Around 20% of the adult population experience chronic pain. Non-pharmacological pain treatments have shown promising effects. Yet, contrary to pharmacological trials, there are no standardised methods for evaluating the effect of non-pharmacological treatments. Studies implementing matched control conditions have found significantly smaller effects than studies comparing treatments to treatment as usual. These findings, which are summarised in this review, emphasise the importance of standardised methods and more precise estimates of how much pain relieve may be obtained from non-pharmacological pain treatment.