conclude that the relationships between alliance characteristics, strategies, and relative success toward establishing local disparity-monitoring systems or implementing programs that aim to reduce disparities are causal. The disparities-related requirements of the AF4Q initiative evolved over time, leaving less opportunity for the alliances to achieve the later objectives. Given this, our thresholds for defining high performers in disparity tracking, and particularly for the implementation of disparity-focused interventions, are low. Thus, our observations regarding the characteristics of high versus low performers in disparities-related domains should be interpreted in this context. Finally, despite the extent of data that resulted from interviews with alliance leaders and staff and triannual reports submitted
to RWJF, some of the disparities-related activities in the AF4Q communities may have been missed, particularly ones that were done by individual health centers or hospitals.
Nonetheless, the experiences of the AF4Q alliances as they worked to address local REL and SES health disparities offer some potential lessons that are relevant to recent health policy and healthcare improvement initiatives. Given the move toward value-based payment systems, concerns regarding the potential for value-based payment reforms to exacerbate healthcare disparities13-15 and early evidence supporting these concerns,16 the response of the AF4Q alliances to their explicit charge to pay attention to advancing healthcare equity while working to advance the quality and value of healthcare in their communities highlights key opportunities and challenges for better incorporating equity into these initiatives. Furthermore, the challenges faced by the AF4Q alliances as they tried to engage community-based organizations offer important lessons for programs that aim to bridge the gap between healthcare and community services like the CMS’ Accountable Health Communities model.17 In particular, the AF4Q experience suggests that initiatives that require cross-sector collaboration may benefit from allowing sufficient time for relationship building. Also, efforts to track disparities should not hold up action to mitigate disparities. Finally, one cannot presume that data documenting disparities will automatically lead to action aimed to eliminate health disparities.
Author affiliation: Center for Healthcare Studies, Northwestern University, Feinberg School of Medicine, Chicago, IL (PD, JH, RK, JCY); Northwestern University, Feinberg School of Medicine, Division of General Internal Medicine and Geriatrics, Chicago, IL (MJJ); Center for Health Care and Policy Research, Penn State University, University Park, PA (YM).
Funding source: This supplement was supported by the Robert Wood Johnson Foundation (RWJF). The Aligning Forces for Quality evaluation is funded by a grant from the RWJF.
Author disclosures: Ms Duckett, Ms Hamil, Dr Jean-Jacques, Mr Kang, Ms Mahmud, and Dr Yonek report receipt of grants from RWJF.
Authorship information: Concept and design (PD, JH, MJJ, RK, YM, JCY); acquisition of data (MJJ, YM, JCY); analysis and interpretation of data (PDD, JH, MJJ, RK, YM, JCY); drafting of the manuscript (PD, JH, MJJ, YM); critical revision of the manuscript for important intellectual content (PD, JH, MJJ, RK, YM, JCY); and administrative, technical, or logistic support (JH).
Address correspondence to: mjeanjac@nm.org.
REFERENCES
1. Agency for Healthcare Research and Quality. 2014 National Healthcare Quality and Disparities Report. www.ahrq.gov/sites/default/files/wysiwyg/research/findings/nhqrdr/nhqdr14/2014nhqdr.pdf. Published May 2015. Accessed July 11, 2016.
2. Scanlon DP, Beich J, Leitzell B, et al. The Aligning Forces for Quality initiative: background and evolution from 2005 to 2015. Am J Manag Care. 2016:22(suppl 12):S346-S359.
3. Bradley EH, Curry LA, Devers KJ. Qualitative data analysis for health services research: developing taxonomy, themes, and theory. Health Serv Res. 2007;42(4):1758-1772.
4. Greene J, Farley DC, Christianson JB, Scanlon DP, Shi Y. From rhetoric to reality: consumer engagement in 16 multi-stakeholder alliances. Am J Manag Care. 2016:22(suppl 12):S403-S412.
5. Yonek JC, Mittler JN, Hasnain-Wynia R. Why and how six aligning forces for quality communities have focused on reducing disparities. Med Care Res Rev. 2014;71(5):435-449. doi: 10.1177/1077558714533826.
6. The Robert Wood Johnson Foundation. Closing the gap: AF4Q hospitals develop ways to improve the quality and equality of care for patients. http://forces4quality.org/af4q/download-document/4311/Resource-11-008%20-%20Closing%20the%20Gap%20EQIC%20Final%20Report%20-%20fnl.pdf. Published November 2011. Accessed July 11, 2016.
7. The Robert Wood Johnson Foundation. Aligning Forces for Quality Hospital Quality Network: what did the collaborative accomplish? http://forces4quality.org/af4q/download-document/5627/Resource-hospital_initiative_key_findings.pdf. Accessed July 11, 2016.
8. Cook S, Mahadevan R, Clarke A, El-Shamaa M. Reducing health and health care disparities: implementation lessons and best practices for health care organizations. http://www.solvingdisparities.org/sites/default/files/Resource-13-058_reducing_health_and_health_care_disparities_-_final.pdf. Published April 2015. Accessed July 12, 2016.
9. Robert Wood Johnson Foundation. Cleveland: Using data to improve blood pressure control in African American patients in Northeast Ohio. http://forces4quality.org/af4q/download-document/6427/Resource-af4q_broch_cleveland_singlepps_web.pdf. Accessed July 10, 2016.
10. Asthma Ready Communities. Asthma Ready.org website. http://asthmaready.org/. Accessed July 11, 2016.
11. Regional Equity Atlas. http://regionalequityatlas.org/. Accessed July 11, 2016.
12. Chin MH, Clarke AR, Nocon RS, et al. A roadmap and best practices for organizations to reduce racial and ethnic disparities in health care. J Gen Intern Med. 2012;27(8):992-1000. doi:
10.1007/s11606-012-2082-9.
13. Jha AK, Zaslavsky AM. Quality reporting that addresses disparities in health care. JAMA. 2014;312(3):225-226. doi: 10.1001/jama.2014.7204.
14. Pollack CE, Armstrong K. Accountable care organizations and health care disparities. JAMA. 2011;305(16):1706-1707. doi: 10.1001/jama.2011.533.
15. Werner RM. Does pay-for-performance steal from the poor and give to the rich? Ann Intern Med. 2010;153(5):340-341. doi: 10.7326/0003-4819-153-5-201009070-00010.
16. Joynt KE, Jha AK. Characteristics of hospitals receiving penalties under the Hospital Readmissions Reduction Program. JAMA. 2013;309(4):342-343. doi: 10.1001/jama.2012.94856.
17. Alley DE, Asomugha CN, Conway PH, Sanghavi DM. Accountable health communities--addressing social needs through Medicare and Medicaid. N Engl J Med. 2016;374(1):8-11. doi: 10.1056/NEJMp1512532.