• Motivation: To understand the current practices in rodent pest control and identify levers to improve the welfare of the rodents.
  • Scope: This paper describes the past and current rodent pest control landscape with a particular focus on harms of rodenticidal poisons and the possibilities for rodent birth control as a cruelty-minimizing alternative. Because rodent birth control is most available in the United States, the focus of this paper is primarily on the U.S.
  • Welfare harms: Rodenticides, and in particular the anticoagulant rodenticides, are cruel to the rodents they are meant to kill as well as dangerous to humans, such as children, who accidentally ingest them and to predators who secondarily ingest the poisons their rodent prey have ingested.
  • Ideal solutions: Controlling a “pest” population can also be achieved through reducing births, greatly reducing the need for lethal control means. Methods for reducing births include:
    • Resource reduction (by e.g. containing food waste)
    • Habitat reduction (by e.g. filling in abandoned burrows or filling in cracks that give access to the interior spaces of walls)
    • Effective birth control agents (such as EPA-approved ContraPest)
  • Practical solutions: Rodent birth control can be used in concert with lethal control methods that are more humane than rodenticides, such as traps and asphyxiants that provide quick deaths. These methods and their relative humaneness and effectiveness are discussed.
  • Ineffective solutions:
    • Most of the known chemicals that reduce rodent fertility have not been developed to be sufficiently effective, cost-effective, or safe to be competitive with lethal control agents.
    • Single-dose sterilants, which would completely abolish fertility in one dose, would come with a higher risk of overdose, bioaccumulation, and the evolution of resistance. 
  • Future directions: A follow-up post will focus on opportunities for impact via advocating the use of rodent birth control and the displacement of cruel pest control methods.

The cruel status quo: rodenticides

鼠目寸光 (shǔ mù cùn guāng): "Rats are shortsighted." A metaphor for people who only care about short-term benefits. – The Bejinger, 2020


Dealing with commensal rodents is a problem as old as civilization. Dogs and cats have long earned their keep with humans in their jobs as rodent exterminators (“ratters”) and (warning: contains links to videos of rat hunting) still do today. Although it’s not a nice word, there’s a reason commensal rodents are called “pests.” [1] They destroy our crops (e.g., rodents cause about 30% of annual damage of rice crops in Asia), eat and contaminate our stored food (rats alone are estimated to be responsible for at least $27 billion in destroyed grain in the U.S. per year), leave droppings and fluids that can spread disease, and sometimes even attack us

Around the world, poison is a frontline rodent control management tool. Typical pest management of rodents consists of trying to reduce obvious food and shelter available to them and then poisoning some of the animals when their numbers get large enough that their presence is noticeable. The chemical agents used to kill rodents are known as “rodenticides.”

There are dozens of modern rodenticides. Many of us think of arsenic when we hear the term “rat poison,” and it’s true that the heavy metals arsenic and thallium used to be sold for rodent control, along with a bevy of other chemicals that are toxic to humans like strychnine and elemental phosphorus. Because most rodenticides are also deadly to humans and other animal species, their sales tend to be restricted. 

Common rodenticides today include the anticoagulants, bromethalin, cholecalciferol (vitamin D3), and zinc phosphide. The anticoagulants are divided into two categories: 

  • First generation anticoagulants (also known as “multi-dose anticoagulants”) were developed before 1970, and include warfarin (which is also prescribed as a human blood thinner) and the “superwarfarins” chlorophacinone and diphacinone. First-generation anticoagulants work best when consumed by a rodent many times over a period of days.
  • Second generation anticoagulants were developed after 1970 in response to rodents evolving resistance to first-generation anticoagulants. Second generation anticoagulants include the stronger superwarfarins brodifacoum, bromadiolone, difenacoum, and difethialone. They can kill a rodent in a single dose.


Both first and second generation anticoagulants usually cause death within 5–7 days of initial ingestion. Humans trying to get rid of pests and not particularly concerned about humaneness tend to prefer that a rodenticide does not cause immediate death because rodents become “bait-shy” when they have observed other rodents eating a bait and subsequently acting strangely or dying. The delay between consumption of the poison and death also gives the unfortunate rodent time to bring back the poison to its nest and die there, perhaps after sharing the poisoned bait with its family. The first-generation anticoagulants take that long in part because the poison accumulates to a toxic dose over several feedings. Second-generation anticoagulants are more potent, such that one feeding usually constitutes a fatal dose, though that dose takes 5–7 days to kill. 

Because of their higher potency, second-generation anticoagulants in particular create an ecological problem. The rodent is able to eat many times the toxic dose before dying, and can pass the poison up the food chain if eaten by a predator (secondary ingestion). To make things worse, rodents who have been poisoned with anticoagulants are slow, confused, and seem to prefer to wander into open areas, where they are more likely to be eaten by predators. A recent study of recovered bald and golden eagles from across the U.S. found that 82% of the animals had anticoagulant rodenticide in their livers

Rodenticides can cause days of agony

The anticoagulant rodenticides interfere with vitamin K-1 metabolism, which interferes with platelet-mediated clotting. Without clotting, the daily wear and tear damage to blood vessels is not repaired, which eventually causes death by blood loss or organ failure due to internal bleeding. This can happen to humans who take the anticoagulant drug warfarin (also a first generation anticoagulant rodenticide) without having their blood clotting carefully monitored– after ingesting an anticoagulant, simple traumas are more likely to lead to internal bleeding, which can can be painful or life-threatening. Bleeding is not usually painful in itself, but the accumulation of blood in closed spaces like the abdomen or around the heart is likely to be agonizing. Bleeding into the lungs or airways can make breathing difficult and lead to air hunger and a feeling of drowning, which, as most humans have some ability to imagine, are highly distressing. One study found that a third of rats poisoned with brodifacoum developed impaired movement followed by paralysis two days before death, and less eating and moving a day before that. Internal bleeding that impairs the function of internal organs can be painful before it is deadly, as humans who have suffered abdominal bleeding describe. 

The non-anticoagulant pesticides receive less attention, but are also cruel. 

  • Bromethalin is a neurotoxin that kills by causing brain and/or nerve damage leading to respiratory distress, which is also likely extremely painful (in this human bromethalin poisoning case, the poisoned person ultimately died of cardiac arrest after suffering brain death due to edema the day before).
  • Cholecalciferol (vitamin D3), now the active ingredient in most direct-to-consumer products by the pest control company d-Con, causes death by acute hypercalcemia. Hypercalcemia causes multi-system failures, but humans who have suffered with it complain most of the pain of the excess calcium causing pancreatitis.Veterinarians have particular concern about cholecalciferol and bromethalin being accidentally ingested by companion animals because, unlike anticoagulant rodenticides, which can be treated by providing vitamin K-1, there is no specific antidote to either of these poisons. Worse, by the time an animal is showing symptoms of poisoning by cholecalciferol or bromethalin, it is generally too late to treat them.
  • Zinc phosphine “liberates toxic phosphine gas” from the lining of the stomach. Immediate death occurs from fluid collecting in the lungs, which causes breathlessness and the sensation of drowning, and delayed deaths are due to organ damage from the phosphide, frequently damage to the heart, which can cause chest pain and symptoms of low blood oxygenation.

Because of the dangers these poisons pose to human consumers and to other animals (and therefore ecosystems) via secondary ingestion, the U.S. Environmental Protection Agency (EPA) restricts the purchase of rodenticides. Second-generation anticoagulants can only be legally purchased by pest management professionals, and all rodenticides sold to consumers must be housed in a bait station and in the form of a block, rather than loose pellets, for the safety of children and pets. Other than consumer safety, the EPA’s greatest concern is the effect of rodenticides on non-target wildlife. In 2018, d-Con switched to using cholecalciferol because the EPA ruled that their anticoagulant-based products “pose unacceptable risks to non-target wildlife.” 

Wildlife who accidentally ingest these poisons may suffer greatly as well. Predators can ingest many times a fatal dose of rodenticides for their size by eating multiple poisoned rodents and suffer the same agony as the target rodents. Audubon magazine reports on necropsy findings from poisoned birds: “There was the great horned owl with a hematoma running the length of its left wing; the red-tailed hawk’s body cavity glistening with unclotted blood; sundry raptors with pools of blood under dissected skin; the redtail with a hematoma that had ballooned its left eye to 10 times normal size; and, “saddest of all,” the redtail with an egg. The well-developed blood vessels in her oviducts had ruptured, and she had slowly bled to death from the inside.”  Erin Ryan shares that, in addition to internal bleeding, birds of prey poisoned with rodenticides even at sublethal doses are more likely to be killed by secondary causes, such as being hit by cars. 

A humane alternative: birth control

There is an alternative to killing to keep rodent numbers down: birth control. Anything that causes a reduction in fertility[2] compared to natural conditions could be used as birth control. I’ll be calling anything that reduces fertility “birth control” or “contraception.” A chemical that abolishes fertility permanently is called a “sterilant.” 

There are many chemicals that reduce the fertility of rodents, some of which are already being developed for use in agriculture or in use in residential pest control. Most of these compounds are toxic at doses near their effective sterilant dose, or would be too expensive or impractical to use as a multi-dose fertility reduction tool. Some work by inducing abortions, which many consider less than ideal. And still others have unacceptable effects on non-target species (for example, gossypol is relatively safe for humans and rodents but toxic to dogs). Even a birth control agent that is safe for target animals to ingest can have a negative impact on an ecosystem if it’s eaten by an endangered or keystone species. However, for rats (and potentially other mammals after obtaining the proper approvals), there is an effective birth control that does not cause harm to rodents or non-target species: ContraPest, which is sold by SenesTech to both licensed pest managers and consumers in the United States. 

Improper exposure to birth control could harm non-target animals or animals of ecosystems by reducing fertility or by bioaccumulating, which seems to be why they are regulated by the EPA in the same class as rodenticides. However, the potential harm of birth control is much less severe than that of rodenticides. Lethal pesticides affect non-target populations by killing or causing physical damage to individuals. A wild animal overdosing on ContraPest at rodent-sized doses would be more likely to simply have reduced fertility than to, for example, suffer internal bleeding and potentially death. The EPA does not properly credit birth control for being more humane than lethal control, both for target and non-target species.

Toward humane integrated pest management

Whether at the scale of city streets or a single residence, pest control efforts are more effective when applied at multiple fronts. To reduce or eliminate a population, there are four levers: reduce births, increase deaths, increase emigration (moving out), and reduce immigration (moving in).[3] [4]“Integrated pest management” is the term used by the pest management industry to describe these multi-front efforts that tackle pest populations from an angle besides poison. 

Table 1: General classification of pest control methods by their means of population control. 

Reduce birthsIncrease deathsIncrease emigrationReduce immigration
  • Reduce resources (food, water)
  • Reduce habitat
  • Provide contraception
  • Poison
  • Lethal traps
  • Ratter dogs and cats
  • Reduce resources (food, water)
  • Reduce habitat
  • Seal off entry routes
  • Relocation methods (humane traps)
  • Reduce resources (food, water)
  • Reduce habitat


Ideal approach: Cruelty-free pest management

Looking at Table 1 above, it’s clear that some methods work on multiple fronts. Reducing rodents’ access to resources (food and water) and habitat work to reduce births, increase emigration, and reduce immigration. (Though they probably also increase deaths for rodents unable to migrate away from resource-deprived environments.) Methods like these, which reduce the resources needed to reproduce, and chemical contraception, which reduces competency to reproduce, are generally the most humane.

Resource reduction 

Reducing resources available to rodents mainly means containing food and garbage, and sealing or removing sources of standing water. Some effective household-level approaches include:

  • Making food containers in the home rodent-proof by replacing the original cardboard packaging with thicker plastic or glass containers.
  • Sealing compost inside the home until it’s ready to go outside, and once it’s outside, covering it as often as possible to prevent it becoming a food source or shelter for outdoor rodents (composters have many tips for keeping rodents out of compost).
  • Frequently cleaning where food has been consumed, as well as cleaning hard-to-reach spots (such as behind appliances and under sinks), as lingering crumbs and scents can attract rodents.
  • Keeping outdoor garbage and recycling bins closed.

Beyond households, containing garbage at every step of the collection process is a bigger task than it may at first appear. Some effective approaches include: 

  • Using smart bins or other covered public trash cans in public areas.
  • Enforcing high garbage collection standards for all residents of a municipality, and paying for greater containment for every step of the way to the garbage’s final destination.
  • Keeping garbage contained during transfers after collection by a solid waste authority.
  • Disposing of garbage by incineration or other method that makes it inaccessible to rodents.

Rodents cannot remain long in a place where they can’t find water. Clearing standing water, especially from interior spaces, is essential. This means fixing leaky pipes and sealing cracks that allow rainwater to pool inside. 

Habitat reduction

There will be fewer rodents if there is no space for them to burrow and hide. A couple of effective approaches here include:

  • Sealing cracks and other rodent entryways in homes, and clearing away junk outside or inside that may provide shelter (as a rule of thumb, a crack you can stick a finger in is large enough for a mouse to fit inside). Sealing cracks is probably more tractable for new construction than for older housing, but even if sealing all the cracks in your home is impossible, you can focus on reducing entryways for rodents into areas of the house with food (such as kitchens and pantries) and water (such as bathrooms, garages, and closets with exposed plumbing).
  • Choosing outdoor plants that aren’t easily converted into rodent bedding and burrows. Plants that obscure the ground beneath them may also obscure entrances to rodent burrows.
  • Leaving a space buffer between vegetation and buildings, which makes it more difficult for rodents to establish routes along walls and into the building. Vegetation should also be far enough away from the building that rodents can’t use it to climb up onto the building and establish a presence on the roof or inside the building via the roof.


If resource reduction and habitat reduction are so effective and seemingly so humane, why are they not more prioritized? It is likely because they are more difficult and require far more diligence than laying out poison to treat a particular rodent population boom. Better containing garbage, for instance, requires enforcing high garbage collection standards for all residents of a municipality and then paying for greater containment for every step of the way to the garbage’s final destination. A single actor can put out poison, but to improve garbage collection, much of the city has to play along. Similarly in the case of habitat reduction, many people have to cooperate to seal off habitats, and in a city where most people rent apartments in aging buildings, landlords may not have the motivation or the money to open up the walls and seal up cracks when they have to close them up according to the building code. Only contraception provides anything close to the convenience of poison. Even here, there are several barriers to product development and adoption. Besides the barriers discussed below, rodent birth control has simply not existed in a commercially viable form for very long– ContraPest, the only EPA-approved rodent birth control, only became available directly to consumers in some U.S. states in 2018. There is every reason to expect it could become much more popular to use contraception to reduce or eliminate the need for rodenticides.

Compromise approach: Harm reduction in integrated pest management

Resource and habitat reduction are unlikely to be practiced well enough to manage rodent populations satisfactorily on their own. And with any rodent birth control, there is a delay of several months before the population is reduced, while older animals die off and new ones aren’t born at the same rate to replace them. When delivering integrated pest management services, SenesTech recommends a combination of rodenticides upfront and ContraPest thereafter to address the initial lag of ContraPest and control the population rebound, caused by low population densities, that follows a die-off from rodenticides. 

Instead of rodenticides, I recommend combining rodent birth control with minimally cruel lethal control measures to handle situations that reproductive and resource control alone cannot or cannot handle quickly or conveniently enough for others to voluntarily adopt them. 

Worse than birth control, better than rodenticides

The following “compromise” lethal pest control methods are weakly ranked– I consider all of them worse than ContraPest but better than rodenticides.

1. Carbon monoxide

Carbon monoxide (CO) is a gas that results from incomplete combustion of organic matter. It is poisonous to all animals that use iron heme as part of respiration. CO causes death by binding heme in the blood irreversibly, so that oxygen (O2) cannot bind it and participate in gas exchange. In humans, CO causes flu-like symptoms as a result of oxygen deprivation in various tissues before causing death.

There are several humaneness concerns about CO as an animal euthanasia method: 

  • Concern: Animals have CO aversion.
    • Response: Animals do seem to demonstrate an awareness of the presence of CO and some avoid it. However, in one of the best studies establishing the fact of CO aversion, not all of the rats in the study avoided CO, even until the point of passing out. The study notes that, “CO is the first agent that was tolerated until [passing out] by a rat,” and ends up recommending CO as a euthanasia agent while urging more study to find non-aversive agents.[5]
  • Concern: CO causes seizures and convulsions and it is not known whether the seizures occur before loss of consciousness.
    • Response: Seizures are not an obvious humaneness concern for the targeted animal, because seizures in humans are only rarely accompanied by pain or suffering. The gruesome appearance of seizures may be a concern for human operators of CO euthanasia apparatuses, but when CO is used to control rodent populations, it is delivered to their hidden burrows and the deaths occur out of sight.
  • Concern: Depending on the method of CO production, irritants may be present, which can cause distress to the animal.
    • Response: This is likely to be a concern with rodent pest control, since it does not use veterinary-grade pure CO gas. The amount of suffering caused by irritants would depend on the level of irritants and the time to asphyxiation.

(Warning: video shows dead moles) BurrowRx is an engine that a professional fills with “smoke oil” to create CO-rich smoke, which is piped into underground rodent burrows outside. Irritants in the smoke may be a concern with BurrowRx. Because it dispenses carbon monoxide, this machine must be operated by professionals and it must be used outside. It is not safe to use indoors or even if fumes could collect indoors, because they could be deadly to humans or pets. The evenness of BurrowRx’s CO delivery over time throughout the burrow may lead to more even concentrations of asphyxiating gas and is probably superior to the more common asphyxiation method of CO2 delivery using dry ice (discussed more below), which may mean quicker, more humane deaths. 

2. Bolt traps

Bolt traps cause instant death by striking a rodent in the head with a piston, which is triggered by the entry of the head into the chamber. Bolt traps are manufactured by Goodnature and marketed for outdoor use, but Erin Ryan, who evaluated the traps in the field in her dissertation, sees no reason the traps shouldn’t be used indoors with one of the company’s shroud caps to keep children and other animals away. Erin found that, when the trap kills, it causes “immediate or near immediate cessation of movement and apparently instantaneous death,” and is therefore very humane. However, in her field trials, it seemed that most interactions do not trigger the bolt (Goodnature clarified to Erin that mice are not meant to trigger the bolt and are usually too light to do so). The bolt trap is marketed for rats, but over 19 weeks observing traps in the field, Ryan only observed eight kills (three house mice, four deer mice, and one grey squirrel), none of them rats. Ryan observed that rats were unwilling to enter far enough into the trap to trigger the bolt mechanism.

3. Snap traps

It may come as a surprise, but many animal advocates (including Charity Entrepreneurship when they evaluated banning glue traps (pdf) as an animal welfare intervention) are in favor of snap traps. This is because, when properly used, snap traps cause quick deaths. 

The U.K.’s Royal Society for the Prevention of Cruelty to Animals (RSPCA) recommends snap traps as “humane” and provides guidance on proper use. It is important to maximize the chance that the rodent enters the trap straight-on, in the proper position for a killing strike when the trap is sprung. To ensure the proper angle of entry, RSPCA and others recommend placing the traps longside against walls, or making a guide wall of stiff paper that directs the rodent to walk straight into the trap. If food is scattered around the trap or if the rodent approaches properly-laid bait from the side, it could be pinned without being killed and left to die in pain of internal bleeding or exposure and dehydration. (Bolt traps seem to solve this approach problem by only being triggered if the animal’s head fully enters the bolt chamber, which is why I prefer them over snap traps.) It is possible to buy snap traps with WiFi monitoring to reduce the need to check the traps. 

One way to improve snap traps is ensuring that the traps are quality-tested. The U.K. requires testing the springs to make sure they are sufficiently strong to provide quick deaths, but in North America this is not required. Some do-it-yourself websites suggest manually tightening the spring on snap traps to make them more effective, although one must be careful when tampering with a spring-loaded device.

4. Electronic traps

In principle, a proper dose of electric shock is a relatively humane, quick death. Because they involve electric shocks, electronic rodent traps are sold for indoor use only. Many of these traps are also quite convenient, with no-touch body disposal and available with WiFi to let users know when a rodent is inside. 

The American Veterinary Medical Association (AVMA) does not recommend euthanasia by electrocution on rodents in veterinary settings because of their small size, stating “[i]t may not result in death in small animals (< 5 kg [11 lb]) because ventricular fibrillation and circulatory collapse do not always persist after cessation of current flow” (p. 46). It’s not obvious that the shock delivered by electronic traps is well-calibrated for the rodent’s size. Negative reviews of the traps often complain that rodents are able to retrieve the bait without being killed

5. Dry ice (CO2)

Dry ice (frozen carbon dioxide, CO2) is used to address outdoor rodent burrows. Dry ice is dispensed into the burrow opening, which is then barricaded. Dry ice is cheap and plentiful, and can be kept cold using a styrofoam cooler. The biggest human safety concern is allowing dry ice to sublimate into CO2 in an enclosed space, which could cause CO2 asphyxiation in humans just as it was intended to in rodents. Despite the ubiquity of dry ice, only a specific brand of dry ice, sold by Bell Labs as Rat Ice (which is just 100% frozen carbon dioxide like any other dry ice) has met EPA approval for use against rats, and it is only available in select northern U.S. cities. Dry ice was being used in the U.S. on outdoor rat burrows before the registration of Rat Ice in 2017, but the EPA viewed it as an unregistered fumigant. Despite laboratory-grade dry ice being available from many suppliers, using unregistered forms of dry ice for rodent control is illegal. 

CO2 asphyxiation is a highly aversive and distressing way to die. Acute exposure to high CO2 concentrations induces air hunger, a sensation triggered by an upset of the balance of dissolved oxygen (O2) and CO2 gases in the blood that has “evolved to be a powerful, emotionally potent signal that can divert attention from other homeostatic warnings, even pain.” Anyone who has merely held their breath for a long time knows the sensation of air hunger, and anyone who has been unable to breathe knows the overriding desperation they felt to breathe again.  

The amount of suffering caused by CO2 asphyxiation depends on time to unconsciousness. The minimal amount of time spent suffering and dying by CO2 asphyxiation depends on the concentration of CO2 and the availability of oxygen. Studies have found that it takes 15–20 seconds (varying between individuals and depending on CO2 concentration) to render rodents unconscious from CO2.

6. Drowning traps

The Ekomille drowning trap from Ekommerce is an enclosed bucket filled with Ekofix, an oil-alcohol solution (a “vinegar-brine solution” is also referred to in different contexts). The Ekomille traps work by luring a rat to climb a baited ladder to the top of the apparatus, where a weight-activated trap door opens, causing the rodent to fall into the solution and drown.[6]

Drowning for a few minutes is more humane than death over several days by anticoagulant pesticides in my estimation, but still a highly distressing way to die, taking two to three minutes as rodents struggle for air. New York City uses the Ekomille drowning trap in the subway, and when giving interviews about the rollout of the new traps, a distributor of the trap, Rat Trap Inc., claimed that, “It knocks them out and they drown eventually." Similarly, their website claims that rodents become unconscious immediately after falling into the liquid. 

If Ekofix really did render rodents unconscious immediately, this would represent a substantial advance in rodent euthanasia that scientists would presumably adopt when using rodents in experiments. One reason for skepticism is that Ekommerce plans to release a new version of Ekomille that uses carbon dioxide, which their brochure claims causes “the death of rodents…in few seconds and even more quickly than Ekomille RC [which uses Ekofix]” (p. 5). There is a tension between the claims that Ekofix renders rodents unconscious immediately but carbon dioxide is nevertheless a more humane agent. My colleague Will McAuliffe confirmed with an export manager at Ekommerce that there is no chemical in Ekofix that is designed to render rodents unconscious more quickly than water. Further, there is a lot of variation in the composition of the solution that is used.

Many products that advertise humaneness aren’t humane or don’t work

The EPA regulates products like rodenticides and ContraPest because they pose a potential risk to the environment. There isn’t a regulatory body that monitors the claims of humane pest control devices or contraceptive herbal supplements. Rodent trap devices are often not regulated for safety, efficacy, or humaneness, allowing marketers to make untested claims. It’s not clear if law enforcement would act on a report to the Federal Trade Commission that someone was lying about the humaneness of a product for killing rodents, though reporting demonstrably false claims may be worth trying. 

Unfortunately, most traps that minimize harm probably don’t work well. Either they are so gentle that rodents aren’t trapped, or each capture requires so much labor (e.g. a human has to release the trapped animal far from the place it was captured) that not many animals are captured and the infestation carries on. 

Other products that claim to be humane birth control may or may not be birth control and they may or may not be humane — without demonstrated efficacy, there’s no reason to trust such claims. For example, Conntraceptol, an alleged contraceptive bait for mice and rats, advertises that it contains many botanical extracts believed to reduce fertility or induce labor/abortions in human females. The makers of Conntraceptol, Viamedex, claim on their website, “The nutrient formula and botanical extracts in CONNTRACEPTOL® FORTE act as a birth control agent, while male rat pheromones serve to attract females to the feeder. Extensive research has proven these botanical extracts to be safe and effective for birth control. They include stone seed root, silphium herbal, blue cohosh, and rue herb.” While some humans (legally) take those botanicals as supplements, it is far from proven that they are safe for humans, let alone for mice and rats. There is some (weak) evidence (not cited by Viamedex) that stone seed reduces fertility in mice. The contraceptive effects of rue and blue cohash (p. 402-404) are via abortion, and both are known to be unsafe for humans if taken excessively or during pregnancy. This product could very well cause suffering to the mice and rats that eat it. But all that concern about safety may be a moot point, because Viamedex does not back up their claims of birth control efficacy anywhere on their website or even list amounts of the active ingredients.

“Compromise” integrated pest management in Newton, Massachusetts

The city of Newton, Massachusetts, a suburb of Boston, uses a multi-pronged approach that balances humaneness, cost, and effectiveness concerns (see also powerpoint presentation in OSF supplementary materials). Like many cities across the U.S., Newton experienced a surge in its visible rat population during the COVID-19 pandemic that required a new approach to rodent management. 

The integrated approach focuses on: 

  • Garbage containment strategies.
  • Eliminating outdoor nests and burrows with BurrowRx (CO).
  • ContraPest for interior spaces.
  • Reducing habitat by increasing the amount of rodent-proofing required in new construction and increasing inspections to improve compliance.

Considering the counterfactual 

Some animal welfare advocates might oppose even Newton's cutting-edge approach to rodent control because they use BurrowRx for existing infestations, insisting that it is simply not necessary to resort to lethal means. While well-intentioned, taking such a hard line fails to acknowledge real differences in how much harm different lethal pest control methods cause, and the practical reality that all municipalities are going to use some method to deal with existing rodent infestations. Indeed, there is a risk that, if ContraPest is not paired with minimally cruel lethal control methods, then jurisdictions may conclude that ContraPest is not sufficiently effective or fast-acting and revert back to using only rodenticides.

It is particularly important to keep counterfactual reasoning front and center when pursuing policy change. It may seem straightforward that a method such as drowning is cruel and should be banned, but the alternative may be even worse, especially if animal advocates do not have the power to enforce a specific solution. In 2013, a Cape Town initiative was paying workers to trap and drown rats in an attempt to reduce non-target ecological impacts of rodenticide use. The program was stopped after the National Council for Societies for the Prevention of Cruelty to Animals of South Africa intervened, citing South Africa’s Animals Protection Act (1962), which does not allow drowning but does allow the use of rodenticides for “pests/vermin.”Death by rodenticide is very likely more cruel than drowning — at the least, death by rodenticides is much longer — but the law was not flexible enough to accommodate this fact. 

Why increase the role of rodent birth control?

Chemical birth control is a very promising intervention because it is potentially completely humane. Unlike poison or traps, if birth control works well, there’s minimal to no contact with rodents. Chemical birth control is also more scalable than the “relatively humane” options like traps and gas asphyxiation, making it a more direct replacement for the current most popular option: rodenticides. Developing safe and effective birth control for liminal animals may also pave the way for using birth control to improve the welfare of wild animal populations.


  • Fewer animals are painfully poisoned to death. Rodenticides, especially anticoagulants and bromethalin, often cause immense suffering. Some birth control chemicals can be toxic in large doses or have unpleasant side effects, but the point of giving an animal birth control is not to damage their body enough that they die.
  • Birth control agents are available that cause no physical[7] pain or suffering to the animal. ContraPest has no discernible negative effect on rats, and if anything, makes females healthier by reducing reproductive expenditure.
  • Rodents treated with birth control have fewer metabolically demanding breeding activities, such as gestation and lactation. Pregnancy and lactation require more calories and resources, leading rodents to cause more disruption through foraging and nesting (p. 3). When rodent populations are reduced with lethal means, their birth rates become higher, so it follows that the rodents that remain may be more driven to seek food, water, and shelter. The use of birth control could also lead to other benefits for the treated rodents, such as reduced sibling competition leading to better health as adults and a population-level increase in per capita resources, which can lead to higher average adult body mass.
  • Less disruption of rodent social life and territories. Territoriality is common for pest rodent species. A small population of sterilized adults is more likely to maintain existing territory boundaries, whereas killing the colony that occupied a territory leaves it open to migrants that may repopulate the territory, proliferate, and become bothersome to humans again. They may also spread disease when they migrate, including diseases communicable to humans. Migration opens up more possibilities for conflict between rodents between rodents from different territories, which can be particularly disruptive and bothersome to humans if it means the rodents spend more time outside of their burrows and in human-occupied spaces.
  • There are effective birth control agents that are much less deadly to humans than rodenticides if accidentally ingested. Because rodents are mammals, it’s likely that what’s poisonous to them will also be poisonous to humans in large enough doses. Birth control for pests may affect human fertility if ingested, however, which might be a newer and scarier fear for some than the familiar risks of keeping rat poison around the house.
  • Bioaccumulation and secondary ingestion of poison by predators are less likely. It’s difficult to speak for all birth control agents, but given that rodenticides make the individuals that ingest them more likely to stagger around in the open and be eaten by predators, I believe birth control agents would be less likely to end up higher in the food chain than rodenticides. On top of that, the painful and damaging effects of birth control agents on non-target and secondary ingesters would most likely be less than those of rodenticides, considering that their aim is to reduce fertility and not to cause death. ContraPest in particular was designed to be processed very quickly by rats so that it wouldn’t bioaccumulate (pers. comm. Nicole Williams).
  • Increasing the use of humane birth control as a viable alternative to rodenticides would pave the way for bans of rodenticides and other cruel lethal control measures. After researching the matter in depth, Charity Entrepreneurship did not recommend a glue trap ban (see also pdf in OSF supplementary materials) because they believed the alternative would be increased rodenticide use: “This intervention does not look promising for a new charity to work on given co-founder counterfactuals [...] This intervention would affect more welfare points on average if we could ensure that rodenticides will not be used as an alternative to glue traps. Therefore looking into banning rodenticides or at least banning the most inhumane rodenticides alongside banning glue traps would make this intervention look even more promising.”


  • There is a time lag of a few months for the lack of births to reduce the population. Rodenticides kill rodents immediately, whereas birth control takes several months to reduce population (as older members of the population die off without being replaced), but both strategies keep the population down with a treatment that is applied at similar intervals. The perception that rodenticides are a quick (true) and permanent (false) solution may make the comparatively careful “plan ahead” solution less appealing.
  • Birth control has to be a long-term, dedicated strategy. This con is a matter of perception. In most pest control contexts, this is basically already the status quo with rodenticides — eradication isn’t usually an option, because it requires killing every rodent and completely preventing recolonization. Attempts to eradicate rodent populations only really work on islands, and even then it only takes one or two migrants catching a ride on a cargo ship to reverse the supposed eradication.
  • Birth control may have higher financial costs. The cost of rodenticides is lower — in the case of ContraPest with its expensive active ingredients, probably intrinsically so (though SenesTech CSO Nicole Williams says that the cost of synthesizing active ingredients will probably go down as the company explores alternative sources and methods). However, the negative externalities of rodenticides are higher in terms of accidental poisonings, secondary ingestion, and suffering of the target rodents.
  • Birth control methods are possibly at odds with other pest control methods preferred for environmental and ecological reasons. Birth control methods could have negative ecological effects through their effects on fertility (e.g., reduced prey for raptors), so environmentalist groups may prefer methods like resource reduction, habitat reduction, traps, or ratter dogs.
  • Subject to extensive, ongoing testing and approval processes. ContraPest is approved by the U.S. EPA for Norway rats and roof rats only. It generally requires further approval at the level of each U.S. state. Not only will ContraPest need to be approved separately in each new country, but every new formulation of ContraPest will generally have to be approved separately. The existing approved ContraPest formulation is designed for Norway rats and roof rats in that the amount of active ingredient is calibrated so that it won’t bioaccumulate in their bodies. That amount might bioaccumulate in the body of a mouse, for example, so ContraPest for mice will need to be a new formulation and will be treated as a separate product by the U.S. EPA, requiring a separate approval process.  


  • For some rodents, both rodenticides and ContraPest could potentially disrupt social structures or hierarchy in addition to their direct effects. Rodent societies are probably quite disrupted by mass lethal poisonings. It’s hard to see how using birth control could disrupt the social hierarchy more than big die-offs, but it’s worth considering that rodent birth control may have some socially disruptive effects, particularly if birth control agents are considered for managing the populations of non-commensal rodents who wouldn’t otherwise be subject to rodenticides, or who depend on breeding suppression for their societies to function (think naked mole rats, for an extreme example).

Ranking birth control options 

1. ContraPest

The most commercially successful rodent birth control agent sold as part of integrated pest management is ContraPest, which holds the patent on “reducing the reproductive capacity of small mammals,” and is EPA-approved for use on Norway rats and roof rats (see also pdf in OSF supplementary materials). ContraPest uses active ingredients 4-vinylcyclohexene diepoxide (VCD) and triptolide to induce rapid aging of sperm and eggs, and has no appreciable side effects on the health and functioning of the rat other than causing temporary reduction or suspension of fertility. In fact, it is possible that females that spend part of their adulthood treated by ContraPest are healthier because they are less depleted from pregnancies.

ContraPest is delivered in multiple doses (usually three or four for full effect). Until recently, ContraPest was a restricted use product (RUP) and had to be administered by a ContraPest technician. In 2018, the EPA removed the federal RUP label from ContraPest and, as of January 2020, ContraPest is a non-RUP in 47 U.S. states. SenesTech now sells ContraPest directly to consumers via their online store, (alternate URL:  

Reasons for ranking:

  1. Completely humane
  2. Highly scalable
  3. Requires compliance of many people to place and replace baits, but works without the level of cooperation and enforcement required for serious resource reduction or habitat reduction
  4. Good evidence base for effectiveness


  1. Expensive
  2. Requires continuous use

2. Resource reduction (i.e. containing garbage, limiting food sources)

Reducing the resources available to commensal rodents to reproduce could be seen as a form of birth control.[8] Though it might seem fruitless to work on containing garbage on one city street when rodents can simply go to another, any reduction of resources where rodents live lowers the carrying capacity of that entire area, and the scale of movement for small rodents can be quite local. Rats in New York City show fine genetic population structure, suggesting highly local territories that are stable over time. This finding was echoed in Vancouver, where rats generally stayed within their own city block and only moved if disrupted by construction or demolition. 

Local resource reduction measures seem to have the potential to reduce local rat populations. For example, Chicago alleyways that had more associated rat complaints to the city were shown to have more uncontained garbage. However, resource reduction is not suitable for all areas that experience rodent infestations. For example, places like farms and food storage warehouses already properly contain their foodstuffs and seal their garbage, yet pest rodents remain a serious problem because they are clever and determined to access that food. Nonetheless, resource reduction could be a successful approach in urban environments where containment could be increased at residences and businesses, and during waste collection and transport.

Reasons for ranking:

  1. Relatively humane, insofar as it works by reducing births and not by thirst and starvation
  2. An “evolution-proof” strategy: rodents would not develop resistance to lack of resources like they could to a chemical
  3. Reduces rodent numbers in the first place by reducing energy available for reproduction, rather than killing rodents with high birth rates to achieve acceptable population size


  1. Requires a lot of cooperation, coordination, and civic-mindedness
  2. Stops working without continuous effort

3. Botanic sterilants

Botanic sterilants are plant extracts with few restrictions on their use that reduce the fertility of rodents. They may have some promise for use in developing countries and by welfare-conscious individual consumers in places without access to products like ContraPest, who would otherwise have to turn to lethal means to curb rodent infestations. In their review of botanic sterilants, Chun-mei and Rui-hong (2002) recommend a combination of Chinese traditional medicine abortifacient trichosanthin (extracted from the herb tian-hua-fen) and gossypol (extracted from cottonseed). However, because of the looser regulations where that study was conducted in China and the demonstrated non-target effects (gossypol, for instance, is safe and effective as a rodent sterilant and safe for humans, but found to be toxic to dogs), I cannot recommend they be widely adopted without more study. 

There is a product in the U.S. that claims to be a botanic sterilant for mice and rats, Conntraceptol. It contains stoneseed, an herb used in Native American traditional medicine for sterility which has weak evidence of reducing fertility in mice. However, the makers have not demonstrated the efficacy of their product and, if it does work, there may be concerns about non-target consumption. 

Reasons for ranking:

  1. It seems likely that, if they are safe, botanic sterilants cause less pain and suffering than rodenticides and are therefore an improvement in terms of welfare
  2. Botanic compounds may be more easily available than ContraPest, especially in developing countries
  3. Many botanic compounds may be easier and cheaper to produce than ContraPest


  1. The safety and efficacy of botanic sterilants are not well established — it’s possible that some compounds in the broad “botanic” class (the abortifacients, for example) cause pain or suffering that isn’t clearly better than rodenticides
  2. Appropriate testing and approval for marketing as birth control would make many of these compounds more expensive

Less promising possibilities for birth control

  • CRISPR and gene drive. These technologies are being explored to control rodent populations. The easiest use case for a gene drive is trying to eradicate isolated populations — given the invasiveness of rodents (rats in particular), there would be no shortage of possible test islands. However, using gene drives to control populations without eliminating them or risking spillover into non-target populations is a much trickier task. Since most rodent infestations are not isolated from other conspecific populations, wild or commensal, eliminating local populations would most likely mean they would eventually be replaced with rodents who had not been exposed to the drive, or that the drive would threaten wild (non-human-commensal) populations.
  • Rodenticides at lower doses. Many of the chemicals that are lethal to rodents at high doses are partially sterilizing at lower doses (notably Diphacinone), but there are other negative side effects from low-dose poisoning. It may perhaps be helpful to make farmers in developing countries aware that rodenticides can be used in this way, so that they can manage pest rodent populations through births as well as deaths — a model and more empirical information would be needed to determine if this would be a net gain or loss in welfare for the rodents, but it seems plausible that it would be a gain. However, it would likely be impossible to enforce the dosage that was used.
  • Aborticides. There are several rodent reproductive control agents that work by inducing abortions, like trichosanthin, terpenes, and sequesterpenes. In developing countries such as China, aborticidal sterilants are more accessible than ContraPest. While an abortion (followed by the eventual natural death of the mother) is clearly better than both mother and fetuses dying early from a rodenticide, advocating for aborticides would potentially damage the reputation of the animal welfare movement.
  • Steroid hormones. We know how to suppress fertility with the right exogenous steroid hormones in humans (“the pill”), and it would be straightforward to do the same for rodents, but we do not yet know how to deliver those hormones to rodents without exposing non-target populations — including pets, human children, and even adult humans.
  • Irradiation. Sterilization by irradiation works — the problem is doing it at scale in a way that contains the radiation. If rodents could selectively be exposed to radiation targeted at the gametes, and they wouldn’t subsequently spread radiation wherever they went, then perhaps this could work. However, it’s very unlikely that the sterile male technique (making sterile males in a lab, usually by irradiation, and releasing them into a wild population to occupy female mates who don’t become pregnant and thereby reduce overall births) could work for rodents, despite being very successful for some insect species. This is for several reasons:
    • It would not be very cost effective because of the cost of rearing the sterile males.
    • The public would probably not appreciate large releases of sterile males into the population they already consider pests.
    • The intervention itself wouldn’t likely be very effective, because pest rodents mate many times over their lifetime and tend not to be monogamous (females actually prefer novel males). Therefore, male rodents are highly substitutable, and the presence of sterile males in the population would not lead to a large reduction in overall pregnancies.
  • Immunocontraceptive vaccines. Immunocontraceptive vaccines work by provoking the animal’s immune system to attack its own reproductive system, leading to temporary sterility (usually up to a few years). Most immunocontraceptive vaccines must be injected to cause the sterilizing immune response, which is utterly impractical as a delivery method for pest rodents, but there are oral delivery versions of vaccinations, like the rabies oral vaccine, that use viral vectors. Perhaps porcine zona pellucida (PZP) or Gonadotropin-releasing hormone (GnRH) immunocontraceptives will one day be available in this form, making immunocontraception a more attractive option for rodent birth control.

Barriers to birth control development and adoption

Cost of development 

Anyone making a new rodent birth control product will face the costs of research – research involving vertebrates, which has a high financial and regulatory overhead –, safety and efficacy testing, developing a brand and customer base, and so on. 

Regulatory barriers

Birth control methods elude regulatory categorization

The EPA considers reproductive control agents to be in the same class as rodenticides, and single-dose rodenticides are restricted mainly because of their effects on non-target wildlife. 

A potent birth control agent like a single-dose sterilant could also be quite ecologically damaging when ingested by the wrong target. However, there are important differences between a poison and a sterilant — and further differences still between a sterilant and other kinds of birth control agents — so the context and absolute level of risk to non-target species should be more fully considered for each agent. ContraPest’s active ingredients have short half-lives and are specifically formulated so that they don’t accumulate in the primary ingester, let alone secondary ingesters. The evaluation of risk of non-target effects should be done on a case-by-case basis, and not lumped together with the dangers posed by rodenticides. 

Given the European Union’s tougher stance on potentially dangerous chemicals compared to the EPA, the difference between birth control and deadly poisons should be clearly stressed when seeking their approval. Until 2018, ContraPest was labeled a restricted use product (RUP) because the original label (see also pdf in OSF supplementary materials) was for pest management professional use. Unhelpfully, the EPA placed ContraPest in the same category as rodenticides at the time it was approved, probably because it’s the only EPA-approved product of its kind. Being erroneously labeled a “rodenticide” and having “for professional use only” on the label restricts ContraPest’s use in LEED-certified buildings and has made it difficult to work with clients like the city of Malibu who want to be poison-free. In 2023, SenesTech plans to petition the EPA to create a new category specific to “reproductive impact” (pers. comm. Nicole Williams). 


Birth control methods need separate approvals by many different governments

Sometimes EPA requirements transfer to other regulatory agencies, but sometimes other countries require more tests. There is usually a 1–2 year minimum to get approved somewhere new, which often includes redoing studies that have already been done in another country to demonstrate efficacy and safety in the new country. Just within the U.S., SenesTech has had to work individually with state governments to have changes to their label recognized


New formulations need to be approved separately as new products

SenesTech has a patent on their method for all small mammals, but ContraPest is only approved for Norway rats and roof rats. The dose and ratio of the active ingredients must be formulated for each species so that a dose is fully metabolized within 15 minutes of ingestion in order to avoid bioaccumulation. That means that formulas for new species would be considered new products by the EPA, and need to be approved separately. Approval of similar products is often accelerated by the company’s previous experience with seeking EPA approval, but even subsequent formulations take a matter of years to be approved.  


Companies need time to meet regulatory requirements

Nicole Williams, Chief Strategy Officer of SenesTech, told me that the biggest regulatory barrier for them is time. At times the company is labor-constrained, but more often their constraint is that tasks such as setting up experiments, observing for 90 days, and waiting for review are serial and take a fixed minimum amount of time. Much of the work is not parallelizable and would not be hastened by more cash at the margin.


Active ingredients are expensive

Integrated pest management services from SenesTech are more expensive than traditional pest management firms. The active ingredients in ContraPest, VCD and especially triptolide, are more expensive than those in popular rodenticides. A bait station of ContraPest sold to a consumer is $32.99 to refill and should have the tank replaced every month. A refillable cholecalciferol bait station from d-Con and 18 baits costs $12.96 and the baits don’t expire. The city of Newton reports spending about five times as much on ContraPest as they had on rodenticide. Nicole Williams of SenesTech told me that the price of triptolide drives the price of ContraPest and that they have a promising avenue for reducing the cost of that ingredient. The original ContraPest will be out of patent around 2033, which may also lead to reductions in price if competitors begin producing their own versions.


Newer products are less able to take advantage of economies of scale

Companies selling new products must discover how to scale up production and distribution as they build a customer base. ContraPest may become less expensive as SenesTech grows and enjoys more economies of scale.

Intrinsic delay in effect of birth control on population size

With any non-aborticidal rodent sterilant, there is a delay of several months in the population reduction while older animals die off and new ones aren’t born at the same rate to replace them. In contrast, single-dose anticoagulant rodenticides will kill most of the rodents that find them in 5–7 days. While both methods require ongoing application to maintain population control, consumers often see the faster initial reduction in the pest rodent population as a point in favor of rodenticides. Large clients often hire pest management companies on an indefinite basis to manage a problem they know will never be eliminated, whereas individual consumers or families may believe (whether true or not) that they only need to exterminate rodents this once, so they might not want to wait to see an effect. 

Lack of awareness and comfort

Most consumers are probably not aware of the option to use birth control for rodent control at all. Newspaper articles on the subject treat the idea of “birth control for rats” as quite novel, for example. The unfamiliarity of the approach may raise the following concerns:

  • Fertility fears. Potential consumers and pest management clients may be concerned about having a chemical in their buildings that reduces fertility, even though it is less dangerous to human health than rodenticide poison. This may be due to particular fears around fertility, or simply because this risk is less familiar to them than the risk of someone in the building accidentally consuming rat poison.
  • Not immediately clear if it’s working. There is a delay before ContraPest causes the population to be reduced, but beyond that, it’s simply less clear when a humane birth control product is working than when a poison is working. When you lay down a rodenticide, you find (or smell) dead rodents. In contrast, when you use ContraPest, it may not be clear what to look for. Even though absence of evidence of rodents around the house was what a consumer may have wanted, they cannot be sure that that’s evidence of absence. 
  • Changing and refilling baits. The residential version of ContraPest is supposed to have its tank changed each month. This requires gloves. It’s a small inconvenience but there are simpler and easier rodenticide bait stations available.
  • Single-dose vs. multi-dose. Applying ContraPest (or any sterilant) well requires observing where the rodents have been congregating and, ideally, placing test baits to ensure that the rodents consume the bait at each location. This requires a fair bit of time from either pest management professionals or end-users. Single-dose sterilants would be easier for consumers to apply themselves, if they could buy them, though they may not work as well and come with greater safety liability and risk of resistance (see Is a single-dose sterilant the answer?).

Is a single-dose sterilant the answer? 

Many animal welfare advocates have told me that they are hoping for a product that could be safely ingested once by a target animal and induce permanent sterility. In this way, populations could be much better controlled much more quickly, which would reduce the need for lethal methods of eliminating existing infestations. The simplicity and power of a single-dose sterilant is attractive in theory, but in reality there are complications.

There are currently no single-dose rodent sterilants being produced at scale. SenesTech is adamant that ContraPest is not a sterilant, but birth control: it maintains the population at a lower number with continuous application, and can be withdrawn to restore full fertility and prevent population collapse. For example, Nicole Williams shared that some clients want to control pest rat populations but maintain rat populations for birds of prey, so they ease up on ContraPest in the months before breeding season to ensure the increased predation does not cause the rat population to collapse.  

I suspect other benefits of delivering birth control in multiple doses include: 

  • Keeping each dose low, which:
    • Reduces the risks of bioaccumulation and secondary consumption (by a predator of the target animal)
    • Reduces the damage of consumption by non-target animals
    • Reduces the risk of the evolution of resistance
  • The chance to improve sampling of the population, with multiple opportunities to assess where they have taken bait and move bait stations accordingly.

Pros of a single-dose sterilant

  • Only one application is needed. This would make a single-dose sterilant feasible where ongoing application of ContraPest is not, such as the Farallon Islands in California, where invasive mice threaten native birds: “Complete eradication is the only solution on the Farallones. ContraPest, or a similar formulation, would require that bait stations be maintained in perpetuity, which is infeasible on these craggy remote islands.”
  • Roughly every rodent that encounters a bait is permanently sterilized, instead of simply having reduced fertility. 
  • The results are faster than ContraPest by a month or more. I am speculating here, because there is no mass-produced rodent sterilant, but if all 3-4 necessary doses of ContraPest could be taken at once, the population would probably start to noticeably decline around a month faster. Fast results are particularly important when considering a product marketed directly to consumers.
  • It may be easier for consumers to use. If a humane and effective single-dose sterilant becomes available directly to consumers, compliance with the instructions is going to be key. A single-dose sterilant removes much of the hassle and difficulty of using the product compared to ContraPest.

Cons of a single-dose sterilant 

  • It seems likely that something strong enough to permanently sterilize in one dose:
    • Could also be acutely toxic to the target animal or others, including humans.
    • Could sterilize humans or pets.
    • Might bioaccumulate and be bad for the environment. The dosage of ContraPest, for example, is tuned such that the active ingredients are fully metabolized by the rat within 15 minutes of ingesting it. A large enough dose to sterilize could bioaccumulate in the rat or be ingested secondarily by a predator of the rat, which could lead to unwanted non-target ecological effects.
  • It could provide strong pressure for the evolution of resistance. 
    • Rodents could evolve resistance to a sterilant by 1) tolerating more of it in their bodies before it has the desired effect, and 2) evolving bait shyness and not eating the sterilant in the first place.
    • If the sterilant had complete efficacy at a single dose, the selection pressure against sensitivity to the sterilant would be extremely strong. SenesTech has published evolutionary models demonstrating the increased chance of the evolution of resistance to lethal agents and complete sterilants. When the goal is to reduce fertility instead of abolishing it, resistance is much less likely to take off in frequency, even when there are resistant phenotypes in the population. This is because there simply isn’t enough reproductive fitness advantage on average to being resistant to contraceptives that merely lower the birth rate, whereas with a sterilant being resistant would be the only way for an exposed individual to reproduce. (Resistance to rodenticides has already been documented, particularly to warfarin, bromadiolone, and difenacoum.)
  • There is increased risk of population collapse due to miscalculation when applying a single-dose sterilant because it would be a less precise and more powerful tool. When one rodent population collapses, its territory is often re-invaded by nearby conspecifics or by another species with a similar niche.
  • An effective and safe single-dose sterilant would probably be subject to more restriction and regulation than ContraPest currently is. Being stronger than a multi-dose sterilant or birth control might make a single-dose sterilant more convenient to deliver, but it will also likely make it subject to stricter regulations. If a single-dose sterilant is more difficult to acquire, that might cancel out the convenience of its delivery.

Direct effects on wild animal welfare of replacing rodenticides with rodent birth control 

How effective of an intervention is getting people, companies, and municipalities to switch from rodenticides to ContraPest? I am writing a second manuscript dedicated to indirect effects of such an intervention. Below, I give a sense of the scale of the direct effects of displacing rodenticides with ContraPest by modeling the number of rats exposed to rodenticides in New York City each year.

 What are the direct effects of displacing rodenticide use with ContraPest?

  • Fewer rodents die painfully from rodenticides[9]
  • Fewer rodents suffer from sub-lethal doses of rodenticides
  • Fewer predators suffer death or sub-lethal poisoning from secondarily ingested rodenticides
  • Less stressful competition among rodents for resources due to decreased population size

Model: How many rats are exposed to rodenticide in New York City per year? 

This Guesstimate model, Rats exposed to rodenticides in New York City per year, estimates the number of rats exposed to rodenticides in New York City[10] each year. Many of the figures involved are quite loose estimates, and I encourage readers to play with assumptions and the values of the loose terms themselves in Guesstimate. Note: Because of the lack of data available about rodenticide and rodent birth control sales and their relative costs, this model is highly speculative. To briefly explain the steps I followed to create this model:

  • I began with data on complaints about rats to New York City’s 311 hotline to estimate the number of exterminator calls (similar to the method used here) in 2019 and 2020 and averaged the numbers from each year as a proxy of the exterminator calls to non-residential locations.
  • From there, I estimated the number of residential rat sightings as a fraction of non-residential real estate, and added the two numbers together to get a total estimate of exterminator calls in New York City per year.
  • Then I estimated the proportion of pest control solutions that employed rodenticides, placing it at approximately half (readers are encouraged to test other values).

Using this method, I estimate ~1.5 million rats are exposed to rodenticide treatments in New York City every year. (Note that Guesstimate outputs distributions, not just point estimates, which I encourage readers to check out themselves.)


Figure 1: Screenshot of from 2022-03-23 at 5.16.49 PM. This image provides a visual snapshot of the terms of the model. Follow the link to the model on Guesstimate to be able to expand the terms, read more detailed explanations of the data and estimates used, and find links to source material. This screenshot may become out-of-date as the model is updated and corrected.


From this rough model, I conclude that, if a densely human-populated city with an entrenched rat population switched entirely to ContraPest, around three-quarters of the rat population would be spared death by rodenticide. The absolute number of rats spared would be rather low compared to other animal interventions, especially for the high price of something like ContraPest. But the adoption of more humane pest control may have many beneficial indirect impacts as well, such as serving as a proof of concept that humans can solve conflicts with liminal animals creatively and humanely. The impact and cost effectiveness of having people switch to ContraPest for pest rodent control will be explored in the followup to this post.  


Commensal rodents have been with humanity since the dawn of civilization, and they can be destructive to human crops, food stores, property, and humans themselves. To reduce any population, one must either increase deaths, decrease births, increase emigration, or reduce immigration. Integrative pest management includes reducing rodents’ access to food, water, and shelter. 

Unfortunately, the status quo for controlling rodent pests today is using poison. Rodenticides cause rodents to die in many horrible ways — particularly the anticoagulant rodenticides, which take 5~7 days to kill as the rodent slowly bleeds to death from the inside. But rodenticides don’t just work on rodents — they are deadly to any animal that consumes them, be it suicidal humans or curious babies, pets, nontarget wild animals, or predators that eat poisoned rodents. Not only are there less inhumane methods, like instant-kill traps and gas asphyxiation, there is an alternative — rodent birth control — that can be applied similarly to rodenticides and play a similar role in reducing the population. 

The only U.S. EPA-approved rodent birth control on the market is ContraPest, from the company SenesTech, which is now approved as a non-restricted use product for use on Norway rats and roof rats in 47 U.S. states. Because it is the first of its kind, it has faced many unique regulatory barriers that continue to slow its development. The active ingredients of ContraPest are expensive, making it more expensive than rodenticides. Because it does not kill the rats, but rather prevents new rats from being conceived, it takes several months for ContraPest to start lowering the rat population, so its initial use is often supplemented with rodenticides. Humane traps that kill quickly, and asphyxiants applied (by professionals) into exterior burrows are more humane ways to supplement ContraPest when combating an existing infestation. 

While many animal advocates hold out hope for a single-dose sterilant, which would permanently sterilize a pest animal in a single dose and supposedly maximize the reach and effectiveness of fertility reduction campaigns, I believe the risk of rodents developing resistance and of non-target consumption would be too great to adopt this approach. 

I made a Fermi model to estimate the amount of rats exposed to rodenticides each year in New York City. I concluded that ~1.5 million, or about 75%, of the 2 million rats in New York City are exposed to rodenticides each year.

Future directions

ContraPest itself may be expensive, but advocating a switch to ContraPest may not be. Advocating for humane pest control at the level of institutions and municipalities could result in policies that cost those groups more money, but do not divert the money from other effective altruism causes. 

A followup to this manuscript is planned, which will expand on how to advocate for bans or restrictions in the use of rodenticides as a cost-effective intervention. This followup will examine the success of similar advocacy campaigns and consider indirect positive effects of promoting rodent birth control as well as direct effects. 


This research is a project of Rethink Priorities. It was written by Holly Elmore. Thanks to David Moss, Marcus Davis, William McAuliffe, and Michelle Graham for supervising. Thanks to Kim Cuddington and Simon Eckerström Liedholm for lending their substantial expertise. Special thanks to our external expert reviewer, Erin Atman Ryan. I appreciated and put to good use comments from Jason Schukraft, Saulius Simcikas, Daniela Waldhorn, Michael St. Jules, and Linch Zhang. 

If you like our work, please consider subscribing to our newsletter. You can see more of our work here.

Papers cited

A bibliography of scientific papers cited, which are also linked as DOI or Pubmed ID links in the text above. A zip file of exported pdfs and metadata is available in the OSF Supplementary Materials in the Zotero RDF format.

“Air Hunger: A Primal Sensation and a Primary Element of Dyspnea - Banzett - - Major Reference Works - Wiley Online Library.” Accessed April 25, 2022.

Asadi-Pooya, Ali A., Marjan Asadollahi, and Michael R. Sperling. “Ictal Pain: Occurrence, Clinical Features, and Underlying Etiologies.” Epilepsy & Behavior 61 (August 2016): 59–62.

Auerbach, Jonathan. “Does New York City Really Have as Many Rats as People?” Significance 11, no. 4 (October 2014): 22–27.

Biedka, Marta, Tamara Kuźba-Kryszak, Tomas Nowikiewicz, and Agnieszka Żyromska. “Fertility Impairment in Radiotherapy.” Contemporary Oncology 20 (2016): 199–204.

Broom, D. M. “The Welfare of Vertebrate Pests in Relation to Their Management.” In Advances in Vertebrate Pest Management, 309–29, 1991.

Chalifoux, A, and A Dalliere. “Physiologic and Behavioral Evaluation of CO Euthanasia of Adult Dogs.” 44, no. 12 (1983): 2412–17. 6686419.

Chun-mei, Zhang, and Wang Rui-hong. “Advance in the Research of Sterilants against Rodents.” Journal of Forestry Research 13, no. 1 (March 2002): 77–81.

Combs, Matthew, Jason Munshi-South, Kaylee Byers, and Chelsea Himsworth. “Harnessing Population Genetics for Pest Management: Theory and Application for Urban Rats (Abstract).” Proceedings of the Vertebrate Pest Conference 28 (2018).

Combs, Matthew, Emily E. Puckett, Jonathan Richardson, Destiny Mims, and Jason Munshi‐South. “Spatial Population Genomics of the Brown Rat ( Rattus Norvegicus ) in New York City.” Molecular Ecology 27, no. 1 (January 2018): 83–98.

“Comparative Effects of Brodifacoum on Rats and Possums | New Zealand Plant Protection.” Accessed April 25, 2022.


“Fatal Bromethalin Poisoning - PubMed.” Accessed April 25, 2022.

Gray, Stephen W. “Rat Survival Time During Drowning After Immersion.” Journal of Applied Physiology 8, no. 3 (November 1955): 245–48.

Horta, Oscar, and Frauke Albersmeier. “Defining Speciesism.” Philosophy Compass 15, no. 11 (November 2020): 1–9.

Isackson, Bobbisue, and Lisandro Irizarry. “Rodenticide Toxicity.” In StatPearls. StatPearls Publishing, n.d.

Jacoblinnert, Kyra, Jens Jacob, and Lyn A Hinds. “The Status of Fertility Control for Rodents—Recent Achievements and Future Directions,” n.d.

Kim, Yong-Hyun, Byung-Joo Park, Hee-Seop Ahn, Sang-Hoon Han, Hyeon-Jeong Go, Joong-Bok Lee, Seung-Yong Park, Chang-Seon Song, Sang-Won Lee, and In-Soo Choi. “Immunocontraceptive Effects in Male Rats Vaccinated with Gonadotropin-Releasing Hormone-I and -II Protein Complex.” Journal of Microbiology and Biotechnology 29, no. 4 (April 28, 2019): 658–64.

Leitschuh, Caroline M., Dona Kanavy, Gregory A. Backus, Rene X. Valdez, Megan Serr, Elizabeth A. Pitts, David Threadgill, and John Godwin. “Developing Gene Drive Technologies to Eradicate Invasive Rodents from Islands.” Journal of Responsible Innovation 5, no. sup1 (January 24, 2018): S121–38.

Lemaître, Jean-François, Vérane Berger, Christophe Bonenfant, Mathieu Douhard, Marlène Gamelon, Floriane Plard, and Jean-Michel Gaillard. “Early-Late Life Trade-Offs and the Evolution of Ageing in the Wild.” Proceedings of the Royal Society B: Biological Sciences 282, no. 1806 (May 7, 2015): 20150209.

Makowska, I. Joanna, and Daniel M. Weary. “Rat Aversion to Carbon Monoxide.” Applied Animal Behaviour Science 121, no. 2 (November 2009): 148–51.

Mason, G, and K E Littin. “THE HUMANENESS OF RODENT PEST CONTROL.” Animal Welfare, 2003, 39.

Mendi, Michael. “The Effects of Litter Size Variation on Mother‐offspring Relationships and Behavioural and Physical Development in Several Mammalian Species (Principally Rodents).” Journal of Zoology 215, no. 1 (May 1988): 15–34.

Moro, Dorian, Margaret Byrne, Malcolm Kennedy, Susan Campbell, and Mark Tizard. “Identifying Knowledge Gaps for Gene Drive Research to Control Invasive Animal Species: The next CRISPR Step.” Global Ecology and Conservation 13 (January 2018): e00363.

Murray, Maureen H. “Public Complaints Reflect Rat Relative Abundance Across Diverse Urban Neighborhoods.” Frontiers in Ecology and Evolution 6 (2018): 10.

Murray, Maureen H., Rebecca Fyffe, Mason Fidino, Kaylee A. Byers, M. Jazmín Ríos, Matthew P. Mulligan, and Seth B. Magle. “Public Complaints Reflect Rat Relative Abundance Across Diverse Urban Neighborhoods.” Frontiers in Ecology and Evolution 6 (November 28, 2018): 189.

Nattrass, Nicoli, Jed Stephens, and Jorich Johann Loubser. “Animal Welfare and Ecology in the Contested Ethics of Rodent Control in Cape Town,” n.d., 10.

Niedringhaus, Kevin D., Nicole M. Nemeth, Samantha Gibbs, Jared Zimmerman, Lisa Shender, Kate Slankard, Heather Fenton, et al. “Anticoagulant Rodenticide Exposure and Toxicosis in Bald Eagles (Haliaeetus Leucocephalus) and Golden Eagles (Aquila Chrysaetos) in the United States.” Edited by Benito Soto-Blanco. PLOS ONE 16, no. 4 (April 7, 2021): e0246134.


“PubChem Compound Summary for CID 3503, Gossypol.” In National Center for Biotechnology Information, 2022.

“PubChem Compound Summary for CID 7833, 4-Vinyl-1-Cyclohexene Dioxide.” In National Center for Biotechnology Information, 2022.

“PubChem Compound Summary for CID 107985, Triptolide.” In National Center for Biotechnology Information, n.d.

Rao, Rama B, and Robert S Hoffman. “Nicotinic Toxicity from Tincture of Blue Cohosh (Caulophyllum Thalictroides) Used as an Abortifacient” 44, no. 4 (2002): 221–22. 12136970.

Rodenberg, HD, CC Chang, and WA Watson. “Zinc Phosphide Ingestion: A Case Report and Review.” 31, no. 6 (1989): 559–62. 2575817.

Romm, Aviva. “Chapter 16: Labor and Birth.” In Botanical Medicine for Women’s Health. Churchill Livingstone, 2010.


Shuster, Stephen M., Brandy Pyzyna, Loretta P. Mayer, and Cheryl A. Dyer. “The Opportunity for Sexual Selection and the Evolution of Non-Responsiveness to Pesticides, Sterility Inducers and Contraceptives.” Heliyon 4, no. 11 (November 2018): e00943.

Tisserand, Robert, and Rodney Young. “The Reproductive System.” In Essential Oil Safety, 147–63. Elsevier, 2014.

“USDA Animal Care: Animal Welfare Act and Animal Welfare Regulations,” n.d., 265.

Ventura-Aquino, Elisa, Jorge Baños-Araujo, Alonso Fernández-Guasti, and Raúl G. Paredes. “An Unknown Male Increases Sexual Incentive Motivation and Partner Preference: Further Evidence for the Coolidge Effect in Female Rats.” Physiology & Behavior 158 (May 2016): 54–59.

“Warfarin‐based Rodenticides: Mode of Action and Mechanism of Resistance - Thijssen - 1995 - Pesticide Science - Wiley Online Library.” Accessed April 25, 2022.

Witmer, Gary W., Stefanie Raymond-Whish, Rachael S. Moulton, Brandy R. Pyzyna, Elissa M. Calloway, Cheryl A. Dyer, Loretta P. Mayer, and Patricia B. Hoyer. “COMPROMISED FERTILITY IN FREE FEEDING OF WILD-CAUGHT NORWAY RATS ( RATTUS NORVEGICUS ) WITH A LIQUID BAIT CONTAINING 4-VINYLCYCLOHEXENE DIEPOXIDE AND TRIPTOLIDE.” Journal of Zoo and Wildlife Medicine 48, no. 1 (March 2017): 80–90.

Wondifraw, Bewketu Takele, Mesele Yihune Tamene, and Afework Bekele Simegn. “Assessment of Crop Damage by Rodent Pests from Experimental Barley Crop Fields in Farta District, South Gondar, Ethiopia.” Edited by Bi-Song Yue. PLOS ONE 16, no. 8 (August 12, 2021): e0255372.

Yamamoto, K., Y. Yamamoto, and H. Kikuchi. “The Effects of Drowning Media on the Lung Water Content: An Experimental Study on Rats.” Zeitschrift Für Rechtsmedizin 90, no. 1 (March 1983): 1–6.

  1. ^

    “Commensal rodent” is a more correct and anti-speciesist term, but for brevity and ease of communication, I will use the term “pest rodents” to refer to commensal rodents that are considered a problem by the humans that live with them.

  2. ^

    There are several ways to reduce fertility, including reducing the number of litters, size of litters, or the frequency of litters, and not all birth control methods will work the same way.

  3. ^

    These categories are not mutually exclusive, of course. What reduces the population through increasing deaths will also mean fewer births immediately following the population reduction, and resource limitation imposed on existing animals may cause them to die of starvation or exposure. Reducing migration to an area will also reduce births there, etc.

  4. ^

    Pest rodents generally belong to large metapopulations that extend beyond the local “infestation” area. Source populations continually replenish the populations of areas where humans apply population control measures.

  5. ^

    CO chambers have been used as a form of veterinary euthanasia, but they are not recommended by the Association of Shelter Veterinarians or the Humane Society of the U.S.. Bear in mind, both organizations are claiming that CO asphyxiation is inhumane compared to lethal injection, which is not an option for rodent pest control. Many of their complaints focus on the CO chamber devices— the noise they make as they bring CO up to the necessary 6% concentration is distressing to the animals, they present a safety risk to human operators (in 2000, a Chattanooga shelter employee died after failing to vent the CO gas from the chamber before removing a dog he had just euthanized), etc. — which is not how CO would be delivered to pest rodent populations.

  6. ^

    Apparently, some of New York City’s rats are too hefty for the machine, as one of the first drowning traps deployed had its spring broken by a single massive rat.

  7. ^

    It is possible that birth control causes some as yet not understood psychological pain to some rodents, for example through depriving them of offspring or changing social dynamics. So, although I feel confident that the possible pain caused by birth control is far better than being subject to lethal population control measures, I don’t know if I can say that birth control would cause no pain.

  8. ^

    Simon Eckerström Liedholm points out that severe resource reduction is a form of lethal control via starvation.

  9. ^

     As opposed to either 1) not being born, or 2) dying of other causes, which I think are likely to be less painful than rodenticides, but may not always be.

  10. ^

    New York City already used ContraPest on a large scale in 2017. The first trial of ContraPest was also in New York City, in 2013. On their website, SenesTech provides a similar case study from Washington, DC (find the case study in the OSF Supplementary Materials), saying: “ContraPest was deployed for 16 months in our nation’s capital city from June 2019 to October 2020. Depending on location, activity was reduced 94-99% with 98-100% reduction in juvenile rodents.”


New Comment
2 comments, sorted by Click to highlight new comments since: Today at 4:30 AM

I really appreciated this in-depth post, both because it seems like a well researched analysis of a problem with some concrete suggestions and because it's easy to read at different levels; the section headers and summaries made it really easy to get a sense of the overall take and then dive deeper into the sub-topics that I was most confused about. 

Thank you! I was hoping it would be useful to people to consult just the heading they were curious about.