Sixty percent. That is the protective effect one team of researchers reported for a spatial repellent tested against malaria on the island of Sumba, Indonesia, in a trial that ran from 2015 to 2018 and was registered at ClinicalTrials.gov under the number NCT02294188.
Twenty-seven point seven percent is the number the same trial produced for roughly the same product, in the same households, against the same disease, over the same two years.
Both figures come from the same published analysis. The difference is not dishonesty. It is arithmetic done on a different slice of the data. The cluster-randomized trial conducted by Din Syafruddin and colleagues tested transfluthrin-treated passive emanators distributed to households across 24 village clusters, with a ten-month baseline period and a 24-month intervention period, monthly blood screening, and biweekly human-landing mosquito catches. It is about as careful a field study as consumer mosquito protection has ever received. And it still yielded a handful of defensible-looking percentages, depending on which question you ask the data.
That gap is the subject of this article. Mosquito treatment is one of the most number-saturated categories in the home-and-garden trade, and most of those numbers arrive stripped of the questions that produced them. Importers, distributors, and procurement teams who cannot reconstruct those questions are reduced to comparing adjectives. Those who can compare designs. That is a real edge, and this guide is built to install it.
First, What Counts as Mosquito Treatment
Before decoding numbers, a map of what they attach to. In professional use, mosquito treatment is any intervention that reduces either the number of mosquitoes or the number of their bites, and the field organizes around the insect’s life position rather than the product aisle:
- Source reduction – draining, filling, or removing standing water so larvae never exist. The oldest layer, and the only one with no resistance problem.
- Larviciding – applying biological agents such as Bacillus thuringiensis israelensis (Bti) or growth regulators to water that cannot be drained.
- Adulticiding – spraying or fogging to knock down flying adults, whether by a municipal truck or a contracted yard service. What the visit physically does, and how quickly the mosquitoes come back, is examined in our breakdown of mosquito spraying.
- Exclusion – screens, netting, and sealed openings that keep adults out of the rooms people occupy.
- Personal protection – skin-applied repellents and treated clothing, which protect a person rather than treat a place.
- Indoor devices – plugged UV-attracted electric units that run continuously inside structures, the layer most relevant to a hospitality or food-retail buyer. Our specification guide for indoor commercial spaces covers placement and sizing.
Program-level purchases stack these layers through contracts and visits, which we unpack separately in what mosquito services actually sell you and in the legal anatomy of a professional mosquito treatment. Every one of those layers generates efficacy numbers. Almost none of them publishes the question the number answers. That is where mispricing, bad specifications, and disappointed end customers begin.
One Trial, Five Numbers: What the Sumba Study Actually Reported
Take the Sumba trial apart slowly, because it is the best free lesson in applied claim-reading available to this industry.
After mass radical cure of infections, the researchers followed 1,341 children aged six months to five years across 24 clusters; households in half the clusters received transfluthrin emanators, half received placebo. During the 24-month intervention, the placebo households recorded 164 first-time infections and 459 cumulative infections among 667 subjects. The treated households recorded 134 first-time and 253 cumulative infections among 665 subjects. Those raw counts are the whole product. Every percentage below is one arithmetic choice away from them:
- 27.7 percent – protective effect against time-to-first-infection, using all 24 clusters. Not statistically significant.
- 31.3 percent – protective effect against total new infections, all 24 clusters. Also not significant.
- 33.3 percent and 40.9 percent – the same two endpoints, but restricted to the 19 clusters that saw at least one infection during baseline. The second figure crossed significance, one-sided, at the 5 percent level.
- 60 percent – effect in the 12 moderate-to-high transmission clusters only, described by the authors as a statistically significant decrease.
- Inconclusive – the authors’ own verdict on the primary entomological analysis, the mosquito-side measurements.
The authors did not cherry-pick 60 percent. They published all of it, flagged the low-incidence problem, and wrote that “additional evidence is required to demonstrate the product class provides an operationally feasible and effective means of reducing malaria transmission.” But imagine each number arriving alone at a trade show booth. “60 percent protection” and “27.7 percent, not significant” are both accurate quotations of a single honest study. A buyer who has never seen the trial cannot tell you which one your competitor’s leaflet is quoting, and therefore cannot tell how much of the effect is real in your market.
The same page of literature adds two historical data points that sharpen the point. A China malaria-prevalence study of coils containing 0.03 percent transfluthrin reported a 77 percent reduction in human Plasmodium falciparum cases. An Indonesian study of metofluthrin-treated coils reported 52 percent protective efficacy against new infections. Both were encouraging. Both were later judged by the World Health Organization’s Vector Control Advisory Group, according to the trial’s own account, to fall short of the evidence bar – too small in scale, or underpowered – to support a policy recommendation. Four technologies’ worth of hope, six defensible percentages, one consistent lesson: a percentage with no description of its trial is an adjective wearing a costume.
The Five Questions That Turn a Percentage Into Information
Every mosquito treatment number – a service guarantee, a trap bag, an Amazon bullet point, a factory test report – becomes decidable once you force it to answer five questions. Each question below has a real study or document attached to it, because in this category the professional literature has already done the demonstrating.
1. What exactly is being counted?
“Mosquitoes reduced” hides at least four different quantities. Professionals count: biting rate (mosquitoes landing on a human per night), trap catch (mosquitoes collected by a specific device in a specific place), larval indices (containers or dips positive in surveyed habitat), and disease endpoints (human infections prevented). A 90 percent drop in one is compatible with a 10 percent drop in another. The human landing catch – the uncomfortable gold standard, a person offering their own legs and counting what lands – remains the reference that devices are validated against, precisely because it is the only one that counts what the buyer actually cares about: bites. A 2015 evaluation of CDC light traps on the Thai-Myanmar border states the logic plainly when it describes the landing catch as focused on “host-seeking mosquitoes that may represent the most relevant proportion of the mosquito population for disease transmission,” which is why trap comparisons are anchored to it (Sriwichai et al., Parasites & Vectors, 2015).
So when a catalog says “catches 90 percent more mosquitoes,” the first question is not how many. It is nine times more of what count, made by what instrument.
2. What instrument made the count?
There is no neutral mosquito counter. Every sampling device has a designed appetite. The CDC publishes a professional Mosquito Trap and Surveillance Tools Guide (fact sheet dated August 15, 2022) that reads like a decoding key for anyone who thinks a “mosquito trap” is one thing:
- The CDC light trap covers a “wide range of mosquito genera and species” of adults, works best set at dusk, and becomes markedly better at host-seeking females when baited with dry-ice carbon dioxide.
- The gravid trap and the autocidal gravid ovitrap (AGO) target egg-laying female Culex and Aedes – they sample a reproductive state, not abundance.
- The ovitrap counts eggs of Aedes species, a different life stage entirely.
- The resting trap samples host-fed females of Anopheles, Culex, and Culiseta in shade, in the early morning.
- The BG-Sentinel trap concentrates on Aedes aegypti, Aedes albopictus, and Culex, and the CDC tips list added attractants – octenol, human scent, carbon dioxide – to raise female catches.
Each row of that guide is an admission that the number produced by a trap is partly a property of the trap. The same fact sheet that lets a professional choose an instrument lets a buyer interrogate a claim: “reduced mosquito counts by 85 percent” is meaningless until you know which instrument, which lure, which placement, which hours of darkness.
3. Compared against what?
The strongest comparison in the category is a concurrent control: treated plots and untreated plots, measured with the same instrument, on the same nights. The weakest is before-and-after – count mosquitoes this week, spray, count again next week. The weak design has a fatal flaw that the professional literature keeps documenting: mosquito numbers move violently on their own. A case study analyzing Arizona’s statewide surveillance data – 71,812 entries collapsing to 42,308 trap nights for 2018 – shows that traps set in response to complaints “produce high estimates of abundance” and, mixed into routine data, inflate population curves; it also records an outlier of more than 100,000 mosquitoes logged for a single trap on a single night (Brown et al., Journal of Medical Entomology, 2021). A before-and-after trial that happens to straddle a natural emergence pulse, or a weather break, has measured the calendar, not the treatment.
The Arizona analysis adds a quieter killer detail: about 47.6 percent of trap entries reported no mosquitoes, and agencies recorded empty traps and trap failures inconsistently. Remove the zeros carelessly and the same dataset tells a different story about when the season starts and how high it peaks. Any percentage built on counts where “no catch” is ambiguously logged already inherits that ambiguity.
4. Over what clock?
A percentage needs a time window, and the window has to survive how mosquitoes refill space. Adulticide deposits degrade with sun and rain; nearby untreated habitat pumps in new adults; larvae that were pupae on treatment day emerge tomorrow. The surveillance literature’s answer is repeated, standardized sampling – at least every other week through the season, and over two or more years, if you want to separate a treatment’s effect from ordinary seasonality (same Brown et al. case study). Single-night demonstrations – the beating heart of many trade-show pitches – sit at the opposite end of the evidence distribution from multi-season surveillance. If a claim’s window is one humid night in June, the claim is a photograph, and photographs can be staged.
5. Replicated how, and where?
One trap in one yard is an anecdote; replicated plots across villages and weeks are data. This is where the mosquito-science literature is most humbling, and where the next section of this guide lives: instruments validated in one setting have a habit of failing in another, and replication across places – not repetitions in one place – is what the honest word “proven” secretly requires.
The Measurement Instrument Itself Can Move the Number
If you remember one demonstration from this article, make it the Mbita trap. It was invented as a safe alternative to the human landing catch: a white conical bednet-like chamber that lets a sleeping human bait attract mosquitoes upward into a collection funnel, so the sleeper is never bitten. Its developers published a validation showing it worked: in semi-field tests at Mbita Point, Kenya, with laboratory-reared Anopheles gambiae, the trap caught 43.2 percent of the number collected by human landing catches. A defensible device. Published. Adopted by programs.
Then Institut Pasteur de Madagascar ran the same comparison in real highland villages. The result, published in Malaria Journal in 2003 (Laganier et al.):
- Across 448 human-nights and 84 trap-nights, human landing collected 6,881 mosquitoes; the Mbita trap collected 85. Overall efficiency: 0.066 – about six percent of what the reference method found.
- Efficiency for the key vector Anopheles funestus ranged between villages from 0.036 to 1.212 – in two of the three villages the unreliable trap out-caught the gold standard, in the third it nearly caught nothing.
- Month-to-month, the trap’s efficiency against that same species swung from 0 to 6.8 with “no clear tendencies that would provide clues to explain this variation.”
- Between the trap and the reference method, no statistically significant correlation could be found at all. Not a worse number – no relationship.
Same trap. Same bait – a sleeping human. Fourfold differences between villages six to seventeen kilometers apart, and an order of magnitude between a laboratory-reared validation and wild populations. The authors concluded the landing catch “remains the reference method for evaluating mosquito density” and warned their finding should not even be extrapolated to other countries. Read that sentence again as a buyer: a peer-reviewed malaria-journal paper says the conclusions about one mosquito instrument may not survive the drive to the next region.
The Thai-Myanmar light-trap study makes the same point with knobs instead of chaos. Whether a trap catches what you care about changed with the trap’s configuration, its position, and the season. Adding a kilogram of dry ice raised capture rates significantly – adjusted capture rate ratios of 1.64 for Culex and 1.38 for Anopheles – but the size of the bait effect itself shifted with seasonality and indoor-versus-outdoor placement. Armigeres preferred outdoor traps; Aedes albopictus trended indoors (1.89, p = 0.07, honestly reported as a tendency, not a result). Six thousand six hundred sixty-eight mosquitoes, 2,880 trap nights, seven villages, and the study’s conclusion is a list of it-depends.
A 2019 evaluation of a UV-LED trap (the Mosclean) from the same research culture gives you the most useful single sentence in all of consumer-device marketing forensics. In semi-field chambers, the new trap caught twice as many Anopheles arabiensis as a CDC light trap. The paper’s own statistics: relative rate 2.1, 95 percent confidence interval 0.97 to 4.56, p = 0.059 (Mwanga et al., Parasites & Vectors, 2019). “Twice as many” is the headline. “The interval includes ‘no difference’ and the test missed the conventional significance threshold by 0.009” is the finding. The paper itself concludes the trap shows promise for surveillance and calls for more field validation – that is not the point. The point is that the honest authors printed both sentences next to each other, and marketing prints neither.
Why Even Professional Mosquito Counts Wobble
If the instrument is partial and the calendar is loud, what is left when an actual mosquito-control program publishes a number? The Arizona surveillance case study in the Journal of Medical Entomology is quietly the most honest document in this literature, because it is a statistics team taking apart their own pipeline. Four findings decide whether a program’s year-over-year graph means anything:
- Why the trap was set. Traps run as routine surveillance at fixed locations give comparable abundance estimates. Traps set because a resident complained sample hotspots and inflate the aggregate. In the state data, one county’s records were 83.1 percent routine; across the other counties, only 59.6 percent were. Two agencies can publish the same species, same year, same trap model, and different realities.
- Whether zero was recorded. An empty trap is a measurement. A trap that malfunctioned and was not logged is missing data. When agencies inconsistently record both, abundance curves bend – the study shows how removing unrecorded zeros shifts both the season start and the peak.
- Whether the pesticide records travel with the trap records. The analysts note that without knowing when and where adulticiding or larviciding happened, a decline in the trap data cannot be attributed to anything – resistance is rising, treatment schedules vary, and the trap cannot tell you which.
- Whether sampling was frequent enough. Every-other-week sampling at stable sites is the working minimum the review cites for separating seasonality from signal; multi-year series are needed for interannual claims.
Read as a buyer, this is a specification for what a real dataset looks like: instrument named, lure stated, placement fixed and logged, zeros recorded, treatments timestamped, cadence disclosed. Any percentage arriving without that metadata is not a weak measurement of the truth; it is a measurement of something else.
What the Regulators Actually Require of a Number
Two American agencies police the numbers in this category, by two different doors, and importers shipping to the US market should know both.
The advertising door: the FTC
Any product that kills, repels, or traps insects with a substance – pesticidal products – carries EPA-registered claims on its label. But devices and personal-protection items live mostly under the Federal Trade Commission’s ordinary advertising law, whose standard is not “the company believes it” but what the FTC calls competent and reliable scientific evidence. The case that made this concrete for mosquito products is the Commission’s action against Viatek Consumer Products Group, marketer of mint-oil “Mosquito Shield Bands.”
The marketing said the wristband created a “vapor barrier” that shielded anyone within five feet and provided 96-120 hours of protection. The FTC’s February 2015 charged the company and its owner with having no competent and reliable scientific evidence for those claims. The May 2016 settlement: a $300,000 payment, a bar on misleading pest-control claims, and an order that future performance claims for any product be backed by substantiation first. “With mosquito-borne illnesses in the news, consumers might be looking for products that protect them from mosquitos,” the director of the FTC’s Bureau of Consumer Protection said when the settlement was announced. “The defendants took advantage of those concerns, and peddled a product without having scientific support that it effectively prevented mosquito bites.”
Note the anatomy: not a dangerous product, not a scam – a product whose numbers outwalked its evidence. That is the failure mode of this entire category, and it is a legal failure mode, not just an ethical one.
The registration door: FIFRA and the testing duty
For products that are pesticides, the claim and the label are legally one object. Federal law defines misbranding to include inadequate directions for use, and the registration data regulations state the performance duty in a sentence worth reading twice: under the product-performance provisions of the data rules, “each registrant must ensure through testing that his product is efficacious when used in accordance with label directions and commonly accepted pest control practices,” with the Agency reserving the right to demand the performance data case by case (40 CFR Part 158, Subpart E, product performance data table, test note 1). The grammar of that obligation is the whole buyer’s checklist in miniature: efficacy is defined relative to label directions and accepted practice – stated frequency, stated placement – not relative to a marketing photograph. A percentage divorced from a frequency and placement is, in the regulator’s own drafting, not even the kind of claim that can be true.
The extension-service floor: “reduction of mosquito annoyance”
University extension entomologists – the people who answer to homeowners, not shoppers – set the rhetorical floor remarkably low, and that floor is a useful calibration tool for anyone specifying this category.
The University of Maine Cooperative Extension’s mosquito management fact sheet (Pest Management Fact Sheet 5110, revised through 2023) opens the non-chemical section with: “Alone, these methods won’t eliminate your mosquito problem. However, using an integrated approach that combines several of these methods together should result in a reduction of mosquito annoyance.” And it names the novelty aisle explicitly: “Beware of novelty approaches to mosquito control, including such things as ‘bug zappers,’ various sound devices, and scented geraniums (‘mosquito plants’)… there is little scientific evidence to support the claims of those who market such products. There is no sure-fire solution to the problems as some would assert.” It also flags a placement rule that any light-based device buyer should notice: homeowners are told to keep outside lighting to a minimum – advice that only makes sense because where light sits decides where insects go, which is exactly the variable the trap-comparison studies keep measuring.
The University of Illinois Extension’s 2026 home-mosquito guide adds the ultrasonic verdict – “Ultrasonic repellent devices claim they mimic predator sounds or male mosquitoes are not supported by research” – and, in an accompanying podcast episode, walks through the bug-zapper and “bucket of doom” myths against the entomology of what actually intercepts a host-seeking mosquito. When the public-institutional literature, the trial literature, and the FTC’s enforcement history all converge on the same sentence – reduction of annoyance, layered, never standalone – that is the ceiling a serious B2B catalog should promise under.
Applying the Same Reading to Your Own Catalog
A reader who has just disassembled six percentages from a malaria-journal trial and a federal settlement is now, deliberately, in a position to interrogate the catalog of the very category we manufacture. Pittar builds indoor electric mosquito killer lamps: plugged-in units that attract with ultraviolet light and intercept with a high-voltage grid, engineered for continuous indoor operation, offered in 10W, 20W, 30W and 40W configurations with an 1800-volt grid. They are sold as the indoor layer of a program, not as a standalone outcome – and after the Sumba trial, you know exactly what that hedge is worth: it is the difference between a claim and a costume.
So run the catalog through the five questions, on purpose, in public – because this is also how to evaluate every competitor sheet crossing your desk:
- Count. A lamp’s honest contribution is measured as indoor catches of the species that actually enter the building, at night, near where people rest – the same “indoor host-seeking” quantity the surveillance literature isolates. It is not a countywide suppression number and should never be advertised as one.
- Instrument. Our units share ancestry with the UV light traps the CDC guide describes: light is an attractant, adults are the life stage, and performance is genus-dependent. A trap family that catches “a wide range of genera” with a light and fan inherits every caveat of the light-trap literature – including that a competing human in the room out-attracts any lamp.
- Comparison. This is why we publish wattage, grid voltage, tube type, and replacement-part supply rather than victory percentages: wattage and grid voltage are verifiable attributes; “74 percent fewer bites” is a trial, and trials are not our product line’s public record. If a supplier does publish a percentage, the comparison column asks: before-and-after in one house, or treated versus control?
- Clock. A device that runs every night occupies a different time axis than a truck that visits every 21 days. Continuous indoor operation is a genuine structural advantage of the device layer – and it is also what makes single-night catch photos meaningless, because the same unit running through a natural population crash looks like heroism.
- Replication. If coverage-area figures appear on any spec sheet you buy against – ours included, from any source – the replication question is: which room, which ceiling height, which climate, which competing lights, which species, who paid for it. A coverage number measured in a 3-by-4-meter test chamber and a hotel lobby are two different products.
The device layer earns its place precisely where the evidence lives: indoors, where mosquitoes that have already entered concentrate; where the units run while the exclusion layer (screens, door discipline) and the outdoor layers (source reduction, larvicide, the contracted visit) do their slower work. For a distributor assembling a portfolio for hospitality, food retail, or property-management customers, that positioning is also the commercially defensible one – a specification you can hand a facilities manager without a lawyer reviewing it.
Six Questions to Ask Any Supplier Who Quotes You a Number
1. What was counted – bites, indoor trap catches, outdoor trap catches, larval indices, or disease cases?
2. What instrument counted it – which trap or method, baited with what, placed where, operating during which hours?
3. Compared against what – a concurrent untreated control, a before-and-after baseline, or nothing at all?
4. Over what clock – one night, the residual window, a full season, and who maintained the traps in between?
5. Replicated where – how many sites, how many nights, how far apart, which species dominate there, and does your export market resemble any of them?
6. Who paid, and where can I read the full report – a named lab with a published protocol, or a marketing department with a number?
A factory that has ever measured anything will answer all six in an afternoon. A factory that has measured nothing will answer with a new adjective. Between the two answers sits your entire exposure profile – the same line the Viatek settlement drew with a $300,000 ruler: no one was pursued for selling a wristband; the settlement was about claims made without competent and reliable scientific evidence. Importers do not stand on the safe side of that line by default – once the carton carries your import, the claims printed on it travel with your name.
If you are building a mosquito-control line for distributors, retailers, or hospitality accounts, the honest specification is also the easiest one to defend: published electrical and physical specifications, replacement tubes and boards, packaging and logo programs under OEM and ODM cooperation, and placement guidance grounded in the indoor-behavior literature rather than in a photograph of a collection tray. Our full product series and the catalog and quote request are the next click; the specification logic above is what we will not hedge when you ask which model fits your market.
Frequently Asked Questions
What is the most effective mosquito treatment?
The one that stacks. Every serious source in this article – extension fact sheets, surveillance statistics, the trial literature, and the CDC’s own trap guide with its six specialized instruments – describes the same structure: eliminate or treat standing water first, exclude with screens, treat the outdoors at the right life stage and time of day, and run the indoor layer continuously. Extension language is blunt about the consequence of skipping layers: no single method “alone” fixes the problem; only the combination “should result in a reduction of mosquito annoyance” (University of Maine Fact Sheet 5110). For the indoor end of that stack, see our room-by-room indoor commercial guide.
Do mosquito treatments actually reduce bites, or just mosquito counts?
Sometimes counts, which is the trap this entire article is built to spring. Counts come from instruments with appetites; bites are a different quantity that requires a human-served reference method to estimate. When those two quantities move together at all, the published correlation is weaker than product pages imply – the Mbita trap’s field efficiency ranged from 0 to 6.8 relative to the human landing catch depending on village and month. The honest sentence for any treatment, spraying included: measure the layer you actually sell, in the units it acts on. Our analysis of spraying covers which life stage each outdoor method intercepts and how fast refill happens.
Are UV electric mosquito lamps a treatment, or an accessory?
They are a monitoring-capable suppression accessory for the indoor layer. The same physics that makes light a partial attractant in the open field – host cues beat photons at close range – is what keeps the device layer honest: indoors, at night, away from competing light, a powered UV unit intercepts the subset of host-seeking insects already inside the envelope. That is a real, defensible service for hotel floors, restaurant back-of-house, and clinics. It is not outdoor area reduction, and no number printed on a lamp carton should imply that it is.
Why do two studies of the same product report 27 percent and 60 percent?
Because the number is an answer, and the analysis population and endpoint are the question. In the Sumba trial, all-24-cluster protection against first infection was 27.7 percent; the same product, tested only in the moderate-to-high transmission clusters, came out at 60 percent; the entomological endpoint came out inconclusive. Every published percentage has exactly this anatomy – pick the endpoint, pick the population, then divide. This is also why lifecycle timing matters when you compare numbers across seasons; the arithmetic of generations is walked through in our lifecycle explainer.
What should I do when a supplier quotes a number but refuses to answer the six questions?
Treat the number as fiction with an apostrophe, and price the risk accordingly. After Viatek, the pattern is predictable: the claim that survives no methodology conversation is the same claim a regulator or a customer’s lawyer will one day read aloud. Ask for the full report, not the headline; if it exists, its methods section will be boring and complete.
What We Could Not Verify
Discipline requires publishing the gaps. First, the Sumba trial figures cited above come from the medRxiv preprint version; if the peer-reviewed publication reports endpoint statistics differently, the published article governs. Second, we found no publicly documented standardized test protocol – in the way the surveillance community standardizes traps – for the “coverage area” or “catch rate” figures consumer UV electric units commonly print; the trap literature cited here is surveillance science, not a consumer-product test code, and it illustrates method fragility rather than certifying any retail device. Third, the effect sizes from Madagascar, Thailand, and Tanzania describe malaria and arbovirus vector ecology under professional sampling; they should be read as lessons about instruments and design, not as predictions for a bedroom in another climate. Anyone who converts them into product claims without new local data is doing exactly what this article warns against.
Sources
- Din Syafruddin et al., “Efficacy of a spatial repellent for control of malaria in Indonesia: a cluster-randomized controlled trial,” medRxiv preprint (v4), trial registered NCT02294188 – full text.
- Centers for Disease Control and Prevention, “Mosquito Control: Mosquito Trap and Surveillance Tools Guide,” fact sheet dated August 15, 2022 – PDF.
- Rémi Laganier, Fara M. Randimby, Voahirana Rajaonarivelo, Vincent Robert, “Is the Mbita trap a reliable tool for evaluating the density of anopheline vectors in the highlands of Madagascar?” Malaria Journal 2:42, November 19, 2003 – full text.
- Patchara Sriwichai et al., “Evaluation of CDC light traps for mosquito surveillance in a malaria endemic area on the Thai-Myanmar border,” Parasites & Vectors 8:636, December 15, 2015 – full text.
- Emmanuel P. Mwanga et al., “Evaluation of an ultraviolet LED trap for catching Anopheles and Culex mosquitoes in south-eastern Tanzania,” Parasites & Vectors 12:418, August 27, 2019 – full text.
- Heidi E. Brown et al., “Understanding Mosquito Surveillance Data for Analytic Efforts: A Case Study,” Journal of Medical Entomology 58(4):1619-1625, February 22, 2021 – full text.
- US Federal Trade Commission, “Marketers of ‘Mosquito Shield Bands’ to Pay $300,000, Barred from Making Misleading Pest-Control Claims under Settlement with FTC,” May 24, 2016 – press release.
- US Federal Trade Commission, “FTC Charges Company, Owner with Deceptively Marketing Mosquito Repellent Wristbands,” February 20, 2015 – press release.
- 40 CFR Part 158, Subpart E – Product Performance Data Requirements Table and Test Notes (product performance for pesticidal claims), Electronic Code of Federal Regulations, current as of November 2026 – eCFR.
- James F. Dill and Griffin M. Dill, “Mosquito Management,” Pest Management Fact Sheet 5110, University of Maine Cooperative Extension (2010, 2019, 2023) – online text.
- Ken Johnson and Christopher Enroth, “Managing mosquitoes at home,” Good Growing blog, University of Illinois Extension, June 12, 2026 – online text.
The Bottom Line
A mosquito treatment number is never a fact; it is a settlement between a counting instrument, a comparison, a clock, a set of repetitions, and a budget – and most of those parties never appear in the advertisement. You now own the six questions that make them appear, in one federal case, one regulation, four field studies, and two extension fact sheets. Ask them of the spray contractor, the trap brand, the marketplace listing, and yes, of us. The suppliers who survive the six questions are the ones worth a purchase order – and the ones whose cartons will still be legal in the market you sell in.
