Pillars of Data Equity

Examine the structural forces that affect data production

Discover how social, institutional, and political forces influence data quality and representation.

The Social Determinants of Data (SDOD) framework examines the structural forces that shape the quality, completeness, and reliability of public health data. By looking upstream at the conditions of data production, we can identify why certain communities are rendered invisible in our public records.

The Pillars of Data Equity represent the key domains through which the Social Determinants of Data operate—and where interventions can improve data quality, completeness, and equity.

Six Pillars of Data Equity

01

Governance

Governance refers to the institutional circuitry—the distributed system of actors (physicians, hospitals, funeral directors, and registrars) involved in the data lifecycle.

  • The Challenge: When responsibility is diffuse and no single institution is held accountable for demographic completeness, data quality suffers.
  • SDOD Perspective: Effective governance requires clear feedback loops and designated accountability for ensuring every record is complete.

02

Policy

Policy dictates the rules and mandates that govern data collection practices.

  • The Challenge: In many systems, ethnicity is treated as a “soft field”—an optional piece of information that can be bypassed without consequences.
  • SDOD Perspective: Data justice requires policy reforms that treat demographic data as essential to the public health mission, mandating its collection as a matter of justice rather than peripheral process.

03

Funding

Funding represents the economic investment in data infrastructure and the people who maintain it.

  • The Challenge: Underinvestment in rural areas or marginalized districts leads to staffing shortages and technological gaps that result in high rates of missing data.
  • SDOD Perspective: Data is a public good. Equitable funding ensures that all regions have the resources to capture high-quality data, preventing geographic “blind spots” in public health.

04

Technology (Infrastructure & Data Standards)

Technology Infrastructure & Data Standards focuses on the design of the digital tools, such as Electronic Death Registration Systems (EDRS), used to capture information.

  • The Challenge: System logic often prioritizes bureaucratic closure (like Social Security numbers) over social equity (like ethnicity), allowing “unknown” entries to pass through unchecked.
  • SDOD Perspective: Technological systems must be re-engineered with built-in audits and logic that prevent the normalization of “missingness”.

05

Community

Community addresses the social and political context–such as language barriers and mistrust–in which data is requested.

  • The Challenge: Fear of surveillance or historical mistreatment can lead to a “paradox of visibility,” where marginalized groups are hyper-surveilled for control but rendered invisible for care.
  • SDOD Perspective: Achieving data equity requires building trust through linguistic inclusion, community data liaisons, and transparent communication about how data is used to support community health.

06

Practices

Practice centers on the frontline workforce–the individuals who collect, verify, and enter data.

  • The Challenge: Data workers are often operating under extreme pressure without formal training on cultural responsiveness or the importance of demographic accuracy.
  • SDOD Perspective: We must treat data workers as a vital part of the public health workforce, providing them with the tools and education needed to conduct respectful, trauma-informed data collection.

How These Pillars Interact

These pillars operationalize the SDOD and do not operate in isolation. For example, a Technological fix (making a field required) will fail without Policy support (mandating the field) or Practice training (teach staff how to ask for the information). The SDOD framework provides a holistic lens to diagnose and repair these broken circuits. Rather than mapping to individual steps, these pillars act as cross-cutting forces that shape the entire data pipeline.