Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jun 1.
Published in final edited form as: J Consult Clin Psychol. 2014 Jan 20;82(3):375–391. doi: 10.1037/a0035491

Effect of Cognitive Behavioral Therapy for Anxiety Disorders on Quality of Life: A Meta-Analysis

Stefan G Hofmann 1, Jade Q Wu 1, Hannah Boettcher 1
PMCID: PMC4029924  NIHMSID: NIHMS556188  PMID: 24447006

Abstract

OBJECTIVE

Although cognitive-behavioral therapy is effective for treating anxiety disorders, little is known about its effect on quality of life. To conduct a meta-analysis of cognitive-behavioral therapy for anxiety disorders on quality of life, we searched for relevant studies in PubMed, PsycINFO and the Cochrane Library, and conducted manual searches.

METHOD

The search identified 44 studies that included 59 CBT trials, totaling 3,326 participants receiving cognitive-behavioral therapy for anxiety disorders. We estimated the controlled and within-group random effects of the treatment changes on quality of life.

RESULTS

The pre-post within-group and controlled effect sizes were moderately strong, Hedges’ g = 0.54 and Hedges’ g = 0.56, respectively. Improvements were greater for physical and psychological domains of quality of life than for environmental and social domains. The overall effect sizes decreased with publication year and increased with treatment duration. Face-to-face treatments delivered individually and in groups produced significantly higher effect sizes than internet-delivered treatments.

CONCLUSION

Cognitive-behavioral therapy for anxiety disorders is moderately effective for improving quality of life, especially in physical and psychological domains. Internet-delivered treatments are less effective in improving quality of life than face-to-face treatments.

Keywords: Quality of Life, Life Satisfaction, Anxiety Disorders, Cognitive Behavioral Therapy


Anxiety disorders are the most prevalent psychiatric disorders, with a lifetime prevalence rate of 28.8% (Kessler et al., 2005). These disorders are associated with high personal and economic costs (DuPont et al., 1996) and low quality of life (Cramer, Torgersen & Kringlen, 2005; Mendlowicz, & Stein, 2000; Olatunji, Cisler, & Tolin, 2007; Rapaport, Clary, Fayyad, & Endicott, 2005). For example, it has been reported that obsessive-compulsive disorder (Koran, Thienemann, & Davenport 1996), panic disorder (Candilis et al.,1999; Rubin et al., 2000) and social anxiety disorder (Safren, Heimberg, Brown, & Holle, 1996–1997; Wittchen & Beloch, 1996) have substantially poorer quality of life than community samples. In some cases, anxiety disorders have an even greater impact on quality of life than chronic medical disorders (Spitzer et al., 1995; Sherbourne, Wells, & Judd, 1996).

Quality of life (QOL) is difficult to define. It includes subjective well-being, life satisfaction, perceptions of social relationships, physical health, economic status, and functioning in daily activities and work (Angermeyer & Kilian, 1997; Mendlowitz & Stein, 2000). Accordingly, the assessment of QOL typically includes subjective views of one’s life circumstances, perceptions of mental and physical health, social and family relationships, and functioning at work and home (DuPont et al., 1996).

Cognitive behavioral therapy (CBT) is an effective treatment for reducing symptoms of anxiety disorders (Hofmann & Smits, 2008). Although symptom reduction is an important goal of treatment, some authors have urged investigators to include QOL as another important indicator of treatment efficacy (Frisch, 1998; Gladis, Gosh, Sishuk, & Crits-Christoph, 1999). The objective of this study was to conduct a quantitative review of the effect of CBT on QOL in patients with anxiety disorders. Although there are different emphases of the various cognitive and behavioral techniques for the range of anxiety disorders, one important commonality of all of these treatment protocols is the premise that cognitions causally influence fear and anxiety, and that the dysfunctional beliefs and cognitive distortions contribute to the maintenance of anxiety disorder (Hofmann, Asmundson, & Beck, 2013). This premise and one of the defining features of a variety of different treatment protocols and include more traditional CBT protocols as well as more modern mindfulness-based cognitive therapy, At the same time, this core assumption is the primary distinguishing feature to other treatments, such as Acceptance and Commitment Therapy and mindfulness-based stress reduction. Our review focuses only on CBT protocols that share the core premise of the centrality of maladaptive cognitions.

We examined controlled and within-group random effect sizes of changes in QOL during the course of CBT for specific phobias, panic disorder with and without agoraphobia, social anxiety disorder, generalized anxiety disorder, obsessive compulsive disorder and posttraumatic stress disorder. To examine whether study quality moderated the effect of CBT on QOL, we quantified study quality by using Effective Public Health Practice Project criteria (EPHPP; Thomas, Ciliska, Dobbins & Micucci, 2004). The effects of study quality and publication year are two important reporting items identified by the PRISMA group (Moher et al., 2009). In a large-scale analysis of 2,400 patients, treatment length has been demonstrated to positively correlate with efficacy of cognitive-behavioral treatment (Howard, Kopta, Krause, & Orlinsky, 1986). Furthermore, internet-based delivery of cognitive behavioral therapy for anxiety disorders has proven to be promising but not uniformly efficacious across extant trials (Andersson, 2009). Therefore, we also examined whether treatment administration modality (individual face-to-face, group face-to-face, and internet-delivered) moderated the effect of the treatment on quality of life.

Methods

Searching

A search was conducted on February 19, 2013 in PubMed, PsycINFO, and the Cochrane Library for all available studies. The following three sets of search terms were used simultaneously: 1) quality of life or quality-of-life; 2) cognitive-behavi* or cognitive behavi* or behavio[u]ral therapy or cognitive therapy; and 3) anxiety or anxious or panic or agoraphobia or social phobia or social anxiety or SAD or generalized anxiety or GAD or obsessive compulsive or obsessive-compulsive or OCD or specific phobia or simple phobia or post-traumatic stress or posttraumatic stress or PTSD or acute stress or ASD. In addition, manual searches for potentially relevant studies were conducted via published papers’ reference lists.

Selection

Studies were selected by two of the authors and three independent trained assessors. Studies were included in the present meta-analysis if: 1) they included at least one cognitive-behavioral intervention, and if this was the primary treatment (i.e. not an adjunct to a primarily pharmacological intervention); 2) they included a sample diagnosed with one or more anxiety disorders; 3) they included a sample of adults at or above the age of 18; 4) they included at least one measure of QOL at pre- and post-intervention; and 5) they provided sufficient data for performing an effect sizes meta-analysis. In case of disagreement regarding a study’s inclusion qualification between assessors, the authors discussed the case until consensus was reached.

Studies were excluded if: 1) a full text or full English translation was unavailable; 2) the study was a qualitative study, a meta-analysis or a review paper; 3) the anxiety disorder studied was secondary to another psychiatric condition or a non-psychiatric medical condition (e.g., cancer patients with anxiety); and 4) the data reported in the study overlapped with those reported in another study considered for inclusion.

As for how we determined the primary anxiety disorder being evaluated in each study, we deferred to the original study’s authors. That is, whichever disorder they specifically recruited (i.e., listed in their inclusion criteria), was the common principal diagnosis among all patients in their trial, and therefore, the principal disorder we considered the trial to evaluate. We took precautions to exclude studies in which anxiety disorders were secondary to non-anxiety and non-psychiatric conditions.

In case two articles reported data from the same trial, the article providing the most complete data was chosen. In case multiple control groups were used alongside the target intervention group, the most active control group was chosen as the comparison condition (e.g., if both a stress management group and a waitlist control group were used, the data of the stress management group was chosen as the comparison to CBT).

Validity Assessment

Two trained independent assessors judged the quality of each trial using the following domains (Thomas et al., 2004): 1) selection bias; 2) study design (i.e., to what extent trials were randomized and/or controlled); 3) confounders; 4) blinding; 5) data collection methods (i.e., self report, assessment, physiological measures); 6) withdrawals and drop-outs; 7) intervention integrity; 8) appropriateness of analysis to study question. For each domain, a score of “Strong,” “Moderate,” or “Weak” was assigned according to quantitative standards issued by the EPHPP, then domain scores were used to generate a global score for each trial that ranged from 1 to 3, with 1 being the best score (Thomas et al., 2004). For example, controlled trials in which assessors and participants were both blind to condition received a score of “strong” on the blinding domain, whereas trials in which neither assessors nor participants were blind to condition received a score of “weak” on the blinding domain. Then, following EPHPP guidelines, a global quality score was computed based on the number of domains with “weak” scores. The global score for each trial could range from 1 to 3, with 1 being the best score and reflecting no “weak” domain ratings; 2 reflecting one “weak” rating, and 3 reflecting two or more “weak” ratings (Thomas et al., 2004). Each study was independently judged by two assessors, who received extensive prior training supervised by the first author. After all studies were assessed, consensus was reached through discussion among assessors and the authors where there was disagreement.

In order to address potential publication bias, a fail-safe N (Rosenthal, 1991; Rosenthal & Rubin, 1988) was conducted. The fail-safe N represents the number of studies not included in the meta-analysis needed to nullify the effect size, and must be greater than 5K + 10 for the effect size to be considered robust. In addition, we constructed a funnel plot to assess publication bias, and further used the Trim and Fill method (Duval & Tweedie, 2000) to examine whether negative or positive trials are under- or overrepresented. This method takes into account sample size of studies and can be used to recalculate the effect size estimate.

Data Extraction

For each selected study, we examined all of the reported measures assessing quality of life and anxiety symptoms. All identified quality of life measures had been previously psychometrically validated. We found no reason to exclude any quality of life measures, with one exception: we chose to exclude data from Marks’ Quality of Life Scale, which assesses QOL specifically in individuals with asthma. Numerical data for QOL and anxiety measures were then extracted from the studies in order to compute pre- to posttreatment changes. One author (JQW) extracted data based on objective criteria (i.e. means and standard deviations, where available, of pre- and post-assessment scores on measures of anxiety symptoms and quality of life. Then, another author (HB) cross checked data from 20% of included studies, with 100% agreement. Data on pre- and posttreatment measures were extracted for CBT treatment conditions, as well as control groups (i.e., subjects who did not receive any treatment, such as in a waitlist control condition) and active comparison groups (i.e., subjects who received clinical care that was not CBT, such as enhance usual care or stress management training). In cases in which relevant data was not reported in the published study, the corresponding authors were contacted and asked to supply the required data.

Study Characteristics

Information about participant characteristics, study design, details of intervention, outcome measures used and other study characteristics was also extracted.

Quantitative Data Synthesis

We used Hedges’ g and its 95% confidence interval as an indicator of effect size for QOL outcomes and anxiety symptom outcomes (Hedges & Olkin, 1985). Hedges’ g is a version of Cohen’s d that takes into account bias from small sample size (Hedges & Olkin, 1985). Within-group effect size were calculated using the following formula: d=(Y1¯-Y2¯SDifference)2(1-r), where Y1¯ is the pretreatment sample mean, Y2¯ is the posttreatment sample mean, SDifference is the standard deviation of the difference, and r is the correlation between pretreatment and posttreatment scores. Hedges’ g can be computed by multiplying d by correction factor J(df)=1-34df-1, where df is the degrees of freedom to estimate the within-group standard deviation.

The controlled effect sizes were computed using the following formula: g=Δ¯CBT-Δ¯CONT(nCBT-1)SDCONT2+(nCONT-1)SDCBT2(ntotal-2)×(1-34(nCBT+nCONT)-9), where Δ is the mean pre- to posttreatment change, SD is the standard deviation of posttreatment scores, n is the sample size, and CONT refers to the control condition.

In order to calculate the effect sizes, pre-post measure correlations were needed. If these correlations were not provided in published reports, we followed recommendations by Rosenthal (1993) to assume a conservative estimation of r = 0.7. Further, because studies included in this meta-analysis were not functionally identical, we calculated the QOL effect size estimates using the random-effects model rather than the fixed-effects model (Hedges & Venea, 1998; Moses, Mosteller, & Buehler, 2002).

Following Cohen (1988), we interpret effects as small (0.2), medium (0.5) or large (0.8). In trials where there were multiple measures of QOL, we averaged effect size estimates across these measures for each trial. Differences in outcome between subsets of trials were evaluated with Cochran’s Q test of heterogeneity.

Moderator Analyses

We examined whether the QOL effect sizes varied as a function of study characteristics (study year, treatment length, study quality, treatment modality), or clinical characteristics (diagnostic category, anxiety symptom improvement). For categorical moderators, we computed separate QOL effect sizes for each group. For continuous moderators, we used meta-regression analyses to compute unstandardized regression coefficients. All analyses were conducted using the program Comprehensive Meta-Analysis, Version 2 (Borenstein, Hedges, Higgins, & Rothstein, 2005).

Results

Trial Flow

Our study selection process is shown in Figure 1. Of 858 hits identified for potential inclusion through database and manual searches (the oldest of which was published in the year 1985), 52 published articles met all inclusion criteria. Of these articles, 14 did not report adequate data for our analysis needs in their published version, so corresponding authors for each were contacted for additional data. Six of these authors were able to provide the necessary data, resulting in a final count of 44 studies included in the present meta-analysis. Across all 44 publications, we identified 59 independent trials. These trials included data for 3,326 patients. The targeted disorders included social anxiety disorder (18), panic disorder (18), posttraumatic stress disorder (5), generalized anxiety disorder (8), obsessive compulsive disorder (7), mixed anxiety disorders (2), and specific phobia (1).

Figure 1.

Figure 1

Flow diagram of study selection process.

aAcross the 44 published papers selected for analysis, there were 59 distinct CBT trials.

Study Characteristics

To evaluate the quality of the studies, we used Effective Public Health Practice Project rating system (EPHPP, Thomas et al., 2004). As seen in Table 1, the scores ranged from 1 to 3 with a median of 1 (M = 1.51, SD = 0.66). The intercoder agreement was 93% after the first round of assessments, after which 100% consensus was reached through discussion between the assessors and the authors. Details of study characteristics are shown in Table 1. The Table further depicts the type of the disorder that was targeted, the number of subjects included in each trial, the percent of female subjects per trial, the mean age of subjects, the type of comorbid diagnoses, information on concurrent medication use, details on the CBT intervention, number of treatment sessions, time duration of reach treatment session, details about the comparison conditions, and measures of quality of life and anxiety.

Table 1.

Study Characteristics

Study Study
quality
Disorder Total
N
%
Female
Age
Mean
(SD)b
Comorbid
Diagnoses
Medication CBT
intervention
No. of
treatment
sessions
or
modules
Time per
session
(min.)
Comparison
condition(s)
QOL
measures
Anxiety
measures
Agdal et al., 2012 2 Specific Phobia (intra-oral injection) 49 78.2 35.5 (12.2) Unknown Unknown CBT, brief CBT 5, 1 180, 60 None QOLI IPS-A, DAS
Andersson et al., 2006 1 SAD 64 56.3 36.4 (9.4) Comorbid diagnoses: 12.5% Allowed stable. Any: 21.9% Internet-delivered self-help n/a n/a Waitlist QOLI LSAS, SPS, SIAS, BAI
Arch et al., 2012 1 Mixed anxiety disorders 128 52.3 37.9 (11.7) Comorbid anxiety: 33.1%
Comorbid MDD: 23.6%
Allowed stable. Any: 48.0% CBT 12 60 Acceptance and Commitment Therapy QOLI ADIS-IV, ASI
Carlbring et al., 2005 1 PD 49 71.0 35.0 (7.7) Comorbid agoraphobia: 51%
Comorbid other anxiety: 49%
Comorbid MDD: 6%
Allowed stable; Antidepressants; 36.7%, anxiolytics: 14.3% ICBT, live CBT 10 Modules 60 None QOLI BAI
Carlbring et al., 2006 1 PD 60 60 36.7 Excluded comorbidities in need of treatment Allowed. Any: 54% Internet-delivered self-help 10 Modules n/a Waitlist QOLI BAI
Carlbring et al., 2007 2 SAD 26 69.2 33.5 (9.3) Comorbidities allowed but unknown Allowed stable. Any: 30.7%, antidepressants: 11.5% Internet-delivered self-help 9 Modules n/a None QOLI LSAS, SPS, SIAS, BAI
Carlbring et al., 2011 1 Any DSM-IV anxiety disorder 54 76 38.8 (10.7) Comorbid diagnoses: 50% Allowed stable. Any: 33% ICBT 6–10 modules n/a Attention control (online discussion group) QOLI BAI
Cordioli et al., 2003 1 OCD 47 51.1 36.5 (12.8) Comorbidities allowed but unknown Allowed stable; 44.7% using anti-obsessionals CBGT 12 120 Waitlist WHOQOL-BREF Y-BOCS, HAM-A
Craigie et al., 2008 2 GAD 23 74 43.4 (13.1) Allowed. Any: 78%
MDD current/remission:69%
Other anxiety: 26%
Allowed stable; 35% using antidepressants or anxioyltics MBCT 9 120 None Q-LES-Q DASS, BAI
de Sousa et al., 2006 1 OCD 56 76.8 38.5 (11.8) No comorbidities Allowed but unknown CBGT 12 120 Sertraline (100 mg/day)a WHOQOL-BREF BAI, Y-BOCS
Diefenbach et al., 2007 2 OCD 70 Unknown 36.8 (11.2) Comorbidities allowed but unknown Allowed stable; 80% using anti-obsessionals CBT 15 Unknown None SDS Y-BOCS
Eng et al., 2001 3 SAD 25 40 35.8 (11.7) Unknown Unknown CBGT 12 Unknown None QOLI SIAS, SPS, LSAS
Eng et al., 2005 3 SAD 40 42 34.4 (14.9) Comorbidities allowed but unknown Unknown CBGT 12 120 None QOLI SIAS, SPS, BFNE
Furmark et al., 2009_trial1 1 SAD 120 67.5 36.1 Comorbidities allowed but unknown Allowed stable; 6.6% using ICBT, Bibliotherapy 9 Modules n/a Waitlist QOLI LSAS-SR, SPS, SIAS, SPSQ, BAI
Furmark et al., 2009_trial2 1 SAD 115 67.8 34.7 Comorbidities allowed but unknown Allowed stable; 14% using ICBT, Bibliotherapy + discussion group 9 Modules n/a Waitlist, Internet-delivered applied relaxation QOLI LSAS-SR, SPS, SIAS, SPSQ, BAI
Gilliam et al., 2011 2 OCD (hoarding) 35 85.7 55.1 (10.6) Allowed. Any: 82%, MDD: 61%, SAD: 30%, GAD: 15%, OCD: 15%, ADHD: 12% Allowed stable. Any: 74%, SSRI: 50%, atypical antidepressant: 35%, stimulants: 24%, benzodiazepine: 18% CBGT 16–20 90 None SDS ADL-H, SIR
Hedman et al., 2011, a 1 SAD 126 35.7 35.3 Comorbidities allowed but unknown Allowed stable. SSRI: 19.8%, SNRI: 4.8% Internet CBT; CBGT 15 Modules 75 None EQ-5D LSAS
Kiropoulos et al., 2008 1 PD 86 72 39.0 (11.1) Allowed. Any: 71.4%, MDE: 11%, dysthymia: 5%, SAD: 16%, GAD: 17%, SP: 11%, PTSD: 3%, Hypochondriasis: 9%, Alcohol abuse: 2% Allowed stable. Any: 48.2% ICBT, CBT 6 Modules or 12 sessions 60 None WHOQOL-BREF PDSS, ASP
Klein et al., 2009 1 PTSD 16 82.40% 39.5 (10.7) Allowed. Any: 50.9%, GAD: 16%, Depression:12%, SAD: 9%, sepcific phobia: 5%, OCD: 2%, dysthymia: 2%, alcohol dependency: 2% Allowed. Any: 38.6% ICBT 10 Modules n/a None WHOQOL-BREF PDSS, DASS
Klein et al., 2010 2 PTSD 22 77.3 43 Allowed. Total unknown. Panic: 27%, MDE/SAD/SP: 23%, dysthymia/GAD: 18%, substance/OCD: 14% Allowed stable but unknown. ICBT 10 Modules n/a None WHOQOL-BREF DASS
Koszycki et al., 2007 1 SAD 52 52.8 38.4 Allowed. Any: 19%, GAD: 11%, Depression: 7%, Dysthymia: 6% Allowed. Any: 28.3% CBGT 12 150 Mindfulness-based stress reduction QOLI LSAS, SIAS, SPS
Ledley et al., 2009 1 SAD 38 57.9 34.87 Allowed. 1 additional: 31.58% 2 additional: 10.5%, dysthymia: 18.4%, GAD: 15.8%, MDD: 13.2%, SP: 7.9% Allowed stable. Any: 26.3%
Paroxetine: 10.5%
Citalopram: 5.3%
buproprion: 2.6%
nefazadone: 2.6%
clonazepam: 2.6%
imipramine: 2.6%
21.1%
CBT 16 60 Delayed treatment SDS, QOLI LSAS, SIAS, SPS
Marchand et al., 2009 1 PD 111 79 38.94 Comorbid anxiety disorders (30%) or MDD (8%) Allowed Stable. Anxiolytic or antidepressant: 62% Brief CBT, CBT, CBGT 7, 14 60 Waitlistc QLSI MI, ASI
Mortberg et al., 2011 1 SAD 67 62 38.5 Allowed. MDD: 27%, PD: 7% Anx NOS: 6%, ED: 5%, substance abuse: 4%, dysthymia: 3%, GAD: 3% No medication allowed. Intensive CGBT, CBT 16 Unknown None SDS LSAS, SPS, SIAS, FNA
Paunovic et al., 2001 1 PTSD 16 18.8 37.9 Unknown Allowed Stable. Any: 75%
An1tidepressants: 31.3%
Benzodiazepine: 12.5%
Antidepressants and Benzodiazepines: 18.8%
Neuroleptics: 6.3%
Muscle Relaxants: 6.3%
CBT 16–20 60–120 None QOLI BAI, STAI-T, STAI-S
Paxling et al., 2011 1 GAD 82 79.8 39.3 Allowed. MDD: 22.5% Allowed Stable. Any: 37.1% ICBT 8 Modules n/a Waitlist QOLI PSWQ, BAI, GAD questionnaire
Pier et al., 2008 1 PD 65 38.5 37.9 Unknown Allowed Stable; Antidepressant: 36.9%
Benzodiazepine: 9.2%
ICBT 6 Modules n/a None WHOQOL-BREF PDSS, ASP, DSASS
Rufer et al., 2010 2 PD 55 62 40 Allowed. Any: 40%, MDD: 28%, OCD: 7.3%, dysthymia: 5.5%, SAD: 3.6%, AN: 1.8%, alcohol abuse: 1.8%, GAD: 1.8%. Allowed. Any: 35% CBGT 5 150 None SF-36 PAS
Schneier et al., 2010 2 GAD 24 45.8 41 Unknown No medication allowed (with the exception of zolpidem for insomnia); 29.2% started
Escitalopram as part of the study
CBT 12–14 90–120 CBT + Escitaloprama Q-LES-Q HARS, PSWQ, GAD severity scale
Schneier et al., 2012 2 PTSD 37 54 50.27 Allowed. Axis I: 70.3%
Axis II: 16.2%
No psychotropic medication allowed 2–4 weeks prior to study (with the exception of zolpidem for insomnia); 51.4% assigned to the Paroxetine condition CBT 10 90 CBT + Paroxetinea Q-LES-Q CAPS
Schnurr et al., 2003 1 PTSD 325 0 50.7 Unknown Allowed stable but unknown. Trauma Focused CBGT 30 90 Present-centered therapy SF-36 CAPS, PCL
Shandley et al., 2008 2 PD 96 79.2 40.9 Allowed. GAD: 22.9%
MDD: 22.9%
SAD: 20.8%
SP:18.8%
Dysthymia: 13.5%
PTSD: 7.3%
Hypochondriasis: 6.3%
OCD: 4.2%
Alcohol dependence: 3.1%
Substance abuse: 1%
Allowed stable. SSRI: 15.6%
Benzodiazepine: 13.5%
SNRI: 7.3%
SSRI + Benzodiazepine: 6.3%
Tricyclic antidepressant: 3.1%
Tricyclic antidepressant + SSRI: 1%
SSRI + SNRI: 1%
Benzodiazepine + SSRI + Antipsychotic: 1%
SSRI + Antipsychotic: 1%
RIMA + Benzodiazepine: 1%
Anticonvulsant + Benzodiazepine + Antipsychotic: 1%
Total: 52.1%
ICBT + psychologist, ICBT + general practitioner 5 Modules n/a None WHOQOL-BREF ASP, PDSS, DASS
Simpson et al., 2008 1 OCD 108 43 39.2 Allowed. MDD: 25%, Other anxiety disorder: 30% Allowed. SRIs were prescribed to all participants. Other: 37% EX/RP 17 90–120 Stress Management Training Q-LES-Q HAM-A, Y-BOCS
Simpson et al., 2010 1 OCD 30 47 39.9 Allowed. Any: 50% Allowed Stable. SRI: 37%
SRI + other: 13%
Other: 3%
EX/RP, EX/RP + MI 18 90 None Q-LES-Q Y-BOCS
Stanley et al., 2009 1 GAD 134 78.4 66.9 Allowed. Any: 40.3%, MDD: 44.8% Allowed. Anxiolytic: 17%
Antidepressant: 31%
CBT 10 Unknown Enhanced Usual Care SF-12 GADSS, PSWQ, HAM-A, BAI
Stanley, Beck, et al., 2003 2 GAD 80 75 66.2 Allowed. MDD: 28%, SAD: 22%, SP: 19%, Affective NOS: 9%, PD: 7%, Dysthymia: 5% No medication allowed. CBT 15 Unknown Minimal Contact Control QOLI, LSIZ PSWQ, WS, STAI, HAM-A
Stanley, Hopko, et al., 2003 1 GAD 134 83.3 70.6 Allowed. MDD: 42%, SP: 17%, SAD: 8% Unknown, CBT 10 Unknown Enhanced Usual Care QOLI, SF-36 GAD severity, PSWQ, BAI
Telch et al., 1995 1 PD 156 68.3 34.8 MDD: 25.4%, SP: 21.8%, SAD: 17.3%, GAD: 10.9%, Dysthymia: 7.3% Allowed stable but unknown. CBGT 12 Unknown Waitlist SAS, SDS TPARF, SPRAS
Tilfors et al., 2008 2 SAD 37 81.1 31.4 Unkown Allowed stable but unknown. Internet-based self-help, exposure 9 Modules 135 None QOLI SPSQ, LSAS-SR, SPS, SIAS, BAI
van Apeldoorn et al., 2012 3 PD 83 54.7 37.5 Allowed. Any: 50% Unknown CBT 21 50 SSRIa, CBT+SSRIa RAND-36 HAM-A, PAI
Wagner et al., 2011 3 PTSD 15 86.7 29.3 Allowed.
MDD: 93%, Anxiety: 93%
Unknown ICBT 10 Essays n/a None EUROHIS-QOL PDS, HCSL-25
Watanabe et al., 2010 2 SAD 40 56.25 34.2 Allowed. Mood: 41.7%, Anxiety: 10.4% Allowed. Benzodiazepine: 25%
Antidepressant: 50%
CBGT 10–20 2 None SF-36 SPS, SIAS
Wetherell et al., 2003 1 GAD 75 80 67.1 Allowed. Any: 52%, SP: 20%
Depressive disorders: 19%
SAD: 15%, PD: 8%, PTSD: 7%
OCD: 5%
Hypochondriasis: 1%
Allowed stable. Any: 40%
Type
Antidepressant: 23%
Benzodiazepine: 15%
Buspirone: 4%
CBGT 12 Unknown Discussion Group, Waitlist RAND-36 PSWQ, HAM-A, BAI
Wetherell et al., 2009 2 GAD 15 83.9 72.2 Allowed. MDD: 13%, PD: 13%, Dysthymia: 6.5%, agoraphobia: 3.2%, SAD: 3.2%, SP: 3.2% Allowed stable. Any: 51.6% CBT 12 Unknown Community Treatment SF-36 HAM-A, PSWQ
Note. Definitions of abbreviations are listed below:
  • Disorders: GAD = generalized anxiety disorder; OCD = obsessive compulsive disorder; PD = panic disorder; PTSD = posttraumatic stress disorder; SAD = social anxiety disorder.
  • CBT interventions: CBT = cognitive behavioral therapy; ICBT = Internet cognitive behavioral therapy; CBGT = cognitive behavioral group therapy; EX/RP = exposure and response prevention; PE = prolonged exposure; MI = motivational interviewing; MBCT = mindfulness-based cognitive therapy
  • QOL measures: EQ-5D = EuroQol; LSIZ = Life Satisfaction Index-Z; Q-LES-Q = Quality Of Life Enjoyment And Satisfaction Questionnaire; QLSI = Quality Of Life Satisfaction Inventory; QOLI = Quality Of Life Inventory; RAND-36 = Medical Outcomes Study short form self-report health survey; SAS = Social Adjustment Scale; SDS = Sheehan Disability Scale; SF-12 = MOS 12-item short-form health survey; SF-36 = MOS 36-item short-form health survey; WHOQOL-BREF = World Health Organization Quality of Life Assessment.
  • Anxiety measures: ADIS-IV = Anxiety Disorders Interview Schedule-IV; ADL-H = Activities of Daily Life for Hoarding; ASI = Anxiety Sensitivity Index; ASP = Anxiety Sensitivity Profile; BAI = Beck Anxiety Index; BFNE = Brief Fear of Negative Evaluation Scale; CAPS = Clinician-Administered PTSD Scale; DAS = Dental Anxiety Scale; DASS = Depression Anxiety Stress Scales; FNA = functional nonadjustment; FQ = Fear Questionnaire; GADSS = Generalized Anxiety Disorder Questionnaire Short Form; HAM-A = Hamilton Rating Scale for Anxiety; HARS = Hamilton Anxiety Rating Scale; HCSL-25 = Hopkins Symptoms Checklist; IPS-A = Injection Phobia Scale - Anxiety; LSAS = Liebowitz Social Anxiety Scale; LSAS-SR = Liebowitz Social Anxiety Scale - Self Report; MI = Mobility Inventory for Agoraphobia; PAS = Panic and Agoraphobia Scale; PCL = PTSD Checklist; PDSS = Panic Disorder Severity Scale; PSWQ = Penn State Worry Questionnaire; SIAS = Social Interaction Anxiety Scale; SIR = Social Impact Ratings; SPRAS = Sheehan Patient Rated Anxiety Scale; SPS = Social Phobia Scale; SPSQ = Social Phobia Screening Questionnaire; SSAI = Spielberger State Anxiety Inventory; STAI = Spielberger Trait Anxiety Inventory; TPARF = Texas Panic Attack Record Form; WS = Worry Scale; Y-BOCS = Yale-Brown Obsessive Compulsive Scale.
a

Comparison groups involving pharmacological treatment data were not used to calculate controlled effect size estimates, as they constitute active treatments instead of an appropriate control condition for CBT; instead, these studies were treated as within-group in statistical analyses.

b

In cases where separate standard deviations were not provided for multiple groups, we report the mean age only across all groups.

c

For this study, data for the waitlist control group were unavailable. Thus, although the study was controlled, we used only data from the active condition in our uncontrolled effect size analysis.

Participant/Patient Characteristics

Participants in the analyzed trials included both males and females (with the exception of Schnurr et al., 2003, which used an all-male veteran sample), with most samples averaging ages in the mid-thirties to mid-forties (except for trials that specifically examined anxiety in elderly populations in their sixties and beyond). All but three studies (Mortberg et al., 2010; Schneier et al., 2010; Schneier et al., 2012) included participants on stable psychotropic medications, including those indicated for anxiety disorders. Among the most common medications used by participants during the course of treatments were selective serotonin reuptake inhibitors and, less frequently, other antidepressants and benzodiazepines. All but one trial (Arch et al., 2012) excluded participants with psychosis, substance use/dependence, and bipolar disorder.

QOL Measure Characteristics

We extracted data from all QOL measures relevant to general physical and psychological health, as well as quality of social and environmental/role functioning. The two most frequently used scales were the Quality of Life Inventory (QOLI; Frisch, Cornell, Villanueva & Retzlaff, 2005) and the World Health Organization Quality of Life Assessment (WHOQOL-BREF; WHOQOL Group, 1998). The QOLI is a 32-item scale that measures quality of life broadly, in domains such as love, work, recreation, self-esteem, surroundings etc. and has been shown to be sensitive to treatment-related change (Frisch et al., 2005). The WHOQOL-BREF is a 26-item questionnaire measuring a similar range of domains, and is used in a wide range of nonpsychiatric medical settings (WHOQOL Group, 1998).

The WHOQOL-BREF was the only QOL outcome measure for which separate domain means were reported. These domains were: 1) Physical (i.e., energy/fatigue, pain/discomfort, sleep/rest); 2) Psychological (i.e., emotions, self esteem, cognition, bodily image/appearance); 3) Social (i.e., personal relationships, social support, sexual activity); and 4) Environment (i.e., finances, safety/security, freedom, access to educational and health resources, participation in recreation, physical environment, transport).

Other QOL measures used in the studies included: Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q; Endicott, Nee, Harrison & Blumenthal, 1993), Life Satisfaction Index (LSIZ; Wood, Wylie, & Sheafor, 1969), Sheehan Disability Scale (SDS; Sheehan et al., 1998), EuroQol (EQ-5D; EuroQol-Group, 1990), MOS 12-item and 36-item short-form health surveys (SF-12, SF-36; Jenkinson et al., 1997; Ware & Sherbourne, 1992), Quality of Life Systemic Inventory (QLSI; Duquette, Dupuis & Perrault, 1994), Social Adjustment Scale (SAS; Weissman et al., 1978), EuroHis-QOL (Schmidt, Muhlan & Power, 2006), and the Medical Outcomes Study short form self-report health survey (RAND-36; RAND Health Sciences Program, 1992). All of these QOL measures are self-report instruments that have been validated in clinical populations with good psychometric properties.

Quantitative Data Synthesis

Pre-post within-group effect sizes for all QOL measures

The random effects meta-analysis yielded an overall QOL effect size of Hedges’ g = 0.54 (95% CI: 0.45–0.63, z = 11.26, p<.0001). Using an alpha level of .01, the fail-safe N for measures of QOL was 5,062 (z = 24.00), indicating that 5,062 studies with effect sizes of zero would be needed to nullify these results. Since this N is greater than 5k + 10 (where k is the number of trials in the analysis), the above effect size is considered statistically robust.

We further examined the funnel plot to assess publication bias (Figure 2). In the absence of publication bias, the studies should be distributed symmetrically with larger studies appearing toward the top of the graph and clustered around the mean effect size and smaller studies toward the bottom. Using the Trim and Fill method (Duval & Tweedie, 2000), we determined that 0 studies would need to fall to the left of the mean (i.e., have an effect size smaller than the mean) and 7 studies would need to fall to the right of the mean (i.e., have an effect size larger than the mean) to make the plot symmetrical, suggesting that our computed effect size is a conservative estimate. A random-effects model for the new imputed mean effect sizes revealed a Hedges’ g = 0.60 (95% CI: 0.51–0.70).

Figure 2.

Figure 2

Funnel plot of precision by Hedges’ g for quality of life measures in the pooled meta-analysis.

Pre-post controlled effect sizes for all QOL measures

Twenty-one of the trials included a control or comparison group. Of these, 13 used active comparison groups, including mindfulness-based stress reduction (1), stress management training (1), applied relaxation (1), minimal contact control (1), enhanced usual care (2), community treatment as usual (1), present-centered therapy (2), discussion group (2), and acceptance and commitment therapy (1). The remaining controlled studies used waitlist or delayed treatment control conditions (8).

The 21 controlled trials together yielded a QOL controlled effect size of Hedges’ g = 0.56 (95% CI: 0.32–0.80, z = 4.54, p < .0001). Using an alpha level of .01, the fail-safe N for measures of QOL was 338 (z = 10.65), indicating that 338 studies with an effect size of zero would be necessary to nullify the controlled effect size results. The Trim and Fill method determined that 0 studies would need to fall to the left of the mean and 6 studies would need to fall to the right of the mean effect size to make the plot symmetrical, suggesting that our computed effect size is a conservative estimate. A random-effects model for the new imputed mean controlled effect sizes revealed a Hedges’ g = 0.75 (95% CI: 0.53–1.00).

To address clinical heterogeneity due to the range of different control/comparison conditions used, we also performed a Cochran’s Q test to compare the effect sizes for trials that used waitlist control conditions (Hedges’ g = 0.78, 95% CI: 0.40–1.15, p < 0.0001) with trials that used active comparison conditions (Hedges’ g = 0.43, 95% CI: 0.10–0.75, p=0.01). This comparison revealed no significant difference (χ2Interaction = 1.88, p> .1).

Pre-post effect sizes for the Quality of Life Inventory

The Quality of Life Inventory (QOLI, Frisch, Cornell, Villanueva & Retzlaff, 2005) was the most common outcome measure for quality of life in our reviewed studies (n = 24). Its purpose is to measure, broadly, general quality of life in a range of domains, and is a widely used instrument to measure QOL in medical and psychiatric research. To ensure that there is no systematic difference between measures of QOL in our meta-analysis, a Cochran’s Q test was conducted to compare the effect size for studies using the QOLI versus those using other measures. The test of heterogeneity revealed no difference between these subgroups of studies (χ2Interaction = 0.76, p = 0.38). Those using QOLI yielded a pre-post effect size of Hedges’ g = 0.48 (95% CI: 0.44–0.63, z = 5.97, p<.0001), whereas the other studies yielded a Hedges’ g = 0.57 (95% CI: 0.45–0.69, z = 9.46, p<.0001).

Pre-post effect sizes for distinct quality of life domains

Among the studies we reviewed, the World Health Organization Quality of Life Assessment (WHOQOL-BREF; WHOQOL Group, 1998) was the only measure with which authors consistently provided data for different domains of QOL (n = 10). The Cochran’s Q test revealed that there is a significant difference among effect sizes in these domains (χ2Interaction= 16.65, p < .001). Specifically, improvement in QOL in the physical domain (Hedges’ g = 0.42, 95% CI: 0.30–0.53, z = 6.94, p < .0001) and psychological domain (Hedges’ g = 0.45, 95% CI: 0.36–0.54, z = 9.83, p < .0001) were greater than that in the environmental (Hedges’ g = 0.25, 95% CI: 0.16–0.33, z = 5.67, p < .0001) and social domains (Hedges’ g = 0.24, 95% CI: 0.15–0.32, z = 5.37, p < .0001).

Pre-post effect sizes for anxiety symptoms

The pre-post within-group random effect size on anxiety symptoms was Hedges’ g = 0.95 (95% CI: 0.80–1.10, z = 12.90, p<.0001). Using an alpha level of .01, the fail-safe N for measures of anxiety was 9,343 (z = 33.40). The Trim and Fill funnel plot determined that 0 studies would need to fall to the left of the mean effect size and 1 study would need to fall to the right of the mean effect size to make the plot symmetric. A random-effects model was assumed for new imputed mean effect sizes of change in anxiety symptoms: Hedges’ g = 0.96 (95% CI: 0.82–1.10).

The controlled effect size of anxiety symptoms was Hedges’ g = 0.69 (95% CI: 0.42–0.97, z = 4.94, p < .0001). Using an alpha level of .01, the fail-safe N for measures of anxiety was a robust 307 (z = 10.41). Using the Trim and Fill method of the funnel plot, 0 studies would need to fall to the right of the mean effect size and 1 study would need to fall to the left of the mean effect size to make the plot symmetric, associated with a new imputed controlled effect sizes of Hedges’ g = 0.70 (95% CI: 0.44–0.96).

Moderator analyses

There was a significant difference among pre-post QOL effect sizes for different disorders (χ2Interaction = 29.11, p <. 05). The diagnosis-specific effect sizes ranged from medium to large. The largest effect size was found for posttraumatic stress disorder (Hedges’ g = 1.12, 95% CI: 0.70–1.53, z = 5.28, p < .0001), followed by obsessive-compulsive disorder (Hedges’ g = 0.80, 95% CI: 0.59–1.00, z = 7.57, p < .0001), panic disorder (Hedges’ g = 0.46, 95% CI: 0.34–0.57, z = 7.62, p < .0001), social anxiety disorder (Hedges’ g = 0.46, 95% CI: 0.31–0.61, z = 5.95, p < .0001), and generalized anxiety disorder (Hedges’ g = 0.43, 95% CI: 0.28–0.58, z = 5.53, p < .0001). There was not enough power to compute effect sizes for trials with mixed anxiety diagnosis samples and for specific phobias.

Of the 59 clinical trials, 22 provided individual CBT delivered face-to-face, 14 provided group CBT delivered face-to-face, and 23 delivered CBT through the Internet. We compared effect sizes for QOL outcome across these three CBT delivery modalities, and found that there was a conventionally significant difference between effect sizes (χ2Interaction = 6.28, p < .05). Specifically, both face-to-face individual CBT (Hedges’ g = 0.61, 95% CI: 0.44–0.78) and face-to-face group CBT (Hedges’ g = 0.65, 95% CI: 0.42–0.88) yielded higher effect sizes than internet-delivered CBT (Hedges’ g = 0.41, 95% CI: 0.31–0.51). Since the trials that delivered Internet-based treatment comprised primarily of SAD (n=10) and PD (n=10) samples, there was a potential disorder type confound for the above results. However, for SAD and PD trials, effect sizes in face-to-face trials were significantly higher than in internet-delivered trials (χ2Interaction= 5.40, p < .05), suggesting that the difference between treatment modalities is not only an artifact of systematic properties of SAD and PD.

Study quality, as assessed by EPHPP scoring, did not moderate improvement in quality of life (B = 0.01, SE = 0.01, p > 0.3). However, QOL improvement in face-to-face CBT trials was moderated by the length of the intervention (B = 0.04, SE = 0.01, p < .001), with longer treatments being associated with larger effect sizes. Effect sizes were moderated by publication year (B = −0.04, SE = 0.01, p < 0.05), indicating that the effect size decreased linearly with time. Improvement in QOL was also significantly moderated by improvement in anxiety symptoms (B = 0.24, SE = 0.03, p < .0001), indicating that larger improvements in anxiety symptoms predicted greater improvement in quality of life. Furthermore, quality of life effect sizes were moderated by publication year (B = −0.04, SE = 0.01, p < 0.05), indicating that the effect size decreased linearly with time. These meta-regression slopes remained significant when removing an outlier (Ledley et al., 2009). The decrease in effect size across time was not accounted for by diminishing study quality, as study quality did not vary as a function of time. Similarly, average publication year did not differ among disorder categories. Later publication year was moderately associated with smaller sample size (Pearson’s r = −.25) and fewer CBT modules or sessions (Pearson’s r = −.17), which may explain the smaller effect sizes found in more recent studies.

Discussion

CBT is an effective treatment for anxiety disorders. As expected, and consistent with a previous meta-analysis (Hofmann & Smits, 2008), we observed that CBT had a large effect on reducing anxiety symptoms (Hedges’ g = 0.93). However, its effect on patients’ quality of life has not yet been examined. To examine the effect of CBT for anxiety disorders on quality of life, we conducted a meta-analytic review. We identified 44 studies that included 59 CBT trials, totaling 3,326 participants.

Our study found solid evidence for the beneficial effect of CBT on quality of life. The pre-post overall and controlled effect sizes of CBT on quality of life were moderately strong. The fail safe N analyses clearly exceeded the critical number (Rosenthal, 1993) for both the within-group and the controlled effect sizes for the QOL measures. The Trim and Fill method suggested that observed effect sizes were a conservative estimate and that the estimated controlled effect size was Hedges g = 0.75. It should be noted that we observed little difference in various analyses (within-group vs. controlled effect size). This, again, suggests that our effect size estimate was relatively robust.

Among those studies that allowed for a subanalysis by quality of life domains (i.e., physical, psychological, social, and environment domains), we observed that the improvements were greater for physical and psychological domains than for environmental and social domains. As with all systematic reviews, the relatively small sample size warrants caution in generalizing our findings.

The effect sizes for CBT delivered face-to-face, whether individually (22 trials; Hedges’ g = 0.61) or in groups (14 trials; Hedges’ g = 0.65) were significantly higher than for Internet-delivered treatments (23 trials; Hedges’ g = 0.41). This points to the limitation of Internet-based treatments and highlights the importance of face-to-face interventions. Furthermore, we observed that the effect size increased with treatment duration for face-to-face treatments, suggesting there might be a limit of how brief a treatment should be. Although anxiety symptoms might respond within only a few sessions (Otto et al., 2012), longer treatments seem to be necessary in order to improve the patient’s quality of life. This may suggest that internet-based CBT is less effective than face-to-face treatment, and that shorter therapy is less effective than longer treatment for improving quality of life. However, it remains to be seen whether these differences are also clinically meaningful. In order to answer this important question, future studies will need to provide a more in-depth analysis of the quality of life construct, which will necessitate new and improved ways to quantify this construct that go beyond self-report measures.

The moderation analyses further revealed that more recently published studies showed smaller effect sizes than older studies. The trend of diminishing effect sizes over time has been discussed elsewhere and described as the decline effect and the cosmic habituation effect (Schooler, 2011; Lehrer, 2010). It is unlikely that this effect is due to differences in study quality, because we specifically examined the quality of the trials and we did not observe that quality of the trial moderated the treatment effect. It is possible that earlier CBT protocols targeted a broader area of the patient’s life and thereby enhancing quality of life in general, whereas newer protocols are more symptom focused.

The largest within-group pre-post CBT effect size in quality of life improvement was found for post-traumatic stress disorder (Hedges’ g = 1.12), followed by obsessive-compulsive disorder (Hedges’ g = 0.80), panic disorder (Hedges’ g = 0.46), social anxiety disorder (Hedges’ g = 0.46), and generalized anxiety disorder (Hedges’ g = 0.43). It should be noted that a direct comparison between the effects of CBT for anxiety disorders on quality of life is problematic because of differences in comorbidities and differences in disease characteristics. For example, anxiety disorders differ in age of onset and it is possible that individuals who have experienced a very early onset of the disorder define the impact of the symptoms on quality of life differently than a person who initially had a high quality of life, which then radically changed after the onset of the disorder (Rapaport, Clary, Fayyad, & Endicott, 2005).

In general, the results are consistent with the notion that cognitive behavioral therapy improves quality of life. Although the effect size estimates were robust, the number of rigorous studies available to us was modest. Moreover, quality of life benefits associated with efficacious treatments may not be unique to cognitive behavioral therapy. We recommend that future studies conduct systematic quantitative reviews to determine whether other active interventions offer similar improvements.

Also, it can be argued that it is inappropriate to pool studies that show a high degree of heterogeneity. However, as displayed in Table 1, the degree of observed heterogeneity was within the range of what can be expected from meta-analytic reviews. For example, the majority of studies reported participants’ mean age to be between 35 and 40, most studies achieved approximately equal gender distribution and virtually all studies excluded serious psychiatric comorbidities (e.g., psychotic disorders). Moreover, because patient characteristics were not systematically reported in most included studies, we were unable to formally assess the role played by such variables as age and gender distribution. Nevertheless, in light of these unknown influences, results should be interpreted with caution, and replication across larger samples is recommended. Future work would be improved by incorporating data on patient characteristics.

Moreover, it is possible that other variables not considered here could have influenced the results. However, our moderator analyses were necessarily restricted to those variables that were reported in the included studies. Moreover, adding post-hoc moderator analyses introduces a significant methodological problem in meta-analyses. Specifically, it has been shown that such analyses are associated with a high Type I error rate (Brookes et al., 2004). Therefore, and in line with the recommendations by Brookes and colleagues (2004), we limited the moderator analyses to the minimally important set of variables that we identified based on our literature review.

Furthermore, we did not have the power to analyze multiple moderators simultaneously in a single model. Therefore, we are unable to speculate about how variables such as study length, publication year and treatment modality might interact in influencing treatment outcome. It is possible that these unknown interactions may explain some of the variation not accounted for by our single-variable moderator analyses. Thus, our results should be interpreted with caution, and future work would be best informed by a model incorporating multiple moderators simultaneously.

Another limitation of the study is related to the assessment of QOL. We examined a number of different self-report instruments that measure QOL. Although each of these measures is commonly used and psychometrically sound, the assessment of QOL rests on the patients’ self-report. Other indicators, such as observer-report or ecological momentary assessments, might provide additional and perhaps more valid measures. Moreover, we were unable to examine the specific QOL domains because studies typically did not report this level of detail. Future studies should examine the individual domains of quality of life, such as physical health, mental health, social activities, work, home, and family (Olatunji, Cisler, & Tolin, 2007). For example, it is possible that CBT for social anxiety disorder primarily enhances QOL by improving social relationships, whereas CBT for panic disorder with agoraphobia might affect mobility and certain areas of work by targeting agoraphobic avoidance.

Despite these limitations, our review suggests that CBT is not only an effective treatment for reducing the immediate symptoms of anxiety disorders, but it also has a beneficial effect on QOL, especially when CBT is conducted over a longer course of time and face-to-face.

Table 2.

Hedges’ g for pre-post QOL effect sizes for each trial.

Disorder Study Name Hedges’ g (95% CI) Z-Value p-value
Generalized Anxiety Disorder Craigie et al., 2008 .53 (.20–.86) 3.16 .002
Paxling et al., 2011 .43 (−.01–.86) 1.94 .052
Schneier et al., 2010 .39 (.07–.70) 2.44 .015
Stanley et al., 2009 .20 (−.14–.55) 1.14 .253
Stanley, Beck, et al., 2003 .58 (.10–1.06) 2.37 .018
Stanley, Hopko, et al., 2003 .94 (−.28–2.15) 1.51 .130
Wetherell et al., 2003 .67 (.05–1.29) 2.13 .033
Wetherell et al., 2009 .35 (−.35–1.04) .98 .328
Subtotal (Generalized Anxiety Disorder) .43(.28–.58) 5.53 .000
Injection Phobia Agdal et al., 2012 .16 (−.05–.36) 1.50 .133
Mixed Anxiety Dx Arch et al., 2012 .00 (−3.81–3.81) 0.00 1.000
Carlbring et al., 2011 .57 (−.04–1.11) 2.09 .037
Obsessive Compulsive Disorder (OCD) Cordioli et al., 2003 .76 (.17–1.35) 2.51 .012
de Sousa et al., 2006 .63 (.33–.94) 4.04 .000
Diefenbach et al., 2007 1.05 (.83–1.28) 9.19 .000
Simpson et al., 2008 .38 (−.01–.77) 1.91 .055
Simpson et al., 2010 – EX/RP .70 (.28–1.12) 3.29 .001
Simpson et al., 2010 – EX/RP+MI .74 (.32–1.17) 3.44 .001
OCD (Hoarding) Gilliam et al., 2011 1.10 (.81–1.38) 7.60 .000
Subtotal (OCD) .80 (.59–1.00) 7.57 .000
Panic Disorder Carlbring et al., 2005 – Internet CBT .36 (.06–.67) 2.35 .019
Carlbring et al., 2005 – Live CBT .50 (.18–.82) 3.06 .002
Carlbring et al., 2006 .74 (.22–1.26) 2.78 .005
Kiropoulos et al., 2008 – Internet CBT .32 (.08–.56) 2.61 .009
Kiropoulos et al., 2008 – Live CBT .36 (.13–.59) 3.09 .002
Klein et al, 2009 – ICBT + frequent contact .29 (−.08–.66) 1.51 .130
Klein et al., 2009 – ICBT + infrequent contact .25 (−.12–.61) 1.31 .191
Pier et al., 2008 – ICBT + GP .40 (.13–.66) 2.95 .003
Pier et al., 2008 – ICBT + psychologist .40 (.13–.68) 2.85 .004
Rufer et al., 2010 .12 (−.08–.33) 1.18 .239
Shandley et al., 2008 – ICBT + GP .29 (.08–.50) 2.73 .006
Shandley et al., 2008 – ICBT + psychologist .42 (.18–.66) 3.45 .001
Telch et al., 1995 .89 (.48–1.31) 4.24 .000
van Apeldoorn et al., 2012 .44 (.23–.65) 4.04 .000
Marchand et al., 2009 - CBGT 1.08 (.76–1.39) 6.63 .000
Marchand et al., 2009 - CBT .87 (.57–1.18) 5.59 .000
Marchand et al., 2009 – CBT Brief .75 (.45–1.05) 4.92 .000
Subtotal (Panic Disorder) .46 (.34–.57) 7.62 .000
Posttraumatic Stress Disorder Klein et al., 2010 .09 (−.23–.40) .548 .584
Pauvonic et al., 2001 – CBT 1.03 (.47–1.60) 3.61 .000
Pauvonic et al., 2001 – Exposure 1.66 (.95–2.38) 4.56 .000
Schneier et al, 2012 .43 (.07–.79) 2.35 .019
Schnurr et al., 2003 1.44 (1.20–1.68) 11.77 .000
Wagner et al., 2011 1.15 (.84–1.45) 7.36 .000
Subtotal (PTSD) .98 (.56–1.41) 4.51 .000
Social Anxiety Disorder Andersson et al., 2006 .47 (−.03–.97) 1.84 .066
Carlbring et al., 2007 .85 (.51–1.19) 4.92 .000
Eng et al., 2001 .45 (.14–.76) 2.86 .004
Eng et al., 2005 .36 (.11–.60) 2.87 .004
Furmark et al., 2009_trial1_Bibliotherapy .40 (−.04–.84) 1.79 .074
Furmark et al., 2009_trial1_ICBT .37 (−.07–.81) 1.65 .099
Furmark et al., 2009_trial2_bibliotherapy .18 (−.32–.69) .71 .476
Furmark et al., 2009_trial2_bibliotherapy+discussion .25 (−.27–.76) .94 .349
Furmark et al., 2009_trial2_ICBT .02 (−.49–.54) .09 .929
Hedman et al., 2011 – Internet CBT .30 (.10–.49) 3.03 .002
Hedman et al., 2011 – Live CBGT .33 (.13–.52) 3.27 .001
Koszycki et al., 2007 .07 (−.52–.66) .24 .810
Ledley et al., 2009 3.38 (2.30–4.45) 6.15 .000
Mortberg et al., 2011 – individual CT 1.00 (.67–1.32) 6.04 .000
Mortberg et al., 2011 – Intensive group CT .60 (.32–.87) 4.27 .000
Tilfors et al., 2008 - ICBT .48 (.13–.83) 2.65 .008
TIflors et al., 2008 – ICBT+EX .16 (−.17–.50) .96 .339
Watanabe et al., 2010 .34 (.14–.55) 3.30 .001
Subtotal (Social Anxiety Disorder) .46 (.31–.61) 5.53 .000
All disorders .56 (.44–.63) 11.28 .000

Acknowledgments

Dr. Hofmann is supported by NIMH grant R01AT007257.

References

Note. All references listed with a * refer to studies included in the meta-analysis.

  • *.Agdal ML, Raadal M, Ost L, Skaret E. Quality-of-life before and after cognitive behavioral therapy (CBT) in patients with intra-oral injection phobia. Acta Odontologica Scandinavica. 2012;70:463–470. doi: 10.3109/00016357.2011.635600. [DOI] [PubMed] [Google Scholar]
  • Andersson GA. Using the internet to provide cognitive behaviour therapy. Behaviour Research and Therapy. 2009;47:175–180. doi: 10.1016/j.brat.2009.01.010. [DOI] [PubMed] [Google Scholar]
  • *.Andersson GA, Carlbring P, Holmstrom A, Sparthan E, Furmark T, Nilsson-Ihrfelt E, Buhrman M, Ekselius L. Internet-based self-help with therapist feedback and in vivo group exposure for social phobia: a randomized controlled trial. Journal of Consulting and Clinical Psychology. 2006;74:677–686. doi: 10.1037/0022-006X.74.4.677. [DOI] [PubMed] [Google Scholar]
  • Angermeyer MC, Kilian R. Theoretical models of quality of life for mental disorders. In: Katschnig H, Freeman H, Sartorius N, editors. Quality of life in mental disorders. New York: Wiley; 1997. pp. 19–54. [Google Scholar]
  • *.Arch JJ, Eifert EH, Davies C, Vilardaga JCP. Randomized clinical trial of cognitive behavioral therapy (CBT) versus acceptance and commitment therapy (ACT) for mixed anxiety disorders. Journal of Consulting and Clinical Psychology. 2012;80:750–765. doi: 10.1037/a0028310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • Borenstein M, Hedges L, Higgins J, Rothstein H. Comprehensive meta-analysis, version 2. Englewood, NJ: Biostat, Inc; 2005. [Google Scholar]
  • Candilis PJ, McLean RY, Otto MW, Manfro GG, Worthington JJ, III, Penava SJ, Marzol PC, Pollack MH. Quality of life in patients with panic disorder. Journal of Nervous and Mental Disease. 1999;187:429–434. doi: 10.1097/00005053-199907000-00006. [DOI] [PubMed] [Google Scholar]
  • *.Carlbring P, Nilsson-Ihrfelt E, Waara J, Kollenstam C, Buhrman M, Kaldo V, Soderberg M, Ekselius L, Andersson G. Treatment of panic disorder: live therapy vs. self-help via the Internet. Behaviour Research and Therapy. 2005;43:1321–1333. doi: 10.1016/j.brat.2004.10.002. [DOI] [PubMed] [Google Scholar]
  • *.Carlbring P, Bohman S, Brunt S, Buhrman M, Westling BE, Ekselius L, Andersson G. Remote treatment of panic disorder: a randomized trial of internet-based cognitive behavior therapy supplemented with telephone calls. American Journal of Psychiatry. 2006;163:2119–2125. doi: 10.1176/appi.ajp.163.12.2119. [DOI] [PubMed] [Google Scholar]
  • *.Carlbring P, Furmark T, Seczko J, Ekselius L, Andersson G. An open study of Internet-based bibliotherapy with minimal therapist contact via email for social phobia. Clinical Psychologist. 2007;10:30–38. doi: 10.1080/13284200500378662. [DOI] [Google Scholar]
  • *.Carlbring P, Maurin L, Torngren C, Linna E, Eriksson T, Sparthan E, Straat M, Marquez von Hage C, Bergman-Nordgren L, Andersson G. Individually-tailored, Internet-based treatment for anxiety disorders: A randomized controlled trial. Behaviour Research and Therapy. 2001;49:18–24. doi: 10.1016/j.brat.2010.10.002. [DOI] [PubMed] [Google Scholar]
  • Cohen J. Statistical power analysis for the behavioral sciences. 2. Hillsdale, NJ: Erlbaum; 1998. [Google Scholar]
  • *.Cordioli AV, Heldt E, Bochi DB, Margis R, de Sousa MB, Tonello JF, Manfro GG, Kapczinski F. Cognitive-behavioral group therapy in obsessive-compulsive disorder: a randomized clinical trial. Psychotherapy and Psychosomatics. 2003;72:211–216. doi: 10.1159/000070785. [DOI] [PubMed] [Google Scholar]
  • *.Craigie MA, Rees CS, Marsh A. Mindfulness-based cognitive therapy for generalized anxiety disorder: a preliminary evaluation. Behavioural and Cognitive Psychotherapy. 2008;36:553–568. doi: 10.1017/S135246580800458X. [DOI] [Google Scholar]
  • Cramer V, Torgersen S, Kringlen E. Quality of life and anxiety disorders: A population study. Journal of Nervous and Mental Disease. 2005;193:196–202. doi: 10.1097/01.nmd.0000154836.22687.13. [DOI] [PubMed] [Google Scholar]
  • *.de Sousa MB, Isolan LR, Oliveira RR, Manfro GG, Cordioli AV. A randomized clinical trial of cognitive-behavioral group therapy and sertraline in the treatment of obsessive-compulsive disorder. Journal of Clinical Psychiatry. 2006;67:1133–1139. doi: 10.4088/JCP.v67n0717. [DOI] [PubMed] [Google Scholar]
  • *.Diefenbach GJ, Abramowitz JS, Norberg MM, Tolin DF. Changes in quality of life following cognitive-behavioral therapy for obsessive-compulsive disorder. Behaviour Research and Therapy. 2007;45:3060–3068. doi: 10.1016/j.brat.2007.04.014. [DOI] [PubMed] [Google Scholar]
  • DuPont RL, Rice DP, Miller LS, Shiraki SS, Rowland CR, Harwood HJ. Economic costs of anxiety disorders. Anxiety. 1996;2:167–172. doi: 10.1002/(SICI)1522-7154(1996)2:4&#x0003c;167::AID-ANXI2&#x0003e;3.0.CO;2-L. [DOI] [PubMed] [Google Scholar]
  • Duval S, Tweedie R. Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Bioometrics. 2000;56:455–463. doi: 10.1111/j.0006-341X.2000.00455.x. [DOI] [PubMed] [Google Scholar]
  • Duquette RL, Dupuis G, Perrault J. A new approach for quality of life assessment in cardiac patients: Rationale and validation of the Quality of Life Systemic Inventory. Canadian Journal of Cardiology. 1994;10:106–112. Retrieved from http://europepmc.org/ [PubMed] [Google Scholar]
  • Endicott J, Nee J, Harrison W, Blumenthal R. Quality of life enjoyment and satisfaction questionnaire. Psychopharmacology Bulletin. 1993;29:321–326. Retrieved from http://psycnet.apa.org/index.cfm?fa=browsePA.home. [PubMed] [Google Scholar]
  • *.Eng W, Coles ME, Heimberg RG, Safren SA. Quality of life following cognitive behavioral treatment for social anxiety disorder: preliminary findings. Depression and Anxiety. 2001;13:192–193. doi: 10.1002/da.1037. [DOI] [PubMed] [Google Scholar]
  • *.Eng W, Coles ME, Heimberg RG, Safren SA. Domains of life satisfaction in social anxiety disorder: relation to symptoms and response to cognitive-behavioral therapy. Journal of Anxiety Disorders. 2005;19:143–156. doi: 10.1016/j.janxdis.2004.01.007. [DOI] [PubMed] [Google Scholar]
  • EuroQol-Group. EuroQol: a new facility for the measurement of health-related quality of life. Health Policy. 1990;16:99–208. doi: 10.1016/0168-8510(90)90421-9. Retrieved from http://www.euroqol.org/ [DOI] [PubMed] [Google Scholar]
  • Frisch MB. Quality of life therapy and assessment in health care. Clinical Psychology: Science and Practice. 1998;5:19–40. doi: 10.1111/j.1468-2850.1998.tb00132.x. [DOI] [Google Scholar]
  • Frisch MB, Clark MP, Rouse SV, Rudd MD, Paweleck J, Greenstone A. Predictive and treatment validity of life satisfaction and the Quality of Life Inventory. Assessment. 2005;12:66–78. doi: 10.1177/1073191104268006. [DOI] [PubMed] [Google Scholar]
  • *.Furmark T, Carlbring P, Heman E, Sonnenstein A, Clevberger P, Bohman B, Eriksson A, Hallen A, Frykman M, Holmstrom A, Sparthan E, Tilfros M, Nilsson-Ihrfelt E, Spark M, Eriksson A, Ekselius L, Andersson G. Guided and unguided self-help for social anxiety disorder: randomised controlled trial. British Journal of Psychiatry. 2009;195:440–447. doi: 10.1192/bjp.bp.108.060996. [DOI] [PubMed] [Google Scholar]
  • *.Gilliam CM, Norberg MM, Villavicencio A, Morrison S, Hannan SE, Tolin DF. Group cognitive-behavioral therapy for hoarding disorder: An open trial. Behaviour Research and Therapy. 2011;49:802–807. doi: 10.1016/j.brat.2011.08.008. [DOI] [PubMed] [Google Scholar]
  • Gladis MM, Gosch EA, Dishuk NM, Crits-Christoph P. Quality of life: Expanding the scope of clinical significance. Journal of Consulting and Clinical Psychology. 1999;67:320–331. doi: 10.1037/0022-006X.67.3.320. [DOI] [PubMed] [Google Scholar]
  • Hedges LV, Olkin I. Statistical methods for meta-analysis. New York: Academic Press; 1985. [Google Scholar]
  • Hedges LV, Vevea JL. Fixed- and random-effects models in meta-analysis. Psychological Methods. 1998;3:486–504. doi: 10.1037//1082-989X.3.4.486. [DOI] [Google Scholar]
  • *.Hedman E, Andersson E, Ljotsson B, Andersson G, Ruck C, Lindefors N. Cost-effectiveness of Internet-based cognitive behavior therapy vs. cognitive behavioral group therapy for social anxiety disorder: Results from a randomized controlled trial. Behaviour Research and Therapy. 2011;49:729–736. doi: 10.1016/j.brat.2011.07.009. [DOI] [PubMed] [Google Scholar]
  • Hofmann SG, Asmundson GJ, Beck AT. The science of cognitive therapy. Behavior Therapy. 2013;44:199–212. doi: 10.1016/j.beth.2009.01.007. [DOI] [PubMed] [Google Scholar]
  • Hofmann SG, Smits JAJ. Cognitive-Behavioral Therapy for adult anxiety disorders: A meta-analysis of randomized placebo-controlled trials. Journal of Clinical Psychiatry. 2008;69:621–632. doi: 10.4088/JCP.v69n0415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • Howard KI, Kopta SM, Krause MS, Orlinsky DE. The dose-effect relationship in psychotherapy. American Psychologist. 1986;41:158–164. doi: 10.1037/0003-066X.41.2.159. [DOI] [PubMed] [Google Scholar]
  • Jenkinson C, Layte R, Jenkinson D, Lawrence K, Petersen S, Paice C, Stradling J. A shorter form health survey: can the SF-12 replicate results from the SF-36 in longitudinal studies? Journal of Public Health Medicine. 1997;19:179–186. doi: 10.1093/oxfordjournals.pubmed.a024606. [DOI] [PubMed] [Google Scholar]
  • Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey replication. Archives of General Psychiatry. 2005;62:593–768. doi: 10.1001/archpsyc.62.6.593. [DOI] [PubMed] [Google Scholar]
  • Kiropoulos LA, Klein B, Austin DW, Gilson K, Pier C, Mitchell J, Ciechomski L. Is internet-based CBT for panic disorder and agoraphobia as effective as face-to-face CBT? Journal of Anxiety Disorders. 2008;22:1273–1284. doi: 10.1016/j.janxdis.2008.01.008. [DOI] [PubMed] [Google Scholar]
  • *.Klein B, Mitchell J, Gilson K, Shandley K, Austin D, Kiropoulos L, Abbott J, Cannard G. A Therapist-Assisted Internet-Based CBT Intervention for Posttraumatic Stress Disorder: Preliminary Results. Cognitive Behaviour Therapy. 2009;38:121–131. doi: 10.1080/16506070902803483. [DOI] [PubMed] [Google Scholar]
  • *.Klein B, Mitchell J, Abbott J, Shandley K, Austin D, Gilson K, Kiropoulos L, Cannard G, Redman T. A therapist-assisted cognitive behavior therapy internet intervention for posttraumatic stress disorder: Pre-, post- and 3-month follow-up results from an open trial. Journal of Anxiety Disorders. 2010;24:635–644. doi: 10.1016/j.janxdis.2010.04.005. [DOI] [PubMed] [Google Scholar]
  • Koran LM, Thienemann ML, Davenport R. Quality of life for patients with obsessive-compulsive disorder. American Journal of Psychiatry. 1996;153:783–788. doi: 10.1016/j.comppsych.2009.10.001. [DOI] [PubMed] [Google Scholar]
  • *.Koszycki D, Benger M, Shlik J, Bradwejn J. Randomized trial of a meditation-based stress reduction program and cognitive behavior therapy in generalized social anxiety disorder. Behaviour Research and Therapy. 2007;45:2518–2526. doi: 10.1016/j.brat.2007.04.011. [DOI] [PubMed] [Google Scholar]
  • *.Ledley DR, Heimberg RG, Hope DA, Hayes SA, Zaider TI, van Dyke M, Turk CL, Kraus C, Fresco DM. Efficacy of a manualized and workbook-driven individual treatment for social anxiety disorder. behavior Therapy. 2009;40:414–424. doi: 10.1016/j.beth.2008.12.001. [DOI] [PubMed] [Google Scholar]
  • Lehrer J. New Yorker. Dec 13, 2010. The decline effect and the scientific method: The truth wears off; p. 52. [Google Scholar]
  • *.Marchand A, Roberge P, Primiano S, Germain V. A randomized, controlled clinical trial of standard, group and brief cognitive-behavioral therapy for panic disorder with agoraphobia: A two-year follow-up. Journal of Anxiety Disorders. 2009;23:1139–1147. doi: 10.1016/j.janxdis.2009.07.019. [DOI] [PubMed] [Google Scholar]
  • Marks GB, Dunn SM, Woolcock AJ. A scale for the measurement of quality of life in adults with asthma. Journal of Clinical Epidemiology. 1992;45(5):461–472. doi: 10.1016/0895-4356(92)90095-5. [DOI] [PubMed] [Google Scholar]
  • Marks I, Swinson R, Basoglu M, Kuch K, Noshirvani H, O’Sulliva G, Lelliot P, Kirby M, McNamee G, Sengun S, Wickwire K. Alprazolam and exposure alone or combined in panic disorder with agoraphobia: a controlled study in London and Toronto. British Journal of Psychiatry. 1993;162:776–787. doi: 10.1192/bjp.162.6.776. [DOI] [PubMed] [Google Scholar]
  • Mendlowicz MV, Stein MB. Quality of life in individuals with anxiety disorders. American Journal of Psychiatry. 2000;157:669–682. doi: 10.1176/appi.ajp.157.5.669. [DOI] [PubMed] [Google Scholar]
  • Moher D, Liberati A, Tetzlaff J, Altman DG The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med. 2009;6(7):e1000097. doi: 10.1371/journal.pmed.1000097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • *.Mortberg E, Clark DM, Bejerot S. Intensive group cognitive therapy and individual cognitive therapy for social phobia: Sustained improvement at 5-year follow-up. Journal of Anxiety Disorders. 2011;25:994–1000. doi: 10.1016/j.janxdis.2011.06.007. [DOI] [PubMed] [Google Scholar]
  • Moses LE, Mosteller F, Beuhler JH. Comparing results of large clinical trials to those of meta-analyses. Statistics in Medicine. 2002;21:793–800. doi: 10.1002/sim.1098. [DOI] [PubMed] [Google Scholar]
  • Olatunji BO, Cisler JM, Tolin DF. Quality of life in the anxiety disorders: A meta-analytic review. Clinical Psychology Review. 2007;27:572–581. doi: 10.1016/j.cpr.2007.01.015. [DOI] [PubMed] [Google Scholar]
  • Otto MW, Tolin DF, Nations KR, Utschig AC, Rothbaum BO, Hofmann SG, Smits JAJ. Five sessions and counting: Considering ultra-brief treatment for panic disorder. Depression and Anxiety. 2012;29:467–470. doi: 10.1002/da.21910. [DOI] [PubMed] [Google Scholar]
  • *.Pauvonic N, Ost L. Cognitive-behavior therapy versus exposure therapy in the treatment of PTSD in refugees. Behaviour Research and Therapy. 2001;39:1183–1197. doi: 10.1016/S0005-7967(00)00093-0. [DOI] [PubMed] [Google Scholar]
  • *.Paxling B, Almlov J, Dahlin M, Carlbring P, Breitholtz E, Eriksson T, Andersson G. Guided internet-delivered cognitive behavior therapy for generalized anxiety disorder: a randomized controlled trial. Cognitive Behaviour Therapy. 2011;40(3):159–173. doi: 10.1080/16506073.2011.576699. [DOI] [PubMed] [Google Scholar]
  • *.Pier C, Austin DW, Klein B, Mitchell J, Schattner P, Ciechomski L, Gilson KJ, Pierce D, Shandley K, Wade V. A controlled trial of internet-based cognitive-behavioural therapy for panic disorder with face-to-face support from a general practitioner or email support from a psychologist. Mental Health in Family Medicine. 2008;5:29–39. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2777548/ [PMC free article] [PubMed] [Google Scholar]
  • RAND Health Sciences Program. RAND 36-item health survey 1.0. Santa Monica, CA: 1992. [Google Scholar]
  • Rapaport MH, Clary C, Fayyad R, Endicott J. Quality-of-life impairment in depressive and anxiety disorders. American Journal of Psychiatry. 2005;162:1171–1178. doi: 10.1176/appi.ajp.162.6.1171. [DOI] [PubMed] [Google Scholar]
  • Rosenthal R. Meta-analytic Procedures for Social Research, revised edition. Thousand Oaks, CA: Publications Inc; 1991. [Google Scholar]
  • Rosenthal R. Meta-analytic procedures for social research. Newbury Park, CA: Sage Publications; 1993. [Google Scholar]
  • Rosenthal R, Rubin DB. Comment: Assumptions and procedures in the file drawer problem. Statistical Science. 1988;3:120–125. doi: 10.1214/ss/1177013014. [DOI] [Google Scholar]
  • Rubin HC, Rapaport MH, Levine B, Gladsjo JK, Rabin A, Auerbach M, Judd LL, Kaplan R. Quality of well being in panic disorder: the assessment of psychiatric and general disability. Journal of Affective Disorder. 2000;57:217–221. doi: 10.1016/S0165-0327(99)00030-0. [DOI] [PubMed] [Google Scholar]
  • *.Rufer M, Albrecht R, Schmidt O, Zaum J, Schnyder U, Hand I, Mueller-Pfeiffer C. Changes in quality of life following cognitive-behavioral group therapy for panic disorder. European Psychiatry. 2010;25:8–14. doi: 10.1016/j.eurpsy.2009.05.003. [DOI] [PubMed] [Google Scholar]
  • Safren SA, Heimberg RG, Brown EJ, Holle C. Quality of life in social phobia. Depression and Anxiety. 1996–1997;4:126–133. doi: 10.1002/(SICI)1520-6394(1996)4:3&#x0003c;126::AID-DA5&#x0003e;3.3.CO;2-E. [DOI] [PubMed] [Google Scholar]
  • Schmidt S, Muhlan H, Power M. The EUROHIS-QOL 8-item index: psychometric results of a cross-cultural field study. European Journal of Public Health. 2006;16:420–428. doi: 10.1093/eurpub/cki155. [DOI] [PubMed] [Google Scholar]
  • Schooler J. Unpublished results hide the decline effect. Nature. 2011;470:437. doi: 10.1038/470437a. [DOI] [PubMed] [Google Scholar]
  • *.Schneier FR, Belzer KD, Kishon R, Amsel L, Simpson HB. Escitalopram for Persistent Symptoms of Generalized Anxiety Disorder After CBT. The Journal of Nervous and Mental Disease. 2010;198:458–461. doi: 10.1097/NMD.0b013e3181da4d77. [DOI] [PubMed] [Google Scholar]
  • *.Schneier FR, Neria Y, Pavlicova M, Hembree E, Suh EJ, Amsel L, Marshall RD. Combined prolonged exposure therapy and paroxetine for PTSD related to the World Trade Center attack: A randomized controlled trial. American Journal of Psychiatry. 2012;169:80–88. doi: 10.1176/appi.ajp.2011.11020321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • *.Schnurr PP, Friedman MJ, Foy DW, Shea MT, Hsieh FY, Lavori PW, Glynn SM, Wattenberg M, Bernardy NC. Randomized trial of trauma-focused group therapy for posttraumatic stress disorder: results from a department of veterans affairs cooperative study. Archive of General Psychiatry. 2003;60:481–489. doi: 10.1001/archpsyc.60.5.481. [DOI] [PubMed] [Google Scholar]
  • *.Shandley K, Austin DW, Klein B, Pier C, Schattner P, Pierce D, Wade V. Therapist-assisted, internet-based treatment for panic disorder: can general practitioners achieve comparable patient outcomes to psychologists? Journal of Medical Internet Research. 2008;10:e14. doi: 10.2196/jmir.1033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, Hergueta T, Baker R, Dunbar GC. The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry. 1998;59:22–57. Retrieved from http://www.psychiatrist.com/default2.asp. [PubMed] [Google Scholar]
  • Sherbourne CD, Wells KB, Judd LL. Functioning and well-being of patients with panic disorder. American Journal of Psychiatry. 1996;153:213–218. doi: 10.1176/ajp.153.2.213. Retrieved from http://psycnet.apa.org/index.cfm?fa=browsePA.home. [DOI] [PubMed] [Google Scholar]
  • *.Simpson HB, Foa EB, Liebowitz MR, Ledley DR, Huppert JD, Cahill S, Vermes D, Schmidt AB, Hembree E. A randomized, controlled trial of cognitive-behavioral therapy for augmenting pharmacotherapy in obsessive-compulsive disorder. American Journal of Psychiatry. 2008;165:621–630. doi: 10.1176/appi.ajp.2007.07091440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • *.Simpson HB, Zuckoff AM, Maher MJ, Page JR, Franklin ME, Foa EB, Schmidt AB, Wang Y. Challenges using motivational interviewing as an adjunct to exposure therapy for obsessive-compulsive disorder. Behaviour Research and Therapy. 2010;48:941–948. doi: 10.1016/j.brat.2010.05.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • Spitzer RL, Kroenke K, Linzer M, Hahn SR, Williams JB, deGruy FV, III, Brody D, Davies M. Health-related quality of life in primary care patients with mental disorders: results from the PRIME-MD 1000 Study. JAMA. 1995;274:1511–1517. doi: 10.1001/jama.274.19.1511. [DOI] [PubMed] [Google Scholar]
  • *.Stanley MA, Wilson NL, Novy DM, Rhoades HM, Wagener PD, Greisinger AJ, Cully JA, Kunik ME. Cognitive behavior therapy for generalized anxiety disorder among older adults in primary care: a randomized clinical trial. Journal of American Medical Association. 2009;301:1460–1467. doi: 10.1001/jama.2009.458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • *.Stanley MA, Beck JG, Novy DM, Averill PM, Swann AC, Diefenbach GJ, Hopko DR. Cognitive-behavioral treatment of late-life generalized anxiety disorder. Journal of Consulting and Clinical Psychology. 2003;71:309–319. doi: 10.1037/0022-006X.71.2.309. [DOI] [PubMed] [Google Scholar]
  • *.Stanley MA, Hopko DR, Diefenbach GJ, Bourland SL, Rodriguez H, Wagener P. Cognitive-behavior therapy for late-life generalized anxiety disorder: Preliminary findings. The American Journal of Geriatric Psychiatry. 2003;11:92–96. doi: 10.1176/appi.ajgp.11.1.92. [DOI] [PubMed] [Google Scholar]
  • *.Telch MJ, Schmidt NB, Jaimez TL, Jacquin KM, Harrington PJ. Impact of cognitive-behavioral treatment on quality of life in panic disorder patients. Journal of Consulting and Clinical Psychology. 1995;63:823–830. doi: 10.1037/0022-006X.63.5.823. [DOI] [PubMed] [Google Scholar]
  • Thomas H, Ciliska D, Dobbins M, Micucci S. A process for systematically reviewing the literature: Providing the research evidence for public health nursing. Worldviews on Evidence-Based Nursing. 2004;1:176–184. doi: 10.1111/j.1524-475X.2004.04006.x. [DOI] [PubMed] [Google Scholar]
  • *.Tilfors M, Carlbring P, Furmark T, Lewenhaupt S, Spak M, Eriksson A, Westling BE, Andersson G. Treating university students with social phobia and public speaking fears: Internet delivered self-help with or without live group exposure sessions. Depression and Anxiety. 2008;25:708–717. doi: 10.1002/da.20416. [DOI] [PubMed] [Google Scholar]
  • *.Van Apeldoorn FJ, Timmerman ME, Mersch PPA, van Hout WJPJ, Visser S, van Dyck R, den Boer JA. A randomized trial of cognitive-behavioral therapy or selective serotonin reuptake inhibitor or both combine for panic disorder with or without agoraphobia: treatment results through 1-year follow-up. Journal of Clinical Psychiatry. 2010;71:574–586. doi: 10.4088/JCP.08m04681blu. [DOI] [PubMed] [Google Scholar]
  • *.Wagner B, Schulz W, Knaevelsrud C. Efficacy of an Internet-based intervention for posttraumatic stress disorder in Iraq: A pilot study. Psychiatry Research. 2012;195:85–88. doi: 10.1016/j.psychres.2011.07.026. [DOI] [PubMed] [Google Scholar]
  • Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36), I. Conceptual framework and item selection. Medical Care. 1992;30:473–83. doi: 10.1097/00005650-199206000-00002. [DOI] [PubMed] [Google Scholar]
  • *.Watanabe N, Furukawa TA, Chen J, Kinoshita Y, Nakano Y, Ogawa S, Funayama T, Ietsugu T, Noda Y. Change in quality of life and their predictors in the long-term follow-up after group cognitive behavioral therapy for social anxiety disorder: a prospective cohort study. BMC Psychiatry. 2010;10:81–91. doi: 10.1186/1471-244X-10-81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • *.Wetherell JL, Gatz M, Craske MG. Treatment of generalized anxiety disorder in older adults. Journal of Clinical Psychiatry. 2003;71:31–40. doi: 10.1037/0022-006X.71.1.31. [DOI] [PubMed] [Google Scholar]
  • *.Wetherell JL, Ayers CR, Sorrell JT, Thorp SR, Nuevo R, Belding W, Gray E, Stanley MA, Arean PA, Donohue M, Unutzer J, Ramsdell J, Xu R, Patterson TL. Modular psychotherapy for anxiety in older primary care patients. The American Journal of Geriatric Psychiatry. 2009;17:483–492. doi: 10.1097/JGP.0b013e3181a31fb5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • WHOQOL Group. Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychological Medicine. 1998;28:551–558. doi: 10.1017/S0033291798006667. [DOI] [PubMed] [Google Scholar]
  • Wittchen HU, Beloch E. The impact of social phobia on quality of life. International. Journal of Clinical Psychopharmacology. 1996;11(suppl 3):15–23. doi: 10.1097/00004850-199606003-00004. [DOI] [PubMed] [Google Scholar]
  • Wood V, Wylie ML, Sheafor B. An analysis of a short self-report measure of life satisfaction: Correlation with rater judgments. Journal of Gerontology. 1969;24:465–469. doi: 10.1093/geronj/24.4.465. [DOI] [PubMed] [Google Scholar]

RESOURCES