Community Trust as Evidence: The e-DICT™ Model for Converting Grassroots Health Engagement into Actionable Research Data

Community health organizations have long served as the primary point of contact between underrepresented populations and the healthcare system. Yet the data generated through that contact — the conversations, the barriers encountered, the outcomes observed — rarely makes it into the research literature that shapes clinical guidelines, algorithmic training sets, or health policy. This brief describes DaisyMe Foundation's e-DICT™ model (Engagement → Data → Insight → Change → Trust), a structured methodology for converting direct community health engagement into research-grade evidence. Drawing on 12,400+ direct program interactions since 2022, we describe the model's five phases, the data infrastructure required to implement it, and the implications for health systems, clinical AI developers, and policymakers seeking community-sourced evidence on health equity gaps.

Community Trust as Evidence: The e-DICT™ Model

DaisyMe Foundation Policy Brief DMF-2026-001 Published July 2026 · Frisco, TX

Abstract

Community health organizations have long served as the primary point of contact between underrepresented populations and the healthcare system. Yet the data generated through that contact — the conversations, the barriers encountered, the outcomes observed — rarely makes it into the research literature that shapes clinical guidelines, algorithmic training sets, or health policy. This brief describes DaisyMe Foundation's e-DICT™ model (Engagement → Data → Insight → Change → Trust), a structured methodology for converting direct community health engagement into research-grade evidence. Drawing on 12,400+ direct program interactions since 2022, we describe the model's five phases, the data infrastructure required to implement it, and the implications for health systems, clinical AI developers, and policymakers seeking community-sourced evidence on health equity gaps.

1. The Problem: A Broken Evidence Pipeline

The standard pathway for health equity research runs from academic institution to clinical trial to peer-reviewed publication to guideline. Community health organizations — the entities with the deepest, most sustained relationships with underrepresented populations — sit almost entirely outside that pipeline.

The consequences are well-documented:

• Clinical AI training datasets systematically underrepresent Black, Hispanic, and Indigenous populations, producing models that perform worse for the patients who most need accurate diagnosis and triage.
• Health policy is shaped by data collected in clinical settings, which excludes the 30–40% of underrepresented community members who avoid or delay care due to cost, distrust, or prior negative experiences.
• The barriers that prevent care access — transportation, language, provider bias, insurance gaps — are experienced and documented at the community level but rarely captured in formats that reach institutional decision-makers.

The result is a structural evidence gap: the populations most affected by health inequity are the least represented in the research that is supposed to address it.

2. The e-DICT™ Model: Five Phases

DaisyMe Foundation developed the e-DICT™ model to close this gap by treating every community health interaction as a data collection opportunity — and every data point as a potential input to systemic change.

Phase 1: Engagement

Direct, sustained contact with community members through culturally responsive programs: health literacy workshops, care navigation sessions, community health events, and peer health educator networks. DaisyMe's engagement model is built on three principles:

1. Meet people where they are. Programs are delivered at churches, barbershops, community centers, and employers — not at clinical facilities that carry the weight of prior negative experiences.
2. Build trust before collecting data. Community members are not research subjects. They are partners. Data collection begins only after a relationship has been established and informed consent has been obtained.
3. Compensate community knowledge. Community members who contribute data, participate in focus groups, or serve as peer educators are compensated for their time and expertise.

Phase 2: Data

Structured capture of the information generated through engagement. DaisyMe's data infrastructure collects:

• Barrier documentation: What prevented this person from accessing care? (Transportation, cost, language, prior negative experience, distrust, lack of information)
• Outcome tracking: What happened after navigation support? Did they access care? Did they receive the recommended treatment? Did they return?
• System failure documentation: Where did the healthcare system fail this person? What did the provider miss? What did the algorithm get wrong?
• Self-reported health status: Pre- and post-program surveys measuring health literacy, confidence in care decisions, and self-reported health outcomes.

All data is collected with explicit informed consent, stored in compliance with HIPAA, and de-identified before any analysis or reporting.

Phase 3: Insight

Analysis of the collected data to identify patterns, gaps, and systemic failures that are not visible in clinical datasets. DaisyMe's research team analyzes:

• Demographic patterns in care barriers: Which barriers are most common for which populations? How do they vary by geography, age, insurance status?
• System failure patterns: Which clinical AI systems produce the most complaints from community members? Which providers generate the most navigation escalations?
• Outcome disparities: Are post-navigation outcomes different for different demographic groups? What predicts successful care access?

This analysis produces the community-sourced evidence base that DaisyMe translates into research briefs, policy recommendations, and institutional partnership proposals.

Phase 4: Change

Translation of insights into action at three levels:

• Individual: Community members receive the information, navigation support, and advocacy they need to access appropriate care.
• Institutional: Health systems and clinical AI developers receive documented evidence of system failures and specific, actionable recommendations for remediation.
• Systemic: Policymakers and funders receive research briefs and advocacy materials grounded in community-sourced evidence.

Phase 5: Trust

The outcome of the full cycle — and the input to the next cycle of engagement. When community members see that their participation produces real change (a health system updates its triage protocol; a clinical AI vendor corrects a bias in its training data; a policymaker cites DaisyMe's research in a committee hearing), trust in the research process increases. That trust enables deeper engagement, richer data, and more powerful evidence in the next cycle.

3. What the Data Shows: Selected Findings from 12,400+ Interactions

The following findings are drawn from DaisyMe Foundation's program data collected between 2022 and 2026. All figures are based on de-identified participant records and self-reported survey data.

3.1 Care Access Barriers

Among the 12,400+ individuals who engaged with DaisyMe programs, the most commonly documented barriers to care access were:

• Cost and insurance gaps (cited by 67% of participants)
• Prior negative experiences with providers (cited by 54% of participants)
• Transportation (cited by 41% of participants)
• Language barriers (cited by 28% of participants)
• Distrust of the healthcare system (cited by 23% of participants)

Notably, distrust was significantly higher among participants who had previously interacted with clinical AI-assisted triage or diagnostic tools and felt their concerns had been dismissed or misclassified.

3.2 Navigation Outcomes

Of the 4,200+ individuals connected to primary care through DaisyMe's navigation programs:

• 79% maintained follow-up care compliance at 90 days
• 1,800+ uninsured participants were enrolled in coverage
• Average time to first appointment: under 7 days (compared to a national average of 24 days for uninsured patients)

3.3 Health Literacy Impact

Among participants who completed DaisyMe's health literacy workshops:

• 91% reported increased confidence in their ability to make care decisions
• 83% completed chronic disease self-management modules
• 87% reported improved health access at 6-month follow-up

3.4 Clinical Research Inclusion

DaisyMe's clinical research inclusion program has enrolled 2,100+ underrepresented participants in clinical trials through 14 partner research institutions, representing a 34% year-over-year increase in BIPOC trial enrollment among partner institutions.

4. Implications for Health Systems and Clinical AI Developers

The e-DICT™ model has direct implications for organizations seeking to address health equity gaps in their clinical operations and algorithmic systems.

4.1 Community Data as a Training Data Source

Clinical AI systems trained exclusively on electronic health record (EHR) data inherit the biases of the clinical system that generated that data — including the systematic underrepresentation of populations who avoid or delay care. Community-sourced data, collected through trusted relationships with underrepresented populations, can supplement EHR data to produce more representative training sets.

DaisyMe Foundation is available to partner with clinical AI developers and research institutions seeking community-sourced data for training set augmentation, bias auditing, and equity validation.

4.2 Navigation Data as a System Failure Signal

When a community member requires navigation support to access care that should have been straightforward, that navigation event is a signal of system failure. DaisyMe's barrier documentation data identifies specific failure points in health systems — the providers who generate the most escalations, the triage protocols that produce the most complaints, the insurance processes that create the most barriers.

Health systems that partner with DaisyMe gain access to this failure signal data as a quality improvement input.

4.3 Trust as Infrastructure

The most expensive problem in health equity research is not data collection — it is trust. Building the relationships required to collect honest, representative data from underrepresented communities takes years. DaisyMe Foundation has built that infrastructure across the Dallas–Fort Worth Metroplex, Houston, and San Antonio. Partner organizations gain access to that trust infrastructure without having to build it from scratch.

5. Policy Recommendations

Based on the evidence generated through the e-DICT™ model, DaisyMe Foundation recommends:

1. Fund community health organizations as research infrastructure. Federal and state health equity funding should explicitly support community-based organizations that collect, analyze, and translate community health data — not just clinical research institutions.

2. Require community-sourced data in clinical AI training sets. The FDA's guidance on AI/ML-based software as a medical device should include explicit requirements for demographic representation in training data, with community-sourced data recognized as a valid supplementary source.

3. Mandate navigation outcome reporting. Health systems that receive federal funding should be required to report navigation escalation rates by demographic group — creating accountability for the system failures that community navigators are currently absorbing invisibly.

4. Compensate community knowledge. Research funding mechanisms should include explicit line items for community participant compensation, recognizing that community members who contribute data are performing research labor.

6. Conclusion

The evidence gap in health equity research is not a data problem. It is a trust problem, a relationship problem, and a structural problem. The data exists — it is generated every day in the conversations between community health navigators and the people they serve. What has been missing is the infrastructure to capture it, analyze it, and translate it into the formats that reach institutional decision-makers.

The e-DICT™ model is DaisyMe Foundation's answer to that gap. It is not a theoretical framework. It is a working methodology, tested across 12,400+ direct community interactions, that produces research-grade evidence from community-level engagement.

We invite health systems, clinical AI developers, research institutions, and policymakers to engage with this evidence — and with the communities that generated it.

About DaisyMe Foundation

DaisyMe Foundation is a 501(c)(3) nonprofit organization headquartered in Frisco, Texas. We advance health equity for underrepresented communities through direct service programs, community health research, and systemic advocacy. Our programs span health literacy education, care navigation, clinical research inclusion, and policy engagement. Since 2022, we have served 12,400+ community members across the Dallas–Fort Worth Metroplex, Houston, and San Antonio.

Contact: info@daisyme.org · daisyme.org

© 2026 DaisyMe Foundation. This brief may be reproduced for educational and advocacy purposes with attribution. All program data is drawn from de-identified participant records collected with informed consent under DaisyMe Foundation's IRB-equivalent community research protocols.