LL1-14 digest — Ch14 Hormones as growth promoters: the precautionary principle or a political risk assessment?#

Late lessons from early warnings (EEA, 2001), pp. 149–156. Authors: Jim W. Bridges and Olga Bridges (University of Surrey). Jim Bridges advised the EU at the WTO on the ban (p. 195). The chapter text does not say this. No panels.

Core story#

Hormones were adopted after 1945 to fatten livestock. DES was the cheap favourite (p. 149).

US. The chapter says DES was banned in 1972 (court records show feed and implant approvals withdrawn in 1972–73), “temporarily reinstated” in 1974 because of “procedural deficiencies”, and banned in 1979 because no safe residue level could be identified (pp. 149–150). Other hormonal promoters stayed in use for economic-efficiency reasons (p. 150).

EU. Both expert committees, the EU’s own and the WHO/FAO one, reached conclusions at odds with the ban: - The EU’s Lamming Committee (1982) reached an interim conclusion that oestradiol and other hormones were safe in cattle. - The committee “was disbanded” in 1987 and “the EU did not publish its interim conclusions” (p. 150). - JECFA (1988) agreed with Lamming. But it assessed only authorised use, single substances and manufacturers’ data (p. 150). - The EU banned six hormones anyway: adopted in 1985, annulled on procedure, re-adopted in 1988, and applied to imports from 1989 (p. 150). The chapter’s chronology is muddled: the 1985 adoption came before JECFA had reported. - Three factors “appear” to have driven the Commission (p. 151): evidence that DES caused vaginal cancer (after prenatal drug exposure, which the chapter blurs), public concern about hormones including the pill, and reports of puberty anomalies in Puerto Rico and Italy of uncertain cause. The conclusions call the ban “principally” a response to public concern (p. 154).

Trade. The US retaliated, then won at the WTO (p. 153).

The authors’ verdict (p. 154): - The committees were never asked to characterise uncertainty, so the original ban was “in reality, a political risk assessment”. - “More recent scientific research … probably justifies” keeping it. p. 154 does not say which research; the likeliest referent is the p. 153 claim that oestradiol is a genotoxic carcinogen. - There is “no good evidence” the ban protected health. Equally, no reliable safety evidence built up where use continued (p. 153).

Key evidence#

JECFA’s three scope limits (p. 150).
The FDA’s safety argument compared residues with oestrogen in pill users and pregnant women. That ignored low-oestrogen children and DES’s structural differences from natural oestrogens (p. 150).
Prepubertal boys’ own hormone levels sit near detection limits, so any added oestrogen is a relatively large share. The chapter calls this “particularly relevant” to the FDA’s rule that food intake stay below 1% of endogenous production (pp. 152–153).
Wildlife warnings from 1970–73 (DDT, catfish, birds) were ignored until the late 1980s. The chapter tentatively ascribes this to drug agencies’ lack of interest in the environment and the assumption of dilution and degradation (p. 152).
Economics. The claimed USD 500m-a-year consumer cost of a DES ban was “probably groundless”. There was “little evidence of a sustained increase” in costs, which “could be explained” by existing substitutes (Table 14.1) and, in part, by wrong cost assumptions. The 1974–79 “breathing space” let industry develop more hormonal promoters (pp. 149–150).
Misuse. JECFA considered only authorised use despite “indications of significant accidental or deliberate misuse”, and the chapter lists six forms of potential misuse (pp. 150–151). The Pimenta Report “found no evidence in the use of oestradiol-17β” but backed the ban because it “facilitated controls and consumer confidence” (p. 151). The chapter reports unproven “claims” that the ban drove illegal use, including of DES (p. 151).

Main mechanisms#

The main mechanisms, all expanded in the insights below, are: - how the scope of an assessment shapes its verdict - baseline errors, with detection limits standing in for safety - “natural” framing - inflated cost claims - misuse and displacement into illegal channels - suppressed advice and remit blind spots - trade-law arbitration - neither side generating the evidence that could settle the question (pp. 150–154)

Transferable insights (technology-neutral)#

“Safe” verdicts are conditional on the scope set for assessors, such as intended use, single agents and sponsor data (p. 150). Strong. The WTO Appellate Body made the same point (AB para 206).
Assessors not asked to characterise uncertainty leave decisions looking political (p. 154). Moderate. Sound logic, but the committees’ terms of reference are not documented.
Averages and high-exposure comparators hide sensitive, low-baseline subgroups (pp. 150, 152–153). Strong. Aksglaede et al. (2006) found children’s oestradiol levels were overestimated.
Detection-based thresholds make “not detected” read as “safe”; better instruments can flip regulatory status without new evidence of harm (pp. 149–150, 152). Strong for the first half, which the chapter documents. The “flip” half comes from US court records (1972–73 withdrawals triggered by new analytical methods, with the FDA disclaiming any health hazard), not from the chapter.
Cost forecasts of restriction can be inflated; substitutes may exist or be developed during delay (p. 150). Moderate. One case, no cost data.
Bans can displace activity into unmonitored channels; without monitoring the net effect is hard to gauge (p. 151). Suggestive. Reported as “claims”.
Remit-bound institutions neglect harms outside their domain, filling gaps with comforting assumptions (p. 152). Moderate. The explanation is tentative in the chapter. Later feedlot and trenbolone studies undercut the degradation assumption but do not test the remit explanation.
Overruling expert advice and not publishing it undermines the legitimacy of the decision (p. 150). Moderate. The legitimacy link is an inference from the chapter’s facts.
Ban versus conditional permission turns partly on how easy each is to enforce (p. 151). Moderate.
After divergent decisions, the outcome evidence that could settle the question may not accumulate on either side, so disputes persist (p. 153). Moderate. In 2003 the EU still could not quantify the risk. But the EU did keep funding research (p. 152), so the gap is decisive evidence, not funding as such.
Benefits flowing only to producers lower the tolerable level of uncertainty; this needs explicit risk–benefit machinery (pp. 153–154). Asserted. The “purely economic” judgement is made of DES specifically.

Caveats#

Standpoint. The lead author was an EU WTO adviser. Both authors presented a paper at the 1995 EC conference they describe, and the WTO Panel cited it (AB fn 107).

Factual errors, checked against primary sources: - Directive 81/602 (1981) already prohibited hormonal substances in farm animals, with DES not among the exemptions. The chapter’s (and Chapter 1’s) “1987” EU DES ban looks wrong. - The import ban covered “third countries”, not “third world countries”. - DES caused clear-cell adenocarcinoma after prenatal drug exposure. The chapter says “adenoma” and implies growth-promoter use. - The 1974 US reinstatement followed a court ruling that no hearing had been held, not a flawed “bill”. - WTO proceedings began in 1996; the panel reported in 1997. The chapter’s “1997” is imprecise. - The chapter acknowledges US success but presents the upheld Appellate Body finding as narrow. It was dispositive: the ban was not based on a risk assessment (also no MGA assessment, no misuse assessment). - Authorised sanctions were US$116.8m plus C$11.3m a year, not about EUR 160m. - The “Jakes” cited for the “1 cancer in 133 years” estimate is T. H. Jukes, a public defender of DES use in cattle. - Internal slips: the 1985 ban predates the JECFA assessment it supposedly overrode; trenbolone is called oestrogenic. - Several citations are mismatched to their claims (they appear shifted in the wildlife paragraph). - Table 14.2 (compiled by the EEA, not the authors) goes beyond the text in places, for example on Pimenta, the 1989 extension and the 2000 workshop’s “confirms impacts”.

Contested claims. “Demonstrated genotoxic carcinogen” (p. 153) overstates a dispute. JECFA in 1999 agreed on genotoxic potential but judged carcinogenicity receptor-mediated, set an ADI, and called residues from good practice unlikely to be a hazard. The EU’s own CVMP also dissented.

Omissions. The chapter ignores Codex standards, the protectionism argument and non-ban alternatives. On protectionism, the Appellate Body cited “the depth and extent of the anxieties” in the EC (AB para 245). The chapter also omits a driver the Appellate Body recorded: harmonising divergent Member State rules.

Editorial amplification. When they reuse this chapter, the editors attribute the 1974 US reinstatement to farm-lobby pressure (p. 179), where the chapter gives procedural deficiencies. They also tie the Puerto Rico episode to growth-promoter misuse (p. 175), where the chapter says its source was not identified.

After 2001: - The EU made the oestradiol ban permanent and the other five provisional (2003). - The dispute was managed through quotas for hormone-free beef (2009 and 2013 MoUs), not resolved. - EFSA (2007, via CRS) said the contribution of residues to cancer risk is unknown. - Environmental concern has grown (Orlando 2004; Qu 2013).