HOW TO MEASURE NO EFFECT. PART III: STATISTICAL ASPECTS OF NOEC, ECx AND NEC ESTIMATES

Abstract

SUMMARYOne of the principal aims of ecotoxicity tests is to determine the concentration level below which the test chemical will have no or at most a negligible eect on the test parameter. Nowadays, the NOEC (no observed eect concentration) is normally used as an estimate for this concentration. The NOEC has, however, several major drawbacks as a summary statistic:(i) it is based on wrong usage of hypothesis testing ± the acceptance of a null hypothesis (no dierence); (ii) the estimate depends on the accuracy of the experimental test ± when the sample error is small, i.e. the test is performed accurately, the NOEC will be lower; (iii) the estimate depends on the sample size ± the larger the sample size, the lower the NOEC will be; (iv) the NOEC is a test concentration; (v) the NOEC depends on the chosen signi®cance level. An alternative is proposed, the ECx, i.e. the concentration causing an eect of x per cent:1. To estimate the ECx a concentration±response model is needed. The value of x in the ECx estimation should be chosen so that the ECx estimate is not too model dependent. 2. In NOEC estimation a certain deviation from the control is accepted. For an ECx estimation to give comparable protection with an NOEC, the x must be chosen somewhat smaller than the maximum non-signi®cant deviation in the case where the test is performed according to normal guideline procedures and the variance in the control is the mean variance for such tests. Another alternative is parametric NEC (no eect concentration) estimation. The NEC is the threshold concentration below which the test chemical will not induce an eect.(a) All NEC estimates will be model dependent. Therefore, they are not appropriate in standard ecotoxicity tests where data will seldom be sucient for model veri®cation.

A toxicity and hazard assessment of fourteen pharmaceuticals to Xenopus laevis larvae

“…These benchmark concentrations were used instead of the NOEC to provide a more meaningful, yet conservative estimate of effect (van der Hoeven, 1997a;van der Hoeven et al 1997b). In the present study, the HQs ranged from 1.4 · 10 -4 to 6.1 · 10 -7 ( Table 2), all of which fall far below the LOC, even with an uncertainty or safety factor of 1,000 applied.…”

Comparison of four different methods for reliability evaluation of ecotoxicity data: a case study of non-standard test data used in environmental risk assessments of pharmaceutical substances

“…Describing why a specific endpoint was used will help clarify the importance of the conducted study. It should be noted that it has been argued that hypotheses should be replaced by dose-response analysis when deriving exotoxicological benchmarks [25][26][27][28].…”

Methods for estimating no-effect toxicity concentrations in ecotoxicology

“…Because the NSEC estimate is based on a significance test relative to the controls (Fisher & Fox, 2023), these estimates will have some level of nonzero observed “effect” associated with them, in the same way that the NOEC can also represent some level of effect (see Mebane et al, 2008). Van Der Hoeven (1997) argues that, because there is a minimum nonzero effect size corresponding to any given NOEC, setting defined effects “ x ” (of similar size to that minimum) should therefore be acceptable. However, the difference is that the “ x ” in EC x is typically arbitrarily defined with no direct link to the biological implications of that effect.…”

“…Although it is true that the probability level considered in the estimation of NOEC is often arbitrary (and perhaps should be more carefully considered with respect to the specific decision context; see e.g., the probabilistic thresholds in Fisher et al, 2018), the basic mechanism of hypothesis testing at the heart of the NOEC (and NSEC) at least provides a clear and coherent framework with which to make decisions under uncertainty. Indeed both Van Der Hoeven (1997) and de Bruijn and Hof (1997) point out that the selection of an appropriate “ x ” is problematic from a policy perspective.…”