Context Matters When Determining “High” Chemical Concentrations – May 2025

When is a chemical level truly “high”?
It’s one of the most common—and most misunderstood—questions in environmental science and health assessments. The answer isn’t as simple as checking against a single guideline or comparing to an average. Without context, even accurate measurements can lead to misleading conclusions. For example, in the Alberta oil sands region, elevated polycyclic aromatic hydrocarbon (PAH) levels in local waterways have raised concern. However, detailed forensic and geochemical studies have shown that natural erosion of bitumen-rich formations—especially during rain events—can release significant amounts of PAHs into surface waters. In such cases, measured concentrations may be high, but their origin is natural rather than industrial. Without understanding the source context, these findings could wrongly be attributed to anthropogenic contamination, potentially leading to unnecessary mitigation efforts or public concern. Similarly, in human biomonitoring, detecting chemicals like PFAS or phthalates in blood doesn’t automatically mean those levels are harmful. Everyone carries trace levels of environmental chemicals, and it’s the concentration, duration of exposure, and toxicological context that define risk—not mere presence. The ‘dose makes the poison’ as they say.

In both biomonitoring and environmental contamination studies, understanding what qualifies as “elevated” means looking beyond raw numbers. Whether you’re measuring chemicals in blood, urine, or soil, the key is comparison—against a well-defined reference population or natural background levels.

In this blog, we explain how statistical tools like percentiles and quantiles help bring clarity to environmental assessments to more accurately determine whether a level is truly above normal. We help regulators, environmental consultants, public health professionals, and community stakeholders who need to make informed decisions about chemical levels in the environment or in people. Whether you’re evaluating sediment data near an industrial site or interpreting biomonitoring results, we show how tools like percentiles and quantiles provide the statistical context needed to distinguish normal variation from actual concern—helping you avoid false alarms or missed risks.

Using NHANES to Define What “High” Means in Biomonitoring

In the case of biomonitoring, chemicals are measured in biological samples like blood or urine to determine exposure levels. A key resource for establishing baseline levels is the National Health and Nutrition Examination Survey (NHANES), an ongoing Centers for Disease Control (CDC) study that collects health and exposure data from thousands of individuals across the U.S.

For example, if polychlorinated biphenyl (PCB) levels in a sample appear elevated, NHANES data can help determine how those levels compare to the broader U.S. population. Rather than relying on a single threshold, NHANES allows comparisons across a distribution, using statistical measures such as percentiles.

Understanding Quantiles, Quartiles, and Percentiles

Statistical tools like quantiles, quartiles, and percentiles help break down a dataset into meaningful segments:

Quantiles divide data into equal-sized groups based on concentration values.

Quartiles split data into four equal parts, referred to as the 1^st, 2^nd, 3^rd, and 4^th quartiles.

Percentiles categorize data based on how much falls below a specific threshold (e.g., the 95^th percentile means 95% of data points are lower than that value).

For PCB data in NHANES, percentiles allow researchers to determine whether an individual’s exposure falls within a normal range or is unusually high. If a sample’s PCB concentration is above the 95^th percentile, it indicates that only 5% of the reference population has higher levels—suggesting potential exposure beyond background levels.

Applying Percentile Analysis in Environmental Studies

The same approach applies when assessing contamination at environmental sites. Instead of using general population data like NHANES, environmental assessments often compare site-specific samples to background levels. If a sample concentration exceeds the upper quantile of background samples, it may indicate contamination rather than natural variation.

For example, at contaminated sites, samples are often compared to background reference areas to determine whether detected levels are naturally occurring or the result of contamination. A concentration above the 95^th percentile of background samples suggests an impact beyond normal conditions.

Key Takeaways

• A single threshold or regulatory limit isn’t always enough—context matters when interpreting chemical data.
• Percentiles and quantiles provide a structured way to compare exposure levels to reference populations or background conditions.
• NHANES data help assess human exposure, while background sampling is crucial for evaluating environmental contamination.

Interested in applying these insights to your data? Whether you’re working with human biomonitoring or environmental contamination assessments, using the right statistical tools can provide clear, defensible conclusions. If you have questions or need help structuring your analysis, reach out. We’d be happy to assist!