A sweeping assessment of environment and dementia
Amid the increasing prevalence of Alzheimer's disease and related dementias (ADRD), a research team led by Harvard Medical School and the Harvard T.H. Chan School of Public Health has completed what it describes as the most extensive systematic evaluation to date of associations between these conditions and nearly 200 environmental exposures. The list reaches from air pollution and noise to diet, medications, social connections and access to green space.
The group also developed a first-of-its-kind exposomic risk score, a tool intended to help estimate a person's risk of these diseases and, in turn, contribute to prevention strategies. The findings were published Sept. 29 in the journal Alzheimer's & Dementia.
Looking beyond genetics
The study's co-senior authors are Chirag Patel, associate professor of biomedical informatics in the Blavatnik Institute at HMS, and Peng Gao, assistant professor of environmental health and exposomics at Harvard Chan School. Both researchers focus on the exposome — the full set of environmental exposures a person encounters across a lifespan — and on how those exposures shape health outcomes.
Genetics are involved in ADRD, but the authors emphasize that most cases cannot be attributed to genetics alone. Previous research has pointed to environmental exposures as another significant contributor, which is why the team set out to assess a broad range of them together rather than studying a single pollutant, behavior or circumstance in isolation.
How the analysis was conducted
The work drew on the UK Biobank, a database that collects health, genetic and lifestyle data from participants over time. The analysis covered nearly 500,000 participants who were followed for a median of 15.3 years.
Using an approach known as an exposome-wide association study, the researchers examined 193 exposures simultaneously, organizing them into five domains. That structure allowed the team to look at how chemical, physical, behavioral, social and ecological factors overlap in the lives of the same people.
The five domains of exposure
- Chemical: includes air pollutants and medications.
- Physical: includes noise.
- Lifestyle: includes diet, physical activity and sleep.
- Social: includes education and socioeconomic status.
- Ecosystems: includes access to green space.
Evaluating these multiple, overlapping exposures at the same time may help researchers identify which factors matter most, and how they interact, in ways that single-exposure studies cannot. A person does not experience polluted air separately from their neighborhood's tree cover, their sleep quality or their income — the study's design acknowledges that reality.
A new exposomic risk score
Beyond cataloguing associations, the study produced an exposomic risk score, which the authors describe as the first of its kind. Such a score aggregates information about a person's environmental exposures to help predict their likelihood of developing ADRD.
If the approach holds up in further research, it could give clinicians and public health officials a way to flag people at elevated risk before symptoms emerge — a step that matters because prevention efforts generally work best earlier in the disease process. The authors position the score as a contribution to prevention strategy rather than a finished clinical tool.
Association is not causation
The researchers are careful to note that an association with ADRD does not necessarily mean a given factor causes the disease. The distinction matters when interpreting a screening effort of this scale.
A clear example comes from medication use. The study found that use of multiple medications was associated with higher ADRD risk. The authors say that finding likely reflects underlying health complexity — people taking several drugs are often managing multiple conditions — rather than the direct effects of the drugs themselves. In other words, the exposure may be a marker of something else happening in the body rather than a driver of dementia.
That caution applies across the board. An exposure that travels alongside dementia risk in the data may be a cause, a consequence, a confounder or a signal of a third factor entirely. Untangling those possibilities requires follow-up work that observational screening alone cannot provide.
What it means for prevention
Even with those caveats, the study's breadth offers a practical benefit. By laying out 193 candidate exposures side by side, the analysis gives researchers a prioritized map of where to look next as they try to understand what drives ADRD in people who do not carry a clear genetic explanation for their disease.
Several of the domains in the study point toward factors that are modifiable at the population level. Air quality, noise levels, green space and socioeconomic conditions are shaped by policy and urban design, not just individual choice. Diet, physical activity and sleep sit at the intersection of personal behavior and the environments that make certain behaviors easier or harder.
The exposomic risk score, meanwhile, points toward a future in which risk assessment accounts for the cumulative environment a person has lived in, not only their genes or a handful of isolated habits.
Reading the findings with care
This is an observational study built on an existing biobank cohort, and its results describe patterns rather than proven mechanisms. The UK Biobank population is not a perfect mirror of the global population, and the median follow-up of 15.3 years, while substantial, cannot capture an entire lifetime of exposure.
What the study does offer is scale and structure: a systematic sweep across hundreds of thousands of participants, a framework for organizing exposures into five domains, and a new scoring approach that other researchers can test and refine. The authors describe these as a foundation for prevention research rather than a final answer.
The bottom line
Alzheimer's disease and related dementias remain conditions that genetics alone cannot explain in most cases. This Harvard-led effort adds a broad, systematic accounting of the environmental side of the equation — nearly 200 exposures, organized into chemical, physical, lifestyle, social and ecosystem domains — and introduces a first-of-its-kind exposomic risk score aimed at prediction and prevention. The headline caution stands: an association in this dataset is not proof of cause, and the polypharmacy finding illustrates how easily a signal can reflect something other than a direct effect. Still, the study gives the field a wider lens and a new instrument for looking through it.
This article is based on reporting by Medical Xpress. Read the original article.
Originally published on medicalxpress.com
