What We Fear Most

You may know the voice. It speaks before the interview, the phone call, the first day back, the message drafted nine times and never sent.

It says the thing will go badly. It says you are not up to it, and that people will see through you. It speaks in the present tense and with total confidence, so it sounds like knowledge. It is a forecast, and forecasts can be checked.

In business, FUD, short for fear, uncertainty and doubt, is the name given to selling by spreading worry about the alternative. The phrase itself is older. The earliest use we could find is in a letter written in Georgia, in the United States, in 1917, which called suspicion “a sign of fear, uncertainty and doubt”. This issue changes one word. Fear, underestimating and doubt.

Then it checks each against the evidence. How often anxious worries come true. Whether other people are as unwilling as the voice predicts. Whether we judge our own ability too harshly, or too kindly. Where the famous figures about worry and impostor syndrome came from. And what a person in this country can ask for when fear has started to run the day.

The answer is not that fear is always wrong. Anxiety in this country is real, measured and more common than it was, and some fears come true. The answer is that the voice goes wrong in predictable directions, and a forecast that keeps going wrong the same way deserves to be checked before it is obeyed.

This company was started by people who met inside the mental health system. We have no wish to make fear sound small.

Every figure here is sourced, dated and named. Where a number could not be traced to the body that published it, it is not printed as fact, and the last page says what was left out.

What this issue is not. It is not a pep talk, and it is not a promise that the thing you dread will not happen. It is a check on the forecasts, and a list of what can be asked for.

The country measures its anxiety every quarter by asking one plain question about the day before: overall, how anxious did you feel yesterday?

Other official measures show a clearer rise. The Office for National Statistics reported in June 2026 that in 2023 and 2024, 24.6 per cent of adults in the UK showed some evidence of depression or anxiety on a standard questionnaire, up from 19.7 per cent in 2018 to 2019.

Children carry it too. In October 2025 the NSPCC reported that Childline counsellors had delivered 12,543 sessions about anxiety in the past year, the most common mental health concern that children bring to the service.

A widely shared line about worry says that 85 per cent of what we worry about never happens, and credits it to a study at Cornell. We went looking for the study.

The students' gut feeling put the chance of a worry coming true at 62 per cent on average, and even their careful, reasoned estimates said 42 per cent. The real rate was 8.6 per cent.

So the honest version is narrower, and still worth having. In the study we found, of students whose worry had become a disorder, most of what they dreaded did not happen, their forecasts ran several times too high, and when the bad thing did come it was often no worse than they feared. It is one study of 29 students. It is not a law about everyone.

None of this means people always expect the worst. When people judge their own risk of ordinary misfortunes, a small study at University College London in 2011 found they took good news on board more readily than bad, though a paper from Cardiff University the same year questions whether people are unrealistically optimistic about their own risk at all. Which way a forecast goes wrong depends on what it is about. That is the reason to check each one.

Fear makes forecasts about other people too: that they will not want to talk, will not want to help, will find us dull. Researchers have tested those forecasts again and again.

Studies published in 2022 by Xuan Zhao and Nicholas Epley, with 2,118 adults in the United States, found that people who needed help underrated how willing others were and how good helping would make them feel, though not on every measure.

Small kindness follows the same pattern. In a study published in 2023, people in a Chicago park who gave away a cup of hot chocolate rated the size of the gesture at 3.76 out of 10. The people who received it rated it 7.00.

Most of these studies were done in the United States, and two of the results above come from the UK. Almost all of them point the same way. The fear lives in the asking, and the asking usually goes better than the fear predicted.

The best known study of self judgement is usually told as a story about fools who think they are clever. It is also a story about the people at the top, who thought they were merely good.

The researchers had a plain explanation for the top quarter's modesty. Those students assumed everyone else had found the test as easy as they had, and when they were shown other people's answers, they raised their estimates. Their modesty was a wrong guess about other people, the same kind of mistake the commuters made.

The famous chart is not in the paper. The curve of confidence that climbs, crashes and slowly recovers, widely shared as the Dunning-Kruger effect, appears nowhere in it: the paper's charts plot four groups by test score, with no time or experience on either axis.

Impostor syndrome began as something two psychologists noticed. In 1978 Pauline Clance and Suzanne Imes, in Atlanta, described what they had seen in more than 150 high achieving women in therapy, groups and classes: “an internal experience of intellectual phoniness”. They called it the impostor phenomenon. It was a clinical observation, and it measured no rate.

The newest pooled estimate, published in 2026, covers medical students: 49 per cent across 34 studies of 9,550 students, 51 per cent of women and 40 per cent of men. Its authors warn that the studies disagree widely and rely on unvalidated cut offs, which weakens the figure.

Clance and Imes thought the feeling was much rarer in men. Later research is split. Of the 33 articles in the 2020 review that compared the sexes, 16 found that women reported significantly more impostor feelings, and 17 found no difference.

Another widely repeated figure about doubt has no published evidence behind it. The claim that men apply for a job when they meet 60 per cent of the criteria, and women only when they meet 100 per cent, is credited to an internal Hewlett-Packard report, and we found no published version of it.

Impostor syndrome is not a diagnosis either. It has no entry in the World Health Organization's international classification of diseases, and the 2020 review says plainly that it “is not a recognized psychiatric disorder”. The feeling is real. The famous number is not a published finding.

The good news is in the NHS's own figures for England, and so is the catch.

Social anxiety and agoraphobia recover least often, but 69.5 per cent of the people who finished treatment for social anxiety still showed reliable improvement. Its courses also run longer: 11 sessions on average, against 7.8 for generalised anxiety.

Fear is good at one thing above all: keeping a person from asking. So here is what can be asked for, and by whom, starting tomorrow morning.

• Refer yourself In England, adults can refer themselves to NHS Talking Therapies for anxiety and depression on the NHS website, and in some areas from 16. You do not need a GP appointment, and you do not need a diagnosis. Its own list includes feeling very anxious about lots of different things, fear of social situations, panic attacks and phobias. In Wales, anyone aged 16 or over can sign up for a free online course based on cognitive behavioural therapy, or CBT, with no referral. In Northern Ireland, Lifeline can arrange counselling in your area within seven days.
• Ask when treatment starts Ask for the date of your second session as well as the first. The first is mainly assessment, and treatment proper begins at the second.
• Ask for the right kind For generalised anxiety, NICE's guideline starts with education and monitoring, then self help, then a choice of CBT or applied relaxation, usually 12 to 15 weekly sessions, or medication, with your preference guiding the choice. For social anxiety it recommends individual CBT and says: “Do not routinely offer group CBT in preference to individual CBT.”
• Help at work, in confidence If you are in paid work, or about to start, Access to Work's mental health support service in England, Scotland and Wales is free to you and your employer, lasts up to nine months, and can be used without your employer knowing. You can use it once. Apply to one of its two providers, Able Futures or Maximus.
• Adjustments If anxiety has a substantial effect on daily life, which in law means more than minor or trivial, and the effect has lasted, or is likely to last, a year or more, or is likely to come back, the Equality Act can count it as a disability. Once an employer knows, or should know, it must make reasonable adjustments and cannot charge you for them. The government's own example: your own desk instead of hot desking for someone with social anxiety. The Act does not apply in Northern Ireland.
• Ten minutes for the worry The NHS suggests setting aside a short worry time each day, 10 or 15 minutes, to write your worries down and look for solutions, and sorting the problems you can solve from the worries that are beyond your control. It costs nothing, and no one has to approve it.

The voice will not go away, and it should not, because some of what it warns about is real. It is wrong more often than it sounds, and in the same places: about the worry that will not come true, the stranger who would be glad to talk, the person who would be glad to help, and, for people who are good at something, where they stand.