The science

Does CBT help with compulsive pornography use?

What CBT involves for compulsive porn use, what the two randomised trials found, and how to get it in the Netherlands and the EU.

By the WINNR team · 17 October 2026 · 11 min read

Someone has told you to get CBT. A GP, a forum, or an article that never explained the letters. You want two things: whether it works for this, and what happens if you walk into a room and start.

CBT is at once the most reasonable psychological treatment to try for compulsive pornography use and one of the least proven for it. Both halves are true, and most pages give you one.

The short answer

Two randomised trials carry most of the weight: a Swedish trial of therapist-delivered group CBT in 137 men with hypersexual disorder [1], and a larger trial of a web-based self-help programme built partly on CBT, in 264 people with problematic pornography use [8]. A preregistered systematic review covering 2000 to 2021 found 24 treatment studies in total for compulsive sexual behaviour disorder (CSBD) and problematic pornography use (PPU), four of them randomised against a waitlist or placebo, plus one randomised comparison of CBT with art therapy [2].

So nobody is lying to you when they recommend CBT. They are quoting a professional consensus rather than trial evidence, and you should know which you have been handed.

What CBT actually involves in the room

CBT is not a conversation about your childhood. It is a set of procedures, roughly in this order. If a therapist says they do CBT and you never see most of these, you are not getting it.

Functional analysis

The engine, and everything else derives from it. You record episodes: where you were, the time, what happened in the hours before, what you felt, what you did, what you got out of it, what it cost you later. Two or three weeks and your pattern becomes data rather than a story you tell yourself. In the systematic review of CSBD and PPU treatments, 18 of the 24 studies used psychotherapy integrating classical and newer CBT components, in which this assessment is the standard first step [2].

Thought records

You write down the thought that arrives just before use. "I have had a brutal week, I have earned this." "Five minutes and then I will work." Then you check it against your own record of what followed.

This is the half of CBT with the weakest component evidence. Longmore and Worrell found little sign that specifically cognitive interventions add value beyond the behavioural parts, and little empirical support for cognitive change being the causal mechanism. They also examined the claim that improvement arrives before the cognitive work begins, and found the evidence for it lacking [4]. That review is from 2007 and still argued about. Take it as a reason not to judge a therapist by how much thought-challenging they do.

Behavioural experiments

A thought record argues with a belief. An experiment collects evidence against it. You make a prediction specific enough to be wrong, then test it. "If I do not look tonight I will not sleep." You do not look, you record when you fell asleep, and the data is yours.

Stimulus control

Changing what is available, where and when. Phone charging in the kitchen, laptop out of the bedroom after a set hour. The Dual Control Model of sexual response holds that arousal reflects a balance of excitation and inhibition [5]. Stimulus control works on the excitation side, removing what presses the accelerator rather than asking you to hold the brake. Our guide to stopping has the practical version.

Relapse prevention planning

The last phase, and the one people skip. You take the high-risk situations the functional analysis found and write what you will do in each, in advance. Relapse prevention has a real meta-analytic base outside this behaviour: Irvin and colleagues pooled 26 studies, 70 hypothesis tests and 9,504 participants and found it generally effective, most clearly for alcohol [6]. The mechanics are in our post on if-then plans, and how you read a slip, which matters as much as the plan, is in our post on streak counting.

What has actually been tested on this behaviour

The group CBT trial

Hallberg and colleagues randomised 137 men diagnosed with hypersexual disorder at a Swedish clinic to seven weeks of group CBT (n = 70) or to a waiting list (n = 67) [1]. The primary outcome was the Hypersexual Disorder: Current Assessment Scale. The CBT group showed significantly greater reductions in hypersexual symptoms and sexual compulsivity, and greater improvement in psychiatric wellbeing, than the men who waited, and the effects held at three and six months.

Now the caveat, in the authors' own words: "The long-term treatment effects are vague due to the low response rate on follow-up measurements" [1]. They also note the study did not include women.

Two further limits. The comparison was a waiting list, so this means CBT beats waiting, not another active treatment. And hypersexual disorder is broader than pornography use.

The larger self-help trial

The biggest randomised test of any self-guided programme for this behaviour is Bőthe and colleagues' "Hands-off" trial [8]: 264 participants randomised to a six-week web-based self-help tool built on motivational interviewing, CBT, mindfulness and social-psychological techniques, or to a waiting list.

The result and its limitation belong together. Completers in the intervention arm reported substantially lower PPU than controls (d = 1.32), on a complete-case analysis in which only 34.5 per cent of participants provided follow-up data at all. Dropout was lopsided too: 89.4 per cent in the intervention arm against 44.7 per cent of controls. A d of 1.32 resting on 13 people in the arm that produced it is not a d of 1.32, because the people who keep answering are not a random sample of those who started.

It is a self-guided programme, not therapist-delivered CBT, so it does not show that CBT works for this. It shows that a structured self-guided programme drawing partly on CBT produced a large effect among the minority who finished. If you are weighing an app against a therapist, that distinction is the whole question.

Everything else

An internet-delivered version of the Swedish programme produced large symptom reductions in 36 men over 12 weeks, holding at three months, but had no control group, so it cannot tell you the treatment caused it [7].

Of the 24 studies in the systematic review, eight focused specifically on PPU, and the authors rated the quality of most as low or very low, with only four scoring above half the available CONSORT points [2]. They conclude that treatment, especially CBT, could be effective, and that the specificity of the effects is questionable.

Two randomised trials with different designs, populations and weaknesses is better than one. It is still not a literature, and a handful across two decades is not one either. Anyone telling you CBT is proven for pornography use has not read them.

What the guidelines assign, and what they do not

The WFSBP guideline on CSBD is the closest thing to an authoritative treatment document [3]. It states that "Psychoeducation and psychotherapy are first-choice treatments and should always be conducted."

Read the title before you read that as an evidence grade. The guideline covers "the assessment and pharmacological treatment" of CSBD, so the graded recommendations it issues, rated for efficacy, safety, tolerability and feasibility, are about medication. Psychotherapy is recommended first with no graded evidence category attached. That is a specialist panel's considered judgment, not a trial result, and the guideline says as much when it notes how few randomised trials exist.

Why clinicians reach for CBT anyway

First, breadth. A meta-review pooled 494 systematic reviews of CBT randomised trials covering 221,128 participants, across 14 of 20 physical and 13 of 20 mental health conditions, and found a pooled effect on health-related quality of life of SMD 0.23 (95% CI 0.14 to 0.33) [9]. Modest, consistent, tested on more conditions than any rival. Two notes travel with it: 351 of the 494 reviews were rated lower quality, and the prediction interval ran from -0.05 to 0.50, so in an untested condition the expected effect includes none at all.

Second, the nearest tested relative is a behavioural addiction rather than a substance one. A Cochrane review of psychological therapies for problem gambling covered 14 randomised trials and 1,245 participants and found benefit from CBT immediately after treatment, with little known about whether it lasts [10].

So the procedures map onto a cue-driven behaviour, and the approach has been tested on more things than anything else. A well-founded extrapolation, and still an extrapolation.

Do not pick a side in the CBT versus ACT argument. The two overlap more than the "third wave" branding suggests [2]. Our science page sets out what the evidence base does and does not support.

Getting CBT in the Netherlands

The route

Start at the huisarts. Many practices have a praktijkondersteuner GGZ who can take several sessions in the practice itself. If that is not enough, the huisarts refers you on: to basis-GGZ for milder or non-complex problems, or to gespecialiseerde GGZ for severe or complex ones [11, 12]. The referral is not a formality. Without a huisarts referral letter, your insurer does not reimburse psychologist care at all [11].

What it costs you

Psychological care sits in the basisverzekering, with no eigen bijdrage and no cap on sessions: the whole trajectory is covered [11]. But GGZ counts towards your eigen risico, which is €385 in 2026 [13], so your first stretch of therapy in a calendar year costs you up to €385 and everything after is covered.

Huisarts and praktijkondersteuner care does not count towards the eigen risico at all [11, 13], which is why the cheapest useful step is also the one you can take this week. One trap: an uncontracted provider may be reimbursed partly or not at all [11], so ask before the first appointment.

Waiting times

The agreed maximums, the Treeknormen, are 4 weeks to a first appointment, 10 weeks from there to the start of treatment, and 14 weeks in total, for basis-GGZ and specialistische GGZ alike [14].

The most recently published national figures are from the first quarter of 2023: 14.1 weeks in total for basis-GGZ and 20.9 weeks for specialistische GGZ, with the public monitor noting that updates are paused while the measurement method changes [15]. We will not quote a current national average we cannot verify, so check your own region and insurer.

While you wait: finding you care faster when the wait is too long, wachtlijstbemiddeling, is a statutory task of your insurer, not a favour, and most people never ask [16].

Sexologists, and the rest of the EU

A seksuoloog NVVS is a GZ-psychologist, psychotherapist or physician with a sexology specialisation; the register also lists consulenten seksuele gezondheid and professionals with a sexology annotation [17]. For a specifically sexual problem, that is often the better-matched clinician. The catch is money: the NVVS's own patient factsheet says care from NVVS professionals is often not yet covered by the basic package [17], so ask about reimbursement before you book.

Elsewhere in the EU the shape is usually the same, a primary-care doctor then a funded psychologist, but entitlements differ by country and we will not guess at yours. What does not differ much is that services built specifically for compulsive sexual behaviour are rare. CSBD has been diagnosable since the ICD-11 release [18], and our post on whether porn addiction is a real diagnosis explains what that code does and does not mean. A code existing does not mean a clinic exists near you.

So ask for someone with experience in compulsive sexual behaviour or sexual health, and take a generalist who will do a proper functional analysis over a specialist you cannot reach. That preference is our reasoning, not a trial finding.

Common questions

Does CBT work for porn addiction?

One randomised trial of 137 men found group CBT beat a waiting list at three and six months, though the authors call the long-term results vague because too few answered at follow-up [1]. A larger self-help trial drawing partly on CBT found a big effect among its completers, a third of those randomised [8]. Across the field there are four randomised trials against a waitlist or placebo, plus one randomised comparison of two therapies, most of low quality [2]. Probably helps, nowhere near proven.

How many sessions of CBT will I need?

The tested formats were short: seven weekly group sessions in the Swedish trial [1], 12 weeks for the internet version [7], six weeks for the self-help tool [8]. Dutch basis-GGZ has no fixed cap [11]. Treat a therapist who cannot say what the plan is by session three as a warning.

Is CBT or ACT better for porn use?

Nobody has run the trial that would answer that, and the two overlap heavily in practice [2]. If you have been losing a willpower fight with urges, our post on ACT for pornography use may fit better.

Can I do CBT on my own?

Parts of it. Functional analysis, stimulus control and if-then planning are self-administrable, and the self-help trial found real differences at six weeks among those who finished [8]. Thought records and behavioural experiments are harder alone, because you are both the experimenter and the person with a motive to fudge the result.

WINNR is a science-based app for men who want to regain control over their pornography use. It works the layer CBT works: your cues, your pattern, and what the behaviour does for you, mapped from your own data rather than an average. The app has not launched yet and is expected in mid-November 2026, and a layer connecting users with psychologists and coaches is in development. The evidence we draw on, and its limits, are on our science page.

A note on what this is

WINNR is not a treatment or therapy. If porn use is seriously affecting your life, speak to a licensed professional; if you are in crisis, contact your local emergency services.

Want a plan for your own pattern? Take the private 4-minute check

Fighting an urge right now? Open the urge page

#

Want a plan for your own pattern? Take the private 4-minute check. Fighting an urge right now? Open the urge page.