Where our claims come from.
Every figure we use on this site, with a link to the study it came from — including the ones with limits worth knowing about.
How common travellers' diarrhoea actually is
The figures behind the numbers we quote on the homepage and in the article.
Supports: Three to seven in ten travellers get sick on a two-week trip
CDC Yellow Book 2026 — Travelers' Diarrhea
The US Centers for Disease Control puts attack rates at 30% to 70% of travellers over a two-week period, depending on destination and season. The same chapter that sets out the standard precautions also concedes that travellers who follow them can still become ill — which is the single most important thing on this page.
Read the source →Supports: The same figures, independently reviewed
Incidence and risk factors for travellers' diarrhoea — systematic review and meta-analysis
A pooled analysis of cohort studies of short-term travellers from high-income countries. Useful as a peer-reviewed check on the CDC's range rather than taking a single source's word for it.
Read the source →Why the bottle is the part nobody thinks about
Supports: The average reusable bottle carries 20.8 million germs
WaterFilterGuru — Swabbing Water Bottles (2022)
Swab study of used reusable bottles found an average of 20.8 million colony-forming units, with spout-top and screw-top lids around 30 million each and squeeze-tops roughly a tenth of that.
Read the source →Worth knowing: this was run by a company that sells water filtration, and it hasn't been peer-reviewed. It's been widely reported, and the finding is consistent with what you'd expect, but we'd rather you knew where it came from than found out yourself.
Supports: Bacteria double every 20 minutes in a warm bottle
USDA Food Safety and Inspection Service — the Danger Zone
Bacteria grow fastest between 4°C and 60°C, doubling in as little as 20 minutes. Bali does not drop below about 24°C at any point in the year, so a bottle in a day bag never leaves that range.
Read the source →This is food-safety guidance, not bottle-specific research. Applying it to a water bottle is our inference, and we think a reasonable one.
What's in bottled water
Supports: Around 240,000 plastic fragments per litre
Rapid single-particle chemical imaging of nanoplastics — PNAS (2024)
Columbia University researchers using a new laser imaging technique found an average of roughly 240,000 detectable plastic fragments per litre of bottled water, about 90% of them nanoplastics — 10 to 100 times more than earlier estimates.
Read the source →This tested five bottles from three brands. It's a paper demonstrating a new detection method, not a survey of the bottled water market — so treat it as an indication of scale, not a figure for any particular brand.
Supports: Plain-English summary of the same study
Columbia University Mailman School of Public Health
The university's own write-up, if you'd rather not read the paper.
Read the source →Does UV-C actually work?
These studies are about the technology, not about our bottle. They establish that UV-C at around 280nm inactivates the organisms we're talking about. They do not test a Boab Bottle — see the section below.
Supports: 280nm UV-C inactivates bacteria, viruses and protozoa
Efficacy of UVC-LED radiation in bacterial, viral, and protozoan inactivation — Water Research (2024)
Measures how effectively UV-C LEDs inactivate a range of organisms, and how much that depends on exposure dose and water clarity. It's also the source for our honest limit: cloudy water casts shadows that organisms hide in, so this doesn't work on a muddy river.
Read the source →Supports: 280nm outperforms conventional UV lamps
Improved disinfection performance for 280nm LEDs over 254nm low-pressure UV lamps — Nature Scientific Reports (2023)
Found that a 279nm UV LED needs roughly half the dose of a conventional 254nm low-pressure lamp to achieve the same reduction in E. coli.
Read the source →Supports: The same technology used in municipal water treatment
Quantitative evaluation of municipal wastewater disinfection by 280nm UVC LED
Assessment of 280nm UV-C LEDs in municipal wastewater treatment. This is what we mean when we say UV-C isn't experimental — it's infrastructure.
Read the source →What we haven't tested yet.
Everything above establishes that UV-C works. None of it tests a Boab Bottle.
We're a new company and we don't yet have our own third-party lab report. When we do, it will be published here in full — the lab, the date, the organisms, the numbers, and anything that didn't go our way.
Until then, we'd rather tell you that than let a list of other people's studies imply something we haven't earned. If that's not good enough yet, we understand. There's a 30-day guarantee for exactly that reason.
Last checked August 2026. If you think we've got something wrong, or you'd like a source we haven't listed, email us and we'll either fix it or explain ourselves.
Nothing on this page is medical advice. Boab Bottle is not a medical device and does not prevent illness. Most travellers' diarrhoea is food-borne.