IMPORTANCE
Chronic noncancer pain (hereafter referred to as chronic pain) is common among older adults and managed frequently with pharmacotherapies that produce suboptimal outcomes. Psychological treatments are recommended, but little information is available regarding their efficacy in older adults.
OBJECTIVE
To determine the efficacy of psychological interventions in older adults with chronic pain and whether treatment effects vary by participant, intervention, and study characteristics.
DATA SOURCES
MEDLINE, Embase, PsycINFO, and the Cochrane Library were searched from inception to March 29, 2017.
STUDY SELECTION
Analysis included studies that (1) used a randomized trial design,(2) evaluated a psychological intervention that used cognitive behavioral modalities alone or in combination with another strategy, (3) enrolled individuals with chronic pain (pain ≥3 months) with a sample mean age of 60 years or older, and (4) reported preintervention and postintervention quantitative data.
DATA EXTRACTION AND SYNTHESIS
Two of the authors independently extracted data. A mixed-model meta-analysis tested the effects of treatment on outcomes. Analyses were performed to investigate the association between participant (eg, age), intervention (eg, treatment mode delivery), and study (eg, methodologic quality) characteristicswith outcomes.
MAIN OUTCOMES AND MEASURES
Pain intensity was the primary outcome; secondary outcomes included pain interference, depressive symptoms, anxiety, catastrophizing beliefs, self-efficacy for managing pain, physical function, and physical health.
RESULTS
Twenty-two studies with 2608 participants (1799 [69.0%] women) were analyzed. Participants’ mean (SD) age was 71.9 (7.1) years. Differences of standardized mean differences (dD) at posttreatment were pain intensity (dD = −0.181, P = .006), pain interference (dD = −0.133, P = .12), depressive symptoms (dD = −0.128, P = .14), anxiety (dD = −0.205, P = .09), catastrophizing beliefs (dD = −0.184, P = .046), self-efficacy (dD = 0.193, P = .02), physical function (dD = 0.006, P = .96), and physical health (dD = 0.160, P = .24). There was evidence of effects persisting beyond the posttreatment assessment only for pain(dD = −0.251, P = .002). In moderator analyses, only mode of therapy (group vs individual) demonstrated a consistent effect in favor of group-based therapy.
CONCLUSIONS AND RELEVANCE
Psychological interventions for the treatment of chronic pain in older adults have small benefits, including reducing pain and catastrophizing beliefs and improving pain self-efficacy for managing pain. These results were strongest when delivered using group-based approaches. Research is needed to develop and test strategies that enhance the efficacy of psychological approaches and sustainability of treatment effects among older adults with chronic pain.