The uncertainty of uncertainty factors in health risk assessment: Navigating the challenges of sectorial differences
Chris Waine, principal toxicologist at Bibra Toxicology Advice & Consulting, tries to untangle the industry-specific differences in the guidance available on applying uncertainty factors in risk assessment
It is the central principle of toxicological assessments that health risk is a function of hazard and exposure. But in a world where the majority of toxicological data available to us as risk assessors was generated at high doses in laboratory animal studies, how do we distinguish between a health hazard under study conditions and one that is relevant to humans?
From lab animals to humans
The point of departure( PoD) in a laboratory animal study is ideally the dose that causes no adverse effects – typically the no-observedadverse-effect level( NOAEL) or a lower confidence limit on a benchmark dose( BMDL).
To understand what this means to humans, however, we have to account for the various uncertainties through the application of one or more uncertainty factors. These can help to account for the inherent differences between species and variability within the human population, as well as extrapolate between routes of administration and study durations( Figure 1).
The human-relevant level that is expected to be without appreciable toxicological effects is variously
Table 1- Standard uncertainty factors in regulatory guidance for different industries
Factor Chemical industry Pharmaceuticals Medical devices Food Inter-individual 10( general population)
5( workers)
Inter-species 10( rats) 17.5( mice) 6( rabbits) 3.5( dogs)
Study duration
[ extrapolating to chronic exposure ]
Route of administration
2( sub-chronic) 6( sub-acute)
10 10
( or 30 for preterm, neonate and very young infants)
5( rats) 12( mice) 2.5( rabbits) 2( dogs)
2( 6-month study in rodents / 3.5-year study in non-rodents)
5( 3-month study in rodents / 2-year study in non-rodents)
10( shorter studies)
2( oral to inhalation) No specific factors for organics but bioavailability is key( 100 % absorption was assumed for all routes in ICH PDE derivations for solvents).
For inorganics – consider oral bioavailability and apply factors of 1, 2, 10 or 100 for inhaled / parenteral drugs
10
10 10
2-6( sub-chronic) 6-10( sub-acute)
10( inhalation to parenteral)
100( oral to parenteral, unless information available on bioavailability)
2( subchronic)
No specific factors
30 SPECIALITY CHEMICALS MAGAZINE ESTABLISHED 1981