The Congregational Care Network: Preliminary Data From a Healthcare/Congregational Partnership for At‐Risk Older Adults

ABSTRACT

Background

Social isolation and loneliness are significant public health crises that can exacerbate stress and diminish health behaviors, leading to overall reductions in well‐being. The effects of systemic upstream social determinants of health (SDOH) can worsen these effects. Partnerships between communities of faith and health systems have the potential to reduce social isolation and loneliness, address unmet social needs, and improve access to healthcare.

Methods

The Congregational Care Network (CCN), a collaboration between a health system and local congregations in neighborhoods with high poverty and gaps in other SDOH, provided 1 h per week of individual, volunteer companionship to older adult patients for 90 days. The health system provided training and professional support from social workers and chaplains. A program evaluation measured loneliness before and after participation and healthcare utilization in the 90 days before, during, and after the program.

Results

CCN recruited 28 congregations representing diverse religious affiliations and 335 patients participated in the CCN program. Patients who received CCN services had a median age of 64.9 years (standard deviation 11.5), were 27.2% male, and 58.8% Black. There were significant reductions in DeJong Gierveld loneliness scores from before to after program engagement (median change score: 1 (interquartile range (IQR) 0–2, p < 0.001)). The proportion with 1+ emergency department visits was significantly lower after CCN compared to before (16.8% vs. 24.6%, p = 0.007); the proportion with inpatient visits was lower during CCN compared to before (12.2% vs. 17.3% vs. p = 0.032). The proportion with outpatient visits was higher during CCN than before (71.0% vs. 63.8%, p = 0.045).

Conclusion

The CCN partnership between congregations and a local health system is a feasible model for at‐risk older adults that may reduce loneliness and shift healthcare utilization from acute to outpatient settings, providing greater continuity of care and fewer burdensome acute care visits.

Summary.

• Key points

  • ○
    The Congregational Care Network (CCN), a collaboration between a health system and local congregations in neighborhoods with high poverty and gaps in other social determinants of health, provided 1 h per week of individual, volunteer companionship to older adult patients for 90 days.
  • ○
    CCN Participants had reductions in loneliness after compared to before the program and also had fewer emergency department visits and more outpatient visits, suggesting more continuity of care and fewer acute care needs.
  • ○
    The CCN partnership between congregations and a local health system has the potential to improve well‐being and healthcare utilization for older adults at risk of poor health outcomes.
• Why does this paper matter?

  • ○
    Social isolation and loneliness are significant public health crises that can exacerbate stress and diminish health behaviors, leading to overall reductions in well‐being.
  • ○
    The Congregational Care Network (CCN), a collaboration between a health system and local congregations in neighborhoods with high poverty and gaps in other social determinants of health, provided 1 h per week of individual, volunteer companionship to older adult patients for 90 days.
  • ○
    This paper provides preliminary evidence that the CCN program has the potential to reduce loneliness and shift healthcare utilization to outpatient care from the hospital and emergency department.
  • ○
    This program could be replicated in many other healthcare settings to provide an effective approach to reducing social isolation and improving healthcare utilization.

1. Introduction

Social isolation and loneliness are significant public health crises [1]. Defined as an objective lack of social connections (social isolation) and a subjective sense of disconnection (loneliness), these conditions can exacerbate stress and diminish health behaviors, leading to reductions in well‐being [2]. Research has demonstrated that loneliness and social isolation are linked to increased all‐cause mortality, comparable to established risk factors like smoking, obesity, and physical inactivity [3]. Older adults are particularly at risk, with 24% of those 65+ socially isolated and 43% of adults 60 and older reporting loneliness [4].

In populations that experience health inequities due to social determinants of health (SDOH), the impact of these conditions can be even more profound [5]. Black and Hispanic individuals face an increased risk of chronic conditions like Alzheimer's disease, cardiovascular disease, diabetes, cancer, and all‐cause mortality [6, 7]. Black and Hispanic older adults also develop multiple chronic diseases at younger ages, beginning in their 50s, due to SDOH such as poverty, systemic discrimination, or access to healthcare [8]. These risks are compounded by social factors, such as limited access to healthcare, systemic discrimination, and the erosion of social support systems, amplifying the impacts of social isolation and loneliness.

In marginalized communities, faith‐based organizations often serve as pillars of social and emotional support. Trusted institutions, many with histories extending decades or centuries, provide more than spiritual guidance; they often play an important role in health promotion, social services, and community engagement. This is especially true in Black communities, where churches have historically played a vital role in supporting families and older adults. By fostering opportunities for social engagement and connection, churches may help mitigate the risks associated with loneliness and isolation [9, 10].

The Congregational Care Network (CCN) builds upon the “patient navigator” model of care developed by the Le Bonheur Methodist Health System in Memphis and the “Connector” model of FaithHealth NC in Winston‐Salem [11, 12, 13]. These models involve partnerships between health systems and local congregations. In the Memphis model, patient navigators are assigned to patients' congregations to help identify and address needs upon discharge from the hospital. Similarly, in the FaithHealth NC model, care coordinators or community health workers connect patients with congregations that can provide the specific social and practical support they require.

Founded in 2020 by Indiana University Health, the CCN represents an evolution of these models with a distinct focus on addressing social isolation and loneliness. The program was informed by an earlier community‐engaged participatory research project with 14 downtown Indianapolis congregations [14]. A key theme that emerged was the need to incorporate congregational knowledge and support systems as key resources in the CCN program. Three congregations that participated in these early discussions were among the first to join the pilot program.

The CCN provides spiritual and emotional companionship for patients by trained volunteers, called Connectors, who live in their neighborhood. Unlike previous models, the CCN integrates more deeply with clinical teams by relying on professional chaplains for patient assessment and referral. CCN volunteers, chaplains, and staff members receive ongoing education and support from Clinical Pastoral Education (CPE) educators to ensure they provide meaningful assistance to patients. The CCN team also includes a licensed clinical social worker (LCSW) to address unmet needs such as food insecurity or behavioral health treatment. The purpose of this manuscript is to describe the CCN model of care and report findings from a program evaluation conducted during the first 29 months of the program to evaluate CCN's impact on loneliness and healthcare utilization.

2. Methods

2.1. Setting

The Indiana University Institutional Review Board determined that this program evaluation was exempt from review. Indiana University (IU) Health, the largest health system in Indiana, includes 15 hospitals and has a total capacity of 2708 beds. In 2018, the system reported 118,019 admissions, 2,992,044 outpatient visits, and 110,445 surgeries.

This pilot program was conducted at the Academic Health Center in Indianapolis, the tertiary referral center for the state that includes primary care clinics serving economically and racially diverse patients, and Bloomington Hospital in a smaller Indiana city with a state university and larger rural population. In 2022, the Indianapolis Metropolitan Statistical Area (MSA) had a population of 2,109,957, with 30% of the total population identified as people of color [15]. The Bloomington MSA included a population of 160,874, with 16.7% of the total population identified as people of color [16].

The IU Health Population Health team identified specific zip codes in both cities that had a high risk of social/health inequities. Study participants were recruited from these zip codes.

2.2. The Congregational Care Network Program

Congregations were selected based on their significant involvement in the health and wellness of their neighborhoods and alignment with community‐based ministries. Congregations that lacked interest or the capacity to support or work with persons of all faiths or of no faith were excluded from participation. Each congregation was provided with funding of $1000 per month from the health system to support the administrative costs of the program. The congregations were given wide discretion in how to allocate the funds. The health system did not ask for information about how the congregations used the funds.

As of January 1, 2023, CCN had recruited 25 congregations representing diverse religious affiliations in Indianapolis. Among these, eight were non‐denominational Christian, six Baptist, two African Methodist Episcopal, two Church of the Nazarene, two Roman Catholic, two United Methodist, one Apostolic, one Muslim, and one Presbyterian/Presbyterian Church U.S.A. (PCUSA). In Bloomington, CCN recruited three congregations: one United Methodist, one African Methodist Episcopal, and one Church of the Nazarene. These congregations are located in a diversity of neighborhoods within identified zip codes, with significant representation from historically Black churches.

The CCN program operates through partnership between each congregation and CCN staff at IU Health (Figure 1). Each congregation identifies volunteers, referred to as “Connectors”, who provide direct support to the patients (Figure 1). A CCN chaplain and LCSW provide key support to each congregation. The chaplain's role is to develop relationships with the clinical team and the patient and assess whether CCN could benefit the patient. Chaplains also connect patients with resources for spiritual support in the community. The LCSW (1) contacts the patient to confirm interest in the program and assigning the patient to a congregation; (2) provides the congregational Connector with information about resources from both the health system and the community to address the patient's health and (SDOH) needs, and (3) offers guidance or assumes care if psychosocial or behavioral health needs exceed the Connector's capacity to provide support. The CCN Program Coordinator manages data collection and oversees the program.

FIGURE 1.

Congregational Care Network program structure at Indiana University Health and within participating congregations.

Connectors are chosen by congregational leadership and include one Lead Connector and at least 5 additional Connectors. The Lead Connector commits to approximately 8 h per week to basic data and maintain communication with CCN staff. Other Connectors commit to 1 h per week for providing companionship to a patient and 1 h for meeting with other Connectors in their congregation to share experiences and reinforce their use of Connector training. Each Connector provides companionship through weekly one‐hour conversations for 12 weeks following the patient's referral to the program. Congregations had the option of using some of the administrative support funds to pay Connectors for their time. Congregations were highly successful in recruiting volunteers for the Connector roles.

Volunteer training is a critical component of the Connector role. CCN partnered with Pathways to Promise, a non‐profit organization that offers “Companionship Training,” a three‐part introductory course on listening skills and supportive care (https://www.pathways2promise.org/). Training consists of three 90‐min sessions led by CCN Team members certified to offer this training. A fourth 90‐min seminar is offered to address referral resources in the community, patient confidentiality, patient rights, HIPAA regulations, and the importance of maintaining boundaries.

Ongoing support is provided following training. Each congregation is assigned a chaplain mentor from the CCN team who connects regularly with the Lead Connector. Each month, Lead Connectors from each of the 26 participating congregations meet on virtual platform for shared learning sessions, led by the coordinator and the LCSW. The chaplain attends this meeting as a content expert, while the LCSW and coordinator are available for consultation and logistical support. Additionally, CCN offers an optional, monthly virtual seminar series, called “Spiritual Care Grand Rounds,” which focuses on spiritual care and health.

2.3. Participants

Participants were health system patients identified from inpatient services and the emergency department beginning on 8/20/2020. In September 2022, the program expanded to include primary care clinics. Eligible participants were IU Health patients aged 50 years or older with one or more chronic illnesses, residing in zip codes identified as having high poverty rates and high risks of SDOH. The program enrolled adults 50 and older because older adults with adverse SDOH have the onset of chronic diseases at a younger age [8]. Exclusion criteria for patients included cognitive impairment severe enough that the patient could not consent to participation, discharge to a long‐term care facility with no plan to move into the community, and lack of a working phone to receive calls from Connectors.

In the outpatient setting, patients were identified as socially isolated or lonely based on subjective assessment by their physician, clinic social worker, or other clinician. Clinicians could either make a direct referral to the CCN program or request that the clinic chaplain introduce the program to the patient. For inpatient settings, the electronic medical record (EMR) produces a census of patients who met the CCN criteria. A referral was then made to the unit chaplain or to the CCN chaplain, and the patient was provided with a brochure about the CCN program. Interested patients signed a release of information form, allowing the CCN team or the congregation to contact them directly.

2.4. The CCN Intervention

First, a CCN staff member contacted each patient to conduct an intake. The patient was then assigned to a congregation, which paired the patient with a CCN Connector (Figure 2). Patients were assigned to a congregation based on geography and on whether Connectors had openings to accept new patients. Assignments were not faith‐based. Congregations had to be willing to serve patients of any faith in order to be part of the CCN program. If a patient requested a congregation of a particular faith, the program tried to honor that request.

FIGURE 2.

Timeline of program enrollment and participation. Abbreviation CCN, Congregational Care Network.

The intervention consisted of approximately 1 h of conversation each week between the Connector and the patient, continuing for a duration of 12 weeks. The goals of the conversations were to: (1) provide support and encouragement for the patient as they navigate their experience with illness; (2) help the patient explore and build a broader network of group or community support; and (3) connect the patient with community or IU health resources to address social impediments to health. While the community support may stem from activities and programs offered by the congregation, it often involved reconnecting patients with family, re‐establishing ties with their community, or encouraging them to pursue vocational or avocational interests they may have neglected. Due to the COVID‐19 pandemic, the original plan for in‐person weekly visits was adjusted, and telephone calls became the primary means of communication between the Connector and patient.

2.5. Data Collection

Upon completion of the 12‐week program, we extracted demographic and healthcare utilization information from the EHR, including information about the visit type and date of clinical visits to any IU Health location. Clinical visits were categorized into the following types: emergency department visits, inpatient stays, observation stays, outpatient in‐aa‐bed visits, and outpatient clinic visits. Information regarding any deaths that occurred during the study observation period was also collected.

The abbreviated De Jong Gierveld Loneliness Scale was administered to assess levels of loneliness and social isolation. This validated six‐item survey includes two subscales: a three‐item scale for social isolation and a three‐item scale for loneliness [17] The scale was administered by a LCSW or a chaplain via telephone during the intake interview and post‐intervention, 12 weeks after the patient's placement with a congregation.

The present analysis includes all participants in the CCN group who were referred to a congregation by 2/1/23.

2.6. Data Analysis

Healthcare utilization data were grouped into the following timeframes: 90 days before enrollment, the period during the CCN program, defined as the 90 days after congregational placement, and 90 days after the program (Days 91–180 after congregational placement; Figure 2). Congregational placement was used to define the period during CCN participation, as the date of first visit with a Connector was not always reported to CCN staff. Scale total scores and subscale scores (pre‐ and post‐intervention) and median number of clinical visits for each visit type were compared between the pre‐program period (Before) and both the program (During) and post‐program (After) periods using Wilcoxon signed‐rank tests, due to the paired aspect and the skewed nature of the data. Chi‐Square tests were also performed, using McNemar's test to account for the participant‐paired aspect of the data on the same participants at different time points. We report both median scores and the proportion of patients with 1 or more visits in the tables but report the latter analysis in the text due to its interpretability, as many median scores were zero. All analytic assumptions were verified. Analyses were performed using SAS v9.4 (SAS Institute, Cary, NC).

3. Results

3.1. Participants

We identified 335 participants who had completed the CCN program and had complete data for all three observation phases (Table 1). CCN participants had a median age of 64.9 years and were 72.8% female (Table 1); 58.8% were Black and 40.6% were white; 1.8% were Hispanic or Latino. A total of 7 (2.1%) CCN participants died during the program observation periods. These participants were removed from further analysis since they did not contribute data to all three observation periods, leaving 328 participants.

TABLE 1.

Characteristics of the 335 congregational care network participants.

	CCN participants number (percent)
Age
< 50	23 (6.9)
50–54	38 (11.3)
55–59	59 (17.6)
60–64	52 (15.5)
65–69	64 (19.1)
70–74	38 (11.3)
75–79	26 (7.8)
80–84	17 (5.1)
85–89	9 (2.7)
90+	9 (2.7)
Gender
Female	244 (72.8)
Male	91 (27.2)
Race
Asian	1 (0.3)
Black	197 (58.8)
NH/PI	0 (0)
Other	1 (0.3)
Unknown	0 (0)
White	136 (40.6)
Ethnicity
Hispanic/latino	6 (1.8)
Not Hispanic/latino	328 (97.9)
Unknown	1 (0.3)
Source of referral
Emergency department	18 (5.4)
Inpatient	101 (30.2)
Observation	16 (4.8)
Outpatient	200 (59.7)
Total died	7 (2.1%)

3.2. Loneliness

There were significant reductions in total DeJong Gierveld loneliness scores from before to during and after program engagement (median change score: 1 (interquartile range (IQR) 0–2, p < 0.001)). Results were similar for the Emotional Loneliness subscale (1, IQR 0–1) and Social Loneliness (0, IQR 0–1) subscales (Figure 3).

FIGURE 3.

DeJong Gierveld Loneliness Scale scores before and after the intervention for the 317 CCN participants who both completed surveys. The X axis shows the score for each subscale (Panels A and B) and the total score (Panel C). The Y axis shows the number of participants with each score.

3.3. Healthcare Utilization

The proportion of patients with 1+ emergency department visits did not differ during the CCN program compared to before but was significantly lower after CCN compared to before (16.8% vs. 24.6%, p = 0.007; Table 2). The proportion of patients with inpatient visits was lower during CCN than before (12.2% vs. 17.3% vs. p = 0.032) but not after. The proportion of patients with observation visits after CCN was lower than before (4.0% vs. 6.4%, p = 0.029) and the proportion with outpatient visits was higher during CCN than before (71.0% vs. 63.8%, p = 0.045).

TABLE 2.

Healthcare Utilization for CCN Participants, by observation period, removing those who died during any observation period (n = 328 participants for each time period).

	Timeline				Before versus during	Before versus after
	Before	Enrollment visit	During	After
Number of visits (Encounter type)
Emergency
Total visits	142	17	101	96
Median (IQR*)	0 (0–1)	0 (0–0)	0 (0–0)	0 (0–0)	0.009	0.004
n (%) with 1+ visits	81 (24.6)	16 (4.9)	62 (18.9)	55 (16.8)	0.054	0.007
Inpatient
Total visits	77	84	58	65
Median (IQR*)	0 (0–0)	0 (0–0)	0 (0–0)	0 (0–0)	0.106	0.364
n (%) with 1+ visits	57 (17.3)	82 (24.9)	40 (12.2)	40 (12.2)	0.032	0.056
Observation visits
Total visits	33	21	27	14
Median (IQR*)	0 (0–0)	0 (0–0)	0 (0–0)	0 (0–0)	0.445	0.008
n (%) with 1+ visits	25 (7.6)	21 (6.4)	22 (6.7)	13 (4.0)	0.602	0.029
Outpatient in a bed
Total visits	25	2	38	23
Median (IQR*)	0 (0–0)	0 (0–0)	0 (0–0)	0 (0–0)	0.102	0.785
n (%) with 1+ visits	22 (6.7)	2 (0.6)	30 (9.2)	21 (6.4)	0.194	0.869
Outpatient/clinic
Total visits	612	68	690	537
Median (IQR*)	1 (0–2)	0 (0–0)	1 (0–3)	1 (0–2)	0.029	0.033
n (%) with 1+ visits	210 (63.8)	62 (18.8)	233 (71.0)	208 (63.6)	0.045	> 0.999

Note: The period before the Congregational Care Network (CCN) Program during is defined as the 90 days before the clinical visit during which enrollment took place; during is the 90 days after congregational placement, and after the CCN program is 91–180 days after congregational placement. Values are: Top row = total overall number of visits per group per time period; middle row = median (Intraquartile Range (IQR)); bottom row = number (percent) with one or more visits.

^*Intraquartile Range.

4. Discussion

The Congregational Care Network is a feasible program that engaged 28 congregations in two Indiana cities to successfully provide support to 335 patients. Furthermore, the program evaluation demonstrated improvements in healthcare utilization patterns and reductions in emotional and social loneliness among participants. We believe that several factors contributed to the success of the CCN program. In outpatient settings, referrals for the program were typically made by the patient's physician or the clinic's LCSW or chaplain. This “warm hand‐off” approach proved to be highly effective, as it built on existing trust between the patient and care team. This trusted connection was then passed on to the patient's congregational volunteer, fostering a seamless support network. Congregations participating in the program agreed to serve patients of any faith or no faith. Connector training included following standards common to chaplain practice in multifaith settings, in which proselytizing about one's own faith or converting patients is a violation of professional standards [18].

Health care institutions seek to be trusted sources of information and care, a concern that is complicated by the history of research and medical practices that have abused the trust of communities of color and persons of limited financial means. By partnering with congregations that have earned the trust of their communities, healthcare systems can address these trust deficits in a meaningful and powerful way. However, for such a partnership to be successful, healthcare systems must earn the trust of the community by valuing the congregation as a true partner, an expert in spiritual wellness, and a vital part of the community's health ecosystem.

This approach has additional merit because clinics and hospitals are often where SDOH are first identified. Navigating both the healthcare system and community resources can be challenging for older adults, especially securing essential resources like transportation, nutritious food, shelter, and medication assistance. Having a trusted relationship with a volunteer from the same neighborhood, someone who is a peer or a companion, can be an effective strategy to assist in this process.

Importantly, communication flows in both directions in this model. The chaplain and the CCN partner act as extensions of the physician, offering continuity of care beyond the clinical setting. The CCN partner, in turn, has a direct line of communication back to the healthcare team through the chaplain, which we believe increases the trust of the community in the healthcare system.

The program evaluation demonstrated preliminary evidence in reported social and emotional loneliness among CCN participants. Older adults with chronic illness are particularly vulnerable to these challenges, which healthcare systems are not always equipped to address [4]. Partnering with congregations and community‐based organizations that have an interest or mission to provide companionship provides an effective avenue to address these psychosocial needs.

One of the most notable findings was the reduction in emergency department (ED) use for CCN participants after the program, compared to the period before their enrollment. A prior systematic review found that of several intervention types, only case management in the ICU reduced ED visits [19]. CCN has the potential to reduce ED utilization at potentially lower cost to health systems.

Additionally, there was an increase in the number of outpatient visits, suggesting that patients were attending regularly scheduled visits rather than seeking acute care. For older adults, frequent ED visits and hospitalizations are both physically and emotionally taxing, and lead to functional decline, nursing home admission, and high mortality [20]. By decreasing ED visits and increasing outpatient visits, the CCN has demonstrated the ability to promote more consistent, proactive management of chronic conditions.

Since this evaluation, the CCN has continued to expand. The program now includes thirty‐nine congregations in four metropolitan areas. The CCN continues to operate via telephone, with plans to expand to in‐person meetings in 2025. Although the program currently relies on telephone calls, the flexibility of this approach allows for adaptation to individual patient needs. Some patients may benefit from one lengthy conversation per week, while others may prefer several briefer conversations spaced throughout the week.

The CCN program and our evaluation have several limitations. Participants and faith communities were primarily Christian and primarily Protestant, limiting generalizability. The intervention was only delivered in English. We learned that two congregations, for whom English was not a first or predominant language—one Muslim and predominantly African immigrants, the other Pentecostal Christian and predominantly Hispanic/Latino/a—were not as successful in connecting with patients who were predominantly English speaking. We note this population is young relative to many geriatrics studies and provides information about adults with a broad age range. However, over half were 65 and older. The recruitment approaches in the inpatient (EMR based approach) and outpatient setting (clinician referral) were slightly different to accommodate the needs and preferences of the clinicians in each setting. This may have led to differences in the type of patients referred from each setting. Finally, social isolation was measured indirectly through the De Jong Gierveld subscale as the perception of missing a social network rather than objectively measuring network size [17].

In conclusion, a health system/congregational partnership could be feasibly delivered in partnership with multiple congregations and was able to improve patterns of healthcare utilization and reduce loneliness for at‐risk, community‐dwelling older adults. The program continues to expand and has the potential to serve a broader population of patients including those with limited English proficiency and patients in rural areas.

Author Contributions

Concept and design: Torke, Foster, Kamwendo. Acquisition, analysis, or interpretation of data: Foster, Torke, Willis, Kamwendo, Slaven, Cheng, Berringer, Howard, Session. Drafting of the manuscript: Foster, Torke. Critical revision of the manuscript: All authors. Statistical analysis: Slaven. Obtaining funding: Foster, Torke, Kamwendo. Administrative, technical or material support: Kamwendo Slaven, Berringer. Supervision: Foster, Torke, Kamwendo.

Conflicts of Interest

The authors declare no conflicts of interest.

Foster J. D., Torke A. M., Willis D. R., et al., “The Congregational Care Network: Preliminary Data From a Healthcare/Congregational Partnership for At‐Risk Older Adults,” Journal of the American Geriatrics Society 73, no. 9 (2025): 2878–2885, 10.1111/jgs.19493.