Sage Journals: Discover world-class research

Abstract

French

Aim

To assess the incremental cost-effectiveness ratio (ICER) of group transdiagnostic cognitive-behavioural therapy (tCBT) added to treatment as usual (TAU) for anxiety disorders compared to TAU only from the healthcare system perspective over a 1-year time horizon.

Methods

Data from a pragmatic multisite randomized controlled trial where adults (18–65 years) with an anxiety disorder were randomized to tCBT + TAU (n = 117) or TAU (n = 114). Group tCBT is a 12-week (2h weekly sessions) community-based intervention. Health service utilization and related costs were captured from medico-administrative data and included those for the intervention, ambulatory visits, hospitalizations and medications. Effectiveness was based on quality-adjusted life years (QALYs). The study included measures at baseline, 4, 8, and 12 months. Intention-to-treat and complete case analyses were carried out. Missing data were imputed using multiple imputation analyses. Seemingly unrelated regression analyses were used to assess the effect of the intervention on total costs and QALYs while also adjusting for baseline confounders. The probability of cost-effectiveness of the intervention was assessed according to different willingness-to-pay (WTP) thresholds using the net benefit regression method.

Results

The ICER of tCBT + TAU as compared to TAU in the intention-to-treat analysis was $6,581/QALY. Complete case analyses showed a similar ICER of $6,642/QALY. The probability at a WTP threshold of $20,000 and $40,000 that tCBT + TAU as compared to TAU is cost-effective is 93.0% and 99.9%.

Conclusion

tCBT added to TAU appears to be cost-effective from the healthcare system perspective for treating adult patients with anxiety disorders. Larger trials including young and older adults as well as a range of anxiety disorders are needed to further investigate the cost-effectiveness of tCBT in different patient populations.

Keywords

anxiety disorders group therapy transdiagnostic cognitive behaviour therapy randomized controlled trial cost-effectiveness

Introduction

Anxiety disorders are the most prevalent common mental disorders with past-year prevalence in adults between 8.4% and 21.3%.^1,2 Anxiety disorders are also associated with increased health care costs.^3,4 Although the long-term effects of the pandemic on the chronicity of anxiety disorders remain to be studied,⁵ the increase in anxiety symptoms and severity⁶ and increased mental health service needs⁷ will most likely add to the strain of limited human resources in the healthcare system.

In Canada, the majority of patients with anxiety disorders consult in primary care.^8,9 Clinical recommendations for anxiety disorders include evidence-informed pharmacological options and psychotherapy such as cognitive-behavioural therapy (CBT) for generalized anxiety disorder, panic disorder, social phobia, and agoraphobia.¹⁰ Although psychotherapy is preferred by patients,¹¹ pharmacological treatments are still the most prescribed option^12,13 due to their wider coverage under public and private drug insurance plans.¹⁴ Barriers to accessing psychotherapy in Canada include insufficient funding, long waiting lists in the public sector and high out-of-pocket costs in consulting in the private sector.^15,16 Sustainable solutions for community-based primary mental health interventions include increasing access to psychotherapy in primary care,^15,16 which has been shown to be cost-effective.^17,18

Among individuals reporting a lifetime anxiety disorder, 56% and 20% will also report at least two and three comorbid anxiety disorders at one point in their lifetime.¹⁹ Transdiagnostic CBT (tCBT) therapy is a promising intervention that can be offered to individuals with a range of anxiety disorders in group therapy. The option of creating groups consisting of individuals with different anxiety disorders can improve timely access to psychological treatment of anxiety disorders by reducing the human resources required for individual anxiety-specific treatment protocols.^20-24 This evidence-based intervention has been developed based on common processes observed across a range of anxiety and depressive disorders and aims to treat multiple anxiety disorders simultaneously by targeting negative affectivity/neuroticism.^20,24 The effectiveness of group tCBT for the treatment of anxiety disorders has been reported with findings showing improvements in anxiety symptoms, social functioning and quality of life as well as reductions in comorbid anxiety and depressive disorders.^25-28 Although the benefits of this intervention have been studied, rare are the studies that have looked at its cost-effectiveness.²⁸

The aim of this study was to assess the cost-effectiveness of group tCBT with treatment as usual (tCBT + TAU) for anxiety disorders compared to TAU from the healthcare system perspective over a 1-year time horizon in community-based primary care settings using data from a pragmatic multisite randomized controlled trial (RCT). Findings from this trial showed improved self-reported and clinician-rated severity of anxiety disorders from baseline to 12 months postrandomization in individuals randomized to group tCBT + TAU as compared to TAU.²⁶ A cost-effectiveness study over an 8-month time horizon using self-reported data from the same trial showed a high probability that tCBT was cost-effective from a societal perspective.²⁹ An earlier study had shown a significant reduction in health service utilization and costs in individuals with anxiety disorders up to 1-year following CBT treatment.³⁰ By linking health survey and medico-administrative data on health service use, and therefore limiting information bias associated with recall and loss to follow-up and the control of a number of confounders, this study included an additional 4 months of follow-up and assesses over a 1-year period the costs and effects of receiving group tCBT + TAU compared to TAU in a community-based sample of adults with anxiety disorders. We expect that the additional follow-up will allow for more precise estimates of the long-term cost-effectiveness of the intervention and inform decision makers as to its cost-effectiveness in the context of a Canadian public healthcare system context.

Methods

The study design is a pragmatic, multicentre, 2-arm parallel, single-blind, superiority RCT conducted in 3 regions in Quebec, Canada (trial registration: https://clinicaltrials.gov/ct2/show/NCT02811458) and has been described elsewhere.³¹ The RCT included 231 participants of which 117 were randomized to tCBT + TAU and 114 to TAU arm. Participants were recruited through advertisements in Google AdWords, Facebook, newspapers, and bulletin ads in the community. Interested adults were invited to complete an online survey on eligibility criteria. Eligible participants were then contacted for a short telephone interview. The final step of verifying participants’ eligibility was assessed with the Anxiety and Related Disorders Interview Schedule for DSM-5 (ADIS-5)³² through a face-to-face interview (baseline, T₀). Participants provided written informed consent.

The study was conducted among French-speaking adults aged 18 to 65 years with an anxiety disorder meeting DSM-5 criteria (social anxiety, panic disorder, agoraphobia, and generalized anxiety disorder) and a score of ≥4 on the clinical severity rating, indicating clinical threshold (disabling disorder) on the ADIS-5 (ranging from 0, no symptoms, to 8, severe symptoms).³² Participants with the following conditions were excluded from the study: cognitive impairment, bipolar disorders, psychosis, substance-related and addictive disorders and active suicidal behaviour. Participants having consulted in the last 12 months with a psychiatrist were also excluded. These patients may reflect more complex clinical profiles that necessitate more specialized mental health care and active follow-up. Although it is common in clinical practice for psychiatrists to refer patients to psychotherapy in various settings (e.g., community, private practice, and hospital), these individuals had already received specialized care reflecting different health service use trajectories. Considering a stepped-care approach, the current study focused on individuals consulting in community and primary care settings with and without physician referrals. At each study site, patients meeting eligibility criteria were randomized (1:1) by blocks of 4 into the intervention (tCBT + TAU) or the control group (TAU).

Data collection and assessments for the tCBT + TAU arm included face-to-face assessments at baseline (T₀) and at 4 months postrandomization, corresponding to posttreatment (T₁), telephone interview at 8 months postrandomization (posttreatment + 4 months, T₂) and face-to-face assessment at 12 months postrandomization (posttreatment + 8 months, T₃). For individuals in the TAU arm, the study included prerandomization face-to-face assessments at baseline (T₀), telephone interviews at 4 months (T₁) and 8 months (T₂), and face-to-face assessment at 12 months (T₃). Information collected from the RCT was linked to provincial medico-administrative data on health and drug service utilization for the 12 months preceding and following baseline (T₀). Of the 231 individuals randomized to either group, 218 participants gave consent to access their provincial medico-administrative data on health service utilization and the remaining 13 participants were considered to have missing information on medico-administrative data.

Measures

Intervention

The 12-week community-based transdiagnostic CBT (tCBT)^31,33 in primary care settings consisted of a 2 h session on a weekly basis and included 4 components: (1) education and self-monitoring of symptoms, (2) specific and (3) generalized cognitive restructuring, and (4) response prevention and graduated exposure. Treatment components are delivered according to a set protocol: education and self-monitoring (session 1), specific cognitive restructuring (sessions 2 and 3), exposure and response prevention (sessions 4–9), and generalized cognitive restructuring (sessions 10 and 11), with the final 12th session focusing on termination and relapse prevention issues. The treatment groups were composed of 10 participants on average. Sessions were conducted by trained PhD-level psychologists assisted by psychologists and psychotherapists from the public health sector. PhD-level psychologists received a 2-day training workshop with the developer of the tCBT protocol (PJN) and a gratis copy of the published treatment manual. Psychotherapists joining the project after the workshop received individual training from a trained PhD-level psychologist (MDP). There were case discussions with all psychotherapists throughout the study period.³¹

The comparator was TAU, which may have included consultations with physicians and mental health specialists in the private or public sector, psychotherapy or psychotropic drug use. A complete description of the intervention and study methods has been previously published.^26,31,33 Additional information on the intervention is in the Supplemental Material.

Economic Evaluation

Health care Costs Related to Health Service Utilization

The economic evaluation was carried out from the healthcare system perspective over a 1-year time horizon. Health care costs were measured from the healthcare system perspective in the year preceding and following baseline (T₀). Residents of Quebec are covered for the majority of their medical consultations under the province's public medical insurance plan. For residents not having private drug insurance coverage, the province's public drug plan will cover drug costs for medications on the provincial formulary. Costs included those incurred for health services during ambulatory visits (cost/visit), hospitalizations (per diem), medications covered under the public drug plan and physician fees paid for consultations and medical acts. Costs not covered by the health system are those incurred by the patient for receipt of services from health professionals in the private sector such as psychologists and medications for individuals covered under a private drug insurance plan.

Health care utilization for any health problem and related costs were ascertained from information found in provincial medical and pharmaceutical administrative databases (Supplemental Table S1). The cost related to the intervention for a training course also included professional training and supervision as well as time spent for intervention and preparation.

Cost estimation was based on previously published methods^34-36 and economic evaluation guidelines^37,38 using the direct allocation approach. The description of unit costs is presented in Supplemental Table S1. The detailed methodology used is described elsewhere.³⁶

Quality-Adjusted Life Years

The EQ-5D-5L was used to assess health-related quality of life on a 5-level rating scale ranging from no problem to extreme problems.³⁹ A summary health utility is calculated by applying Canadian weights to each response level to the 5 questions.⁴⁰ The total 1-year quality-adjusted life year (QALY; details in the Supplemental Material) is based on the EQ-5D-5L health utilities⁴⁰ measured at T₀, T₁, T₂, and T₃ and by calculating the area under the curve during the 1-year observation period [(((T₀ + T₁)/2) × 4months) + (((T₁ + T₂)/2) × 4months) + (((T₂ + T₃)/2) × 4months)/12months].³⁸

Other Covariables

Study covariables included sociodemographic and clinical characteristics defined in the Supplemental Material.

Statistical Analysis

Nonparametric and parametric descriptive (Mann–Whitney test, Student's t-test and chi-square statistics) and bivariate analyses (generalized linear models with log link and gamma distribution) were conducted to assess the association between the intervention and health care costs, health care utilization, QALYs, and study variables.

The current economic evaluation included an intention-to-treat analysis and additional complete case analyses that considered individuals with complete follow-up information. Challenges in analyses of data and the valid conclusions in economic evaluations include the importance of accounting for differences in baseline characteristics and potential important confounders, missing data throughout follow-up, skewed distribution of costs and the correlation between costs and QALYs.^41,42 It has been shown that seemingly unrelated regression analyses correcting for confounders with multiple imputations address all 4 challenges.⁴²

In order to estimate the cost-effectiveness of tCBT + TAU as compared to TAU, in the intent-to-treat analysis, we first carried out multiple imputations using chained equations for missing data on total healthcare system costs and QALYs and potential study confounders.^43-45 There were 45 imputed datasets that corresponded to the percentage of individuals with incomplete data,⁴⁶ and the mean and standard error of the mean (SEM) of the means of each multiple imputation was estimated with Rubin's rule.⁴⁷

The total healthcare system costs and QALYs associated with the intervention were studied with seemingly unrelated regression analysis, to account for the correlation between the costs and QALYs, where 2 separate regression models are specified simultaneously, that is, one for total costs and QALYs.^48,49 Total costs were adjusted for past-year total healthcare system costs and baseline primary diagnosis of anxiety disorder, depression symptom severity, and psychotropic drug use. QALYs were adjusted for baseline preference-based index value, primary diagnosis of anxiety disorder, depression symptom severity, and psychotropic drug use.

The incremental cost-effectiveness ratio (ICER) was calculated using the difference in mean point estimates in costs divided by the difference in mean QALYs, associated with tCBT + TAU as compared to TAU, with the following formula: ICER = ΔC/ΔE = (C_tCBT+TAU−C_TAU)/(QALY_tCBT+TAU–QALY_TAU).³⁸ The cost-effectiveness of tCBT + TAU as compared to TAU was assessed with the net benefit regression method^50,51 across a range of willingness-to-pay (WTP) thresholds relevant to the Canadian context,⁵² with uncertainty explored by cost-effectiveness acceptability curves (CEAC).⁵³ These are presented by plotting the probability (y-axis) that tCBT + TAU as compared to TAU is cost-effective for a range of WTP values ranging up to $140,000 (x-axis).^53,54 For this, we calculated the net monetary benefit (NMB) for each individual, which combines economic and health outcomes such as the QALY in a dollar amount⁵⁵ with the following formula: [NMB = (WTP×QALY)–costs]. For different WTP thresholds, multiple regression analyses were carried out to study the NMB as a function of the intervention (tCBT + TAU versus TAU) controlling for baseline primary diagnosis of anxiety disorder, depression symptom severity, psychotropic drug use, health-related quality of life at baseline and past-year total healthcare system costs. Regression analyses were carried out on imputed data for intent-to-treat and complete case analyses. All analyses were carried out with SAS version 9.4.

Results

The flowchart of the study is presented in the Supplemental Material. Among the 231 participants randomized, 218 (94%) gave access to their medico-administrative data of which, n = 112 were randomized to the tCBT + TAU and n = 106 to the TAU groups had similar sociodemographic and clinical characteristics except for the primary diagnosis and comorbid diagnoses (Table 1). At the baseline, the 2 groups had similar scores on health-related quality of life (HRQOL) on the EQ-5D-5L and the EQ-Visual Analogue Scale. Retention rates at follow-up were 86% (T₁), 74% (T₂) and 72% (T₃). In general, the tCBT + TAU group had on average better HRQOL than the TAU group at T₁, T₂, and T₃ (p < 0.05) (Table 2). The difference in EQ-5D-5L index scores between tCBT + TAU and TAU at T₁, T₂, and T₃ was 0.08 (p < 0.001), 0.04 (p = 0.057), and 0.04 (p = 0.032), respectively. The difference in visual analogue scores between tCBT + TAU and TAU at T₁, T₂, and T₃ was 6.3 (p = 0.004), 5.8 (p = 0.004), and 5.4 (p = 0.025), respectively. The difference in patient healthcare questionnaire-9 scores between tCBT + TAU and TAU at T₁, T₂, and T₃ was −3.4 (p < 0.001), −2.1 (p = 0.011), and −2.1 (p = 0.010), respectively. The difference in Beck anxiety inventory scores between tCBT + TAU and TAU at T₁, T₂, and T₃ was −8.2 (p < 0.001), −7.9 (p < 0.001), and −6.3 (p < 0.001), respectively.

Table 1.

Characteristics of the Study Sample According to Treatment Group.

	tCBT + TAU(n = 112)	TAU(n = 106)	Missing data	p-value
Sex (women), n (%)	97 (86.6)	89 (84.0)	0	0.581
Age, means (SD)	37.7 (12.4)	36.5 (12.0)	0	0.443
Marital status, n (%)
Married/living with a partner	62 (55.4)	56 (52.8)	1	0.911
Single	42 (37.5)	40 (37.7)
Separated/divorced	8 (7.1)	9 (8.5)
Education, n (%)
Secondary or less	8 (7.2)	14 (13.2)	1	0.338
Postsecondary/vocational	56 (50.5)	49 (46.2)
University	47 (42.3)	43 (40.6)
Occupation, n (%)
Work full time	69 (61.6)	60 (56.6)	0	0.718
Student	21 (18.8)	24 (22.6)
Other	22 (19.6)	22 (20.8)
Economic situation (enough and higher), n (%)	83 (74.1)	84 (79.2)	2	0.359
Physical health (excellent/good), n (%)	44 (39.3)	32 (30.2)	0	0.159
Clinical characteristics
Principal diagnosis, n (%)
Generalized anxiety disorder	67 (59.8)	47 (44.3)	0	0.033
Social anxiety disorder	26 (23.2)	38 (35.9)
Panic disorder	13 (11.6)	19 (17.9)
Agoraphobia	6 (5.4)	2 (1.9)
Comorbid diagnoses, n (%)
Generalized anxiety disorder	29 (25.9)	30 (28.3)	0	0.689
Social anxiety disorder	43 (38.4)	33 (31.1)	0	0.261
Panic disorder	38 (33.9)	22 (20.8)	0	0.030
Agoraphobia	32 (28.6)	27 (25.5)	0	0.607
Specific phobia	22 (19.6)	16 (15.1)	0	0.376
Posttraumatic stress disorder	3 (2.7)	4 (3.8)	0	0.716
Obsessive-compulsive disorder	8 (7.1)	4 (3.8)	0	0.376
Persistent depressive disorder	11 (9.8)	6 (5.7)	0	0.252
Major depressive disorder	29 (25.9)	25 (23.6)	0	0.693
Sheehan disability, means (SD)	14.8 (6.3)	13.8 (6.2)	6	0.252
Psychologist/psychotherapist in the last 12 months, n (%)	56 (50.0)	42 (39.6)	0	0.124
Self-reported psychotropic medication use in the last 12 months, n (%)
SSRI/SNRI	60 (53.6)	58 (54.7)	0	0.865
Benzodiazepine	41 (36.6)	30 (28.3)	0	0.191

Note. tCBT = transdiagnostic cognitive-behavioural therapy; TAU = treatment as usual; SSRI = selective serotonin reuptake inhibitor; SNRI = serotonin and norepinephrine reuptake inhibitor.

P-value based on chi-square and t-test statistics.

Table 2.

Effects Associated With the Treatment Group.

	tCBT + TAU	TAU	Missing data	p-value
Pretreatment (baseline, T₀)	n = 112	n = 106
EQ-5D-5L index, means (SD)	0.69 (0.16)	0.72 (0.15)	2	0.284
Visual analogue scale, means (SD)	66.3 (16.2)	67.9 (14.5)	0	0.433

Four months postrandomization (posttreatment, T₁)	N = 90	N = 97
EQ-5D-5L index, means (SD)	0.81 (0.14)	0.73 (0.17)	8	0.001
Visual analogue scale, means (SD)	73.8 (15.0)	67.5 (13.8)	8	0.004

Eight months postrandomization (posttreatment, T₂)	N = 77	N = 85
EQ-5D-5L index, means (SD)	0.82 (0.11)	0.78 (0.13)	1	0.057
Visual analogue scale, means (SD)	75.6 (12.7)	69.8 (12.6)	1	0.004

Twelve months postrandomization (posttreatment, T₃)	N = 72	N = 85	0
EQ-5D-5L index, means (SD)	0.82 (0.12)	0.78 (0.14)	2	0.032
Visual analogue scale, means (SD)	74.9 (14.9)	69.5 (15.2)	0	0.025

Note. tCBT = transdiagnostic cognitive-behavioural therapy; TAU = treatment as usual.

P-value based on t-test statistics.

Health care utilization according to the intervention group is presented in Supplemental Table S2. In general, no significant differences were observed in the type of health care utilization between intervention groups. The related health care costs are presented in Table 3. No significant difference was observed in total costs. In the year preceding baseline, individuals randomized in the TAU incurred higher day surgery-related costs than those in the tCBT + TAU arm and in the year following baseline, individuals randomized in the tCBT + TAU arm incurred higher emergency department costs than those in the TAU arm.

Table 3.

Unadjusted Costs (CAD) (mean per person) According to Treatment Groups.

	tCBT + TAU(N = 112)	TAU(N = 106)	p-value
12 months prior to T₀
Outpatient visits
Mental health reasons	524	469	0.740
Physical conditions	584	523	0.589
Psychosocial services from CLSC	526	307	0.173
Total	1,634	1,299	0.159
Emergency department visits	8	9	0.860
Inpatient stays
Hospitalizations	75	72	0.913
Day surgeries	19	57	0.001
Total	94	129	0.420
Physician fees	590	560	0.757
Medication costs paid under the public drug plan	117	129	0.770
Total costs	2,443	2,126	0.339
12 months after T₀
Outpatient visits
Mental health reasons	388	281	0.370
Physical conditions	536	558	0.853
Psychosocial services from CLSC	699	495	0.369
Total	1,623	1,334	0.266
Emergency department visits	20	5	<0.001
Inpatient stays
Hospitalizations	141	199	0.417
Day surgeries	72	76	0.888
Total	213	275	0.552
Physician fees	611	794	0.174
Medication costs paid by the public drug plan	245	125	0.067
Total costs	2,712	2,533	0.679
Program costs	410
Total costs including program costs	3,122

Note. tCBT = transdiagnostic cognitive-behavioural therapy; TAU = treatment as usual; CLSC = Centre Local de Services Communautaires.

P-values are based on generalized linear models with log link and gamma distribution.

The ICER point estimates from the imputed intention-to-treat and complete case analyses are presented in Table 4. The results showed that the ICER associated with tCBT + TAU as compared to TAU in the intention-to-treat analysis was $20,000/QALY and $6,581/QALY, based on unadjusted and adjusted point estimates, respectively. Complete case analysis showed similar results with an ICER of $15,556/QALY and $6,642/QALY, based on unadjusted and adjusted point estimates, respectively. The CEAC according to analyses is presented in Figure 1. Results from the adjusted NMB regression, based on intention-to-treat analyses, showed that at a WTP threshold of $20,000 and $40,000, the probability that tCBT + TAU as compared to TAU is cost-effective is 93.0% and 99.9%, respectively. Results from the complete case analyses showed that at a WTP threshold of $20,000 and $40,000, the probability that tCBT + TAU as compared to TAU is cost-effective is 82% and 97%, respectively. Results from the unadjusted NMB regressions showed that in the intention-to-treat and complete case analyses and at a WTP threshold of $60,000, the probability that tCBT + TAU as compared to TAU is >85%.

Figure 1.

Cost acceptability curves from the healthcare system perspective.

Table 4.

Unadjusted and adjusted incremental QALYs, costs, and cost-effectiveness ratio from the health care system perspective over the 12-month period.

	Intention-to-treat analysis with imputed data for missing information on study variables n=117 in tCBT+TAU group; n=114 in TAU group				Complete case analysis n=60 in tCBT+TAU group; n=71 in TAU group
	tCBT+TAU	TAU	tCBT+TAU versus TAU		tCBT+TAU	TAU	tCBT+TAU versus TAU
	Mean (SEM*)		Unadjusted Beta estimates (SEM*)	Adjusted Beta estimates (SEM*)	Mean (SEM*)		Unadjusted Beta estimates (SEM)	Adjusted Beta estimates (SEM)
Costs	3120 (310)	2526 (273)	594 (414)	281 (385)	3311 (482)	2638 (375)	672 (602)	275 (546)
QALYs	0.7901 (0.0095)	0.7604(0.0104)	0.0297(0.0141)	0.0427(0.0083)	0.8051(0.0142)	0.7619(0.0147)	0.0432 (0.0206)	0.0414(0.0137)
ICER			$20 000/QALY	$6 581/QALY			$15 556/QALY	$6 642/QALY

CAD=Canadian dollars; ICER: Incremental cost-effectiveness ratio

SEM*: standard error of the mean from multiple imputation

Unadjusted and adjusted estimates obtained from seemingly unrelated regressions

Discussion

This economic evaluation contributes to the rare studies on the cost-effectiveness of group tCBT for patients with anxiety disorders using information from provincial medico-administrative data to assess health care utilization and related costs, reducing recall and social desirability bias. This study also included a methodology and analytic approach that addressed missing information, skewness of data, confounders and the correlation between costs and QALYs,⁴² which increased the validity of conclusions. Furthermore, this study was carried out within primary care settings where the majority of anxiety disorders are treated, and the healthcare system perspective taken provides important information for decision makers given the public healthcare system context where residents are covered for the majority of their medical consultations.

The findings showed that on average the total 1-year health care costs incurred by individuals randomized to receive group tCBT + TAU and TAU were similar. The average QALYs were, however, on average higher in those randomized to receive group tCBT + TAU as compared to TAU. The results also showed that group tCBT + TAU was more cost-effective than TAU from the healthcare system perspective for treating adult patients with anxiety disorders over a 1-year time horizon. The CEAC from the current study showed that tCBT + TAU was cost-effective as compared to TAU with high probability and this is from findings observed in the intention-to-treat and complete case analyses.

To our knowledge, very few studies have assessed the cost-effectiveness of group tCBT.^29,31,56 In the current study, the ICER adjusting for potential baseline confounders was $6,581/QALY in the intention-to-treat analysis, which was similar to the ICER of $6,642/QALY observed in the complete case analysis. These estimates are similar to those reported on the effectiveness and cost-effectiveness of CBT for anxiety disorders. In Australia, the cost-effectiveness ratio for CBT compared to TAU was $6900/QALY AUD (CAD $6500/QALY) for the treatment of generalized anxiety disorder and panic disorder from the public healthcare system perspective.⁵⁷ CBT has also been shown to be cost-effective in specialized mental health settings as part of routine care for patients with anxiety disorders.⁵⁸

Although specific WTP thresholds have not been systematically used in studies, earlier benchmarks suggested a threshold of CAD $50,000/QALY to CAD $100,000/QALY.^59-61 A recent meta-regression analysis of a systematic review predicted a WTP threshold in the province of Quebec of CAD 127,985/QALY.⁵² Based on these WTP thresholds, the results of the current study, suggest that group tCBT + TAU as compared to TAU has a high probability of being cost-effective. In fact, the probability was 93% and 99% that tCBT + TAU as compared to TAU was cost-effective at a WTP of $20,000 and $40,000, respectively. Using self-declared health service utilization data from the same trial over an 8-month period, the probability that tCBT + TAU as compared to TAU was 97% at a WTP of $50,000 from the healthcare system perspective.

The results of this study should be interpreted with the following considerations. First, among the 231 participants randomized and who completed the baseline interview, 13 did not consent to access their medico-administrative data. Loss to follow-up was 14% at T₁, 26% at T₂, and 28% at T₃. Of the individuals randomized, 131 had complete information on all study variables. There were no differences in losses to follow-up with regard to sociodemographic factors and anxiety symptoms. Losses were, however, more likely to have a principal diagnosis of agoraphobia and panic disorder. Second, costs were measured from the healthcare system perspective and health service utilization was assessed through information found in the provincial medico-administrative data, which may be subject to data entry errors. Studies however have shown that these provincial medical and pharmaceutical registries are valid sources for health service utilization⁶² and therefore the potential information bias leading to an underestimation of health system costs is reduced. Furthermore, we do not expect that this information bias was differential between groups and therefore results presented are more likely to be conservative. Third, although the use of the EQ-5D has been recommended in economic evaluation guidelines,⁶³ other instruments such as the Assessment of Quality of Life-8 dimensions scale (AQoL-8D) may be more sensitive in assessing changes related to the treatment of mental disorders such as anxiety⁶⁴ and should be considered in future studies assessing the cost-effectiveness of tCBT. Finally, the findings can be generalized to adults with anxiety disorders consulting in community settings. Individuals having consulted in the past year with a psychiatrist were excluded from the study. Studies have shown that individuals with anxiety having been prescribed medications from a psychiatrist, as compared to a primary care physician, were twice more likely to receive psychotherapy.⁶⁵ Although patients consulting psychiatrists may represent different referral and health service trajectories, research should also assess the cost-effectiveness of group tCBT in patients that have received specialized mental health services such as those from psychiatrists.

Conclusion

The study findings suggest that, for anxiety disorders, group tCBT + TAU as compared to TAU has a high probability of being cost-effective. This group intervention has the potential to increase access to psychotherapy treatment for anxiety disorders in Canada, where barriers such as long waiting lists in mental health services persist. The implementation of group tCBT within a primary care context is also aligned with increasing delivery of community-based mental health interventions, a sustainable solution to improve access to mental health services in Canada.¹⁶ Future studies including younger and older adults and different clinical patient populations with different anxiety disorders are needed to further investigate the cost-effectiveness of tCBT.

Supplemental Material

sj-docx-1-cpa-10.1177_07067437231187459 - Supplemental material for Cost-Effectiveness of Group Transdiagnostic Cognitive Behavioural Therapy for Anxiety Disorders in Primary Care Settings: Economic Evaluation From the Healthcare System Perspective Over a 1-Year Time Horizon

Supplemental material, sj-docx-1-cpa-10.1177_07067437231187459 for Cost-Effectiveness of Group Transdiagnostic Cognitive Behavioural Therapy for Anxiety Disorders in Primary Care Settings: Economic Evaluation From the Healthcare System Perspective Over a 1-Year Time Horizon by Helen-Maria Vasiliadis, Catherine Lamoureux-Lamarche, Alexandra Chapdelaine, Martin D. Provencher, Peter J. Norton, Djamal Berbiche and Pasquale Roberge in The Canadian Journal of Psychiatry

Supplemental Material

sj-doc-2-cpa-10.1177_07067437231187459 - Supplemental material for Cost-Effectiveness of Group Transdiagnostic Cognitive Behavioural Therapy for Anxiety Disorders in Primary Care Settings: Economic Evaluation From the Healthcare System Perspective Over a 1-Year Time Horizon

Supplemental material, sj-doc-2-cpa-10.1177_07067437231187459 for Cost-Effectiveness of Group Transdiagnostic Cognitive Behavioural Therapy for Anxiety Disorders in Primary Care Settings: Economic Evaluation From the Healthcare System Perspective Over a 1-Year Time Horizon by Helen-Maria Vasiliadis, Catherine Lamoureux-Lamarche, Alexandra Chapdelaine, Martin D. Provencher, Peter J. Norton, Djamal Berbiche and Pasquale Roberge in The Canadian Journal of Psychiatry

Footnotes

Data Access

The authors are not legally authorized to share or publicly publish data. Participants were not requested to give informed consent for data sharing. Requests for access to the data should be addressed to the ethics boards who approved the study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Ethical Considerations

The research protocol was approved by the principal ethics review boards of the Integrated Health and Social Services Centers in Estrie (No. MP-22-2016-570), Québec City (No. 2017-166), and Laval (No. 2016-2017-C54). All participants provided written informed consent.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The current study was funded by the Canadian Institutes of Health Research funded this study (CIHR grant nos. 334091 and 340410). The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

ORCID iDs

Helen-Maria Vasiliadis

Catherine Lamoureux-Lamarche

Alexandra Chapdelaine

Martin D. Provencher

Djamal Berbiche

Supplemental Material

Supplemental material for this article is available online.

References

Bandelow

Michaelis

. Epidemiology of anxiety disorders in the 21st century. Dialogues Clin Neurosci. 2015;17(3):327-335.

Somers

Goldner

Waraich

, et al. Prevalence and incidence studies of anxiety disorders: a systematic review of the literature. Can J Psychiatry. 2006;51(2):100-113.

Horenstein

Heimberg

. Anxiety disorders and healthcare utilization: a systematic review. Clin Psychol Rev. 2020;81:101894.

Konnopka

König

. Economic burden of anxiety disorders: a systematic review and meta-analysis. Pharmacoeconomics. 2020;38(1):25-37.

Vasiliadis

H-M

Spagnolo

Bartram

, et al. Factors associated with change in moderate or severe symptoms of anxiety and depression in community living adults and older adults during the first wave of the pandemic across Canada. Can J Public Health. 2022. Submitted.

COVID-19 Mental Disorders Collaborators. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. Lancet. 2021;398(10312):1700-1712.

Statistics Canada. Self-perceived mental health and mental health care needs during the COVID-19 pandemic. Statistics Canada; 2021.

Young

Klap

Sherbourne

, et al. The quality of care for depressive and anxiety disorders in the United States. Arch Gen Psychiatry. 2001;58(1):55-61.

Roberge

Normand-Lauzière

Raymond

, et al. Generalized anxiety disorder in primary care: mental health services use and treatment adequacy. BMC Fam Pract. 2015;16:146.

10.

Katzman

Bleau

Blier

, et al. Canadian Clinical practice guidelines for the management of anxiety, posttraumatic stress and obsessive-compulsive disorders. BMC Psychiatry. 2014;14(1):S1.

11.

Marcus

Olfson

. National trends in the treatment for depression from 1998 to 2007. Arch Gen Psychiatry. 2010;67(12):1265-1273.

12.

McManus

Bebbington

Jenkins

, et al. Mental health and wellbeing in England: The adult psychiatric morbidity survey 2014. NHS Digital. 2016.

13.

Clark

. Realizing the mass public benefit of evidence-based psychological therapies: the IAPT program. Annu Rev Clin Psychol. 2018;14:159-183.

14.

Government of Canada. Prescription drug insurance coverage. 2020. Accessed November 23, 2022. https://www.canada.ca/en/health-canada/services/health-care-system/pharmaceuticals/access-insurance-coverage-prescription-medicines.html.

15.

Vasiliadis

Spagnolo

Lesage

. Public funding of evidence-based psychotherapy for common mental disorders: increasing calls for action in Canadian provinces. Healthcare Policy. 2021;16(3):16-25.

16.

Moroz

D’Angelo

. Mental health services in Canada: barriers and cost-effective solutions to increase access. Healthc Manage Forum. 2020;33(6):282-287.

17.

Vasiliadis

H-M

Dezetter

Latimer

, et al. Assessing the costs and benefits of insuring psychological services as part of medicare for depression in Canada. Psychiatr Serv. 2017;68(9):899-906.

18.

Health Quality Ontario. Psychotherapy for major depressive disorder and generalized anxiety disorder: a health technology assessment. Ont Health Technol Assess Ser. 2017;17(15):1.

19.

Goldstein-Piekarski

Williams

Humphreys

. A trans-diagnostic review of anxiety disorder comorbidity and the impact of multiple exclusion criteria on studying clinical outcomes in anxiety disorders. Transl Psychiatry. 2016;6(6):e847.

20.

Norton

Philipp

. Transdiagnostic approaches to the treatment of anxiety disorders: a quantitative review. Psychotherapy (Chic). 2008;45(2):214-226.

21.

Guimond

A-J

Provencher

Bélair

. Traitements cognitifscomportementaux transdiagnostiques des troubles anxieux: recension des écrits. Rev Québéc psychol. 2013;34(2):265-293.

22.

Norton

Roberge

. Transdiagnostic therapy. Psychiatr Clin North Am. 2017;40(4):675-687.

23.

Barlow

Allen

Choate

. Toward a unified treatment for emotional disorders - republished article. Behav Ther. 2016;47(6):838-853.

24.

Talkovsky

Norton

. Mediators of transdiagnostic group cognitive behavior therapy. J Anxiety Disord. 2014;28(8):919-924.

25.

Jiang

MYW

Kandris

El-Masry

, et al. Effectiveness of transdiagnostic group cognitive-behavioural therapy for anxiety disorders in a naturalistic clinical setting. Clin Psychologist. 2021;25(1):69-77.

26.

Roberge

Provencher

, & Gaboury

, et al. Group transdiagnostic cognitive-behavior therapy for anxiety disorders: a pragmatic randomized clinical trial. Psychol Med. 2020;52(13):1-11.

27.

Norton

Barrera

Mathew

, et al. Effect of transdiagnostic cbt for anxiety disorders on comorbid diagnoses. Depress Anxiety. 2013;30(2):168-173.

28.

Andersen

Toner

Bland

, et al. Effectiveness of transdiagnostic cognitive behaviour therapy for anxiety and depression in adults: a systematic review and meta-analysis. Behav Cogn Psychother. 2016;44(6):673-690.

29.

Chapdelaine

Vasiliadis

Provencher

, et al. Cost-effectiveness of transdiagnostic group cognitive behavioural therapy for anxiety disorders versus treatment as usual: economic evaluation of a pragmatic randomized controlled trial over an 8-month time horizon using self-reported data. Psychol Med. 2023:1-13.

30.

Roberge

Marchand

Reinharz

, et al. Healthcare utilization following cognitive-behavioral treatment for panic disorder with agoraphobia. Cogn Behav Ther. 2005;34(2):79-88.

31.

Roberge

Provencher

Gosselin

, et al. A pragmatic randomized controlled trial of group transdiagnostic cognitive-behaviour therapy for anxiety disorders in primary care: study protocol. BMC Psychiatry. 2018;18(1):320.

32.

Brown

, & Barlow

. Anxiety and related disorders interview schedule for DSM-5 (ADIS-5)-adult and lifetime version: Clinician manual. New-York, NY: Oxford University Press; 2014.

33.

Norton

. Group cognitive-behavioral therapy of anxiety: A transdiagnostic treatment manual. New York, NY: Guilford Press; 2012.

34.

Vasiliadis

H-M

Dionne

Preville

, et al. The excess healthcare costs associated with depression and anxiety in elderly living in the community. Am J Geriatric Psychiatry. 2013;21(6):536-548.

35.

Reinharz

Lesage

Contandriopoulos

. Cost-effectiveness analysis of psychiatric deinstitutionalization. Can J Psychiatry. 2000;45(6):533-538.

36.

Lamoureux-Lamarche

Berbiche

Vasiliadis

. Health care system and patient costs associated with receipt of adequate mental health care for depression and anxiety disorders in older adults. BMC Psychiatry. 2022;22:175.

37.

Canadian Agency for Drugs and Technologies in Health. Guidelines for the economic evaluation of health technologies: Canada. 4th ed. ed. Ottawa, Ontario: CADTH; 2017.

38.

Drummond

Sculpher

, & Claxton

, et al. Methods for the economic evaluation of health care programmes. New-York, NY: Oxford University Press; 2015.

39.

van Reenen

Janssen

. EQ-5D-5L user guide: basic information on how to use the EQ-5D-5L instrument. Rotterdam: EuroQol Research Foundation; 2015.

40.

Xie

Pullenayegum

Gaebel

, et al. A time trade-off-derived value set of the EQ-5D-5L for Canada. Med Care. 2016;54(1):98-105.

41.

Faria

Gomes

Epstein

, et al. A guide to handling missing data in cost-effectiveness analysis conducted within randomised controlled trials. Pharmacoeconomics. 2014;32(12):1157-1170.

42.

Mutubuki

El Alili

Bosmans

, et al.

The statistical approach in trial-based economic evaluations matters: get your statistics together!

BMC Health Serv Res. 2021;21(1):475.

43.

Ramsey

Willke

Briggs

, et al. Good research practices for cost-effectiveness analysis alongside clinical trials: the ISPOR RCT-CEA task force report. Value Health. 2005;8(5):521-533.

44.

Simons

Rivero-Arias

L-M

, et al. Missing data in the health related quality of life EQ-5D-3L instrument—should we impute individual domains or the actual index. Paper presented at: Research paper presented at the 83rd Health Economists’ Study Group Meeting, 26–28 June 2013.

45.

National Institute for Health and Care Excellence. Guide to the methods of technology appraisal 2013. National Institute for Health and Clinical Excellence. 2013.

46.

White

Royston

Wood

. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377-399.

47.

Rubin

. Multiple imputation for nonresponse in surveys. New York, NY: Wiley; 2004.

48.

Willan

Briggs

Hoch

. Regression methods for covariate adjustment and subgroup analysis for non-censored cost-effectiveness data. Health Econ. 2004;13(5):461-475.

49.

Willan

. Statistical analysis of cost-effectiveness data from randomized clinical trials. Expert Rev Pharmacoecon Outcomes Res. 2006;6(3):337-346.

50.

Hoch

Rockx

Krahn

. Using the net benefit regression framework to construct cost-effectiveness acceptability curves: an example using data from a trial of external loop recorders versus Holter monitoring for ambulatory monitoring of “community acquired” syncope. BMC Health Serv Res. 2006;6:68.

51.

Hoch

Dewa

. Advantages of the net benefit regression framework for economic evaluations of interventions in the workplace: a case study of the cost-effectiveness of a collaborative mental health care program for people receiving short-term disability benefits for psychiatric disorders. J Occup Environ Med. 2014;56(4):441-445.

52.

Kouakou

CRC

Poder

. Willingness to pay for a quality-adjusted life year: a systematic review with meta-regression. Eur J Health Econ. 2022;23(2):277-299.

53.

Fenwick

O'Brien

Briggs

. Cost-effectiveness acceptability curves–facts, fallacies and frequently asked questions. Health Econ. 2004;13(5):405-415.

54.

van Hout

Gordon

, et al. Costs, effects and C/E-ratios alongside a clinical trial. Health Econ. 1994;3(5):309-319.

55.

Briggs

Sculpher

Claxton

. Decision modelling for health economic evaluation. Oxford: Oxford University Press; 2006.

56.

Aguilera-Martín

Gálvez-Lara

Cuadrado

, et al. Cost-effectiveness and cost-utility evaluation of individual vs. Group transdiagnostic psychological treatment for emotional disorders in primary care (PsicAP-costs): a multicentre randomized controlled trial protocol. BMC Psychiatry. 2022;22(1):99.

57.

Heuzenroeder

Donnelly

Haby

, et al. Cost-effectiveness of psychological and pharmacological interventions for generalized anxiety disorder and panic disorder. Aust N Z J Psychiatry. 2004;38(8):602-612.

58.

Romijn

Batelaan

Koning

, et al. Acceptability, effectiveness and cost-effectiveness of blended cognitive-behavioural therapy (bCBT) versus face-to-face CBT (ftfCBT) for anxiety disorders in specialised mental health care: a 15-week randomised controlled trial with 1-year follow-up. PLoS One. 2021;16(11):e0259493.

59.

Griffiths

Vadlamudi

. Cadth’2019’s $50000 cost-effectiveness threshold: fact or fiction? Value Health. 2016;19(7):A488-A489.

60.

Neumann

Cohen

Weinstein

. Updating cost-effectiveness—the curious resilience of the $50,000-per-QALY threshold. N Engl J Med. 2014;371(9):796-797.

61.

Laupacis

Feeny

Detsky

, et al. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. CMAJ: Can Med Assoc J. 1992;146(4):473.

62.

Tamblyn

Lavoie

Petrella

, et al. The use of prescription claims databases in pharmacoepidemiological research: the accuracy and comprehensiveness of the prescription claims database in Quebec. J Clin Epidemiol. 1995;48(8):999-1009.

63.

Brazier

Longworth

. NICE DSU Technical Support Document 8: an introduction to the measurement and valuation of health for NICE submissions. 2011.

64.

Mihalopoulos

Chen

Iezzi

, et al. Assessing outcomes for cost-utility analysis in depression: comparison of five multi-attribute utility instruments with two depression-specific outcome measures. Br J Psychiatry. 2014;205(5):390-397.

65.

Weisberg

Dyck

Culpepper

, et al. Psychiatric treatment in primary care patients with anxiety disorders: a comparison of care received from primary care providers and psychiatrists. Am J Psychiatry. 2007;164(2):276-282.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.22 MB

0.20 MB