Nobody Reads the Efficacy Data: A Waiver, and the Assumption Underneath It

1. Introduction: returning to a question we have asked before

Our earlier article What the Regulator Actually Saw examined the federal efficacy data waiver, the 1979 marketplace rationale behind it, and the Canadian requirement that every product have value. This article returns to the American waiver with material that article did not have.

1.0a And we should say plainly that this overlaps it

The first draft of this article was written without reference to the earlier one, repeated its central argument as though new, and stated that we had not established the Canadian position when we had. Those errors are corrected in the text and recorded in §29.

1.0b And the distinction from safety is the one to hold

Nothing in this article concerns whether products are dangerous, which is examined separately and at length.³

1.1 Efficacy meaning whether the product controls the pest

Described in the guidance as product performance.¹

1.1b And the exceptions include the work we do

Structural pests such as termites sit inside the category requiring submission, which is a point we return to in §27.¹

1.2 What this article adds

Four things the earlier article lacked: the agency's own statement that performance review exists to prevent exposure from ineffective products, a 2023 petition to move the boundary, the state layer, and the softer vocabulary states permit when a claim outruns its evidence. Sections 13, 16 and 21 are the new material.

2. What has to be submitted

Performance data are required where a product makes claims against public health pests, pests that affect building structure such as termites, or certain invasive pests.¹

2.0a And antimicrobials sit in their own regime

With their own subparts and their own performance standards, which is the most heavily reviewed corner of the system.²³

2.0b And the categories are defined by claim rather than by chemistry

What triggers submission is what the label says the product does, not what is in it.¹

2.1 The structural inclusion is notable

Termites appear alongside public health pests rather than with ordinary insect claims, which places building damage in the category requiring proof.¹

3. And what does not

Everything else.

The Agency generally does not require submission of efficacy data to support non-public health-related claims.¹

3.0a And the phrasing is generally rather than never

Which leaves room for the case-by-case power in §8 and stops the waiver being absolute.¹

3.0b The examples given are revealing

Odour-causing bacteria, weeds and plant pathogens, all cases where a user can see or smell whether something changed.¹

3.1 Which covers most agricultural products

One legal summary notes that most agricultural pesticides fall outside the groups requiring submission.⁵

4. The waiver is the default

Which is how the rule is written.

The agency has waived the requirement to submit efficacy data unless the product bears one of the named claims, with the exceptions following the waiver.²

4.0b And it appears in the regulation rather than in guidance

Which places it in binding text rather than in an agency's description of its own practice.²

4.1 And the drafting order matters

A rule reading waived unless carries a different presumption from one reading required except, even where the two would cover the same products.²

5. Which is the opposite of the intuition

Our observation.

A reasonable person assumes a regulator approving a product for sale has satisfied itself the product does what the label says. The structure is the reverse: proof of performance is the exception and the waiver is the rule.

5.0b And the assumption is not unreasonable

Approval for sale carries an implication of fitness for purpose in most regulated markets, which is why the structure surprises.

5.1 And safety review is not affected

Toxicology, exposure and environmental data requirements are extensive and separate, which our registration articles have described.³

5.1a And it is a real difference in kind

Our registration articles described toxicology, metabolism and environmental fate packages running to many separate studies, against nothing at all on whether the product works.³

5.1b And the two reviews serve different people

Safety review protects bystanders and the environment. Efficacy review would protect the buyer, who is the party the structure leaves to themselves.

5.2 So the gap is specific

Whether the product is safe is examined closely. Whether it works is not.

6. The data still has to exist

Each registrant must ensure through testing that his product is efficacious when used in accordance with label directions and commonly accepted pest control practices.²

6.1 And the standard is use as directed

Efficacy when used in accordance with label directions and commonly accepted pest control practices, which sets the conditions the testing must model.²

6.1b And commonly accepted practices is undefined here

A phrase doing considerable work in a sentence that sets the whole testing standard.²

6.2 Which is a narrower claim than it sounds

Nothing requires the product to work under the conditions a householder will actually create, which our label compliance articles found are frequently not the directed ones.

7. Held by the party it favours

And available on request.

The applicant remains responsible for developing and keeping efficacy data on file,¹ with one state-level document putting it plainly: industry is required to have the data available upon request.⁶

7.0a Which is why calling it an absence would be wrong

The testing duty is real and enforceable. What is absent is an independent reader.

7.0b And the obligation is stated in the same sentence as the waiver

Which makes the rule a transfer of responsibility rather than a removal of it.²

7.1 Which is a different thing from review

Data nobody has read has not been checked for design, sample size, control conditions or selective reporting, all of which our own methods articles treat as the whole question.

7.1b And selective reporting is invisible from outside

A registrant holding ten trials and a file containing three has met the obligation as written, since the rule requires data rather than all data.²

7.2 And the party holding it chose the design

Which our resistance ratio article found was decisive even in peer-reviewed work.

8. And the agency may ask

A reserved power.

The agency reserves the right to require, on a case-by-case basis, submission of efficacy data for any pesticide product registered or proposed for registration.²

8.0a And the power covers registered products too

Not only pending applications, so a product already on sale can be asked to prove itself.²

8.0b And the trigger for asking is not specified

Nothing in the text we read describes what would prompt the agency to exercise it.²

8.1 We did not establish how often it does

Which §29 records as a gap, since a reserved power exercised routinely and one exercised never produce very different regimes from the same text.

9. The clause that explains the boundary

The exception covers claims to control microorganisms that pose a threat to human health and whose presence cannot readily be observed by the user.²

9.0b The clause is a qualifier rather than a separate condition

It narrows the microorganism exception rather than standing beside it, which is what makes observability part of the test rather than a coincidence.²

9.1 And the vertebrate exception has its own logic

Naming rodents, birds, bats, canids and skunks that may directly or indirectly transmit disease, which is a harm the buyer cannot see either.²

10. Review attaches to invisibility

Our reading.

The qualifying clause is not about severity. A disinfectant claim is reviewed because the buyer cannot see whether the bacteria died. The logic implies that where the buyer can see, the buyer is the check.

10.0b And the clause is doing real work

Without it the exception would read as covering every claim against organisms that threaten human health, which is a far wider category.²

10.1 Which explains the shape of the whole list

Disease vectors and disease-carrying vertebrates are reviewed because the harm is a transmission risk the buyer cannot observe either, and structural pests because the damage is inside a wall.²¹

10.1a And the principle is about who is positioned to notice

Which is a reasonable thing for a regulator with limited capacity to think about, and a questionable thing to rely on without testing whether the noticing actually happens.

10.1b And it explains the odour example too

A product for odour-causing bacteria is named as waived, and odour is the one property of a microbial problem a user can detect without equipment.¹

10.2 It is a coherent principle

And a defensible one, given finite review capacity.

11. Which is a theory of verification

Stated in a regulation rather than in an argument.

The rule assumes that a purchaser who can observe an outcome will notice a product that does not work, and that this observation does the job regulatory review would otherwise do.

11.0b Which is how most consequential assumptions travel

Not as a claim somebody defends, but as a qualifier nobody notices.

11.1 And it is never stated as one

The assumption sits inside a subordinate clause rather than in any explanatory passage we found.

12. And it is the theory our own work disputes

Directly.

Our credence good article argued that a pest control outcome is not observable in the way this assumes: a population that declines seasonally looks like a success, a population that was never there looks like a success, and a failure looks like a reinfestation.

12.0a And the observation is usually made once

A buyer forms a verdict from a single use in a single setting with no control, which is the weakest possible evidential design.

12.0b And the professional case is not much better

Our callback rate article found the trade itself lacks an agreed measure of whether a treatment worked.

12.1 The consumer product case is worse

Our essential oil and repellent articles found buyers attributing outcomes to products in situations with no comparison and no measurement.

12.1b And the observation has to be repeated to mean anything

A single application with no control tells a buyer nothing, and nobody applies a product twice to find out.

12.2 So the substitute check is weak where it matters most

Not in a laboratory where a disinfectant is invisible, but in a kitchen where a person decides whether the spray worked.

Sections 10 to 12 restate an argument our earlier article made from the agency's 1979 notice. The route here is different, through the qualifying clause in the regulation itself, and it arrives at the same place.

13. The agency's own rationale

Which we had not expected to find.

Requirements to develop performance data provide a mechanism to ensure that pesticide products will control the pests listed on the label and that unnecessary pesticide exposure to the environment will not occur as a result of the use of ineffective products.³

13.0b And it is written as a justification

Explaining why performance requirements exist at all, which makes it the agency's own account of what is lost where they do not apply.³

13.1 Two purposes in one sentence

That the product controls what the label says, and that ineffective products do not produce exposure for nothing. The second is the one that travels.³

14. Which cuts against the waiver

Our argument.

If an ineffective product causes unnecessary exposure, then the exposure argument for reviewing efficacy applies to every product, not only to ones whose failure is invisible. A person spraying twice because the first application did nothing has doubled their exposure whether or not they could see the result.

14.0aa And it is the stronger of the two objections

An argument from the buyer's interest can be answered by saying buyers can look. An argument from exposure cannot.

14.0a The argument does not depend on visibility at all

Exposure from an ineffective product happens at the moment of application, before anybody could have observed whether it worked.

14.0b And the environmental half applies equally

An ineffective agricultural product is applied to a field either way, which is the exposure the sentence names.³

14.0c And it matches what the earlier article found in Canada

Where a regulator's own fact sheet credited efficacy review with lower label rates, which is the same connection between performance and exposure, reached from the other side of the border.

14.1 And the sentence is the agency's

Appearing on its own data requirements page rather than in a critique.³

15. The boundary has moved before

Which shows it is a choice rather than a constant.

A rulemaking record notes that the agency decided to rescind the proposed efficacy data waiver with respect to vertebrate control agents intended for control of pests that directly or indirectly transmit disease to humans, having previously proposed to waive it.⁴

15.0b And the direction was toward more review rather than less

The withdrawal added a category to the list requiring submission, which is the opposite of the drift one might assume.⁴

15.1 Which means somebody argued both ways

A waiver proposed and then withdrawn is a decision that was contested inside the process rather than an oversight.⁴

16. The 2023 petition

An attempt to move it again.

An environmental responsibility organisation, a bird conservancy and other groups petitioned to require submission of efficacy data for neonicotinoid and other systemic insecticides, for existing and future registrations, and to review registered products in light of it.⁵

16.0b And the target class is one this journal has written about

Our neonicotinoid article examined the selectivity argument for these compounds without asking whether their field performance had ever been reviewed.

16.1 And it asks for retrospective review as well

Covering existing registrations rather than only future ones, which is the harder half of the request.⁵

17. What the petitioners think the data would show

Stated by the legal summary rather than by them.

The petition suggests the petitioners believe efficacy data will at best show that neonicotinoids and other systemic insecticides are overused and at worst that they are not effective at all.⁵

17.0b And the groups are not the usual petitioners on efficacy

An environmental responsibility organisation and a bird conservancy are bodies whose ordinary concern is harm rather than performance.⁵

17.1 Which is an unusual reason to want data disclosed

Not that a product is dangerous, but that it may be useless, and that its use is therefore exposure bought for nothing.

17.2 That framing is ours

The summary reports the belief without drawing the connection.⁵

18. And the objection to asking

Also from the legal summary.

The requested change would be a paradigm shift in how EPA evaluates a large percentage of the pesticide products it registers and would create significant administrative burdens.⁵

18.0b And it concedes the current position

Describing the change as a shift in how the agency evaluates products is an acknowledgement of how they are evaluated now.⁵

18.1 Which is an argument about capacity rather than about merit

And a serious one, since a requirement nobody can staff is a requirement in name only.

18.1b And the burden argument has a counterpart

The cost of reviewing the data sits with the agency, while the cost of products that do not work sits with everybody who buys them.

18.2 It also confirms the scale

A large percentage of registered products is the quantity currently going through without performance review.⁵

19. The state layer

Where some of the missing scrutiny sits.

Exempt products are subject to registration by individual states, which are not required to permit sale simply because a product is federally exempt, and each state may have different label or data requirements.⁷

19.0a And the layer exists mainly for exempt products

The guidance we read concerns minimum risk registrations, so how far state scrutiny extends to conventional products is not something we established.⁷

19.0b Which creates a patchwork rather than a backstop

A product unexamined federally may be examined in one state and not in the next, and a buyer has no way to know which applies to the tin in front of them.

19.1 With states weighing whether to accept partial data

One document records discussion of whether raw data rather than complete reports might be accepted, provided it is competent and reliable.⁶

19.2 And the state route is discretionary

States are not required to permit sale of a federally exempt product, which makes the second layer a power rather than a duty.⁷

20. The worked example

Where states evaluate efficacy claims, the label should not exaggerate the extent of control, and if mosquito or bed bug populations were reduced by only half while the label claimed control, the claim might be treated as overblown.⁶

20.0b And half is a real effect

A product halving a population has done something, which is what makes the wording question harder than a simple case of a product that does nothing.

20.1 And the pests chosen are telling

Mosquitoes and bed bugs, both in categories where a buyer has strong feelings and weak measurement, which is the combination our own articles keep finding.⁶

21. Soft claims

A registrant may be required to drop false or unsubstantiated claims, or some states may allow the option of a softer claim to restate an overblown one, with the listed options including reduce, diminish, lower, weaken, shorten, suppress and subdue.⁶⁷

21.0b And the finding sits with the state rather than the registrant

It is the state evaluating the claim and requiring the change, which makes the softer word an outcome of scrutiny rather than a voluntary hedge.⁶

21.1 And the alternative is removal

Dropping the claim entirely is the other option described, so a softer word is the outcome a registrant would prefer.⁶

22. A regulated vocabulary nobody explains

Our point.

The difference between a label saying controls and one saying suppresses is a regulatory finding about whether the evidence supported the stronger word. Nothing on the package says so.

22.0b And the softer word replaces rather than explains

The option described is restating an overblown claim, not adding the finding that produced the restatement.⁶

22.1 And the words are not intuitive

Subdue and weaken are not ordinary consumer vocabulary, and a shopper comparing two products has no reason to read them as a graded scale.

22.1a And the list is graded without being ordered

Seven words offered as alternatives, with nothing saying whether suppress is stronger than reduce.⁷

22.1b And the strong word remains available elsewhere

A competing product with the data can say control, so the shelf holds both words with no way to tell which difference is real.

22.2 Which makes it a disclosure that does not disclose

The same pattern our hygiene grading and aircraft disinsection articles found.

23. Which is where this lands for a reader

Practically.

A softer verb on a label is worth noticing, because in at least some cases it records that a stronger claim could not be supported.⁶

23.0b And it is one of very few label signals with any content

Most of what a package says is marketing. A hedged verb may be the residue of an actual determination.⁶

23.1 With the obvious caveat

A softer verb may also reflect an honest description of a product that genuinely suppresses rather than controls, and nothing on the label distinguishes the two cases.

24. How this relates to our two earlier articles

Which we should be explicit about.

Our exempt products article concerned minimum risk products outside federal registration and found their efficacy unexamined. What the Regulator Actually Saw had already shown that most fully registered products are not reviewed for performance either. This article adds the soft claim material, which bears on both.²⁷

24.0b And the exempt category is at least visible as a category

A buyer can be told a product is exempt. Nothing marks a registered product as one whose performance data was never submitted.

24.1 So the exempt products article read narrower than it should

It presented a gap peculiar to an exemption, and the later article had already shown the gap is closer to the general position.

24.1b And both articles now say the same thing

That the question of whether a product works is generally left to the seller to answer and to the buyer to check.

24.2 And the correction is worth making plainly

An article that treats a general condition as a peculiarity of one category misleads its reader about how common the condition is.

25. What we take from it

Three things.

The waiver is the default and submission is the exception. Section 4.²

The exceptions track what a user cannot observe. Section 10.²

And the agency's own exposure rationale applies more widely than the rule. Section 14.³

25.0b And a fourth, about checking our own archive

An article about unexamined evidence was drafted without examining this journal's own earlier work, which is the failure it describes, committed by the people describing it.

25.1 And one thing we cannot take from it

Whether the unreviewed products work, which is the question the whole structure leaves open.

26. The measurement we attempted

And did not publish.

We tried to classify this journal's own references by source type, to state what proportion of our evidence base is trade material rather than peer-reviewed or official. The classification left two-thirds of references unresolved, which is not a number worth reporting.

26.0b The failure was in our own method

Our reference descriptions name source types in prose rather than in a fixed vocabulary, so a keyword pass could not classify them reliably.

26.1 We record the attempt rather than the result

Because an article about unexamined evidence should not introduce a statistic it cannot stand behind.

26.1b Which is the discipline this article is about

A claim whose evidence nobody has examined should not be published merely because it is convenient, and that applies to us as much as to a registrant.

26.2 And the impression stands without the number

Our recent limitations sections have repeatedly recorded articles resting largely on trade material, which is an observation about particular articles rather than a measurement.

27. Our own position

The disclosure.

We choose products from labels and from what the distributor tells us, and this article says that for many products nobody outside the manufacturer has examined the evidence behind either. We have never asked a supplier for efficacy data, which §6 says they are required to hold.

27.0b And the products we use sit under a different rule

Registered under the Canadian system, where value is a statutory condition for every product, so the waiver described here is not the regime behind our own labels.

27.1 Which is a thing we could start doing

And the only concrete action this article generates for us.

27.2 And we should say what it would prove

Not much on its own, since we are not equipped to appraise a trial design, which is the point §7.1 makes about the agency's position rather than a reason not to ask.

28. The Manitoba position

28.1 The regime described here is American

Every source in this article is a United States federal or state document. The Canadian position was set out in our earlier article and differs in structure: value is a statutory condition of registration for every product rather than a data requirement that can be waived.

28.1b But the difference has narrowed

That article also found that since a 2010 proposal, information other than efficacy studies may be used to satisfy the value condition, so the Canadian requirement is mandatory in form and flexible in what meets it.

28.2 What we still could not find

Whether any Canadian equivalent of the soft claim vocabulary exists, how often the value flexibility is used for structural products, and whether any Canadian body has considered the exposure argument in §14.

28.3 Which leaves the new material here untested at home

The petition, the state guidance and the softer verbs are all American, and none of them has been checked against the system our own products are registered under.

29. Limitations and open questions

This article substantially overlaps an earlier one. What the Regulator Actually Saw covered the waiver, its 1979 rationale, the argument about buyer observation and the Canadian value requirement. The first published version of this article repeated that ground without citing it and stated wrongly that we had not established the Canadian position.

That is the most important limitation because it is an error of the kind this journal criticises in others, and a reader of both articles would have found us contradicting ourselves. The text now credits the earlier article and restricts its own claims to the material the earlier one lacked.

The Canadian comparison is not re-examined here. Section 28 relies on the earlier article's reading of the Canadian position rather than on fresh sources.

We do not know how often the reserved power is used. Section 8 describes a case-by-case right to demand data, and without a count of how often it is exercised the practical effect of the waiver is unclear.²

Two sources are state-level guidance documents. The worked example and the soft claim vocabulary come from association guidance rather than from statute, and describe what states may do rather than what any state does.⁶⁷

The petitioners' beliefs are characterised by a law firm. Section 17 reports what a legal alert says the petitioners believe, not what the petition says, and we did not read the petition.⁵

And the soft claim example concerns exempt products. The guidance is aimed at minimum risk registrations, and we have not established that the same vocabulary applies to conventional registered products.⁷

Sections 10 to 12 restate the argument of our earlier article by a different route, and sections 5, 7.1, 14, 22, 24 and 26 are our reasoning. The argument that review tracks observability, the objection from the agency's own exposure rationale, the reading of the soft claim vocabulary and the correction to our earlier article are ours rather than sourced positions.

30. Conclusion

Our earlier article What the Regulator Actually Saw set out the waiver, its 1979 marketplace rationale and the contrasting Canadian value requirement, and this one adds to it rather than replacing it. The federal data requirements say the agency has waived the requirement to submit efficacy data, and then list the exceptions: antimicrobial claims, public health pests, structural pests such as termites, certain invasive pests, and vertebrates such as rodents, birds, bats and skunks that may transmit disease.²¹ Everything else is waived, which a legal summary notes covers most agricultural products and a large percentage of what is registered.⁵ The safety review is untouched by this and remains extensive. The gap is narrow and specific: whether a product is dangerous is examined closely, and whether it works is generally not.

The data is not absent. Each registrant must ensure through testing that the product is efficacious when used according to label directions, must keep that data, and must produce it on request, with the agency reserving a case-by-case right to ask.²⁶ What is missing is review, which is a different thing: nobody outside the registrant has examined the design, the controls or the reporting, and our own articles treat those as the whole question. The reason the boundary sits where it does is written into the rule. The microorganism exception applies to pests whose presence cannot readily be observed by the user, which makes mandatory review a function of invisibility rather than of stakes, and leaves buyer observation as the substitute check everywhere else. That is a coherent principle and a defensible allocation of finite capacity. It is also a theory of verification that this journal's credence good work disputes, because a pest control outcome is precisely the thing a buyer cannot reliably read: seasonal decline looks like success, an absent population looks like success, and failure looks like reinfestation.

The agency's own stated rationale for performance data is that it ensures products control the pests on the label and that unnecessary exposure does not occur through the use of ineffective products.³ That argument does not stop at the boundary of the waiver: a person who applies a product twice because the first application achieved nothing has doubled their exposure regardless of whether they could see it. The boundary has moved before, with the agency rescinding a proposed waiver for disease-carrying vertebrates, and a 2023 petition sought to move it again toward systemic insecticides, with the response that doing so would be a paradigm shift creating significant administrative burdens.⁴⁵ Meanwhile state guidance records what happens when a claim outruns its evidence: a label saying control where the population fell by half may be treated as exaggerated, and a softer word permitted in its place, from a list running reduce, diminish, lower, weaken, shorten, suppress and subdue.⁶⁷ Which means a verb on a label can record a regulatory finding that the stronger claim was not supported, in a vocabulary no buyer has been given.

References

1. Guide for applicants new to the pesticide registration process, published by the United States federal environmental agency. Official government guidance. Source for the statement that efficacy or product performance data are required if a product will make claims against public health pests, pests that affect building structure such as termites, or certain invasive pests; for the statement that the agency generally does not require submission of efficacy data to support non-public health-related claims, giving control of odour-causing bacteria, weeds and plant pathogens as examples; for the statement that the applicant remains responsible for ensuring that such products perform as intended by developing and keeping efficacy data on file; and for the note that the agency retains the right to require submission of efficacy data for non-public health related claims on a case-by-case basis. https://www.epa.gov/pesticide-registration/how-register-pesticide-guide-applicants-new-process
2. Part 158 of Title 40 of the United States Code of Federal Regulations, data requirements for pesticides, accessed through the official electronic code. Primary regulatory text and the strongest source in this article. Source for the statement that the agency has waived the requirement to submit efficacy data unless the product bears a claim to control pest microorganisms that pose a threat to human health and whose presence cannot readily be observed by the user, including microorganisms infectious to man in any area of the inanimate environment, or a claim to control vertebrates such as rodents, birds, bats, canids and skunks that may directly or indirectly transmit diseases to humans; for the requirement that each registrant must nonetheless ensure through testing that the product is efficacious when used in accordance with label directions and commonly accepted pest control practices; and for the agency's reservation of the right to require, on a case-by-case basis, submission of efficacy data for any pesticide product registered or proposed for registration. https://www.ecfr.gov/current/title-40/chapter-I/subchapter-E/part-158
3. Data requirements for pesticide registration, a page published by the same federal environmental agency. Official government material. Source for the statement that requirements to develop data on product performance provide a mechanism to ensure that pesticide products will control the pests listed on the label and that unnecessary pesticide exposure to the environment will not occur as a result of the use of ineffective products; and for the description of specific performance standards used to validate efficacy data in public health areas, covering disinfectants controlling microorganisms infectious to humans in the inanimate environment and pesticides used to control vertebrates and invertebrates that may directly or indirectly transmit diseases to humans. https://www.epa.gov/pesticide-registration/data-requirements-pesticide-registration
4. Federal Register notice concerning pesticide product performance data requirements for products claiming efficacy against certain invertebrate pests, published in 2021. Official rulemaking record. Source for the account that the agency had proposed an efficacy data waiver for vertebrate control agents and subsequently decided to rescind that proposed waiver with respect to agents intended for control of pests that directly or indirectly transmit disease to humans; for the reproduction of the resulting test note describing the scope of the waiver; and for the reference to the statutory provision under which the agency may waive data requirements. https://www.federalregister.gov/documents/2021/03/22/2021-05137/pesticide-product-performance-data-requirements-for-products-claiming-efficacy-against-certain
5. Legal alert on a petition asking the federal agency to require submission of efficacy data for systemic insecticides, published by a law firm in 2023. Professional commentary rather than a primary document, which we flag because the petitioners' position reaches us only through it. Source for the account that an environmental responsibility organisation, a bird conservancy and other groups petitioned in November 2023 for a rule requiring submission of efficacy data for neonicotinoid and other systemic insecticides for existing, pending and future registrations and review of registered products in light of that data; for the statement that the agency currently requires submission only for a subset of pesticides, notably antimicrobials and products claiming to control public health pests, with most agricultural pesticides outside those groups; for the characterisation that registrants are expected to possess efficacy data but need not submit it unless required case by case; for the summary that the petitioners believe efficacy data would at best show such insecticides are overused and at worst that they are not effective at all; and for the assessment that the requested change would be a paradigm shift in how a large percentage of registered products are evaluated and would create significant administrative burdens. https://www.wiley.law/alert-EPA-Considering-Petition-to-Require-Submittal-of-Efficacy-Data-for-Neonicotinoids-and-Other-Systemic-Insecticides
6. Questions and answers document on efficacy data guidance for minimum risk pesticides, published by an association of state pesticide control officials. Quasi-official guidance describing what states may do rather than binding law, which we flag. Source for the statement that where states evaluate efficacy claims the label and labelling should not exaggerate the extent of control; for the worked example that if mosquito or bed bug populations were reduced by only fifty per cent and the success was overstated on the label as control, the claim might be considered exaggerated; for the statement that a registrant may be required to drop false or unsubstantiated claims or that some states may allow the option of a softer claim to restate an overblown efficacy claim; for the statement that the federal agency does not require industry to submit data for non-public health pest claims and that industry is required to have the data available upon request; and for the record of state discussion about whether raw data rather than complete reports might be accepted provided it is competent and reliable. https://aapco.org/wp-content/uploads/2022/02/25b_QA-Document-for-Data-Efficacy-Guidance-11.29.21.pdf
7. Product efficacy data guidance for minimum risk pesticides, published by the same association of state pesticide control officials. Quasi-official guidance material. Source for the list of softer claim wording including reduce, diminish, lower, weaken, shorten, suppress and subdue; and for the statement that products exempted from federal regulation under the relevant statutory section remain subject to registration by individual states, that states are not required to permit sale of an exempted product simply because it is federally exempt, and that each state may have different label or data requirements. https://aapco.org/wp-content/uploads/2019/03/efficacy-data-guidelines-1-22-19.pdf