Eight Places Exposure Evidence Fails Under Scrutiny, and How to Check Yours in Six Minutes

Ask most organisations whether their occupational exposures are under control and the answer is a document: the occupational hygiene survey done three, five or eight years ago. That report answered a question on the day. It was not written to be read line by line by an inspector, a claims assessor or an expert witness looking for its weakest point.

A survey is an event. Evidence is a chain: the legal duty, the hazard inventory, the sampling design, the field sheet, the laboratory certificate, the statistics and the health record of the person affected. Many occupational diseases, including silicosis, noise induced hearing loss and occupational cancers, surface years after the exposure, so records made today will be read long after their authors have moved on. These are the eight places where the chain most often breaks.

01Legal duties, records and retention

Which statutory duty were you working to, and where is the record that shows you discharged it? In Botswana the relevant instruments commonly include the Factories Act (Cap 44:01), the Mines, Quarries, Works and Machinery Act (Cap 44:02), the Public Health Act, the Radiation Protection Act 2006 and the Workers' Compensation Act (Cap 47:03). Which duties apply to your operation should be confirmed with legal advisers.

The evidence fails when there is no written legal register, or when the exposure record of a worker who left five years ago cannot be found. The claimant's own account may then go largely unchallenged. Good looks like records indexed by worker and exposure group, retrievable the same day and kept for decades. The South African Regulations for Hazardous Chemical Agents 2021, a common regional benchmark and not Botswana law, set 40 years.

02Hazard inventory, exposure groups and control verification

Without a written health hazard inventory and similar exposure groups, meaning workers who do comparable tasks with comparable exposure, each result represents nobody but the person who wore the pump. Two gaps recur. Maintenance, shutdown and cleaning tasks, often the highest exposures on a site, are missing from the routine survey. And controls are assumed, not verified. A reviewer wants the measurement that showed the extraction worked, not the invoice for its installation. Good looks like task based groups reviewed by a registered occupational hygienist, local exhaust ventilation tested against design values on a defined interval, and respirator fit test records per worker.

03Sampling strategy and field records

Area readings and handheld spot checks describe a room. They do not establish what a worker inhaled. Defensible airborne measurement is personal sampling in the breathing zone, for a duration that matches the limit: the full shift for an eight hour time-weighted average, 15 minutes for a short-term limit.

One or two samples per group cannot describe an exposure that varies from day to day. Recognised strategies point the same way. EN 689 requires at least six measurements per group for its statistical test and accepts three to five only where every result is a small fraction of the limit. The AIHA strategy recommends six to ten. Fewer than that needs a written justification. Each sample also needs its field sheet: pump calibration before and after, flow rate, start and stop times, the task observed and the controls in use. Without the calibration the sampled volume is unknown, and the concentration becomes an estimate.

A survey tells you what was measured on the day. Evidence is what you can still produce, explain and connect to a named worker ten years later.

04Laboratory analysis and chain of custody

A well taken sample can still be discounted at the laboratory. Which laboratory analysed it? Is that specific analysis within the scope of its ISO/IEC 17025 accreditation, granted by a recognised body such as SADCAS or SANAS? A certificate alone is not enough, because accreditation is granted per method and scopes change. Who signed for the sample at each handover between the pump and the bench, and do the sample numbers on the field sheet, the custody form and the laboratory report agree? Good reports also show field blanks, detection limits and measurement uncertainty. A result reported as not detected against a high detection limit may conceal an exposure that matters.

05Statistical interpretation and exposure decisions

Exposure is a distribution, usually right-skewed and close to lognormal, so an average below the limit says little about the days above it. The defensible question is where the upper tail sits: the 95th percentile of the group's exposure, with its confidence limits, compared against a stated limit under a stated decision rule such as the AIHA exposure categories.

Three weaknesses are found quickly. The limit used is not stated. Botswana publishes limited occupational exposure limits, so practitioners commonly benchmark against the South African Regulations for Hazardous Chemical Agents 2021 and the ACGIH TLVs, and the choice should be written down. Results are marked above or below the limit one at a time, with no conclusion for the group. And non-detects are recorded as zero or left out, when a recognised method for censored data should be used.

06Noise, heat and vibration

A hearing loss claim turns on personal dose. A noise map shows loud zones, not what reached a worker's ear over the shift. Good evidence is personal dosimetry per exposure group, reported as the exposure level normalised to eight hours (LAeq,8h, also written LEX,8h) with a 3 dB exchange rate and the C-weighted peak level, with calibration checks before and after. The usual comparison is the 85 dB(A) rating limit in the South African noise regulations, a common regional benchmark and not Botswana law.

Heat fails in the same way: air temperature and a water point are not an assessment. A recognised index such as WBGT, measured at the task and read with work rate, clothing and acclimatisation, is what supports a work and rest regime. For vibration, the evidence is a daily exposure value calculated from site measurements, not a manufacturer's figure.

07Ionising radiation and NORM

Many sites hold radiation sources without realising it: density and level gauges on conveyors and filling lines, X-ray inspection units, handheld XRF analysers, and naturally occurring radioactive material (NORM) in ore, scale and sludge. Under the Radiation Protection Act 2006, authorisation and dose records are among the first things an inspector is likely to request. Confirm requirements with the Radiation Protection Inspectorate.

The evidence fails when authorisation has lapsed or sits only with a contractor, or when dose records are held only by the dosimetry supplier. Where NORM or unsealed sources are handled, part of the dose may be internal, so a badge alone may not give the full dose of record. The international benchmark, IAEA GSR Part 3, expects dose records to be kept until the worker reaches, or would have reached, age 75, and for at least 30 years after the work ends. The retention period that applies to your authorisation should be confirmed with the Radiation Protection Inspectorate.

08Medical surveillance linkage

A claim asks whether the employer knew a worker's exposure and health status together. Annual medicals for everyone, unrelated to exposure, show that a doctor was seen, not that the risk was known. Surveillance should be prescribed by an occupational medical practitioner from the exposure assessment, per exposure group, with baseline, periodic and exit examinations and tests matched to each agent.

Two further tests follow. Can one worker's exposure history and surveillance results be shown side by side? And when an audiogram or lung function test showed a change, what was done about the exposure? A finding that was filed and not acted on is very hard to explain later, because it shows the employer held the information and did nothing with it. That is never a reason to stop looking or to avoid surveillance. It is a reason to define the response before the result arrives, act on findings and record the action.

None of these eight points calls for more sampling as a first step. Most call for knowing which links are held, which are partial and which are missing.

The Exposure Evidence Defensibility Check is a self-reported screening, written by registered occupational hygienists and scored automatically from the answers given. It is not an occupational hygiene survey, an audit or a legal opinion, it does not prove compliance or defensibility, and it cannot predict the outcome of any inspection, insurance decision or legal matter. This article is general information, not legal advice. Legislation is cited in general terms, and the duties that apply to a specific operation should be confirmed with qualified legal advisers.

References

1. American Industrial Hygiene Association (AIHA), 2015. A Strategy for Assessing and Managing Occupational Exposures, 4th edition. AIHA Press, Falls Church, Virginia.
2. European Committee for Standardization (CEN). EN 689:2018+AC:2019. Workplace exposure: measurement of exposure by inhalation to chemical agents, strategy for testing compliance with occupational exposure limit values.
3. International Organization for Standardization, 2017. ISO/IEC 17025:2017. General requirements for the competence of testing and calibration laboratories. iso.org
4. International Atomic Energy Agency, 2014. Radiation Protection and Safety of Radiation Sources: International Basic Safety Standards. IAEA Safety Standards Series No. GSR Part 3. IAEA, Vienna.
5. Republic of South Africa, Department of Employment and Labour, 2021. Regulations for Hazardous Chemical Agents, 2021. Government Notice R. 280, Government Gazette No. 44348, 29 March 2021.
6. American Conference of Governmental Industrial Hygienists (ACGIH). TLVs and BEIs. Published annually. acgih.org
7. Republic of South Africa, Department of Employment and Labour. Noise Exposure Regulations, 2024, made under the Occupational Health and Safety Act, 1993. Government Notice 5953, Government Gazette No. 52226, 6 March 2025.
8. International Organization for Standardization, 2017. ISO 7243:2017. Ergonomics of the thermal environment: assessment of heat stress using the WBGT (wet bulb globe temperature) index.
9. Republic of Botswana. Factories Act (Cap 44:01); Mines, Quarries, Works and Machinery Act (Cap 44:02); Workers' Compensation Act (Cap 47:03); Public Health Act; Radiation Protection Act, 2006 (Act No. 22 of 2006) and Radiation Protection Regulations, 2008. Laws of Botswana, Government Printer, Gaborone.