Hand washing
Checked against 2 peer-reviewed sources, most recently on 10 October 2026.
When researchers pooled randomised trials of programmes that got people cleaning their hands more often - with soap and water or with hand sanitiser - fewer people caught a respiratory infection such as a cold than in the comparison groups. The trials did not show a clear effect on flu. The free fix: make getting home your cue to wash with soap and water before you touch anything else.
The one honest fix
Every time you get home, wash your hands with soap and water before you touch anything else — your phone, the fridge, your face.
What the evidence says
In a Cochrane review of randomised trials, programmes that promoted hand hygiene led to fewer people getting an acute respiratory infection than in comparison groups that kept their usual routine or received less, such as written advice alone (risk ratio 0.86, 95% CI 0.81 to 0.90 - about 14% fewer people; 9 trials, 52,105 participants)
−14% · RR 0.86, 95% CI 0.81-0.90
Limitations & how to read this number
What the trials tested was a hand-hygiene programme - teaching, reminders and often free soap or hand sanitiser - not one hand wash at one moment of the day. Three of the nine programmes were built around alcohol hand sanitiser rather than soap and water, and the pooled result does not separate the two, so it is not a measure of soap and water alone. The nine trials ran in very different places: households in England, New York, Boston and Mumbai, childcare centres in Spain and Colombia, villages in Bangladesh and Kenya, and an army training camp in the United States, and several of them counted infections in young children rather than adults. The trials counted respiratory infections in different ways - symptoms reported by participants or parents over the study, symptoms in the past week at a survey visit, or clinic records - and mostly without a laboratory test. In most trials the participants knew which group they were in, and the reviewers noted that adherence to the interventions was relatively low. When the same trials looked only at flu-like illness or at flu confirmed by a lab test, they did not show a clear difference. The reviewers rated this result as moderate certainty.
SourceJefferson T et al. (2023) · Cochrane Database of Systematic Reviews · 10.1002/14651858.CD006207.pub6
In the same review's summary of findings, the typical (median) risk of an acute respiratory infection in the comparison groups was 380 in 1,000; at the pooled effect, that would be 327 in 1,000 with a hand-hygiene programme (95% CI 308 to 342) - about 53 fewer people per 1,000
327 per 1,000 vs 380 per 1,000 · 95% CI 308-342
Limitations & how to read this number
This is a worked example from the review, not a count from one trial: 380 in 1,000 is the typical (median) risk in the trials' comparison groups, and 327 is what a 14% reduction would mean at that starting level. The trials measured infections over different periods - from a few months to more than a year - and in different ways, so it is not a figure per year or per winter. It counts people who had an infection, not how many colds each person had.
SourceJefferson T et al. (2023) · Cochrane Database of Systematic Reviews · 10.1002/14651858.CD006207.pub6
How to actually do it
When you get home, go to the sink before your phone, the fridge or the sofa, and wash your hands with soap and water. The trials tested programmes that helped people clean their hands more often, not hand washing at one particular moment of the day - getting home is simply an easy, fixed cue to start with, after a day of touching shared surfaces.
Supporting findings (4)
In the same review, a subgroup analysis across all 19 hand-hygiene trials - each contributing whichever respiratory outcome had the most cases - found no evidence that the effect differed between trials the reviewers grouped as adult and trials in children; in the 8 trials grouped as adult, fewer people got a respiratory illness with a hand-hygiene programme (risk ratio 0.84, 95% CI 0.78 to 0.91; 41,951 participants)
−16% · RR 0.84, 95% CI 0.78-0.91
This is a subgroup of a broader combined outcome - any respiratory illness, flu-like illness or flu, whichever had the most cases in each trial - which the reviewers rated as low certainty because of how the trials were run and because their results varied. The review does not list in its text which trials it grouped as adult, so this split cannot be checked trial by trial, and some household trials included children as well as adults. It shows that the benefit was not limited to trials in children, but it is a less precise measure than the main result above.
SourceJefferson T et al. (2023) · Cochrane Database of Systematic Reviews · 10.1002/14651858.CD006207.pub6
In the same review, hand-hygiene programmes did not clearly reduce flu-like illness (risk ratio 0.94, 95% CI 0.81 to 1.09; 11 trials, 34,503 participants)
no significant difference · 95% CI 0.81-1.09
Flu-like illness was defined more strictly than a respiratory infection in general - each trial set its own definition, usually requiring a fever together with respiratory symptoms - so there were far fewer cases to compare, and the trials' results varied. The reviewers rated this as low certainty: it shows no clear difference, not proof that there is none.
SourceJefferson T et al. (2023) · Cochrane Database of Systematic Reviews · 10.1002/14651858.CD006207.pub6
In the same review, hand-hygiene programmes did not clearly reduce flu confirmed by a laboratory test (risk ratio 0.91, 95% CI 0.63 to 1.30; 8 trials, 8,332 participants)
no significant difference · 95% CI 0.63-1.30
Only some of the trials tested people for flu in a laboratory, so this comparison rests on far fewer people and the range of possible effects is wide. The reviewers rated it as low certainty: it shows no clear difference, not proof that hand washing does nothing against flu.
SourceJefferson T et al. (2023) · Cochrane Database of Systematic Reviews · 10.1002/14651858.CD006207.pub6
In a randomised trial in England of adults who shared a household, an online programme that encouraged more frequent hand washing was followed by fewer people reporting a respiratory infection over 16 weeks - 51% in the programme group against 59% without it (risk ratio 0.86, 95% CI 0.83 to 0.89; 16,908 participants followed up)
−14% · RR 0.86, 95% CI 0.83-0.89
This is one trial, and it is one of the nine behind the main result above, so it is not separate confirmation. Participants knew which group they were in and reported their own infections. The programme asked people to wash their hands more often with soap or hand gel and to plan for themselves when they would do it; it did not test washing at any one particular moment. It also recorded slightly more minor skin irritation in the programme group.
SourceLittle P et al. (2015) · The Lancet · 10.1016/S0140-6736(15)60127-1
Go deeper
Open what you care about
What the trials tested
The Cochrane review pooled nine randomised trials that compared a hand-hygiene programme - teaching, reminders and often free soap or hand sanitiser - with comparison groups that kept their usual routine or received less, such as written advice alone. They ran in homes, childcare centres, villages and an army training camp, in adults and in young children. One of them, in England, gave adults sharing a household an online programme that encouraged more frequent hand washing.
How big the difference is
In the comparison groups, the typical risk of a respiratory infection was 380 in every 1,000 people. At the pooled effect, that would fall to about 327 in 1,000 - around 53 fewer people out of every 1,000. It is a modest difference, for something that costs nothing.
What it did not show
When the same review looked only at flu-like illness, or at flu confirmed by a laboratory test, the trials did not show a clear difference. The benefit on this page is for respiratory infections in general, which are mostly colds.
Questions people ask
Does hand washing protect against flu?
The trials pooled on this page did not show a clear reduction in flu-like illness or in flu confirmed by a laboratory test. The clearer benefit was for respiratory infections in general, such as colds.
Do I need hand sanitiser?
No. Some of the programmes in these trials used hand sanitiser and others used soap and water. The pooled result does not separate the two, so this page makes no claim that one works better; soap and water at your own sink is simply the free way to do the fix.
Sources
- Jefferson T et al. (2023). Physical interventions to interrupt or reduce the spread of respiratory viruses. Cochrane Database of Systematic Reviews. Meta-analysis of randomised trials. doi:10.1002/14651858.CD006207.pub6
- Little P et al. (2015). An internet-delivered handwashing intervention to modify influenza-like illness and respiratory infection transmission (PRIMIT): a primary care randomised trial. The Lancet. Randomised controlled trial. doi:10.1016/S0140-6736(15)60127-1
Published 10 October 2026 · Evidence last verified 10 October 2026 · adversarial fact-check + human science gate — how we check.
Corrections are published openly on the corrections page. Educational content, not medical advice — talk to a professional about your own situation.