Cost-Effective and Convenient Exposure Assessment: Could It Be on a Path Toward Big Data?
Sponsored by Assay Technology
Industrial hygienists responsible for workplaces that use chemicals typically do chemical risk assessments using combinations of three methods: monitoring (air sampling), modeling, and professional judgment. Professional judgment (so-called “qualitative exposure assessment”) has traditionally been popular and cost-effective (though not precise), as OSHA rules generally permit an IH to decide whether employees are significantly exposed (or not) without monitoring or modeling.
Qualitative Exposure Assessment and Professional Judgment
When an IH does decide to make a “quantitative” exposure assessment, she has the opportunity to select monitoring or modeling (or a combination) to assess exposures to specific chemicals in each worker’s personal breathing zone. The existence of occupational exposure limits for many hazardous chemicals used in workplaces allows chemical exposure assessment to be a proxy for an overall risk assessment. Since exposure at or below an OEL is believed to cause no harm, an assessment that typical (representative) personal exposures are below an OEL suggests that an employee is not subject to unacceptable chemical risks.
Reducing Uncertainty in Exposure Assessments
Since the International Organization for Standardization has reminded IH practitioners about the uncertainty in exposure estimates—see ISO 17025, part 7.6—IH practitioners are thinking more about how to evaluate and reduce that uncertainty. It is long-established truth that hourly, daily, and weekly variation of air concentrations in workplaces is a greater source of uncertainty in exposure assessment than variations in monitoring methods. Accordingly, an efficient way to reduce assessment uncertainty is to collect more data and continuously reanalyze all exposure data together using valid statistical methods, including exposure models.
Efficient Methods for Reducing Uncertainty
To be cost effective, efficient monitoring and data collection methods are required, with measurements classified by employee, location, job, date, time, etc. Modern computers and software allow sharing of data across devices and permit the IH to send, store, update, and analyze data against reasonable exposure models.
This combination of efficient samplers with data collection received from an instrument or an electronic lab report allows increased data collection leading to a reduction in uncertainty in our answer to the question: What is the estimated average and range (average plus uncertainty) of exposure for each chemical agent to each employee as a fraction of the OEL?
Current Examples and a Look at the Future
For a few chemicals, wearable instruments can measure and log data, which is transmitted or downloaded to an onsite device in communication with an exposure data system (on a server or in the cloud). Exposure data systems, once available only to large organizations, can now be shared online. Apps that communicate data to exposure data systems are also becoming more available, and custom apps can also be developed.
While most chemicals are collected on personal samplers and sent to a lab for analysis, labs increasingly offer electronic reporting systems that send exposure data to your device or system sorted by employee, location, time, etc.
Some labs offer personal device apps that allow you to log in samples and receive data on your personal device, where it can be reviewed and downloaded to your exposure data system.
In science, things get better as uncertainty is reduced, and we are moving in that direction. The Synergist recently discussed whether “big data” would come to IH anytime soon. That question will be answered as we see how much useful data can be collected and shared across our networks of sensors, smart phones, tablets, PCs, and servers.