Cognitive behavioural therapy, or CBT, has long been considered the standard treatment for generalised anxiety disorder. It works well for many people, though roughly half of those treated go on to make a full recovery, leaving clear room for improvement.
A study by Nordahl and colleagues compared CBT with a newer approach, metacognitive therapy (MCT), in adults with long-term generalised anxiety disorder. Of 246 people assessed, 81 were randomly assigned to CBT, MCT or a wait-list control group. Both treatments helped, but MCT produced significantly better outcomes than CBT (mean difference 9.762, 95% CI 2.679 to 16.845, P = 0.004). Recovery rates were 65% for MCT compared with 38% for CBT, and this gap remained at two-year follow-up.
The two therapies take different approaches to worry. CBT typically examines whether anxious thoughts are realistic. MCT instead looks at a person's relationship with worry itself: the beliefs that keep it going, such as thinking that worrying keeps you safe, or that worry cannot be controlled. Working with these beliefs directly appears to make a meaningful difference for many people.
CBT remains a well-supported treatment, and this is one study, not the final word. But it does suggest that when one approach hasn't helped as much as hoped, another may be worth discussing. If anxiety is affecting your daily life, speaking with a psychologist about your options is a reasonable and worthwhile step.
Background: two therapies, one target
Generalised anxiety disorder is a condition in which worry becomes chronic, excessive and difficult to control, often spreading across many areas of life. Cognitive-behavioural therapy has long been the standard treatment. It works by identifying anxious thoughts and testing how realistic or useful they are, alongside strategies for managing physical tension and avoidance.
Metacognitive therapy takes a different angle. Rather than examining the content of worried thoughts, it targets metacognition: a person's beliefs about their own thinking, such as the conviction that worrying keeps them safe, or that the process cannot be switched off. The proposition is that these beliefs, rather than the worries themselves, keep the cycle running.
What this study did
Nordahl and colleagues ran a randomised controlled trial, a design in which participants are assigned at random to different treatments so the groups are on average similar apart from the therapy received. That helps rule out other explanations for any difference found. Of 246 adults with long-term GAD who were assessed, 81 were randomised into three groups: CBT with 28 people, metacognitive therapy with 32, and a wait-list control group of 21 who received no treatment during the study period and acted as a baseline. Outcomes were measured before treatment, after treatment, and again at a two-year follow-up.
What they found
Both therapies produced real improvement compared with the wait-list group, confirming that treatment rather than the passage of time was responsible for the change. Metacognitive therapy outperformed CBT, with a mean difference of 9.762 on the outcome measure, a 95% confidence interval of 2.679 to 16.845, and a p value of 0.004. A confidence interval gives a plausible range for the true difference between treatments, and because this range sits entirely above zero, the advantage is unlikely to reflect chance.
In practical terms, 65% of people receiving metacognitive therapy met criteria for full recovery, against 38% of those receiving CBT. That gap was still present two years later, which suggests the benefit was not temporary.
Why this matters
The authors note that both treatments were delivered with equal therapeutic care and expectation, so the difference in outcomes cannot simply be attributed to non-specific factors such as therapist warmth or a patient's expectation of improvement, which can influence results regardless of technique. That strengthens the case that something in the method of metacognitive therapy, targeting beliefs about worry rather than its content, is doing genuine work.
Limits worth knowing
This was a single trial with modest group sizes of 28, 32 and 21 people, so the precise figures could shift if the study were repeated elsewhere. The authors who developed the treatment manuals used in the study also have financial ties to books and materials on these therapies. That is disclosed transparently, and it is worth bearing in mind when weighing the results. Replication by other research groups and in other settings would help establish how widely the findings apply.
If worry has begun running your life, both approaches are established and evidence-based, and a psychologist can help you work out which might suit you.