FQHC data trust gaps that break HRSA UDS reporting
HRSA requires 1,400+ health centers to submit Uniform Data System reports annually, covering clinical quality, demographics, financial performance, and social determinants. Each UDS table depends on data integrity standards that most EHR systems do not enforce natively, creating a trust gap between what gets reported and what actually happened at the point of care.
HRSA's Bureau of Primary Health Care requires every Section 330 grantee to submit a Uniform Data System report each calendar year. That mandate covers approximately 1,400 health centers operating across more than 15,000 service delivery sites, serving over 31 million patients as of 2023 UDS data. Each submission spans dozens of tables. Every table depends on data that originates in an EHR, a billing system, or a manual workflow. The gap between what HRSA requires and what health centers actually produce is where FQHC data trust breaks down.
What UDS reporting actually covers
| | Value (%) | |---|---| | Demographics (3A/3B) | 25 | | Services & Utilization (4/5/6A) | 25 | | Clinical Quality (6B/7) | 25 | | Financial (8A/9D/9E) | 20 | | Staffing & Other | 5 |
The Uniform Data System is not a single report. It is a structured collection of tables that quantify nearly every operational and clinical dimension of a health center's performance. The 2024 UDS Manual specifies reporting across patient demographics (Tables 3A and 3B), staffing and utilization (Tables 5 and 6A), clinical quality measures (Tables 6B and 7), financial data (Tables 8A and 9D through 9E), and social risk factors.
Table 3B alone requires granular breakdowns by race, ethnicity, language, and insurance status for every patient who had at least one visit or service during the reporting year. Table 6B requires clinical quality measure data aligned with HRSA's selected measures, many of which map to UDS clinical benchmarks for hypertension control, diabetes management, depression screening, and cancer screening rates.
Table 4, often overlooked, captures patients by specific service type: medical, dental, mental health, substance use disorder, vision, and enabling services. Getting this table right requires that every encounter is coded to the correct service category, a process that depends on CPT and revenue code accuracy at the point of registration.
The 19 clinical quality measures and their data dependencies
For the 2024 reporting year, HRSA requires health centers to report on clinical quality measures that include hypertension control (blood pressure < 140/90), diabetes HbA1c poor control (> 9%), depression screening and follow-up, BMI screening, tobacco use screening and cessation intervention, colorectal cancer screening, cervical cancer screening, childhood immunization status, HIV screening, and prenatal care measures, among others.
Each measure has a numerator and denominator definition that HRSA specifies down to the eligible population, the exclusion criteria, and the evidence period. Getting the denominator wrong inflates or deflates performance rates. Getting the numerator wrong misrepresents clinical quality.
The denominator problem is severe. A health center that cannot reliably identify which patients had a qualifying visit during the measurement year will miscount the eligible population for every clinical measure. This is not a hypothetical concern. The National Association of Community Health Centers (NACHC) has noted that EHR configuration differences across sites within the same health center can produce inconsistent denominator counts for the same measure.
Where EHR data breaks the submission
Most FQHCs use one of a handful of EHR systems. Epic, eClinicalWorks, athenahealth, and NextGen dominate the FQHC market, but the way each system stores and exports data for UDS reporting differs significantly. Structured fields for race, ethnicity, preferred language, and insurance status are often incomplete or inconsistently populated.
A 2022 OIG report on HRSA health center oversight found that data quality issues in health center reporting were widespread enough to limit HRSA's ability to assess grantee compliance. The report did not quantify an error rate across all tables, but it identified systematic gaps in how health centers captured and validated demographic and clinical data before submission.
The problem compounds at multi-site health centers. A health center operating 12 sites across three counties may have different registration workflows, different EHR configurations, and different staff training standards at each location. When UDS submission rolls up data from all sites into a single report, inconsistencies become invisible at the aggregate level but remain present in the underlying records.
Demographic data quality and the Table 3B problem
Table 3B is the demographic core of every UDS submission. It requires patient counts by age, sex, race, ethnicity, and insurance status. HRSA's reporting categories for race follow OMB standards but also include a "More than one race" option and an "Unreported/Refused to report" category.
The unreported category is where data trust erodes fastest. A health center that shows 25% of patients with unreported race data is not necessarily serving a population that refuses to self-identify. More often, registration staff are not asking the question, the EHR field is buried in a workflow, or the field defaults to blank when skipped.
Insurance status reporting in Table 3B requires classification into Medicaid/CHIP, Medicare, other public, private, and uninsured categories, broken out by sliding fee scale status. A patient whose Medicaid eligibility lapsed mid-year but was not updated in the EHR will be counted in the wrong category. According to KFF analysis of Medicaid enrollment data, over 25 million people were disenrolled from Medicaid during the 2023-2024 unwinding period. FQHCs serving high-Medicaid populations saw insurance status churn at rates that outpaced their ability to update records in real time.
Financial tables and the cost-per-patient calculation
Tables 8A, 9D, and 9E require health centers to report total costs, revenue by source, and costs allocated by service category. These tables feed HRSA's cost-per-patient benchmarks, which influence grant award calculations and operational assessments.
The cost allocation methodology requires health centers to assign expenses to medical, dental, mental health, substance use, and enabling service lines. Many health centers lack the accounting infrastructure to perform precise cost allocation. When a behavioral health provider splits time between two service lines, the allocation often follows a rough percentage estimate rather than a time-tracked calculation.
Financial data integrity in UDS submissions has direct consequences. HRSA uses cost-per-patient data to compare health centers and to flag outliers for further review. A health center that underreports costs in one service line and overreports in another may trigger an audit or lose credibility in future grant applications.
Social determinant screening data and Table 6B
HRSA has steadily expanded expectations for social determinant of health screening data in UDS reporting. The 2023 UDS data showed that health centers reported screening rates for depression, tobacco use, and other behavioral risk factors, but screening for housing instability, food insecurity, and transportation barriers remains inconsistent across the network.
The challenge is not just whether screening happens. It is whether screening results are captured in structured EHR fields that can be extracted for UDS reporting. A clinician who documents a patient's housing instability in a free-text note has performed the screening but has not created a reportable data point. Unless the finding is coded using ICD-10 Z-codes (Z59 for housing, Z59.4 for food insecurity), it will not appear in the UDS submission.
Z-code capture rates remain low across the health system. CMS reported in 2023 that fewer than 2% of Medicare fee-for-service claims included Z-codes for social determinants, and FQHC rates, while likely higher due to HRSA's emphasis on SDOH, still reflect inconsistent capture practices.
The 2025 reporting changes health centers must track
HRSA updates UDS requirements annually. For the 2025 reporting year, HRSA has removed the requirement to report Sexual Orientation and Gender Identity (SOGI) data, reversing a policy direction that had been expanding since 2016. This change means health centers that invested in SOGI data collection workflows now face a reporting landscape where that data is no longer federally required but may still be required by state programs or accreditation bodies.
Additional changes for 2025 include updated clinical quality measure specifications, revised table instructions, and potential modifications to enabling services reporting. Health centers that rely on prior-year report templates without reviewing the current UDS Manual risk submitting data under outdated definitions.
Why standard data quality checks are not enough
HRSA provides a set of built-in validation checks through the Electronic Handbooks (EHBs) submission system. These checks catch obvious errors: negative patient counts, percentages that exceed 100, and tables that do not cross-reference correctly.
But validation checks only catch structural errors. They do not catch substantive data quality failures. A health center can submit a UDS report that passes every EHB validation check and still contain data that misrepresents clinical performance, demographic composition, or financial position.
The difference between passing validation and achieving data trust is the difference between a file that can be uploaded and a file that tells the truth. Validation confirms format. Trust confirms meaning.
What FQHC data trust actually requires
| | Value (%) | |---|---| | Provenance | 25 | | Recency | 15 | | Concordance | 10 | | Quality | 10 | | Consent | 20 | | Validation | 10 | | Breadth | 5 | | Stability | 5 |
Data trust for UDS reporting requires verification across multiple dimensions. Provenance: can the health center trace every number in the submission back to a source record in the EHR or financial system? Recency: does the data reflect the current state of each patient's demographics, insurance status, and clinical conditions as of the reporting period close? Concordance: do the numbers in Table 3B match the numbers implied by Tables 4 and 5? Quality: are clinical measures calculated using the correct denominators and numerators per HRSA's specifications?
These are not aspirational standards. They are the minimum conditions for a UDS submission that can withstand scrutiny during a site visit, an OIG audit, or a Service Area Competition review.
Health centers that treat UDS as a once-a-year data export project will continue to produce reports that pass validation but fail trust. Health centers that build data trust infrastructure, scoring every record for completeness, recency, and provenance before it enters the submission pipeline, will produce reports that reflect what actually happened in their exam rooms.
The downstream cost of low-trust UDS data
UDS data does not stay inside HRSA. It feeds national benchmarks, state-level primary care planning, health equity research, and congressional funding justifications. The HRSA Health Center Program data portal makes UDS data publicly available, meaning every submission becomes a public record of a health center's performance.
When UDS data is wrong, the consequences compound. Researchers who use UDS data to study health center effectiveness draw conclusions from flawed denominators. Policymakers who compare health center performance across states are comparing numbers with different underlying data quality standards. HRSA itself, when it uses UDS data to justify the $2 billion annual appropriation for the Health Center Program referenced in HRSA's FY2024 budget justification, is building its case on data whose integrity varies widely from grantee to grantee.
The fix is not more validation checks. It is scoring the data before it enters the reporting pipeline, at the record level, across every dimension that determines whether a number is trustworthy or just plausible.
The DTI Engine scores every record 0 to 100 across eight dimensions before your AI model or reporting system sees it. For FQHCs preparing UDS submissions, that means provenance verification, recency checks, and concordance scoring on every patient record before it rolls up into a table. If your team is building data infrastructure for UDS reporting or evaluating data quality across multi-site health center operations, talk to the SuperTruth commercial team. Schedule a conversation or call (215) 918-4140.
Further reading:
Sources
Jason Alan Snyder
Co-founder of SuperTruth and Artists & Robots, and an inventor on the Data Trust Index patents. Twenty-plus years building technology inside Interpublic Group. He writes here nearly every day on data trust, provenance, and what AI should be allowed to act on, and publishes essays on his Substack.
About SuperTruth · LinkedIn · Substack · jasonalansnyder.com
See it in practice
DTI scores the record, not the patient.
8 dimensions. 0 to 100. Travels with every record permanently.