Contact email

hrod@berkeley.edu

Overview

• A group of community health centers implemented team-based models of care that integrated a new medical assistant and/or community health worker-leds to aid in diabetes care management program.  The trial examines studies differential outcomes based on the role of the primary care team member leading the intervention integrated onto the primary care team to aid in diabetes care management..

Collaborators

Mark W. Friedberg, MD
Arturo Vargas-Bustamante, PhD, MA, MPP
Xiao Chen, PhD
Ana E. Martinez, MPH
Dylan H. Roby, PhD

Status/Stage of Development

Planning

Funding Sources

The Agency for Healthcare Research and Quality through the American Recovery and Reinvestment Act funded this project. The publication was partially supported by the Berkeley Research Impact Initiative.

Practice Setting

Community Health Centers

National/Policy Context

• Team-based models that include pharmacist-led medication management and registered nurse care management improve outcomes for diabetes management. However,  community health centers often do not have adequate financial resources for these models.  
• Team-based models that include community health workers and medical assistants in patient management are more financially feasible, but their effectiveness at improving diabetes related health outcomes has yet to be studied.

Local/Organizational Context

• The California Primary Care Association was leading a statewide diabetes quality improvement initiative, Accelerating Quality Improvement through Collaboration (AQIC), among community health center members and wanted to advance the testing of interventions to improve diabetes outcomes among socioeconomically vulnerable patients of CHCs.  The researchers and practice partners worked together to prioritize promising interventions that were financially feasible for CHCs to implement.

Patient Population Served and Payor Information

• Community health centers with predominantly low-income, Latino Spanish-speaking  and Chinese patients who were both insured and uninsured. 

Leadership

• The study had an executive committee comprised of the PI (Hector Rodriguez, PhD, MPH), project manager for the study (Ana, Martinez, MPH), Medical Director of the community health center network (Kent Imai, MD), CEO of the network (Dolores Alvarado, MSW), QI director of the network (Elena Alcala, MPH), research data leader (Dylan Roby, PhD), and health disparities researcher (Arturo Vargas-Bustamante, PhD).  The executive committee made decisions about CHC interventions, data submission requirements, and relationships with participating CHC sites. Three subcommittees, a data management subcommittee, a research analysis subcommittee, and an intervention subcommittee involved a broader set of clinicians and researchers, and made decisions and recommendations that were then vetted/approved by the executive committee.

Project Research + Planning

• Community Health Center Randomization
  • 16 community health center sites all located in three Northern California counties and affiliated with the same regional community clinic association were selected for this cluster randomized trial. 
    • The participating CHCs served primarily low-income, Latino and Chinese patients and patients. 
    • All CHC sites studied have bilingual and bicultural staff that can provide care in patients’ language preference.
  • Before this sites were randomized, the research team conducted practice surveys of each site director, adult primary care clinicians, and staff members from each participating community health center.  
    • These survey assessed baseline practice dynamic and organizational factors that could impact the effectiveness of implementing team-based models of care. The site director survey additionally assessed the diabetes care management capabilities of the community health center sites. 
  • Based on the response of the site directors and mean responses to the clinician and staff surveys, health centers were dichotomized by practices into “high” (top eight) vs. “low” (bottom eight) based on the following composite measures: 
    • Diabetes structural capabilities
    • Primary care team functioning
    • Practice size
• Clustered randomization 
  • Using cluster analyses, community health centers were grouped practices into three sampling strata. Each stratum then had community health centers randomly assigned to one of three groups:
    • Medical assistant interventions
    • Community health worker intervention 
    • Control 
• Forming the Care Team
  • In most cases, existing medical assistants were promoted to take new responsibilities for diabetes care management. 
  • All community health centers hired replacements for the medical assistants taking on the project roles. 
  • All intervention personnel received training on motivational interviewing. 
  • Medical assistants and community health workers were trained over three in-person six-hour sessions on health coaching and panel management for diabetes care by expert trainers prior to the intervention.

Tools or Products Developed

• Role-specific diabetes care management protocols 
  • These protocols were used by intervention team personnel for care administration. 
  • The protocols were developed by the participating clinics’ quality improvement teams in conjunction with the regional community clinic association.

Training

• All intervention personnel received training on motivational interviewing. 
• Medical assistants and community health workers were trained over three in-person, six-hour sessions on health coaching and panel management for diabetes care by expert trainers prior to the intervention.
  • The medical assistant training included a brief module on health coaching strategies and emphasized office-based panel management activities for diabetic patients. The training included tasks such as:
    • maintaining the diabetes registry
    • reviewing the panel at regular intervals
    • following up on primary care clinician instructions
    • assisting with medication reconciliation
    • targeting patients for appropriate interventions and referral to community resources
  • The community health workers received a brief, less extensive module on panel management activities. Community health worker training emphasized broad range of patient health coaching skills, including: 
    • helping patients set their agendas for clinicians visits
    • making sure patients understand what their clinicians want them to do
    • determining whether patients agreed with their care plans
    • providing support to patients’ efforts in adopting healthy behavior
    • assisting patients to improve medication understanding and adherence

Tech Involved

• Electronic medical record
• Telephone

Team Members Involved

• Community Health Worker
• MAs
• Physicians
• Support Staff

Workflow Steps

• Implementation 
  • Following training, intervention team personnel followed role-specific diabetes care management protocols.
  • Site administrators and frontline personnel received ongoing technical assistance from a regional community clinic association. This assistance included:
    • Managing and reporting clinical and administrative data central to assessing diabetes care processes and intermediate outcomes
    • reinforcing motivational interviewing skills among medical assistant and community health worker staff
• Patient and Medical Team Experience
  • The early intervention patient experience survey of adult diabetics was taken for two months approximately mid-way through the intervention period. A post-intervention survey was also taken. These surveys asked about care experiences during the prior six months. 
  • Mid-intervention interviews of community health center clinician and staff key informants were conducted. Post-intervention interviews were also conducted. All interviews were audio recorded and transcribed, done by telephone or in-person using an interview guide.

Outcomes

• Clinical Processes and Intermediate Outcomes
  • Community Health Workers
    • Glycated hemoglobin: Patients in the community health worker intervention arm had improved annual glycated hemoglobin testing (18.5% points, p < 0.001), but the differences were comparable to improvements in HbA1c testing among patients in the community health worker control group.
    • LDL-C: Testing for LDL-C improved significantly more over time for the community health worker control group as compared to the community health worker intervention group (9.8% points vs. 0%, p < 0.01).
  • Medical Assistants
    • LDL-C: The medical assistant intervention group patients improved LDL-C control (8.4% points, p < 0.05) over time, but the improvements were no different than patients in the medical assistant control group. 
    • HbA1c: Testing for HbA1c improved more for the medical assistant control group compared to the intervention group (5.8% points vs. 1.0% points,p < 0.05). 
• Patient-Reported Measure Results
  • There were no differential changes in communication and access to care composite scores over time across the study arms. 
  • Self-reported hypoglycemic events did not differentially change over time by study arm.
• Key Informant Interview Results
  • There was a positive influence of data collection activities associated with project participation on improved documentation, outreach, and in-reach efforts for diabetes care management.
  • All new personnel in the intervention sites considered health coaching their most important responsibility.
  • Post-implementation interviews supported that the project personnel were fully integrated into the diabetes care management workflow.

Benefits

• Community health workers and medical assistants identified health coaching as their most important responsibility, highlighting the value of a system which provides space for them to serve this role.
• Diabetes care improved in community health centers which integrated community health works and medical assistants in primary care teams.
  • The improvement could be even greater if practice leaders minimize use of community health workers and medical assistants to cover shortages arising in busy primary care practices. This would allow community health workers and medical assistants more time to focus on aspects of care such as health coaching.
• The process of data collection within an integrated care team improved documentation outreach, and in-reach efforts for diabetes care management.

Intervention-Specific Challenges

• Analysis
  • The study team was unable to control for race/ethnicity in regression models due to issues with collinearity. 
  • The small sample size of the study lead to unbalanced groups in the randomization process. 
  • Generalizability of this study is limited by patient sample attrition and survey non-response.
  • There were notable pre-intervention differences in the overall quality of diabetes care at the different care centers. These differences underlie patient characteristics that were not ascertained before randomization.
• Intervention
  • Due to the time required for regular responsibilities of a medical assistant, the medical assistants at intervention sites did not have enough dedicated time to conduct health coaching activities. Thus, the busy schedule of floor medical assistant duties may interfere with the health coaching, especially when demand is high and staffing is low. 
  • This intervention may require scaling up the number of community health workers and medical assistants in order to most effectively meet the health care needs of patients and improve health outcomes. This would create a financial burden to hospitals, providing a major setback for community health centers with limited resources for which this intervention is intended.

Sources

• Rodriguez, H. P., Friedberg, M. W., Vargas-Bustamante, A., Chen, X., Martinez, A. E., & Roby, D. H. (2018). The impact of integrating medical assistants and community health workers on diabetes care management in community health centers. BMC Health Services Research, 18(1), 875. doi: 10.1186/s12913-018-3710-9