How Do Area Agencies on Aging Build Partnerships With Health Care Organizations?

Abstract

Background and Objectives

Partnerships between health care and social service organizations may contribute to lower health care use and spending. Such partnerships are increasing, including Area Agencies on Aging (AAAs) working and contracting with health care organizations. Nevertheless, knowledge about how AAAs establish and manage successful collaborations is limited. We sought to understand how AAAs establish and manage partnerships with health care organizations.

Research Design and Methods

We conducted an explanatory sequential mixed-methods study using a positive deviance approach. We used national-level data to identify AAAs with multiple health care partners serving areas with low utilization of nursing homes by residents with low-care needs (n = 9) and AAAs with few health care partners and high utilization for comparison (n = 3). We conducted in-depth interviews with key informants from these 12 AAAs and their partner organizations (total n = 130). A 5-person multidisciplinary team used the constant comparative method of analysis, supported by Atlas.ti software.

Results

Highly partnered AAAs were characterized by 3 distinctive features of organizational culture: (a) attention to external environments, (b) openness to innovation and change, and (c) risk-taking to learn, improve, and grow. AAAs and partners describe a broad set of organizational strategies and partnership development tactics, depending on their local contexts. These features were underdeveloped in AAAs with few health care partnerships.

Discussion and Implications

While federal and state policies can create more favorable environments for AAA–health care partnerships, AAAs can also work internally to foster an organizational culture that allows them to thrive in dynamic and challenging environments.

Area Agencies on Aging, Cross-sectoral collaborations, Organizational culture, Positive deviance

Background and Objectives

Policymakers and health and social care providers are directing substantial attention toward cross-sectoral partnerships that address nonmedical influences on health (Alley et al., 2016; American Hospital Association, 2017; Kreuter et al., 2021; National Academies of Sciences, 2019). Evidence regarding the impacts of health and social care collaborations has been limited by measurement challenges (Alderwick et al., 2021; Amarashingham et al., 2018; Siegel et al., 2018; Winters et al., 2016; Woulfe et al., 2010), though several studies point to the potential for partnerships to improve mortality (Mays et al., 2016) and help avoid unnecessary health care utilization (Brewster, Brault et al., 2018; Brewster et al., 2019). Partnerships between health care organizations and Area Agencies on Aging (AAAs)—which coordinate and provide social services for older adults in communities across the United States—appear to reduce avoidable health care use and spending among older adults (Brewster, Brault et al., 2018; Brewster, Kunkel et al., 2018).

Challenges and facilitators of partnerships to integrate health care and social services have been well described (Alderwick et al., 2021; Amarashingham et al., 2018; Kunkel et al., 2018, 2020; Winters et al., 2016). Major constraints to partnering include inadequate financing infrastructure and payment mechanisms, misaligned goals, inability to measure or demonstrate return on investment, technical and strategic barriers to shared information technology platforms (Cartier et al., 2020; National Association of Area Agencies on Aging, 2015), and lack of awareness by health care staff of community and social care resources (Byhoff et al., 2019). Potential supports of partnerships include a common mission (Durfey et al., 2021), evidence of impact of AAA services on health outcomes or return on investment (Durfey et al., 2021), mutually beneficial and sustainable financing arrangements, inclusive governance structures to ensure joint responsibility and full representation (Brewster, Kunkel et al., 2018; Miller et al., 2017; Woulfe et al., 2010), trusted neutral conveners (Loehrer et al., 2015), and investments in relationship building across organizations (Byhoff & Taylor, 2018).

Several large reviews of research on health and social care partnerships have noted limitations to existing literature (Gottlieb et al., 2017), recommending research at the cross-organizational level (Winters et al., 2016), and greater attention to the role of social, economic, and political context (Woulfe et al., 2010). Although partnerships between AAAs and health care organizations are growing (Kunkel et al., 2013, 2018; Valluru et al., 2019), and success stories described (Aging and Disability Business Institute, n.d.-a), to our knowledge, there has been no in-depth study of best practices among AAAs that have partnered successfully with health care organizations published in the empirical literature. Positive deviance methodology (Marsh et al., 2004) has been particularly fruitful to understand patterns of collaboration across diverse contexts (Brault et al., 2018; Brewster, Brault et al., 2018) and recommended as a promising strategy for further inquiries (Alderwick et al., 2021).

Accordingly, we used a positive deviance methodology to understand how AAAs establish partnerships with health care organizations, with a focus on organizational factors that engender partnership success. We aimed to develop a taxonomy of empirically derived, theoretically informed organizational approaches used by AAAs with health care partnerships. Findings may be useful to AAAs and community-based organizations (CBOs) seeking collaborations with health care entities, as well as policymakers interested in fostering such relationships.

Research Design and Methods

Study Design and Sample

We used an explanatory sequential mixed-methods study design (Curry & Nunez-Smith, 2015; Figure 1) with deviant case sampling (Marsh et al., 2004; Patton, 2002). We chose mixed methods in order to develop a comprehensive understanding of AAAs through generating quantitative evidence about factors associated with partnerships, followed by a deeper exploration of those and other unmeasured factors, using in-depth qualitative methods. We used deviant case sampling because it allowed us to compare experiences across AAAs with distinct levels of health care partnerships.

Figure 1.

Explanatory sequential mixed-methods design. AAA = Area Agency on Aging.

To identify AAA sites with high and low levels of health care partnerships, we used data from the 2019 National Survey of Area Agencies on Aging to obtain a measure of AAAs’ formal partnerships with health care organizations (USAging, 2020). In the survey, AAAs were presented with a list of 10 health-related partners (e.g., hospitals, behavioral health providers, home health care) and asked to indicate if the AAA had a working partnership “that is formalized with a contract or memorandum of understanding.” We calculated the percentage of health care organizations with which the AAA reported partnering, out of the total possible health care partners and assigned quartile scores. We merged the data with the 2018 CBO–Health Care Contracting Survey to obtain additional information about the AAA’s contracting behavior (Kunkel et al., 2018).

To identify AAAs whose Planning and Service Areas (PSAs) had high and low levels of potentially avoidable nursing home use, we used data from the LTC Focus database maintained by Brown University to measure the proportion of nursing home residents with low-care needs in AAAs’ PSAs. We chose low-care needs nursing home use as our measure of potentially avoidable health care utilization because this measure aligns with AAAs’ focus on supporting older adults to age in place and has been documented as sensitive to AAA activities (Brewster et al., 2021, 2020). Nursing home data were available at the county level; the PSAs for most (88%) AAAs align with county boundaries so we calculated population-weighted averages by PSA and excluded AAAs whose PSAs did not align with county boundaries. We assigned quartile scores for low-care nursing home use to each AAA.

Study inclusion criteria were completion of the 2019 National Survey of AAAs and 2018 CBO–Health Care Contracting Survey, a PSA defined at the county level, and available low-care needs nursing home use data. A total of 304 AAAs met these criteria. Of these, AAAs were eligible for selection as highly partnered AAAs if they scored in the highest quartile for partnerships with health care organizations and the lowest quartile for low-care nursing home use (19 AAAs). AAAs were eligible for selection as low-partnered AAAs if they scored in the lowest quartile for partnerships with health care organizations and the highest quartile for low-care nursing home use (18 AAAs). We assigned random numbers to agencies in each eligibility group and selected them in numerical order, moving directly to the next site within the rank ordering. We skipped over agencies if they (a) duplicated geographic location, area served, and organizational structure to ensure sample diversity; (b) refused to participate; or (c) did not continue with follow-up recruitment.

We sent an invitation to participate in the study to the AAA director; those who agreed worked with the study team to generate a list of potential key informants (Patton, 2002) from the AAA staff and partnering organizations. Of the AAA directors invited to participate, two refused due to competing priorities (the coronavirus disease 2019 [COVID-19] pandemic and natural disasters in the region), and two were lost to follow-up recruitment. We recruited sites iteratively with data collection until we achieved thematic saturation (Morse, 1995) both in terms of sites with high and low levels of health care partnerships (nine sites with high levels of health care partnerships and three with low levels of health care partnerships) and within each site (n = 130 participants; Table 1). The study was exempt from internal research review board review under 45CFR46.104; all study participants provided their verbal informed consent to participate in this study.

Table 1.

Site Characteristics

Category	% of Formal health care partnerships	% of Low-care nursing home usea	Governance structure	Census region	Median household income ($)	Population	% of Population older than 65	Number of interviewees (n = 130)
Highly partnered	40	3.1–5	City/county government	West	88,131	846,006	16.2	6
Highly partnered	40	3.1–5	RPDA/COG	South	69,709	4,656,875	12.2	13
Highly partnered	40	5.1–10	Independent nonprofit	Midwest	48,858	549,683	18.9	12
Highly partnered	80	3.1–5	City/county government	South	108,820	1,050,688	16.1	14
Highly partnered	60	0–3	Independent nonprofit	Northeast	55,120	322,763	22.8	12
Highly partnered	50	3.1–5	RPDA/COG	West	53,695	252,042	19.9	11
Highly partnered	50	5.1–10	City/county government	South	92,224	528,898	10.3	13
Highly partnered	50	3.1–5	Independent nonprofit	Northeast	62,573	112,663	20.1	13
Highly partnered	70	0–3	RPDA/COG	West	57,643	687,972	14.9	11
Low partnered	10	>20	RPDA/COG	West	65,859	764,636	13.7	15
Low partnered	0	>20	City/county government	Midwest	43,896	300,576	13.1	6
Low partnered	0	10.1–20	City/county government	Northeast	71,189	672,391	12.2	4

Note: AAA = Area Agency on Aging. AAAs in this sample may be a part of Regional Planning and Development Agencies (RPDA) or Council of Governments (COG), city or county governments, or be independent nonprofit agencies.

^aPercentages of low-care nursing home use reported as a range to maintain site anonymity.

Data Collection and Analysis

Interviewers were not blind to the partnership levels of the sites. We designed the study to include ethnographic fieldwork; however, this was dropped at the onset of the COVID-19 pandemic. Instead, interviewers reviewed available background information (e.g., characteristics of the AAA and region) to provide a useful context within which to position organizations and participants, drawing from practices recommended in anthropology (Trotter et al., 1998). The interview guide (Supplementary File 1) was designed to elicit descriptions of processes of creating and sustaining partnerships, grounded in respondents’ direct experiences. Probes explored the potential roles of AAA leadership, organizational structure, and policy environment in facilitating effective partnerships, as well as concrete strategies for a partnership that could be replicated in other communities. Interviews were originally planned to be conducted in-person, but due to the COVID-19 pandemic were changed to virtual data collection via Zoom. Interviews were approximately 1 h in duration, were audio-recorded, and professionally transcribed.

Data analysis was conducted by a five-member multidisciplinary team using the constant comparison method (Glaser & Strauss, 1967; Miles & Huberman, 1994), which is “The method of comparing and contrasting is used in … forming categories, establishing the boundaries of the categories, assigning the segments to categories, summarizing the content of each category, finding negative evidence, etc. The goal is to discern conceptual similarities, to refine the discriminative power of categories, and to discover patterns” (Tesch, 1990). We began with a “start list” (Miles & Huberman, 1994) of code categories derived from prior research and findings from the Phase 1 quantitative analyses: community features, AAA organizational features, partner organizational features, partnership initiatives, relational aspects of partnerships, and external influences. Analysis was done iteratively with data collection. Analysts coded transcripts independently, meeting regularly to arrive at consensus on divergent views, revising the code structure as needed.

This process of refining codes and describing properties of each continued until no new concepts emerged. The structure was shared at regular intervals with the full team to determine if concepts were consistent with findings from the Phase 1 quantitative data, empirical literature and their own expertise, and was finalized after 13 rounds of revision (Supplementary File 2). We compared themes by performance status at the final stage of analyses. We used three techniques to minimize the effect of researchers’ preconceived biases: (a) researcher reflexivity throughout the process; (b) a five-person multidisciplinary team trained to critically analyze transcripts, with a focus on identifying disconfirming cases; and (c) data collection strategies to encourage respondents to provide both positive and negative comments (Mays & Pope, 2000). We used ATLAS.ti Scientific Software, version 8.4 (ATLAS.ti, Berlin Germany) to facilitate organization, coding, and retrieval of quotations.

Research Team Reflexivity

The research team was diverse with regard to professional background, organizational affiliation, and expertise. Most members have over a decade of experience studying long-term care and home and community-based services. Interviewers did not have prior relationships with interviewees before the study (except one AAA staff person was known to the interviewer) and the motivations for conducting the study were described during the consent process.

We assessed the trustworthiness of the data using Guba’s 4 constructs (Guba, 1981). In addressing credibility, we attempted to generate detailed descriptions of the phenomenon (AAA health care partnerships) and assessed the degree to which the findings cohere with what is known. To support transferability of findings, we aimed to describe the context, as well as the process of establishing partnerships, to allow others similarly situated to determine whether findings are relevant to their settings. In terms of dependability, we used several techniques: (a) a five-member multidisciplinary analysis team, (b) a detailed audit trail (memos regarding analytic decisions, all versions of the code structure and code reports), and (c) maintenance of the entire data set in Atlas.ti. To achieve confirmability, we regularly sought feedback on the findings as they evolved from both the quantitative analysis and dissemination teams (Expanded Methods- see Supplementary File 3).

Results

We interviewed a wide range of participants from AAAs and various partner organizations, with an average of 11 at each site (Table 2).

Table 2.

Characteristics of Study Participants (n = 130) Among AAAs and Partner Organizations

Variables	n
Role/job functions of AAA participants	69
AAA executive director/chief executive officer	9
ADRC staff, senior center directors, coordinators	10
Case management/behavioral and physical health	12
Business operations	10
Community programs/resources	8
Home- and community-based services	12
Integrated care/care transitions	8
Organization type of partner participants	61
Hospital
Social worker, community liaison, case manager, administrator	9
Nonhospital health care
Community health care provider, clinics, visiting nurse agency, long-term care facility	6
Mental/behavioral health	4
Department of Health/Public Health, Medicaid	5
Social service organizations
Nutrition and housing	4
Emergency preparedness	5
Educational	5
Elder services, elder justice, age-friendly communities, civic groups (other)	23

Note: AAA = Area Agency on Aging; ADRC = Aging and Disability Resource Center.

While we set out to identify specific strategies and tactics related to partnerships, we also found deeper and more generalized features of organizational culture that were common across highly partnered sites. These underlying features of culture supported the AAAs’ ability to build partnerships with health care organizations: (a) attention to external environments, (b) openness to innovation and change, and (c) risk-taking to learn, improve, and grow. We also identified a set of broad strategies and a wide array of tactics that AAAs employed, depending on their unique local and state context.

In AAAs with few health care partnerships and high levels of utilization, these features of organizational culture, strategies, and tactics that characterized highly partnered sites were underdeveloped. In these sites, participants described positive relationships with partners in sectors other than health care (to be expected, because that was the basis for their inclusion). Some successes of community collaboration were described, but AAAs appeared to play a less central role in those initiatives. In addition, these AAAs did not report investment in learning deeply about partner pressures and goals, adapting to meet partner goals, and embracing risk-taking. They described challenges such as persistent lack of political will among state/local governments to support older adults, funding constraints, and limited public awareness of the AAAs.

We developed a taxonomy of distinctive features of AAA organizational culture and identified broad strategies and specific partnership development tactics related to those features as reported by AAAs and their partners (Table 3). The taxonomy is illustrated by brief vignettes and quotations from a wide range of participants from AAAs and partnering organizations.

Table 3.

Distinctive Features of AAA Organizational Culture, Organizational Strategies, and Partnership Development Tactics Among Highly Partnered AAAs

Distinctive features of AAA organizational culture	Organizational strategies	Partnership development tactics
Attention to external environments	Monitor trends (e.g., demographic, regulatory, market) and opportunities	Dedicated planning units; robust data capacity; coalition meetings; explicit investment in partner relationships; MOUs including role differentiation; embedding staff in partner organizations; credentialing for data access; shared FTE across organizations; processes to support timely communication
	Actively seek to understand and align with partners’ goals, ways of working, and language
	Define and market AAA expertise
	Ensure close coordination with partners
Openness to innovation and change	Adapt roles, programs, and operations	Value and support multifaceted staff training (e.g., incentives/awards); educational professional development/info exchange forums; VA benefit fairs and informal mentors
	Develop workforce capacity to innovate
Risk-taking to learn, improve, and grow	Embrace uncertainty	Managerial risk-taking (both latitude and accountability for managers); unfunded or cofunding grant-writer position
	Balance calculated risks with pragmatism
	Accept short-term losses/take actions to anticipate the future market

Note: AAA = Area Agency on Aging; MOU = Memorandum of Understanding; FTE = full-time employee; VA = Veterans Affairs.

Attention to External Environments

Highly partnered AAAs directed attention to their external environments, using four broad strategies: (a) monitoring trends and opportunities, (b) actively seeking to understand and align with partners’ goals and ways of working, (c) defining and marketing expertise, and (d) ensuring close coordination with partners.

Monitoring trends

Highly partnered AAAs routinely studied trends at the local and national levels and proactively pursued partnerships in response to these trends. One AAA director described conducting research on hospitals in order to frame their negotiation strategy:

You have to meet their deficit needs. You can’t just impose a service on them that they don’t see the value of … you’ve got to look at the latest reports. You’ve got to look at the star ratings. And it won’t be a one size fits all. For one organization it’s going to be, “they might listen to us in this and can we deliver something to fill that space?” 06_ID10

Actively seeking to understand partners

Highly partnered AAAs also invested in understanding health care partners’ organizational ways of working. One AAA created a case manager role dedicated to working with the Veteran’s Administration (VA) to leverage VA benefits to serve their large number of veteran clients. This relationship began as part of a pilot program to bring Veteran resources to rural communities, with AAA staff becoming accredited as Veteran’s Service Organization (VSO) representatives. The AAA’s VSO described the evolution of the relationship over 5 years; they overcame early challenges, such as the AAA’s difficulties in accessing information at the VA and the VA being “a little distrustful” of working with the AAA. Building the relationship required the AAA to “prove themselves” (e.g., “we could bring benefits to the table that their veterans could use, we would not transport their records, we would follow their recommendations [on confidentiality]”). Over time, the VA grew to value the breadth of screening and referral done by the AAA that could benefit older veterans (e.g., weatherization, caregiver support). The AAA VSO connected with a VSO within the VA who became an informal mentor who helped her learn the VA’s “unspoken culture”.

The VA has a culture. One way that you can build trust with them is you are early to any event … you show that you understand their language. There’s a certain culture of hierarchy. You make sure that you don’t go over someone’s head if you’re not supposed to. 03_ID02

In another example, in 2012, one AAA partnered with a sister agency and several hospitals in the state on CMS’s (Centers for Medicare & Medicaid Services) Community-based Care Transitions Program aimed at reducing hospital readmissions among Medicare beneficiaries. As the pilot ended, the AAA CEO leveraged a relationship she had developed with one of the hospital executives to create a AAA staff position (“Hospital Options Counselor”) embedded in the hospital to address readmissions by arranging community supports during the hospitalization. Overcoming initial confusion about the role and its value, hospital staff welcomed the counselor’s knowledge of community resources and, importantly, saw the role as making their own jobs easier:

Just the ease of not having to send a paper form … [the AAA staff person] has access to our EMR [electronic medical record] … that’s really helpful because she can see their insurance, what’s going on with them medically …. So that is less legwork on our staff, which is always a plus. When you can have somebody who knows what they’re looking for and we don’t have to kind of figure all that out … that was a big benefit to our staff. 07_ID12

Defining and marketing expertise

Even as they committed to a shared goal (“the care of the commons”), these AAAs were deliberate about marketing their services. Through formal coalition meetings and informal interactions with various organizations, they learned about existing resources and defined their role accordingly (“how do we prioritize where we want to innovate … what is our lane?”). One AAA caregiver coordinator reported: “Collectively we’re identifying gaps, strengths, how between us we could actually share resources to adapt service delivery.” One AAA director saw formal coalition meetings as a place where organizations assembled a comprehensive mapping of community services to facilitate referrals, avoid duplication, and identify opportunities:

Collectively we’re looking at our services … the worst thing is to duplicate services …. We could focus our energy on something that’s not being provided. 15_ID01

Ensuring close coordination was an ongoing priority for AAA and health care partners. One large health system and AAA were “working in [their] own siloed organizations, doing the same classes” for evidence-based programs such as fall prevention. Attendance was unpredictable and finding certified teachers was difficult. Several years ago they met to explore how they could share volunteers, resources, and advertising “rather than try to independently go it alone.” They established a Memorandum of Understanding (MOU) for working with Retired Senior Volunteer Program volunteers, as this hospital-based community specialist described:

The MOU really helped eliminate some of the overlap that was happening …. It was a benefit to them [AAA] because then we can have a shared pool of volunteers with a clear understanding of what they can and can’t do … we set that rather than a contract [at our health system] it’s like an Act of Congress times ten to get a contract put in place. An MOU is an easier way for us to be able to legally justify some of the agreements made. 03_ID12

Highly partnered AAAs used several tactics for monitoring the external environment, including dedicated planning units, investing in data capacity to facilitate collaboration, and understanding the needs of hospital partners (“we created a program development unit … not only because we know environments change and the competition is changing, but the need is changing”). Hosting or participating in meetings with coalitions and partners was also a powerful tool (“building trust … and guidelines for decision making as a coalition with partners from different agencies … was time consuming, but well worth it in the end”). While AAAs also pursued formal contracts, MOUs were seen as flexible yet formal agreements for working together. For one hospital transitions partnership, the MOU defined roles, protocols for data sharing and electronic health record access (data security and HIPAA compliance was a challenge in some sites), regular program review, and performance-based metrics for accountability. In addition to MOUs, other structures that supported coordination included (a) a single interagency website hosted by the state, (b) dedicated point staff at the health system and AAA, and (c) embedding AAA or health care partner staff in each other’s systems (e.g., daily rounds, physical office, huddle calls). Finally, partners developed policies and practices that facilitated open and timely communications (e.g., encrypted email, walkie talkie/Vocera; shared drive for resources).

Openness to Innovation and Change

AAA staff and partners described AAAs with robust partnerships as receptive to innovation and change, using two broad strategies: (a) adapting roles, programs, and operations and (b) developing workforce capacity for innovation.

Adapting roles, programs, and operations

The importance of AAAs’ flexibility in responding to dynamic environments was commonly highlighted. An AAA director of community resources: “we can’t just keep doing the same thing over and over again. Whatever may have made sense 5 years ago doesn’t necessarily make sense today … what elements can we keep and where do we need to tweak?” One AAA CEO reflected how a care transition demonstration program helped them establish a relationship with several insurers that evolved into a larger partnership. The insurers had identified pregnant women and behavioral health as priorities and invited the AAA to expand hospital transition services beyond older adults to also serve these areas:

[The insurers] said, “we want to help with people leaving the hospital of any age”…. We thought, “okay, let’s build on this.” The next thing you know, we have an agreement where we deal with people leaving the hospital … It’s very much turned into selling social determinants of health expertise as much as selling it with older adults. 07_ID01

Developing workforce capacity to innovate

AAAs recognized the imperative to develop workforce capacity and invested heavily in skill building for both their staff and for partners. Continuous professional development was regarded as a core value (“I’m a lifelong learner and I expect everybody I’ve hired on my team to be”). Staff in these AAAs described peer learning that is explicitly encouraged by leadership (“they strengthen each of us so that we can mentor each other”). AAAs also pursued cross-organizational learning; as one AAA home care director shared, she and her hospital partner “were teaching each other.” Other skills AAAs honed for engaging effectively with health care included technical expertise and business development skills such as “negotiating with confidence.” Importantly, one AAA emphasized specifically tending to capacity building among middle managers (“it’s key to make sure your middle managers are comfortable with networking”). The AAA CEO highlighted the consequences of underdeveloped middle managers who resisted moving in new directions:

[It’s] the middle manager piece where things can still fall off the rail … that’s where we’re seeing the most tension in developing partnerships … not at the strategic level and not necessarily at the service delivery level. It’s at the operational level, inside the middle management of all partnering organizations. 10_ID01

AAAs employed a range of tactics to support workforce professionalism and capacity development. These included valuing and investing in staff training (e.g., offering incentives/awards); active engagement with educational professional development forums; strengthening technology skills. One AAA manager reflected on pushing herself and team to develop new expertise:

Before the pandemic, I was pushing people to think, we have technology, let’s embrace it … I’m of a generation where we were not digital natives. I’ve learned to use technology. It’s not my most comfortable place and a lot of my case managers are in that same place … people who were really kind of technophobic in the past have started to realize that maybe there’s a place for this. 09_ID03

Risk-Taking to Learn, Improve, and Grow

AAAs with robust partnerships viewed risk-taking as intrinsic to surviving and thriving in a highly dynamic environment. They used several approaches to risk-taking: (a) embracing uncertainty, (b) balancing calculated risk with pragmatism, and (c) accepting short-term losses in favor of strategic positioning for the future.

Embracing uncertainty

AAAs expressed being comfortable with complexity as they pursued innovative programs, rather than “simple funding with a very clear job.” This AAA Assistant Director described the early stages of pioneering a new health care relationship in unfamiliar terrain. Their agency was invited to participate in launching a new program with a large health care company to manage a commercial population because the AAA was “the best in community service delivery.”

In the beginning you have to jump into something you’re not as familiar with …. We have to understand different language for the health care system …. We are going to facilitate relationships with a health care partner. So why don’t we pioneer it along with everyone else and let’s get comfortable with that …. We kind of went into this relationship with, this is a safe way for us to practice, because when the big dogs come talking to us, we better be more ready. 06_ID10

Balancing risks with pragmatism

Notably, highly partnered AAAs were deliberate about identifying risks (“we are not risk averse. We just want to know what the risks are and make calculated risks for development”). At the same time, these AAAs were pragmatic in assessing when losses were not sustainable or advantageous. For instance, this AAA manager described:

Our AmeriCorps funding had been reduced from the federal level … so we lost some of the partnerships. It was also getting challenging to find AmeriCorps members in certain areas … we made the decision that we couldn’t keep it going financially. So we didn’t apply for a renewal. 03_ID06

Accepting short-term losses

Risk-taking included investing resources in services that were not yet profitable as part of proving their value proposition. Taking actions to position themselves for a future market, one AAA manager described short-term trials to demonstrate the value of their services that lead to paid contracts with health care entities:

We’re an entrepreneurial social enterprise that specializes in aging and disabilities … to prove our value, we step out and do some things for free for a little bit of time just to say, look at what we can do. Let’s see if this is something you want. We’ll do this for six months. We won’t charge you anything. We’ll collect data. We’ll show you what happens. So we’ve had a lot of success with that. 06_ID03

One AAA director summarized her agency’s investment in providing care transitions for a local hospital, noting it took several years before the partnership was financially viable. She considered the investment worthwhile because it helped them understand the hospital’s challenges and develop services to fit that need:

We had an agreement with the local hospital [for care transitions], but we were not being reimbursed …. We had a really good opportunity to learn what challenges the hospital was having and also how could we modify how we deliver our services … then when this really took off, when we were getting enough reimbursements for the services we were providing, we have been able to add admin support … collectively, we’re about three to four years in to actually receiving compensation for the care transition. 15_ID01

Tactics that highly partnered AAAs used for risk-taking included allowing managerial risk-taking (balancing latitude for decisions and accountability), investing resources for pilot programs, and dedicating staff for grant-writing. One AAA CEO described how their board endorsed AAA executives to pursue negotiations with insurers:

I look at it as an opportunity to prove ourselves …. There are some of us who are much more like, “… let’s put money into this. Let’s really build it.” And there are some who are much more like, “I don’t know. Are you sure?” … sometimes you have to kind of pull the rest along. We had a board meeting where … they said, “okay, officers, we trust you. Go negotiate with the insurance companies and then let us know what happens.” 07_ID01

Discussion

We found that AAAs engaged in multiple partnerships with health care entities exhibit features of organizational culture that distinguish them from those who have not yet established such partnerships. Although as with AAAs nationally (Kunkel, 2019), those in our study were diverse in notable ways (i.e., geographic region, area served, and governance structure), the highly partnered AAAs focused on external environments, were open to change, and engaged in risk-taking in order to learn and grow. These shared features shed light on what may facilitate successful partnerships among AAAs and health care organizations. In addition, highly partnered AAAs employed a “toolkit” of strategies and partnership development tactics that were largely undeveloped in the sites with low levels of health care partnerships.

Our findings are consistent with theories of open systems (Katz & Kahn, 1966), which posit that organizations exist in constantly changing environments and must understand and adapt effectively in order to survive. Participants at the sites with high levels of health care partnerships were intentional in tuning in to the external environment in a number of ways (e.g., regulatory conditions, health care market changes, and social and demographic trends). Our results also provide further empirical support that adaptive organizational culture (Constanza et al., 2016) represents one important mode by which organizations can adjust to suit changing environmental pressures. Two features of adaptive organizational culture, attention to external environments and openness to innovation and change, differentiated the AAAs with the greatest and least number of health care partnerships. First, highly partnered AAAs are oriented to scanning and understanding their environments, working to know health care partners’ pressures, incentives, operations, and cultures, including speaking their language (Taylor & Byhoff, 2021). Second, risk-taking, which has been underdeveloped in prior conceptualizations of adaptive organizational culture (Constanza et al., 2016; Schein, 2010), was apparent among highly partnered AAAs. Participants described in detail how the organization would assess risk/reward and intentionally engage in risk-taking based on this calculus. Supports for AAAs seeking to develop these capacities include the Aging and Disability Business Institute (Aging and Disability Business Institute, n.d.-b), led by USAging. The Business Institute provides resources, training, and technical assistance to CBOs to enhance business acumen, including development of many of the ways of working and skills demonstrated by highly partnered AAAs in this study.

Highly partnered AAAs pursued health care partnerships through focusing on several conditions of collective impact theory (Kania & Kramer, 2013). First, setting a common agenda with partners was described as a core activity that occurred early in the partnership as AAAs worked to understand and align with their environment. Second, partners invested heavily in defining and marketing their unique expertise to engage in distinct but coordinated activities, using tools such as MOUs to ensure role clarity and support accountability. Finally, AAAs and partners placed a high value on timely communication, facilitated by tactics such as embedding staff in the partner organization or providing AAA staff access to electronic health records.

There are several limitations to this study. First, the criteria for selecting AAAs from high and low quartiles may be influenced by other factors, such as overall funding levels for AAAs, pressures/incentives faced by health care organizations, and features of the health care system that contribute to low-care nursing home use outside AAA control. Second, our sample included 12 AAAs and partners that were purposefully selected to provide rich information; findings may not be transferable to a wider range of AAAs. Third, social desirability and recall bias may have occurred. However, to minimize these effects, we interviewed multiple key informants from diverse organizations at each site and encouraged participants to share both positive and negative experiences (Patton, 2002). This study is hypothesis-generating; full positive deviance studies include subsequent phases of quantitative measurement on nationally representative samples (Bradley et al., 2009), an important next step.

Conclusion

While external factors such as state and federal policy (such as the CHRONIC Care Act) can create an environment that can promote these partnerships, AAAs and other CBOs can also strengthen their internal capacity for this work by accessing USAging’s Aging and Disability Business Institute resources to strengthen their organizational culture in the areas of openness, adaptability, and risk-taking. CBOs can assess their readiness for health care contracting in domains such as strategy and planning, market awareness and orientation, and management and operations. Business Institute tools, such as a market assessment and competitor analysis, orient CBOs to the external environment and help them develop value propositions for services using health care appropriate metrics. Training on organizational culture change and risk management prepares staff for contracting with health care organizations by encouraging a mindset of openness. Learning collaboratives on financial acumen also help CBOs to determine their costs and manage financial risk.

We studied a set of highly partnered AAAs from diverse settings to identify common features of organizational culture and strategic and tactical approaches that may engender success in partnering with health care entities. While changing organizational culture takes time (Schein, 2010), evidence suggests it is possible (Curry et al., 2018). AAAs seeking to become more highly partnered may consider cultivating openness, adaptability, and risk-taking within their organizational cultures, enabling them to thrive in a dynamic environment.

Acknowledgments

We would like to thank USAging and the USAging-led Aging and Disability Business Institute for sharing their expertise on AAAs and health care partnerships and Terrie Wetle, PhD, Brown University Center for Gerontology and Healthcare Research, for her guidance on this project.

Funding

This work was supported by RRF Foundation for Aging and the Donaghue Foundation. The 2019 AAA National Survey was supported, in part, by grant number 90PPUC0001 from the U.S. Administration for Community Living, Department of Health and Human Services. Grantees undertaking projects with government sponsorship are encouraged to express freely their findings and conclusions. Points of view or opinions do not, therefore, necessarily represent official ACL policy.

Conflict of Interest

None declared.