Briya AIRE Study of 32 Years of U.S. Data Reveals Worsening Self-Reported Health in Midlife but Improvement at Older Ages

Presented at GSA 2026, research by Briya, NYU and Johns Hopkins used AIRE to harmonize 6.4 million U.S. health survey responses, revealing opposing age-related trends obscured by the overall average

NEW YORK, Nov. 5, 2026 /PRNewswire/ -- Briya, developer of AIRE, the scientific AI platform turning fragmented healthcare data into defensible evidence, today announced research showing how Briya AIRE helps researchers overcome one of the most persistent obstacles in population health research: data collected under different rules over long periods of time. The findings were presented at the Gerontological Society of America (GSA) 2026 Annual Scientific Meeting in National Harbor, Maryland.

For the study, "AI-Assisted Harmonization of 32 Years of BRFSS Data Reveals Divergent Age-Specific Health Trends in the United States," researchers used AIRE to turn 32 years of changing Behavioral Risk Factor Surveillance System (BRFSS) data into a single, analysis-ready dataset. This made it possible to answer a key question: How has self-rated health changed across age groups over three decades?

The problem extends well beyond any one survey. Whether evidence spans survey waves, health systems or registries, the rules for how data is collected, coded and weighted shift over time. Across BRFSS's history, question wording, response coding, age variables, sampling frames and weighting methods have all changed. Reconciling those changes is slow, specialized work that must happen before analysis can begin, and it determines whether an apparent trend is real or an artifact of how the data was collected. For many research teams, that operational burden can limit the research questions teams are able to pursue.

Key highlights

• The challenge: Long-running health datasets change how data is collected, coded and weighted over time. Reconciling those changes is slow, specialized work that must happen before analysis can begin.
• The scale: Briya AIRE harmonized 6,447,646 survey responses from U.S. adults collected through BRFSS between 1993 and 2024, reconciling changes in question wording, response coding, age variables, sampling frames and weighting methods.
• The finding: Between 1993 and 2024, fair or poor self-rated health rose from 13.8% to 21.6% among adults aged 50 to 54 and fell from 36.2% to 26.1% among adults 80 and older. The overall rate of 23.3% did not capture this divergence.
• Why it matters: Reducing the friction of harmonizing disparate data gives research teams capacity for more complex, longer-horizon studies.

The team applied survey-weighted analyses and conducted sensitivity analyses around major methodological changes, including the 2011 transition to cell-phone sampling and raking.

The analysis also found prevalence falling from 29.9% to 22.7% among adults aged 70 to 74. The age-prevalence curves crossed near age 64, suggesting morbidity is expanding in midlife while being postponed at older ages. The findings reflect combined period and cohort changes, rather than changes in individual health span.

"The headline number hides two opposite stories: Americans over 70 report better health than a generation ago, while people in their early 50s report worse," said Dr. Or Shaked, Executive Director, Real-World Evidence and Medical Research at Briya. "If poorer health is arriving earlier in life, policy, prevention and clinical development built around age 65 may need to start a decade sooner."

"Without AIRE, reconciling 32 years of changing survey data would have required a substantial amount of manual work before the analysis could even begin," said Professor Jordan Weiss of NYU Grossman School of Medicine. "By dramatically reducing that burden, Briya gives researchers the ability to work across larger, more complex datasets and ask questions that would otherwise be far more difficult and time-consuming to pursue."

Reducing the time spent compiling and reconciling disparate data can give research teams greater capacity to take on more complex studies: longer time spans, more disparate sources, finer subgroups.

"The hardest part of ambitious research often isn't the analysis. It's the operational work of getting disparate data into a form you can trust," said Jason Hein, Chief Commercial Officer at Briya. "AIRE takes that friction off researchers' plates while keeping them in control of every scientific decision. That's what lets teams take on more complex studies and go deeper on the questions that matter."

The research was conducted by Dr. Or Shaked, Dr. April Chan-Rubinstein and Ayelet Basson of Briya; Prof. Silvia Koton of Johns Hopkins Bloomberg School of Public Health and Tel Aviv University; and Prof. Shoshana Ballew, Prof. Josef Coresh and Prof. Jordan Weiss of NYU Grossman School of Medicine. The abstract will be published in Innovation in Aging. The project was not funded.

About Briya

Briya is the company behind AIRE, the scientific AI platform that serves as the missing piece between fragmented healthcare data and defensible evidence. Purpose-built for health data and regulated research, AIRE unifies institutional, client, Briya, and third-party data into a single research environment, supported by a global real-world data network covering 120 million patient journeys across four continents. By bringing secure AI with advanced NLP and no-code analytics into research workflows, backed by full compliance, traceability, and human oversight, AIRE enables research service firms, life sciences teams, and healthcare institutions to turn underused clinical data into transparent, reproducible evidence. The result: researchers spend less time on data preparation and operational complexity, leaving more room for science.

Visit www.briya.com and follow Briya on LinkedIn.

Media Contact:
Ellie Hanson
FINN Partners for Briya
[email protected]

SOURCE Briya