Interventions for generalized anxiety disorder in older adults: Systematic review and meta-analysis

Abstract

Introduction

Generalized anxiety disorder (GAD) is one of the most common anxiety disorders in later life, with widespread consequences for individuals and society.

Objective

To perform a systematic review of the efficacy of controlled interventions for GAD in adults aged 55 years and older.

Method

Direct search of digital databases and the main publications on aging and iterative searches of the references from retrieved articles.

Results

Twenty-seven trials (14 pharmacological, 13 psychotherapeutic) fulfilled the inclusion criteria, reporting results from 2373 baseline participants. There were no differences between trials in their overall quality. Pooled treatment effects for pharmacological (OR = 0.32, 95% CI: 0.18, 0.54) and psychotherapeutic (OR = 0.33, 95% CI: 0.17, 0.66) trials were similar, with findings in each case favoring active interventions over control conditions.

Conclusions

Older adults with GAD benefited from both pharmacological and psychotherapeutic interventions. Future studies should investigate combined treatment with medication and psychotherapy.

Introduction

Recent research on late-life anxiety disorders has failed to confirm some previously held beliefs, including the notions that anxiety disorders rarely have their onset in later life or usually occur in comorbid relation to other disorders (Wetherell, Maser, & van Balkom, 2005). A recent review indicated that the prevalence of anxiety disorders ranged from 1.2 to 15% in community dwelling adults aged 55 years and older, and was almost twice as high in clinical settings (Bryant, Jackson, & Ames, 2008). In a national representative sample of community dwelling adults the prevalence of anxiety disorders was twice as high as the prevalence of mood disorders (Byers, Yaffe, Covinsky, Friedman, & Bruce, 2010). Generalized anxiety disorder (GAD) is one of the most common anxiety disorders in older adults, with surveys conducted with representative older cohorts obtaining 12-month rates between 1.0 (Grant et al., 2005) and 7.3% (Beekman et al., 1998).

Anxiety symptoms and disorders have widespread consequences for the person and the society. At an individual level, anxiety symptoms and disorders have been found to be associated with increased sleep disturbance (Brenes et al., 2009), intake of anxiety related medication (Gonçalves, Pachana, & Byrne, 2011), disability (Brenes et al., 2005), mortality (Ostir & Goodwin, 2006), and decreased cognitive functioning (Beaudreau and O’Hara, 2009, Mantella et al., 2007). At a societal level, late-life anxiety was related to increased use of health services, namely longer appointments with physicians (Stanley, Roberts, Bourland, & Novy, 2001), and inability to perform daily functions (Trollor, Anderson, Sachdev, Brodaty, & Andrews, 2007).

GAD has a low likelihood of spontaneous remission (Lenze, Mulsant, Mohlman, et al., 2005). An observational longitudinal study of community dwelling older adults found that at a 3-year follow-up more than half of those diagnosed with GAD at baseline fulfilled criteria for a mental health disorder (Schoevers, Deeg, van Tilburg, & Beekman, 2005). Despite the high prevalence, widespread impact and chronic course of the disorder, individuals with this disorder seldom seek the help of mental health professionals (Hunt, Issakidis, & Andrews, 2002), and there is a two-year mean delay between the onset of the disorder and the initiation of treatment (Grant et al., 2005).

Older adults are even less likely than young and middle aged people to seek help from a mental health professional, with the primary care physician being the first choice of the majority of older people with anxiety disorders (De Beurs et al., 1999). However, contrary to what has been previously believed, older adults seem to display positive attitudes towards mental health professionals and would be willing to accept help if offered (Arean et al., 2002, Mackenzie et al., 2008). A recent survey conducted with a sample of healthy older adults showed that when confronted with a hypothetical scenario of an anxiety disorder, three quarters of the participants chose psychotherapy as the preferred treatment, with the remaining participants selecting either pharmacotherapy or combined treatment (Mohlman, in press).

In comparison with other anxiety disorders, GAD has been neglected by researchers, with a relative paucity of published research devoted to this disorder (Byrne and Pachana, 2010, Dugas et al., 2010). There is an even more limited literature on the treatment of late-life GAD, in comparison with reports of treatment trials in other age groups and for other disorders. As older adults have an increased likelihood of comorbid physical disorders and polypharmacy, pharmacological trials have often used older age as an exclusion criterion (e.g., Enkelmann, 1991, Lecrubier et al., 1993). Even in those trials in which no age ceiling was defined, the proportion of older adults has been relatively small, with studies reporting an average sample age of approximately 40 years (e.g., Gelenberg et al., 2000, Rickels et al., 1997).

The limited recognition given to anxiety disorders in later life, along with the reluctance of older people to seek consultations with mental health professionals, might partially explain the sparseness of psychotherapeutic trials (Wetherell, Lenze, & Stanley, 2005). However, over the past decade there has been a gradual change in this trend, with an increasing number of treatment trials on anxiety disorders in older people being reported. Recent studies have reported on the efficacy of both pharmacological and psychotherapeutic interventions, with published reports generally favoring treatment over a control condition (Nordhus & Pallesen, 2003). Effect sizes for both clinician-rated and self-rated anxiety were found to be moderate to large for behavioral interventions and large for pharmacological interventions (Pinquart & Duberstein, 2007). In older people with a variety of different anxiety disorders, cognitive behavioral therapy (CBT) has been found to be significantly more effective in worry reduction than a waiting list control condition, although not more so than an active comparator (Hendriks, Voshaar, Keijsers, Hoogduin, & van Balkom, 2008). Conversely, another meta-analysis that included both controlled and uncontrolled trials found no additional gains to older adults with a variety of anxiety disorders when other CBT components were added to relaxation training alone (Thorp et al., 2009).

Notwithstanding the previously published results, it should be noted that the comparison of treatment trials has been constrained by the range of different anxiety disorders studied and the diversity of control conditions employed. Against this background, we undertook a systematic review and meta-analysis of controlled interventions for GAD in older adults, namely pharmacological, psychological, behavioral and alternative or life-style therapies. The main goal was to identify studies that had been conducted with the purpose of treating GAD in older adults. Only studies that reported a control condition, either active or passive, were included. Studies were assessed using qualitative and quantitative parameters. The quality of studies was assessed through the quality rating system developed by the Cochrane Collaboration Depression, Anxiety and Neurosis Review Group (CCDAN, Moncrieff, Chruchill, Drummond, & Mcguire, 2001). Quantitative analyses were conducted through random-effects analysis (DerSimonian & Laird, 1986), whereas publication bias was assessed using the funnel plot technique (Begg & Mazumdar, 1994). Because participant attrition can inflate results and reduce generalizability, we adopted an intent-to-treat approach to the analysis. This means that all baseline participants were included in the analysis, with drop-outs being rated as non-responders (Hollis & Campbell, 1999).