What we do
Three programmes, and the limits of each
Prevention, rehabilitation and information. They are one path rather than three services: a child reached by the first never needs the second, a child who does need it comes back through work funded by the third. What follows is what each consists of, and what each will not do.
Prevent
Recruitment happens somewhere specific, to someone specific, through someone the child already knows. Prevention means being in those places before it does.
What this consists of
- Working with the adults already around a child: teachers, sports coaches, youth workers, the people who notice a change first and usually do not know what to do with it.
- Sessions in schools about what grooming actually looks like, taught in the terms a young person recognises rather than in the terms an adult would use.
- Following where recruitment moves. It is a market and it changes platform faster than any curriculum does.
- Being reachable by an adult who is worried about a specific child and does not yet want to make a report.
What this will not do
We do not run our own reporting line and we do not assess suspicions. Sweden has 114 14 for the police, 116 111 for BRIS, and 112 when a child is in immediate danger. An organisation that collects reports it has no mandate to act on delays the people who do have one.
Give to PreventSupported
School programmes reliably increase what children know about abuse: a standardised mean difference of 0.61 across 18 trials, and the knowledge was largely still there six months later at 0.69.
Systematic review · 24 randomised trials, 5 802 children · Walsh, Zwi, Woolfenden och Shlonsky, Cochrane Database of Systematic Reviews 2015, CD004380.pub3
Programmes running more than three sessions worked better than shorter ones, and children from about eight years old gained more than younger ones. That is why our sessions are a series and not a visit.
Meta-analysis · 31 studies, 9 049 participants · Lu, Barlow, Meinck, Walsh och Wu, Research on Social Work Practice 2023, 33(4), 390–412
A warning shown at the moment someone reaches for material cut the share who continued from 75 to 35 per cent. Interrupting at the point of decision does something that education months earlier does not.
Randomised trial · 474 men aged 18 to 30 · Prichard, Wortley, Watters, Spiranovic och Scanlan, randomiserat försök
Not supported
The claim that these programmes get children to tell someone does not hold. The unadjusted odds ratio is 3.56, but children sit in classes rather than as individuals, and Cochrane's own adjustment for that drops it to 3.04 with an interval of 0.75 to 12.33, no longer significant.
Systematic review · 3 trials, 1 788 children · Walsh m.fl., Cochrane 2015, CD004380.pub3, känslighetsanalys för klustring
Nobody has shown that these programmes reduce abuse. Cochrane's own summary says the long-term benefit in terms of incidence has not yet been adequately measured. What is established is knowledge, simulated skills, and an uncertain effect on telling someone.
Systematic review · The whole review · Walsh m.fl., Cochrane 2015, CD004380.pub3, plain language summary
Rehabilitate
Getting out is an event. Recovering is not. The work that matters here is measured in years, and most of it is unglamorous and administrative.
What this consists of
- Trauma-informed psychological support, delivered by licensed psychologists. We do not provide therapy ourselves and we do not employ people to practise outside their licence.
- Getting a child back into school, which usually means arguing with an administrative system rather than doing anything therapeutic.
- Practical help with the paperwork that decides whether someone can stay: applications, appointments, and being in the room when it matters.
- Support for the adult who is caring for the child, because that adult is usually the reason the child stays somewhere safe.
What this will not do
We are not a clinic and we do not hold a treatment mandate. Where a child needs care, the care comes from licensed practitioners and from the health system that is responsible for it. Our part is making sure the door to that system is actually open.
Give to RehabilitateSupported
Trauma-focused cognitive behavioural therapy is the best-supported treatment there is for children after trauma, with a Hedges g of 1.06 against no treatment at the end of therapy.
Meta-analysis · 70 randomised trials, 5 528 children and young people · Hoppen, Wessarges, Jehn m.fl., nätverksmetaanalys, publicerad online december 2024, årgång 2025
NICE recommends individual trauma-focused CBT for those aged 7 to 17 presenting more than three months after the event, typically over 6 to 12 sessions. This is a defined course of treatment, not open-ended support.
Clinical guideline · UK national guideline · NICE, Post-traumatic stress disorder, NG116, rek. 1.6.11 och 1.6.12
Not supported
Measured against another active treatment rather than against nothing, the effect roughly halves: 0.55 rather than 1.06, and in a second meta-analysis 0.52 rather than a before-and-after 1.14. The larger figure contains time passing and children who would have improved anyway.
Meta-analysis · 28 randomised trials and 33 uncontrolled studies, 4 523 participants · Hoppen m.fl. 2024/2025, och Thielemann, Kasparik, König, Unterhitzenberger och Rosner
For the youngest children the evidence barely exists. SBU found seven studies of psychological treatment for children aged seven or under after sexual violence, of which two had merely moderate risk of bias, both from the United States and both over twenty years old. Sweden has no national care programme for these children.
Register study · 7 studies, 2 usable · SBU, Psykologiska och psykosociala behandlingsinsatser för barn 7 år eller yngre som utsatts för sexuellt våld
Recovery is not what the numbers after care look like. Among Swedes aged 19 to 25 who had been in out-of-home care, 38 per cent had completed upper secondary education, against 82 per cent of their peers.
Register study · National register study · Socialstyrelsen 2026, registerstudie om skolgång hos placerade barn
Inform
Most of what circulates about this subject is wrong, and the wrong version is what people act on. Publishing what the law and the sources actually say is the cheapest useful thing we do.
What this consists of
- A public reference on what the law says, in five chapters, with every claim carrying the document it comes from.
- The figures that circulate in this field and do not survive being checked, published with the reason each one fails.
- Training for professionals who meet children and are expected to recognise something they have never been taught to see.
- Answering journalists properly, including when the honest answer is that no defensible number exists.
What this will not do
We do not publish a figure we cannot source, even when a sourced one would be less striking, and even when everybody else in the sector is publishing the striking one.
Give to InformSupported
Public campaigns do shift behaviour, by a little. Across 48 campaigns the average effect was r = 0.09, which works out at roughly eight per cent of a population changing something.
Meta-analysis · 48 campaigns, 168 362 people · Snyder, Hamilton, Mitchell, Kiwanuka-Tondo, Fleming-Milici och Proctor
A national helpline campaign measurably raised the number of people reaching for help: 11 190 unique callers across 20 235 calls, and 109 432 visitors to the site.
Cohort study · National campaign, measured against helpline traffic · Utvärdering av Stop It Now, Storbritannien
Not supported
Awareness work can make things worse. The US National Youth Anti-Drug Media Campaign spent close to a billion dollars over six years and produced no favourable effect; greater exposure predicted weaker intention to avoid cannabis.
Cohort study · National campaign, six years · Hornik, Jacobsohn, Orwin, Piesse och Kalton, utvärdering 1998–2004
Frightening young people works in reverse. Scared Straight and programmes like it increased offending across nine randomised trials, with an odds ratio of 1.68.
Systematic review · 9 randomised trials, 946 young people · Petrosino, Turpin-Petrosino, Hollis-Peel och Lavenberg, Campbell systematisk översikt
Training the adults around children has not been shown to produce reports. Across eleven randomised trials of professional training, not one measured actual referrals to social services or police. What was measured was what participants said about themselves.
Systematic review · 11 randomised trials, 1 484 participants · Walsh, Eggins, Hine, Mathews m.fl., Cochrane-översikt
The paperwork that decides whether someone can stay
A person trafficked into Sweden from outside the EU has, on paper, a route to a residence permit. In practice the route runs through other people. The application is not made by the person it concerns, and a refusal is not theirs to appeal. Helping someone through that is unglamorous, slow, and among the most consequential things we do.
EU Directive 2011/36/EU obliges member states to assist victims unconditionally: assistance may not be made to depend on a willingness to testify. The gap between that obligation and how a case actually moves is where this part of the work sits.
The exact provisions of the Swedish Aliens Act are not quoted here yet. They were researched but the verification pass on them did not complete, and this site does not publish a legal reference it has not checked.
None of this runs on goodwill
Prevention costs staff time in places that do not generate a story. Rehabilitation costs years. Publishing a reference costs the work of reading every source it cites. Regular funding is the only kind any of it can be planned against.