Community Health Worker-based Chronic Care Management Program

Evidence-level

Effective (About evidence-level criteria)

Description

Americans who live in the rural areas of Appalachia have long-experienced health challenges influenced not just by healthcare access and geospatial factors, but by factors that also include many of the social determinants of health, such as socioeconomic status and access to education. These regions have high prevalence rates of certain medical conditions and are often referred to as the "diabetes belt" or the similarly described "stroke belt." To address these health conditions and many others, public-private partnerships have been exploring new healthcare delivery models. One such program is the community health worker-based Chronic Care Management (CCM) program.

When a 2012-2015 public-private grant activity — based at the Williamson Health and Wellness Center in Mingo County, West Virginia — demonstrated positive results from its CCM model, participating organizations recognized the potential value of using the outcomes associated with its fee-for-service structure to create a pay-for-performance model that could financially sustain the model's use. This interest prompted scaling of the effort to make sure that similar results could be seen with a larger program enrollment in other areas of Appalachia.

To expand the program outside Mingo County, philanthropy organizations served as convenors and enlisted 11 Federally Qualified Health Centers (FQHCs) and 3 rural hospitals in Ohio, Kentucky, and West Virginia. Lead organizations also envisioned the benefit of parallel involvement of other interested organizations in observing or participating in the program's quarterly meetings: insurance carriers, health policy experts, and public health research experts. This approach allowed for early recognition of how quality outcomes were found to be associated with significant cost savings.

The program model is based on a care coordination team that consist of an advanced practice provider, a nurse, and the core team member, a community health worker (CHW). The role of the CHW is to work with patients in their homes and in their communities. The team meets weekly to discuss cases and update patients' care plans.

The CHWs are full-time employees of the participating FQHCs. Training and job requirements are dependent on individual state laws and the employee policies of participating healthcare organizations.

CHW caseloads often begin at 25 to 30 patients and assessments are completed during initial weekly home visits. Visit frequency decreases as health conditions stabilize. Because a patient's program enrollment does not necessarily expire as acuity decreases and self-management skills increase, with improvement comes less frequent visits and program maturity allows a CHW's caseload to increase to 40 to 50 patients.

Program enrollment is ordered by healthcare providers who identify patients on the basis of their diagnosis and medical condition severity. Insurance carriers were also allowed to suggest patients for enrollment.

Although the original model's qualifying condition was diabetes — the most common condition of enrollees — the scaled model noted similar program benefits for patients with heart disease and chronic obstructive pulmonary disease.

Support of the original model in Mingo County was provided by several public and private grants. Subsequent scaling efforts involved financial support from multiple organizations, including national foundations, small private and family foundations, and hospital conversion foundations as well other government grants with grant funding usually covering program startup costs. Specific philanthropy organizations included: the Appalachian Regional Commission; the Claude Worthington Benedum Foundation; the Greater Kanawha Valley Foundation. the Merck Foundation's, Bridging the Gap: Reducing Disparities in Diabetes initiative; the Pallottine Foundation of Buckhannon; and the Sisters Health Foundation.

Services offered

Evidence-based practices used to provide services:

• Huddles
• Chronic Care Management programs

CHW-based CCM team consist of a lead advanced practice provider, a nurse, and a CHW who meet for weekly team meetings, or "huddles"

Specific team member roles and services:

Nurse:

• Provides clinical case management elements: medication reconciliation, appointment scheduling, referrals and coordination with primary care provider

CHW:

• Weekly in-home visits involve care plan review, medication adherence review, and update of self-management progress and goals
• During the assessment, discusses other issues impacting health and overall well-being

Results

Original Mingo County single-site results during 12-month interval for cohort of 137 patients:

• Average HbA1C reduction from 10.2% to 8.5% in 12 months
• Emergency room visits decreased by 22%
• Hospitalizations decreased by 30%

Data from expanded 3-state implementation cycle from May 2017 through September 2019:

• The model's strength was demonstrated by a high adoption rate by the participating healthcare organizations in 3 Appalachian states
• Most prevalent condition: diabetes
• Cohort of 729 high-risk patients:
  • 446 had both baseline and 6-month follow-up HbA1c test
  • 63% (282) lowered HbA1c: mean decrease of 2.4 percentage points
  • 96 patients decreased HbA1c below 10%
  • For the cohort of 96 patients who lowered their HbA1c below 10%, using an assumption based on 1 less hospitalization per year, a $384,000 annual cost savings could be realized

• Based on an average of $45,000 per CHW, cost savings found by one carrier neared $5,000 per patient over a 4-month period.

As a result of this analysis, some payers engaged with FQHCs in shared savings arrangements.

The most common issues discovered to be impacting health status were social situations, literacy, and economic barriers. For example:

• Insulin expired because a refrigerator failed. The patient's financial situation precluded replacement, a situation remedied by CHW securing social service agency and faith-based organization funding for replacement.
• Missed opportunity for housing assistance caused by literacy issues, a situation remedied by CHW-assisting patient in completion of a complex form.

An important outcome of the CHW-based program study is providing data and information that has convinced other health centers throughout Appalachia to implement the program and employ CHWs "beyond the scope of the grant funding."

Further results for this program:

Crespo, R., Christiansen, M., Tieman, K., Wittberg, R. 2020. An Emerging Model for Community Health Worker-Based Chronic Care Management for Patients With High Health Care Costs in Rural Appalachia. Prev Chronic Dis. 17, E13.

In this short video, a Clay County, West Virginia CHW shares some positive outcome stories associated with of some of her work:

Replication

Important to consider using these two hiring guidelines for CHWs: community-based residence and ability to communicate and relate to patients with respect and empathy.

• CHWs must come from the local community since, as a model, it fits in the Appalachian culture with its strong kinship networks and traditions of neighbor helping neighbor. It would be unlikely that patients would accept non-community individuals into their homes and allow frank discussions concerning health issues.

Anticipate that strong relationships will develop between the CHWs and their patients and sometimes it is best for the patient to stay permanently enrolled with decreased home visit frequency equitable to the medical condition's severity.

Anticipate that the most intensive CHW training will occur during a shadowing experience with a nurse or experienced CHW on home visits, in addition to weekly continuing education in team huddle participation.

Using health center administration and physician champions will have a positive influence on program model adoption and patient enrollment.

If possible, consider engaging interested philanthropy partners as conveners. They often have key relationships with other change agents such as insurance carrier leadership, health policy experts, and public health researchers, especially for data analysis.

Consider building early relationships with payers. From the beginning of this project, these relationships were critical. The presence of insurance carriers at quarterly meetings contributed to early development of pilot payment models. Several began using their own data to create payment models.

Contact Information

Richard Crespo, PhD Marshall University School of Medicine, Huntington, West Virginia
304.634.6706
crespo@marshall.edu