Efficacy and Safety of Pharmacological, Physical, and Psychological Interventions for the Management of Chronic Pain in Children

A WHO Systematic Review and Meta-analysis

Emma Fisher; Gemma Villanueva; Nicholas Henschke; Sarah J. Nevitt; William Zempsky; Katrin Probyn; Brian Buckley; Tess E. Cooper; Navil Sethna; Christopher Eccleston


Pain. 2022;163(1):e1-e19. 

In This Article


We conducted a systematic review and meta-analysis to determine the efficacy and safety of pharmacological, physical, and psychological therapies for the management of chronic pain in youth, used to support the World Health Organisation guidelines on the management of chronic pain in children.[40] We found each modality of intervention separately showed some benefits of reducing pain intensity posttreatment, but these effects were not maintained at follow-up. Some physical and psychological interventions also reduced functional disability posttreatment, and effects were maintained for psychological interventions at follow-up. Pharmacological and physical therapies presented very little data that could be analysed for other outcomes, meaning any absence of effects and beneficial effects should be interpreted with caution. Most data were available for psychological therapies; however, we found no beneficial effects for improving HRQOL, emotional functioning, role functioning, or sleep in the meta-analyses. A beneficial effect of psychological interventions was found for global satisfaction with treatment. Adverse events were poorly reported, particularly in physical and psychological trials.

We also conducted secondary analyses investigating equivalence trials of pharmacological and physical studies. These analyses did not elucidate one type of intervention being more beneficial than another. We were able to conduct a number of subgroup analyses for psychological treatments by control type, pain condition, treatment duration, route (face-to-face vs remotely delivered), therapy type, and size. The estimates of effects for these subgroup analyses were similar to the primary analyses and driven by the subgroup within most studies. For example, most studies delivered CBT and the estimate of effect for all outcomes was similar to those presented in the primary analyses. We are unsure of the efficacy of other types of therapies (eg, hypnosis) because of lack of individual studies contributing to analyses. One notable subgroup analysis we conducted separated studies delivered face-to-face and remotely; remotely delivered therapies seem to be a viable mode of delivery for children with chronic pain.

Overall, certainty of evidence ranged from high to very low; we downgraded outcomes due to limitations in study design, indirectness, inconsistency, and imprecision. Imprecision was a particular problem with pharmacological and physical interventions due to a small number of studies and participants contributing to the analyses. Emotional functioning outcomes were most consistently rated as higher certainty of evidence despite the lack of benefit for pharmacological, physical, or psychological interventions. However, many of these interventions do not aim to improve emotional functioning. Overall, the study quality as assessed by risk of bias was unclear or high risk of bias. Incomplete outcome data and selective reporting bias were particular areas across the 3 modalities that raised particular concerns and contributed to lower certainty of evidence judgements.

As highlighted in a recent Lancet Child and Adolescent Health Commission on paediatric pain, one key goal is to "make pain better."[11] The most common intervention given to children with chronic pain in practice is pharmacological; however, there is more evidence (and higher certainty evidence) for psychological therapies. Interdisciplinary care is recognised as the optimal standard for the management of chronic pain, which include a combination of pharmacological, psychological, and physical therapies.[17,40] However, this may not be feasible in many parts of the world, and more work is needed to understand how to manage chronic pain in low-income and middle-income settings.

We used Cochrane methodology to conduct this systematic review, but despite this, there are a number of limitations. First, it is possible that a small number of studies were missed. Second, although our search criterion was broad, we excluded complementary and alternative therapies and passive forms of physical therapies such as massage and manipulation.

This systematic review provides the most up-to-date evidence for the management of chronic pain in youth. These findings support previously published systematic reviews[10,13,14] but importantly expand the scope in terms of population and outcomes.

There are some clear gaps in the literature, which are important to address in future research. We found few studies of chronic pain management in lower-income settings. Children and adolescents report chronic pain in low-income and middle-income countries, but these countries face unique challenges when implementing pain services.[36] We also found no studies of children with palliative conditions or intellectual disabilities. Although the RCT design may not be the most appropriate to investigate efficacy in children with palliative conditions, more research is needed to determine the harms and benefits of these interventions for this population. We found one trial registration describing a study completed in 2018 that included children with cancer, but we received no response from the authors when requesting an update. Research is needed in children with cancer diagnoses and chronic pain. Most studies included older children and adolescents. The mean age for all 3 intervention classifications was 12 to 13 years. We did not find any studies of infants with chronic pain. Finally, we were unable to present data for some critical outcomes. In particular, AEs are poorly reported in psychological and physical trials, and there were surprisingly few events in pharmacological trials. Future trials across all 3 modalities, particularly pharmacological and physical therapies, should evaluate all critical outcomes, following the established PedIMMPACT recommendations.[30]

There are potential avenues for future funding and research. In particular, we encourage funding in and research the following areas:

  1. Investigation into how interventions can be best delivered in low-income countries.

  2. Future trials in high-income countries should include much larger sample sizes (ie, more than 200 participants/arm). Small studies inflate the estimate of effect and reduce our confidence in the beneficial effects of interventions.

  3. Adverse event reporting must be mandatory in any trial delivering any intervention to children with chronic pain. Funding bodies, policy makers, and journal editors can help to enforce this.

  4. Evidence for pharmacological and physical interventions is currently small and should be improved. Creative solutions that do not rely on the RCT design, particularly for pharmacological interventions, should be explored.

Regarding clinical practice recommendations, there is most evidence that psychological therapies, namely CBT, should be used in clinical practice. Physical and pharmacological therapies may also be used to manage chronic pain, but beneficial effects were limited to posttreatment data and there were few AEs, although AEs were poorly reported in physical interventions. The WHO guidelines on the management of chronic pain in children conditionally recommend the use of physical, psychological, and pharmacological interventions, and psychological therapies can delivered face-to-face or remotely.[40]

In conclusion, this systematic review and meta-analysis provides the most up-to-date evidence for the management of chronic pain in youth. Pharmacological, physical, and psychological therapies may reduce pain intensity in various chronic pain conditions, with the most robust evidence available for psychological therapies.

Deviation From Protocol

  1. We had initially planned to exclude trials that compared physical therapies with other physical therapies; however, for comprehensiveness, we included these and summarise them in a separate comparison.

  2. In our protocol, we stated that we would preferentially extract depression and anxiety to other emotional functioning outcomes. However, because of the many studies reporting data on both outcomes, we performed separate analyses of depression and anxiety outcomes.