The Strategy Everyone Recommends: What the Evidence Actually Says About Rotating Insecticides

1. Introduction: a correction to our own advice

This journal has told readers to rotate modes of action in articles on cockroaches, bed bugs, stored product beetles, rodenticides and head lice. We have never examined whether the advice is supported, and it is the sort of thing a journal should check on itself.

1.1 What this paper concludes in advance

That rotation is probably worth doing, for a reason different from the one usually given, on an evidence base that is almost entirely modelled because the experiment cannot be afforded. Sections 19 and 21 are where that lands.

2. The four strategies

The options, which are more distinct than casual use suggests.

Resistance management can take many forms, including the use of insecticide mixtures, mosaics or alternations and rotations.⁶ A recent model allows insecticides to be deployed as sequences, rotations, being periodic switching of insecticides, or full-dose mixtures, being two insecticides in one formulation.²

2.1 Why the distinctions are not pedantic

They differ in what an individual insect experiences. Under a sequence or a rotation, an insect meets one compound in its lifetime. Under a mosaic it meets one, unless it moves. Under a mixture it meets both at once.

Selection acts on individuals, so what an individual encounters is the whole question. Strategies that look similar from a purchasing perspective are entirely different from the population's point of view.

2.2 What defines a mixture specifically

The fundamental strategy concept of a mixture is defined by the simultaneous use of insecticides and their overlapping exposure.⁴

Overlapping exposure is the operative phrase. An insect encountering a mixture meets both actives; an insect under a rotation meets one, then later meets the other, and may not be the same insect.

3. What a sequence is

The benchmark, and the thing most of this trade actually does.

A sequence is continuous deployment until a defined withdrawal threshold, termed insecticide lifespan, as indicated by resistance diagnosis in bioassays.² It is described as the nonstrategy, a switch in the solo use of insecticides when they fail due to resistance evolution.⁴

3.1 Why the word nonstrategy matters

It is what happens by default when nobody manages anything. A product is used because it works, continues to be used, stops working, and gets replaced.

That is the history this journal documented for warfarin, for pyrethroids in bed bugs, and for phosphine in stored grain. None of those sequences were chosen; they occurred, and in each case the next product was reached for after the previous one had already failed rather than before.

3.2 The uncomfortable implication

If rotation gains little over the nonstrategy, then decades of resistance management advice has been recommending something close to what would have happened anyway.

Sections 8 and 19 qualify that considerably. It is the right place to start because it is where the modelling literature started.

4. What the modelling found

The history of the question.

In the 1980s, a series of mathematical models based on population genetics sought to disentangle early speculation on when different approaches to using multiple insecticides would delay the evolution of resistance, and that modelling of mosaics and rotations tended to show minimal gains on sequences.⁴

4.1 What the models were comparing

Time to resistance, under stated assumptions about the genetics, the selection pressure and the deployment pattern. A strategy wins by delaying the point at which control fails.

That is a reasonable objective and it is not the only one, which §9 develops into the main qualification of this article.

4.2 The scale of the literature

That statement is supported by a string of citations spanning several decades, alongside work on the tactical manipulation of immigration, refugia and dominance studied in their own right.⁴

This is not one dissenting model. It is the general direction of a mature theoretical literature.

5. Why that result is uncomfortable

Because rotation is stated as settled practice.

Trade and industry framing describes rotating of modes of action as the most used and effective resistance management approach.⁶ A 2024 perspective opens by noting that rotations have been the cornerstone of insecticide resistance management for many decades.¹

5.1 The gap between practice and theory

An approach described as the most used and effective, sitting on a modelling literature that found it gaining little over doing nothing in particular.

That gap is the article. It does not resolve into rotation being useless, and §17 to §19 are why, but it is a real divergence between what is recommended and what was demonstrated.

6. The monogenic assumption

A limitation of the older models that matters.

Previous models assumed resistance to have a simple, monogenic basis, whereas in natural populations resistance will often be a complex polygenic trait determined by multiple genetic variants.²

6.1 Why the assumption was made

Because single-gene resistance is tractable. A model with one locus and two alleles can be solved; one with many loci of small effect requires a different formalism.

6.2 Why it matters for the conclusion

Single-gene resistance behaves differently from quantitative resistance. A major-gene mutation is present or absent; a polygenic trait shifts gradually and can respond to selection in smaller increments.

Which means the older conclusion in §4 might be an artefact of the assumption rather than a finding about the world, and §7 is the test of that.

7. The polygenic model

The recent work, and its design.

A quantitative genetics model was developed to model resistance as a polygenic trait, assumed to be encoded by a large number of genes, each with very small effect on the phenotype.² Strategies were compared on their strategy lifespan, capped at 500 generations, with statistical techniques used to identify and quantify parameters driving the evolution of resistance.²

7.1 What the polygenic assumption buys

Modelling resistance as many genes of small effect means genetic variance does not change substantially over selection,² so the population responds continuously rather than by fixing a single allele.

That is closer to what the metabolic resistance literature describes, where detoxification capacity rises gradually rather than switching on. It also means resistance can keep increasing rather than reaching a ceiling.

7.2 Five hundred generations

For a German cockroach at roughly two months per generation that is over eighty years. The model is asking a question no observation could.

8. What it found about rotation

The headline result, which largely confirms the older literature.

Deploying single insecticides as sequences or rotations usually made little overall difference to their strategy lifespan.²

8.1 The word usually

Usually made little overall difference.² Which implies conditions under which it made a substantial one, and the parameters in §20 are presumably where those conditions live.

8.2 So the older conclusion survives

Relaxing the monogenic assumption did not rescue rotation on this measure. A model built specifically to be more realistic about the genetics reached the same place.

That is a stronger result than either study alone, because the two use different formalisms and different assumptions.

9. The distinction between lifespan and level

The qualification in the same sentence, which is where rotation earns something.

Sequences and rotations differed little in lifespan, though rotations displayed lower mean and peak resistances.²

9.1 Two different things to want

How long a strategy remains usable, and how resistant the population gets along the way.

Rotation does not extend the first much. It does reduce the second, which means at any given moment the population under rotation is less resistant than the population under a sequence, even if both arrive at failure at similar times.

9.2 Why the two measures came apart

Strategy lifespan is set by when the last usable product fails. Mean and peak resistance describe the state of the population throughout.

A strategy can keep the population less resistant on average while still arriving at exhaustion at much the same time, which is exactly what the model reports and is the pattern §19.1 explains.

9.3 Why lower peak resistance is worth having

Because control quality in the intervening years is a function of resistance level, not of whether the threshold has been crossed.

A programme that keeps resistance lower throughout is delivering better control throughout, which is what a client experiences. The strategy lifespan is what the industry experiences. That distinction is ours.

10. What it found about mixtures

The result that has revived the argument.

Deploying two insecticides in a full-dose mixture formulation was found to extend the strategy lifespan when compared to deploying each in sequence or rotation. This pattern was observed regardless of the level of cross resistance between the insecticides or the starting level of resistance.²

10.1 The strength of that claim

Regardless of cross resistance is unusually strong. One would expect mixtures to lose their advantage where the two actives share a resistance mechanism, and the model reports the pattern holding anyway.

10.2 The resurgence it belongs to

A 2024 perspective notes that in recent years, there has been a resurgence of interest in the use of insecticide mixtures, particularly based on new theoretical models.¹ The use of mixtures for resistance management has been a controversial topic for many decades.⁴

11. The contradiction with extension guidance

Where this article stops being tidy.

Extension guidance for tree fruit states that mixtures rest on the concept that insects resistant to one pesticide will be killed by the other components of the mixture and that few pests will be resistant to the entire mixture, but that though mixtures of fungicides have been used for years to combat resistance, both field experience and models have shown that mixtures should be avoided whenever possible with insects and spider mites.⁵

11.1 The fungicide comparison in that passage

Notice that the same guidance accepts mixtures for fungicides and rejects them for insects and mites.⁵ So the objection is not to mixtures in principle.

Something about the biology of the organisms, or the resistance mechanisms available to them, is being treated as the reason the same tactic works in one case and not the other. The source does not say what.

11.2 The direct conflict

A recent model finds mixtures extend strategy lifespan beyond rotation. Extension guidance says field experience and models show mixtures should be avoided with insects.

We are not able to resolve that, and we are not going to pretend otherwise. Both are reputable sources making confident statements in opposite directions.

12. How to hold both

What a reader should do with a live disagreement.

The possibilities are that the extension guidance predates the recent modelling and reflects the earlier consensus; that the model's assumptions favour mixtures in ways field conditions do not; that mixtures behave differently in agriculture and in public health; or that the disagreement concerns half-dose against full-dose mixtures, which are different propositions.

12.1 The half-dose question specifically

The model tested full-dose mixtures.² A mixture at full rates of both actives is a different object from two products at half rates each, and the objection that mixtures expose a population to two compounds at sub-lethal levels applies to the second and not the first.

That is our hypothesis for reconciling them and we have not verified it. We flag it as the most likely explanation rather than the answer.

12.2 The practical consequence for us

Tank mixing is not something this trade does casually, and label conditions govern whether it is permitted at all, which the label article in this journal set out. The disagreement is therefore mostly academic for structural work, and we would not act on the mixture result.

13. The cage experiment

The closest thing to direct evidence we located.

In greenhouse cage experiments, a laboratory-selected population of diamondback moth resistant to spinosad, indoxacarb and Bt was used to compare population growth and resistance evolution if these three insecticides were rotated or used in a mosaic fashion.³

The average population density through nine generations was lowest in the treatment in which the insecticide was rotated every generation, at 20.7, compared with 41.4 where the insecticide was rotated every third generation or 41.8 where the insecticides were applied as a mosaic.³

Results of both population density and resistance development indicated that insecticide rotation every generation was better for resistance management than if the insecticide was rotated every third generation or if the three insecticides were applied as a mosaic.³

13.1 Why this design is unusually informative

The population was already resistant to all three products before the experiment began.³ So the question was not whether a strategy prevents resistance arising, but how it performs against a population that already has it.

That is the situation most of this trade is actually in, which makes the design closer to practice than a study starting from a susceptible population would be.

13.2 The two outcomes measured

Population density and resistance development, and they agreed.³ An agreement between a control outcome and an evolutionary outcome is worth noting, because the two can diverge and the previous article in this journal described a case where a falling count meant a failing instrument rather than a falling population.

14. What frequency does

The variable the cage study isolates, which the modelling discussion tends to skip.

Rotating every generation produced roughly half the population density of rotating every third generation.³

14.1 Why this is the most actionable result in the article

It does not ask whether to rotate. It asks how often, and finds a large difference between two schedules that would both be described as rotation.

A programme switching products annually and one switching every service visit are both rotating. If the cage result transfers, they are not equivalent.

14.2 What generation-scale rotation would mean here

A German cockroach generation is a matter of months under warm indoor conditions, which would put the switch at roughly every second or third service visit on a monthly programme.

For a bed bug it is longer; for a rodent longer still. The instruction rotate every generation is therefore species-specific, and translating it is a step the sources do not take. This paragraph is our extrapolation and should be treated as such.

14.3 The caution

One experiment, in cages, on a laboratory-selected population, with one pest and three specific products.³ Generation-scale rotation is also operationally demanding for an insect whose generation time is measured in weeks.

14.4 Why the mosaic performed poorly

A mosaic applies different products in different places simultaneously, so an individual experiences one product and its offspring may experience another only if they move.

In a cage, movement between patches is easy, which should favour mosaics if anything. That it performed no better than infrequent rotation is therefore a reasonably hard result, and the reasoning is ours.

15. Why field evidence is so thin

The structural reason this question is settled by models.

Field trials of resistance management strategies are usually prohibitively expensive with long timeframes, and mathematical modelling is often used to evaluate alternative options.² Evaluation is challenging due to the number of potential strategies, the need for replication over different ecological and epidemiological settings, and the long study durations needed to observe phenotypic changes in resistance.⁸

Cluster randomised controlled trials used for evaluating vector control tools generally last no longer than 2 to 3 years, which is unlikely to be of sufficient duration to detect the benefit of any one strategy over another, while modelling can simulate the response to strategies over decades.⁸

15.1 Why replication makes it worse

It is not enough to run one long trial. The comparison needs replication across ecological settings, because a result from one site under one set of conditions would not generalise, and the number of candidate strategies multiplies the arms required.⁸

A properly powered field evaluation would therefore be many decade-long trials running in parallel, which is why nobody has attempted it.

15.2 The arithmetic of the problem

The best-funded trials in this area run for two to three years. The phenomenon being managed unfolds over decades. The experiment that would answer the question is longer than the careers of the people who would run it.

16. What that means epistemically

How much weight a modelled answer should carry.

Models are not weak evidence in general. They are the only available evidence here, which is a different situation from being the preferred evidence.

16.1 What a model can and cannot settle

It can establish what follows from a set of assumptions about genetics, selection and deployment. It cannot establish that those assumptions describe a cockroach population in an apartment block.

Section 6 is the illustration: a whole literature's conclusion turned partly on an assumption about the genetic architecture, and changing that assumption was a research programme in itself.

16.2 The comparison with the previous article

The insect light trap article found an instrument whose degradation was invisible and recommended a calendar rule in place of measurement. Here the whole subject is beyond measurement, and the substitute is simulation.

Both are cases of acting on a proxy because the direct observation is unavailable, and in both the honest position is to say which one you are using.

16.3 Why we are not dismissing it

Because the alternative is nothing. A field of practice that refused modelled evidence where trials are impossible would be a field with no basis for any recommendation at all, and this journal has criticised exactly that absence in other contexts.

17. The three explanations for rotation

Why rotation should work, according to the literature.

Three explanations have been offered for how rotations delay resistance evolution: counterselection from resistance cost, the relaxation of selection and intergenerational redundant kill. The 2024 perspective shows that all three explanations can make sense of the comparison of rotations with another resistance-management strategy but have failed to elucidate the principle at work.¹

17.1 Why three explanations is itself a warning sign

A practice with three competing mechanistic accounts, none of which the recent review finds sufficient, is a practice adopted before it was understood.

That is not unusual and not disqualifying. This journal's article on horizontal transfer described a technique that worked for years before the mechanism was characterised. But it does explain why the models disagreed: they were testing different things while using the same word.

17.2 The first explanation and its problem

Counterselection assumes resistance carries a fitness cost, so a withdrawn product allows susceptibility to return. This journal's cockroach article found fitness costs real but reversion timescales exceeding service intervals, which limits how much the mechanism can deliver.

17.3 The third explanation

Intergenerational redundant kill is the rotation version of what a mixture does simultaneously: survivors of one product meet a different one later.

It works only if the survivors and their descendants are still present when the second product arrives, which makes it dependent on rotation frequency in the way §14 measured. A rotation with a long interval gives the survivors time to become the whole population before anything else happens to them.

18. The resolution offered

What the recent perspective concludes.

The principles of resistance management have previously been reduced to moderation, saturation and multiple attack, with rotations presented alongside mixtures and mosaics as a strategy of multiple attack in using more than one insecticide.¹

The authors argue instead that rotations work by moderation, delaying resistance to insecticides through the use of each insecticide less over time, and suggest the principles are recast as moderation, saturation and redundancy.⁷

18.1 Why the old classification misled

Grouping rotation with mixtures and mosaics as multiple attack implied they work the same way and should be compared on the same measure. That framing made rotation look like a weak mixture, since it delivers the multiple compounds separated in time rather than together.

If rotation is not a multiple attack strategy at all, the comparison was never the right one, which may be much of why forty years of models found it unimpressive.

18.2 The reclassification

Rotation has been filed under the wrong heading. It is not a way of hitting the population with multiple things; it is a way of using each thing less.

19. Why moderation is a better account

The explanatory advantage, which we think is substantial.

Under moderation, rotation's benefit follows directly from the quantity of selection applied. Use a product for a third of the time and it experiences a third of the selection pressure, which delays resistance to it by roughly that factor.

19.1 It explains the modelling result

If rotation delays resistance to each product by using each less, then across the whole set the total time until everything fails need not change much, which is §4 and §8.

Each product lasts longer; there are the same number of products; the arsenal empties at a similar time. What changes is the resistance level at any given moment, which is §9.

19.2 It explains the frequency result

Moderation says the benefit depends on how little each product is used and how evenly, which makes frequent switching better than infrequent, matching §14.

That connection is ours; the sources do not join the frequency experiment to the moderation argument.

19.3 What moderation implies about doing nothing

If the mechanism is using each product less, then reducing total insecticide use achieves part of the same effect without any rotation at all.

That points back at every non-chemical measure this journal has covered: exclusion, sanitation, moisture control and monitoring all reduce how much product a building needs, and on this account they are resistance management as well as control. That inference is ours.

19.4 The cooperative conclusion

When rotations and mixtures are not conceptualised as competing methods of multiple attack, these strategies can more obviously work together through the complementary principles of moderation and redundancy, and whether solo products or a mixture of products are used, rotations are an effective method of risk management, preserving the arsenal of all effective insecticides for longer.⁷

20. What the parameters that matter are

The model's own account of what drives the outcome.

Statistical analysis highlighted the importance of insecticide coverage, cross resistance, heritability, and fitness costs for selecting an appropriate strategy.²

20.1 The one an operator controls

Coverage. How much of the population is exposed to the treatment is an operational variable, and this journal's article on attached housing argued that partial treatment leaves an untreated reservoir. On the moderation account in §18, coverage and moderation pull in opposite directions, since high coverage means more of the population selected and low coverage means a refuge. The sources do not reconcile that and neither can we.

Cross resistance is a property of the chemistry, heritability of the genetics, and fitness costs of the organism. Coverage is the one determined by how the work is done.

20.2 Why heritability appears on that list

Heritability governs how much of the variation in resistance passes to the next generation, and therefore how fast selection can move the population.

It is not something an operator influences, but it explains why some species develop resistance quickly and others do not, and why the answer to should I rotate may differ between a German cockroach and a carpenter ant. That connection is ours.

20.3 Cross resistance is the one that voids the strategy

Two products in different named groups that share a detoxification mechanism are not a rotation in any meaningful sense. The population experiences continuous selection on the same trait.

That is why this journal has repeatedly insisted on reading the active ingredient rather than the trade name, and it is the single easiest way for a rotation to be nominal.

21. What survives for this trade

Keep rotating. It lowers mean and peak resistance even where lifespan is similar.²

Rotate more often than you think. Every generation beat every third generation.³

Check the group, not the label. Cross resistance is named as a driving parameter.²

Treat coverage as a resistance variable. It is the parameter you control.²

Understand it as using less. Moderation rather than multiple attack.⁷

Do not expect proof. The trial has never been run and probably will not be.⁸

Count non-chemical measures as part of it. On the moderation account, anything that reduces how much product a building needs is resistance management.

22. Limitations and open questions

No structural pest evidence at all. The literature here is agricultural and public health. We located nothing evaluating rotation strategies for cockroaches, bed bugs or rodents in buildings, which is the setting this journal is about.

The evidence base is models. Stated in §15 and §16, and it is the central limitation of the whole subject rather than of this article.²⁸

The cage experiment is a single study. One pest, three products, laboratory selected population, greenhouse cages, nine generations.³

We have not resolved the mixture disagreement. Section 11 sets out the conflict and §12.1 offers an unverified hypothesis.²⁵

We have read abstracts and summary sections. The models' assumptions, particularly about fitness costs and heritability, are where their conclusions live and we have not assessed them.

Rotation frequency for slow-generation pests is untreated. The frequency result concerns generations, and translating that into service intervals for organisms with different generation times is not addressed by anything we found.

Sections 3.2, 5.1, 9.2, 12.1, 14.3, 16, 19.2 and 20.1 are our reasoning. The nonstrategy implication, the practice-theory gap framing, the argument for lower peak resistance, the half-dose hypothesis, the mosaic reasoning, the epistemic assessment, the link between moderation and frequency, and the coverage point are ours rather than sourced positions.

Our commercial position. This article examines advice we have given repeatedly and finds the evidence weaker and the mechanism different from what we implied. It concludes that the advice was right for the wrong reason, which is a less comfortable result than either confirming or abandoning it.

23. Conclusion

Rotation has been the cornerstone of resistance management for decades and is described as the most used and effective approach.¹⁶ The modelling literature from the 1980s onward found mosaics and rotations gaining little over sequences, where a sequence is simply using each product until it fails.⁴ A recent model built specifically to treat resistance as a polygenic trait reached the same conclusion on strategy lifespan.²

What rotation does deliver, in that model, is lower mean and peak resistance,² which is what a client experiences as control quality even where the arsenal empties at a similar time. A cage experiment found rotating every generation produced half the population density of rotating every third, so frequency matters more than the binary question.³ And the best current account of why any of it works is not multiple attack but moderation: each product simply gets used less.⁷

None of this is settled, the field trial that would settle it runs longer than anyone funds, and extension guidance and recent models still disagree about mixtures. We are going to keep recommending rotation, and we will describe it differently: not as hitting the population with several things, but as spending each thing more slowly. It is a smaller claim and it is the one the evidence supports.

References

1. Madgwick, P. and colleagues (2024). What is the value of rotations to insecticide resistance management? Pest Management Science. doi:10.1002/ps.7939. Used for the statements that rotations have been the cornerstone of insecticide resistance management for many decades; that in recent years there has been a resurgence of interest in the use of insecticide mixtures, particularly based on new theoretical models; that the principles of resistance management have previously been reduced to moderation, saturation and multiple attack, with rotations presented alongside mixtures and mosaics as a strategy of multiple attack; and that three explanations have been offered for how rotations delay resistance evolution, being counterselection from resistance cost, the relaxation of selection and intergenerational redundant kill, all three of which can make sense of comparisons between strategies but have failed to elucidate the principle at work. https://scijournals.onlinelibrary.wiley.com/doi/10.1002/ps.7939
2. Insecticide resistance management strategies for public health control of mosquitoes exhibiting polygenic resistance: a comparison of sequences, rotations, and mixtures. PubMed Central PMC10130562. Principal modelling source. Used for the statements that field trials of resistance management strategies are usually prohibitively expensive with long timeframes so that mathematical modelling is often used to evaluate alternative options; that previous models assumed resistance to have a simple monogenic basis whereas in natural populations it will often be a complex polygenic trait determined by multiple genetic variants; for the development of a quantitative genetics model treating resistance as a polygenic trait encoded by a large number of genes each of very small effect; for the definitions of sequences as continuous deployment until a defined withdrawal threshold termed insecticide lifespan as indicated by resistance diagnosis in bioassays, rotations as periodic switching, and full-dose mixtures as two insecticides in one formulation; for the comparison of strategies on strategy lifespan capped at 500 generations; for the findings that deploying single insecticides as sequences or rotations usually made little overall difference to strategy lifespan though rotations displayed lower mean and peak resistances, and that deploying two insecticides in a full-dose mixture extended strategy lifespan compared with deploying each in sequence or rotation regardless of the level of cross resistance or the starting level of resistance; and for the statistical finding highlighting the importance of insecticide coverage, cross resistance, heritability and fitness costs in selecting an appropriate strategy. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10130562/
3. Zhao, J. and colleagues. Testing insecticide resistance management strategies: mosaic versus rotations. PubMed record 20552665. Used for the account that the diamondback moth has demonstrated an ability to develop resistance to many different classes of insecticides including proteins produced by Bacillus thuringiensis, and more recently to spinosad and indoxacarb; for the design in which greenhouse cage experiments used a laboratory-selected population resistant to spinosad, indoxacarb and Bacillus thuringiensis to compare population growth and resistance evolution under rotation or mosaic use; for the result that average population density through nine generations was lowest where the insecticide was rotated every generation at 20.7 plus or minus 3.20, compared with 41.4 plus or minus 17.6 where rotated every third generation and 41.8 plus or minus 6.53 under a mosaic; and for the conclusion that results of both population density and resistance development indicated rotation every generation was better for resistance management than rotation every third generation or mosaic application. https://pubmed.ncbi.nlm.nih.gov/20552665/
4. Beyond redundant kill: a fundamental explanation of how insecticide mixtures work for resistance management. PubMed Central PMC10092901. Used for the statements that the use of insecticide mixtures for resistance management has been a controversial topic for many decades; that the fundamental strategy concept of a mixture is defined by the simultaneous use of insecticides and their overlapping exposure; that a role exists for redundant kill acting against resistant individuals, linked to overlapping exposure; that in the 1980s a series of mathematical models based on population genetics sought to disentangle early speculation on when different approaches to using multiple insecticides would delay resistance; and that modelling of mosaics and rotations tended to show minimal gains on sequences, often viewed as the nonstrategy benchmark where there is a switch in the solo use of insecticides when they fail due to resistance evolution, alongside study of the tactical manipulation of immigration, refugia and dominance. https://pmc.ncbi.nlm.nih.gov/articles/PMC10092901/
5. Managing Pesticide Resistance. Washington State University Tree Fruit extension. Extension guidance. Used for the statements that resistance manifest in the field results in measurable reductions in the relative efficacy of a pesticide and that estimating this reduction is an important step in managing resistance; that management by multiple attack involves using either mixtures or rotations to thwart resistance; that the use of mixtures such as tank mixes of two or more pesticides is based on the concept that insects resistant to one will be killed by the other components and that few pests will be resistant to the entire mixture; and that though mixtures of fungicides have been used for years to combat resistance, both field experience and models have shown that mixtures should be avoided whenever possible with insects and spider mites. https://treefruit.wsu.edu/crop-protection/opm/resistance/
6. Testing insecticide resistance management strategies: mosaic versus rotations, research repository record with associated citing commentary. Used for the statements that resistance management can take many forms including the use of insecticide mixtures, mosaics or alternations and rotations; that rotating of modes of action is described as the most used and effective resistance management approach; and that the importance of effective and proactive resistance management to maintain the efficacy of current and future insecticides has long been recognised by the agrochemical industry. https://www.researchgate.net/publication/44677621_Testing_insecticide_resistancemanagement_strategies_Mosaic_versus_rotations
7. What is the value of rotations to insecticide resistance management? Repository abstract record. Used for the authors' conclusions that rotations work by moderation, delaying resistance to insecticides through the use of each insecticide less over time; that the principles of resistance management should be recast as moderation, saturation and redundancy; that when rotations and mixtures are not conceptualised as competing methods of multiple attack these strategies can more obviously work together through the complementary principles of moderation and redundancy; and that whether solo products or a mixture of products are used, rotations are an effective method of risk management, preserving the arsenal of all effective insecticides for longer, with a successful resistance-management plan making appropriate use of all the principles. https://ngdc.cncb.ac.cn/openlb/publication/OLB-PM-38173134
8. Insecticide resistance management strategies for public health control of mosquitoes exhibiting polygenic resistance, preprint full text. Used for the statements that evaluating resistance management strategies in the laboratory and field is challenging due to the number of potential strategies, the need for replication over different ecological and epidemiological settings, and the long study durations needed to observe phenotypic changes in resistance; that cluster randomised controlled trials used for evaluating the epidemiological effectiveness of malaria vector control tools generally last no longer than two to three years, which is unlikely to be of sufficient duration to detect the benefit of any one strategy over another; and that mathematical modelling and computer simulations can simulate the resistance response to strategies over decades and therefore provide a valuable evaluation tool. https://www.biorxiv.org/content/10.1101/2022.10.28.514211.full.pdf