Gould RL, Coulson MC, Howard RJ
J Am Geriatr Soc 2012 Feb;60(2):218-29
DESIGN: Electronic literature databases and the Cochrane Trials Registry were searched for articles. A systematic critical review, random-effects meta-analysis, and meta-regression of randomized controlled trials were conducted.
SETTING: Community outpatient clinics.
PARTICIPANTS: People with diagnoses of anxiety disorders.
MEASUREMENTS: Outcome measures of anxiety and depression.
RESULTS: Twelve studies were included. CBT was significantly more effective than treatment as usual or being on a waiting list at reducing anxiety symptoms at 0-month follow-up, with the effect size being moderate, but when CBT was compared with an active control condition, the between-group difference in favor of CBT was not statistically significant, and the effect size was small. At 6- but not 3- or 12-month follow-up, CBT was significantly more effective at reducing anxiety symptoms than an active control condition, although the effect size was again small. Meta-regression analyses revealed only one factor (type of control group) to be significantly associated with the magnitude of effect sizes.
CONCLUSION: The review confirms the effectiveness of CBT for anxiety disorders in older people but is suggestive of lower efficacy in older than working-age people. The small effect sizes in favor of CBT over an active control condition illustrate the need to investigate other treatment approaches that may be used to substitute or augment CBT to increase the effectiveness of treatment of anxiety disorders in older people.