Depression in older adults tends to be more chronic and, as a consequence, psychological interventions addressing it tend to be less efficient.

In a recent article published in the journal Maturitas, Cuijpers and collaborators conducted a meta-analysis of trials for psychological treatments for depression in older adults. They noted that the rapid increase in the number of such trials over the recent years could make it possible to tackle research questions not approached in previous meta-analyses, for instance regarding the efficiency of previously less studied therapies, such as life review treatments.

Methods

The authors included randomised controlled trials in older adults (over 50 years of age), examining psychological interventions for depression. Depression was defined according to a diagnostic interview or through a cut-off on a self-report scale. In included trials, psychological treatments were compared to a control group, another psychological treatment, or pharmacotherapy. Studies in which a combination of psychotherapy and pharmacotherapy was compared with psychotherapy or pharmacotherapy alone were also included.

Included studies were assessed for methodological quality, using four criteria of the Risk of Bias Assessment tool developed by the Cochrane Collaboration: adequate sequence generation, allocation concealment, blinding of outcome assessors, incomplete outcome date (this was coded as positive if analysis were conducted following the intent-to-treat principle, meaning all randomised participants were included in the analysis).

Effect sizes (ES) were calculated as standardized mean differences for each comparison between psychotherapy and a control group or comparison group at post-test and at follow-up of 6 months or longer. They were then transformed in Hedges’ g in order to adjust for bias due to small size (.20, .50 and .80 correspond to small, moderate and respectively large ES). The standardized mean difference was also transformed in an index more amenable to a clinical interpretation, the NNT, indicating the number of patients that have to be treated to generate one additional positive outcome.

Results

44 studies met the inclusion criteria and were included in the meta-analysis.

• The quality of the included studies was very variable:
  • Only 13 studies out of 44 meeting all 4 quality criteria
  • Another 13 meeting 2 or 3 quality criteria
  • And the remaining 18 showing a low quality (0 or 1 of the 4 criteria)
• The effect of psychotherapy versus a control group post-intervention was medium, with a mean ES (g) of 0.64 (95% CI 0.47 to 0.80), corresponding to an NNT of 2.86. However, the authors also reported high heterogeneity and identified 5 studies as potential outliers. It is interesting to note that 4 of the 5 outliers examined a life review intervention. With the exclusion of these studies, the overall ES decreased to 0.43 (95% CI 0.33 to 0.52)
• The authors documented significant publication bias, using three different methods for examining this aspect. Moreover, with the adjustment for publication bias, the ES decreased to 0.35 (95% CI 0.16 to 0.54)
• The authors also computed the mean ES for the effect of psychotherapy versus control at follow-up (6 months or longer), which was small (g= 0.27, 95% CI 0.16 to 0.37) and there was no evidence of heterogeneity or publication bias in this case
• Effect sizes were larger in studies comparing psychotherapy with waiting list control than in the ones where care as usual or another type of control group was used
• Also,  studies with a higher quality score resulted in lower effect sizes than studies with a lower quality score
• The authors did not find evidence for a dose response, as there was no significant association between the number of sessions and ES
• Both cognitive behavior therapy and problem-solving therapy were more effective than the other psychotherapies
• On the other hand, non-directive supportive counselling was significantly less effective than other therapies

Conclusions

The authors concluded:

In this updated meta-analysis of psychological treatments of depression in older adults, we could confirm that these treatments have moderate to high effect on depression, which were maintained at 6 months or longer post-randomization.

However they also emphasized that:

The effects are probably overestimated because of publication bias and because of the low quality of several of the included studies.

While cognitive behavior therapy and problem solving therapy were found to be more efficient than other therapies investigated, and non-directive counseling less efficient, the authors caution that, along with study quality, research allegiance in favor of therapies other than counselling could have been a factor in these results, as it is a common phenomenon in this field of research. However, this cannot be established at the moment due to the reduced number of studies for the comparisons among different psychotherapies.

Limitations

• The methodological quality of many of the studies included in the meta-analysis was fairly low
• The number of effect sizes for different types of interventions included was relatively small, implying that results for the comparisons between different types of psychotherapies should be considered with caution
• It is not clear whether the included studies are representative of older adults in general, as while most studies included older adults over 60 or 65, it was not clear whether elderly individuals, over 75 or 80, were also part of the respective samples
• Most studies included participants scoring over a cut-off on a self-report measure or with subthreshold depression, and the number of studies targeting older adults with a diagnosed depressive disorder was relatively small

Link

Cuijpers P, Karyotaki E, Pot AM, Park M, Reynolds CF. (2014) Managing depression in older age: Psychological interventions. Maturitas. doi:10.1016/j.maturitas.2014.05.027 [Pubmed abstract]