Community Models of Care
Models of care are unique structures and processes of organizations or provider groups that deliver health service(s). Community models identified were: outpatient clinic (n = 11, 32%), home-based (n = 9, 26%), geographically defined (n = 8, 24%), and community-engaged clinic (n = 6, 18%) (Table A).
Outpatient Clinics. Outpatient clinics (n = 11) were freestanding or an outpatient arm of acute care hospital systems. Outpatient clinics most frequently provided diagnosis-driven health care services for geriatric conditions (n = 4) or all-inclusive services for individual older adults (n = 7). As a group, out-patient clinic models attempted to address care issues for older adults by creating a coordinated, one-stop-shop for complex care services.
Three of the four specialty/diagnosis-driven outpatient clinic models were evaluated on the effects of early detection and intervention for specific geriatric syndromes: Primary Care Research in Substance Abuse and Mental Health for Elderly (PRISM-E; Azar, Chopra, Cho, Coakley, & Rudolph, 2011; Domino et al., 2008; Kaskie & Buckwalter, 2010; Lee, Mericle, Ayalon, & Areán, 2009; McIntyre et al., 2008; Wiley-Exley, Domino, Maxwell, & Levkoff, 2009), Aging Brain Care Model (Callahan et al., 2011), and Improving Mood-Promoting Access to Collaborative Treatment (IMPACT; Unützer et al., 2008). Upon detection of dementia (Aging Brain Care Model), depression (PRISM-E and IMPACT), and/or substance abuse in older adults (PRISM-E), interdisciplinary teams tailored individual interventions and lessened illness acuity (i.e., measured by fewer neuropsychiatric symptoms).
Four of the seven all-inclusive services were evaluated and the Guided Care model was found to be the most extensively studied for its effects on: (a) the quality of primary care experiences in older adults with multimorbidities (Boyd et al., 2008), (b) provider satisfaction with care (Marstellar et al., 2013), (c) cost utilization (Sylvia et al., 2008), and (d) family caregivers (Wolff et al., 2010). The Guided Care model improved quality of care and quality of physician–patient communication; increased physician satisfaction with communication and management of chronic care; and reduced insurance expenditures, hospital admissions, hospital days, and emergency department (ED) visits. UAMS Senior Health Clinics reported similar outcomes except for financial viability (McAtee, Crandall, Wright, & Beverly, 2009). Program of All-Inclusive Care for the Elderly (PACE) and PACE's academic counterpart, Living Independently For Elders (LIFE) in Pennsylvania and New Jersey, report positive outcomes and continue to recruit older adults and improve satisfaction among older adults and caregivers (Sullivan-Marx, Bradway, & Barnsteiner, 2010).
Patient-centered medical homes (PCMHs) provide support and education for older adults and their families to manage and organize their own care at the level they choose (Agency for Healthcare Research and Quality, n.d.). Fishman et al. (2012) demonstrated the effectiveness of PCMHs; however, Hoff (2010) suggested a one-size-fits-all model could potentially miss specific patient populations.
The Nursing Model for Anticoagulation Management Services and the Heart Failure Resource Center are described as innovative with demonstrated effects in reducing health care costs (Health Workforce Solutions LLC, 2008); however, evaluation data are not reported.
As a group, the outpatient clinic models have tapped into unmet needs for older adults including: (a) refocusing care from provider- to person-centered, (b) accounting for the local context of care, (c) addressing the individual, (d) improving communication and understanding of older adults, and (e) effecting collaborative care for geriatric syndromes and other high incidence chronic conditions.
Home-Based. Home-based models provided care in the home. Nine models delivered geriatric services, including transitional care services (n = 4), primary care (n = 4), or telemedicine (n = 1). Eight of the nine models were systematically evaluated. There were three descriptive studies and one each of the following: randomized controlled trial, retrospective chart review, quasi-experimental study, case study, cost analysis, and post-hoc analysis.
Transitional care services addressed the transition of older adults with multiple chronic conditions from one setting to community-based care, typically from hospital to home. The four transitional care services, After Discharge Management of Low Income Frail Elderly (AD-LIFE), Care Transitions, Transitional Care Model (TCM®), and Chronic Care Coordination models reported reduced acute care use, reduced lengths of stay, and fewer ED visits. AD-LIFE showed improvement in both health and patient satisfaction for those with chronic health conditions (Allen et al., 2011). TCM, led by an advanced practice RN (APRN), was qualitatively evaluated and found older adults and caregivers had the necessary information and knowledge, and care coordination and caregiver experience were improved (Bradway et al., 2012; Naylor & Keating, 2008).
Primary care in the home was reported in the House Call-Home Care Program, the Veterans Administration Home-Based Primary Care (VA-HBPC) Program, Geriatric Resources for Assessment and Care of Elders (GRACE), and the Weydt model. Hospitalization and skilled nursing facility use patterns decreased in older adults served by the VA-HBPC Program (Wajnberg, Wang, Aniff, & Kunins, 2010). The GRACE model improved acute care use and documented improvements in health-related quality of life but no difference in physical functional outcomes (Counsell, 2011; Counsell, Callahan, Tu, Stump, & Arling, 2009). Weydt's model was nurseled and aimed to better address issues of poverty, social isolation, and loneliness, suggesting that a client-oriented rather than provider-oriented model increased providers' and older adults' positive perceptions of care and clinical outcomes (Weydt, 2010). Although telemedicine is on the rise, the one reported telemedicine model (n = 1) was diagnosis-driven for older adults with Parkinson's disease (Joseph, 2009). No evaluation of this model has been reported.
Geographically Defined. Geographically defined models included health care providers (e.g., a group of individuals or health systems) partnering with a geographically located community to offer health care services across the care continuum. Two of the eight geographically defined community models have been evaluated: a Community-Based Participatory Research (CBPR) Lower Mississippi Delta program and the Racial and Ethnic Approaches to Community Health (REACH). Similar to traditional public health models, the recipient of services is a population with a health care problem who reside in a geographically defined area: REACH focuses on diabetes and the CBPR Lower Mississippi Delta program focuses on nutritional needs. These models partner with community members of the designated area and collaboratively develop culturally tailored health programming delivered in community-determined locations. The CBPR Lower Mississippi Delta (Kennedy et al., 2011) and REACH (Hill & Jenkins, 2010; Hossler, Jenkins, & King, 2010; Jenkins, Myers, Heidari, Kelechi, & Buckner-Brown, 2011; Magwood, Jenkins, & Zapka, 2009) models reported high community member engagement with a positive impact on their health choices.
The Samaritan Center (Keller, Brown, & Koerner, 2010), Values Driven System (Gottlieb, Sylvester, & Eby, 2008), Layde model (Layde et al., 2012), Healthy Ager (Matthews, Parker, & Drake, 2012), Collaborative Patient Care Management (Health Workforce Solutions LLC, 2008), and Comprehensive Rural Care Collaborative (Health Workforce Solutions LLC, 2008) partner with communities at the outset of community work to improve buy-in, uptake, and sustainability, and to combine health care and non-health care services, such as retail or skills training centers. To improve community ownership of the services, the local hubs provide a visible community presence. The majority of geographically defined models were not evaluated, but on their websites, they report improved health outcomes and increased satisfaction.
Community-Engaged Models. Engaged models included a hybrid of community input and population statistics to guide program implementation (n = 6). There are six community engaged models that range from being fully engaged (i.e., 11th Street Clinic, Project Dulce, Michener model) to minimally engaged (i.e., Bauer model, Carolina Health Net, and Rebholz model). Fully engaged included a community assessment coupled with community-determined needs to guide health care services. Minimally engaged included provider selection of community services (Table A).
Four of the six community-engaged models were evaluated. Older adults' depression improved (Bauer model; Bauer et al., 2011), and student-run community clinics served the urgent and primary care health needs of an underserved population, improving health outcomes (Rebholz model; Rebholz et al., 2013). Project Dulce demonstrated improved clinical outcomes of diabetes and an economically significant reduction in hospital expenditures for this specific geographic and ethnic community (Philis-Tsimikas et al., 2012). In the Michener model, after assessing the community, neighborhood clinics were developed to meet needs and reduce the number of ED visits, which was a significant savings to the local hospital (Michener, 2013). As a group, fully engaged models were implemented in underserved communities that had pronounced resource needs and poor access to care.
The 11th Street Clinic (fully engaged; Gerrity, 2010) and Carolina Health Net (minimally engaged; Denham, Hay, Steiner, & Newton, 2013) described high levels of local uptake, improved community member health outcomes, and decreased ED visits.