Advertisement

Cost-Effectiveness of Repetitive Transcranial Magnetic Stimulation versus Antidepressant Therapy for Treatment-Resistant Depression

Open ArchivePublished:June 11, 2015DOI:https://doi.org/10.1016/j.jval.2015.04.004

      Abstract

      Background

      Repetitive transcranial magnetic stimulation (rTMS) therapy is a clinically safe, noninvasive, nonsystemic treatment for major depressive disorder.

      Objective

      We evaluated the cost-effectiveness of rTMS versus pharmacotherapy for the treatment of patients with major depressive disorder who have failed at least two adequate courses of antidepressant medications.

      Methods

      A 3-year Markov microsimulation model with 2-monthly cycles was used to compare the costs and quality-adjusted life-years (QALYs) of rTMS and a mix of antidepressant medications (including selective serotonin reuptake inhibitors, serotonin and norepinephrine reuptake inhibitors, tricyclics, noradrenergic and specific serotonergic antidepressants, and monoamine oxidase inhibitors). The model synthesized data sourced from published literature, national cost reports, and expert opinions. Incremental cost-utility ratios were calculated, and uncertainty of the results was assessed using univariate and multivariate probabilistic sensitivity analyses.

      Results

      Compared with pharmacotherapy, rTMS is a dominant/cost-effective alternative for patients with treatment-resistant depressive disorder. The model predicted that QALYs gained with rTMS were higher than those gained with antidepressant medications (1.25 vs. 1.18 QALYs) while costs were slightly less (AU $31,003 vs. AU $31,190). In the Australian context, at the willingness-to-pay threshold of AU $50,000 per QALY gain, the probability that rTMS was cost-effective was 73%. Sensitivity analyses confirmed the superiority of rTMS in terms of value for money compared with antidepressant medications.

      Conclusions

      Although both pharmacotherapy and rTMS are clinically effective treatments for major depressive disorder, rTMS is shown to outperform antidepressants in terms of cost-effectiveness for patients who have failed at least two adequate courses of antidepressant medications.

      Keywords

      Introduction

      Major depressive disorder (MDD) is a significant burden to many health care systems. It is a chronic and debilitating disease, and it significantly decreases quality of life [
      • Lepine J.P.
      • Briley M.
      The increasing burden of depression.
      ,
      • Mrazek D.A.
      • Hornberger J.C.
      • Altar C.A.
      • et al.
      A review of the clinical, economic, and societal burden of treatment-resistant depression: 1996–2013.
      ]. It is one of the most common of all psychiatric disorders and ranks among the leading causes of disability worldwide [
      • Greden J.F.
      The burden of recurrent depression: causes, consequences, and future prospects.
      ]. Although many patients with depression respond to first-line medication and psychotherapy treatments, an estimated 20% to 40% of the patients are unable to tolerate pharmacotherapy or do not benefit from these treatments after repeated attempts [
      • Greden J.F.
      The burden of recurrent depression: causes, consequences, and future prospects.
      ]. In addition, even with successful acute treatment outcomes, the long-term durability of response among treatment-resistant patients is poor. A recent review reported that the rate of recurrence of MDD, treated in specialized mental health settings, was very high: 60% after 5 years, 67% after 10 years, and 85% after 15 years [
      • Lepine J.P.
      • Briley M.
      The increasing burden of depression.
      ]. Patients with treatment-resistant depression (TRD) contribute to a disproportionately high burden of illness than do patients who respond to treatment; they are twice as likely to be hospitalized and have higher treatment costs [
      • Mrazek D.A.
      • Hornberger J.C.
      • Altar C.A.
      • et al.
      A review of the clinical, economic, and societal burden of treatment-resistant depression: 1996–2013.
      ].
      Repetitive transcranial magnetic stimulation (rTMS) therapy is a noninvasive, nonsystemic therapeutic device offering treatment that uses pulsed magnetic fields at magnetic resonance imaging strength to induce an electric current in a localized region of the cerebral cortex. During the rTMS session, the patient is conscious and there is no requirement for an anesthetic or muscle relaxants. A treatment session usually lasts approximately 40 minutes and is normally performed three to five times a week over a period of 4 to 6 weeks. After each session, patients may continue with their daily work or other routines. rTMS produces a clinical benefit without the systemic adverse effects typical of oral medications and appears to have no adverse effects on cognition [
      • Carpenter L.L.
      • Janicak P.G.
      • Aaronson S.T.
      • et al.
      Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice.
      ,
      • Daskalakis Z.J.
      • Levinson A.J.
      • Fitzgerald P.B.
      Repetitive transcranial magnetic stimulation for major depressive disorder: a review.
      ,
      • Janicak P.G.
      • Nahas Z.
      • Lisanby S.H.
      • et al.
      Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month, multisite, open-label study.
      ,
      • Wall C.A.
      • Croarkin P.E.
      • McClintock S.M.
      • et al.
      Neurocognitive effects of repetitive transcranial magnetic stimulation in adolescents with major depressive disorder.
      ]. rTMS involves an electromagnetic coil and is not suitable in patients with metal items such as cochlear implants and implanted electrodes. It is also not recommended for patients who are at risk of epileptic seizures, who are withdrawing from drugs or alcohol, or who have drug or alcohol dependence [
      • Daskalakis Z.J.
      • Levinson A.J.
      • Fitzgerald P.B.
      Repetitive transcranial magnetic stimulation for major depressive disorder: a review.
      ].
      Clinical practice guidelines stipulate that current options for TRD are antidepressant medications, rTMS, and electroconvulsive therapy (ECT). In dozens of small trials, however, rTMS has been compared only with ECT mostly because they are both classified as physical therapies. ECT, however, is commonly used in emergency settings for psychotic patients, whereas rTMS is indicated for a wider range of mild to severe MDD. In addition, many patients with MDD are unable to tolerate the adverse effects of ECT or refuse to have ECT because of the associated stigma or fear about potential adverse effects (e.g., cognitive impairment). As such, it has been suggested that ECT is complementary, rather than a replaceable treatment, to rTMS and standard pharmacotherapies [
      • Fitzgerald P.
      Repetitive transcranial magnetic stimulation and electroconvulsive therapy: complementary or competitive therapeutic options in depression?.
      ].
      Simpson et al. [
      • Simpson K.N.
      • Welch M.J.
      • Kozel F.A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
      ] is the only study that compared pharmacotherapies and rTMS in the treatment of MDD. It concluded that rTMS provided a net cost saving compared with antidepressant medications [
      • Simpson K.N.
      • Welch M.J.
      • Kozel F.A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
      ]. Two studies compared rTMS with ECT and arrived at different conclusions [
      • Knapp M.
      • Romeo R.
      • Mogg A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation vs. electroconvulsive therapy for severe depression: a multi-centre randomised controlled trial.
      ,
      • Kozel F.A.
      • George M.S.
      • Simpson K.N.
      Decision analysis of the cost-effectiveness of repetitive transcranial magnetic stimulation versus electroconvulsive therapy for treatment of nonpsychotic severe depression.
      ]. Kozel et al. [
      • Kozel F.A.
      • George M.S.
      • Simpson K.N.
      Decision analysis of the cost-effectiveness of repetitive transcranial magnetic stimulation versus electroconvulsive therapy for treatment of nonpsychotic severe depression.
      ] (US-based) suggested that rTMS would be a cost-effective treatment for patients with MDD compared with ECT alone over 12 months, whereas Knapp et al. [
      • Knapp M.
      • Romeo R.
      • Mogg A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation vs. electroconvulsive therapy for severe depression: a multi-centre randomised controlled trial.
      ] (UK-based) found that rTMS has a very low probability of being a cost-effective alternative to ECT over 6 months.
      Because of the paucity of health economic assessments of rTMS for health service decision making, the purpose of this study was to investigate the cost-effectiveness of rTMS compared with pharmacotherapies for patients with TRD within the context of the Australian health system. We sought to use updated evidence to populate our economic model and ascertain whether our results corroborated the short-term findings from Simpson et al. [
      • Simpson K.N.
      • Welch M.J.
      • Kozel F.A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
      ].

      Methods

       Overview

      A hypothetical health state transition (Markov) model was used to combine data on the health care costs and health effects of rTMS and antidepressants over 3 years. The study population was patients with MDD who have failed two adequate medication trials from two different classes of drugs. Using a health system perspective, the key outcome was the incremental cost per quality-adjusted life-year (QALY), which represents the additional cost of rTMS per additional QALY compared with antidepressants. The stability of the results was thoroughly tested in sensitivity analyses.

       Treatment Strategies

      Two main treatment strategies were considered: rTMS and pharmacotherapies (standard antidepressant medications). The model considers a mix of pharmacotherapies because in practice, a large variety of antidepressants are prescribed for patients with MDD depending on previous treatments, medication tolerance, and resistance. The antidepressant mix includes selective serotonin reuptake inhibitors, which account for a large share of MDD medications in Australia, followed by serotonin and norepinephrine reuptake inhibitors, tricyclics, noradrenergic and specific serotonergic antidepressants, reversible inhibitor of monoamine oxidase A, and monoamine oxidase inhibitors [
      • Stephenson C.P.
      • Karanges E.
      • McGregor I.S.
      Trends in the utilisation of psychotropic medications in Australia from 2000 to 2011.
      ].
      The model assumed that other standard psychotherapies (i.e., talking therapies) were maintained during both treatment options. ECT, augmentation (e.g., lithium and atypical antipsychotic medications), and hospitalizations were available for those who failed the two main treatments.

       Model Structure

      A Markov microsimulation model was constructed and analyzed in TreeAge Pro 2014 software. The model duration was 3 years with 2-monthly cycles. Eight health states were used to account for acute or continuing treatments, combinations of responsiveness to treatment and relapse options, and deaths (see File 2 in Supplemental Materials found at doi: 10.1016/j.jval.2015.04.004). The MDD health states were based on the 17-item Hamilton Depression Rating Scale (HAM-D17), which is one of the most widely used and accepted measures for rating the severity of depression symptoms. In the absence of long-term clinical data, the duration of 3 years was chosen to track several courses of treatment per patient, which is typical of clinical practice.
      Patients entered the model and moved between the various health states according to their treatments, their response to therapies, and their chance of remission or relapse. The probability of gaining remission or regressing varied according to the strategy under analysis (either rTMS or antidepressant). After this point, the model for both strategies was identical in incorporating the probabilities of receiving salvage treatments and their efficacy outcomes (ECT, augmentation, and hospitalization) and the probability of having adverse events during treatment.

       Data Inputs and Sources

      To identify relevant evidence to populate the model, the Cochrane Library and Medline databases were searched. In both instances, a basic search strategy was used with key words (and their combinations) such as major depressive disorder, major depression, rTMS, antidepressant, ECT, treatment resistant. A manual search of the references of each identified article of interest was also completed for further information. Other sources of information included national epidemiological reports and hospital cost reports (Table 1).
      Table 1Summary of variables used in the economic model.
      DescriptionBase caseDistributionLowHighSources and assumptions
      Transition probabilities
       rTMS
        Remission: First treatment21.5%Beta19.7%31.2%Meta-analysis (File 1 in Supplemental Materials)
        Response: First treatment37.5%Beta33.2%48.7%Meta-analysis (File 1 in Supplemental Materials)
        Start maintenance10.0%Beta5.0%15.0%Expert opinion
      The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.
        Lose remission (no maintenance)16.6%Beta10.0%20.0%
      • Janicak P.G.
      • Nahas Z.
      • Lisanby S.H.
      • et al.
      Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month, multisite, open-label study.
      ,
      • Rush A.J.
      STAR*D: what have we learned?.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
      ,
      • Mantovani A.
      • Pavlicova M.
      • Avery D.
      • et al.
      Long-term efficacy of repeated daily prefrontal transcranial magnetic stimulation (TMS) in treatment-resistant depression.
      ,
      • Sackeim H.A.
      • Haskett R.F.
      • Mulsant B.H.
      • et al.
      Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial.
        Lose remission with maintenance12.0%Beta8.0%16.0%
      • Janicak P.G.
      • Nahas Z.
      • Lisanby S.H.
      • et al.
      Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month, multisite, open-label study.
      ,
      • Rush A.J.
      STAR*D: what have we learned?.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
      ,
      • Mantovani A.
      • Pavlicova M.
      • Avery D.
      • et al.
      Long-term efficacy of repeated daily prefrontal transcranial magnetic stimulation (TMS) in treatment-resistant depression.
      ,
      • Sackeim H.A.
      • Haskett R.F.
      • Mulsant B.H.
      • et al.
      Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial.
       Antidepressant medications
        Remission: First treatment13.6%Beta13.0%36.8%
      • Rush A.J.
      STAR*D: what have we learned?.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
        Response: First treatment16.8%Beta16.0%48.6%
      • Rush A.J.
      STAR*D: what have we learned?.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
        Lose remission28.1%Beta20.0%40.0%
      • Rush A.J.
      STAR*D: what have we learned?.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
       ECT (after failing rTMS or antidepressants)
        Remission: First treatment46.3%Beta20.0%70.0%Meta-analysis (File 1 in Supplemental Materials)
        Response: First treatment60.9%Beta40.0%80.0%Meta-analysis (File 1 in Supplemental Materials)
        Lose remission22.3%Beta15.0%35.0%
      • Sackeim H.A.
      • Haskett R.F.
      • Mulsant B.H.
      • et al.
      Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial.
       For all treatment arms
        Hospitalization10.4%Beta8.0%12.0%Calculated from literature
        Gaining REM after hospitalization35.0%Beta20.0%50.0%Assumption
        REM: % decrement for each subsequent treatment20.0%Triangular15.0%25.0%Assumption
        RESP: % decrement for each subsequent treatment15.0%Triangular10.0%20.0%Assumption
        Retreatment after relapse36.2%Beta25.0%45.0%Expert opinion
      The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.
        Relapse from partial remission50.0%Beta40.0%71.0%
      • Rush A.J.
      STAR*D: what have we learned?.
        Relapse: % increase for each subsequent treatment10.0%Triangular8.0%12.0%
      • Rush A.J.
      STAR*D: what have we learned?.
        Getting ECT after failing the main treatment25.0%Beta20.0%30.0%Assumption
        Having adverse events during treatment5.80%Beta4.0%8.0%
      • Rush A.J.
      STAR*D: what have we learned?.
      Health utilities
       Remission (HAM-D17 score <8)0.860Beta0.7500.900
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ,
      • Hawthorne G.
      • Cheok F.
      • Goldney R.
      • et al.
      The excess cost of depression in South Australia: a population-based study.
       Partial remission (mild-moderate 8 ≤ HAM-D17 score <20)0.710Beta0.6500.820
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ,
      • Hawthorne G.
      • Cheok F.
      • Goldney R.
      • et al.
      The excess cost of depression in South Australia: a population-based study.
       No response (severe-very severe HAM-D17 score ≥20)0.520Beta0.2500.580
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ,
      • Hawthorne G.
      • Cheok F.
      • Goldney R.
      • et al.
      The excess cost of depression in South Australia: a population-based study.
       Disutility for antidepressant treatment0.066Triangular0.0400.100
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ,
      • Hawthorne G.
      • Cheok F.
      • Goldney R.
      • et al.
      The excess cost of depression in South Australia: a population-based study.
       Disutility for rTMS treatment0.101Triangular0.0500.150Estimation
       Disutility for ECT treatment0.104Triangular0.5000.150Estimation
       Hospitalization (severe-very severe with suicidal risk HAM-D17 score ≥20)0.300Beta0.0900.400
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ,
      • Hawthorne G.
      • Cheok F.
      • Goldney R.
      • et al.
      The excess cost of depression in South Australia: a population-based study.
      Resources and cost components (2013–2014 AUD)
       rTMS
        Number of acute sessions28.3Triangular20.030.0
      • Carpenter L.L.
      • Janicak P.G.
      • Aaronson S.T.
      • et al.
      Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice.
      ; Expert opinion
      The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.
        Number of maintenance sessions4.0Triangular3.05.0Expert opinion
      The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.
        Cost per session$150.00Gamma$120.00$180.00Assumption ± 20%
       Antidepressant
        Months per course of treatment3.0Triangular2.06.0Expert opinion
      The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.
        Cost per month$17.27Gamma$13.82$20.72Calculation ± 20%
       ECT
        Number of sessions10.0Triangular6.012.0ECT literature
        Cost per session$814.00Gamma$651.20$976.80AR-DRG v.6 ± 20%
       Augmentation
        Cost per course treatment$235.19Gamma204.1359.0Expert opinion
      The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.
      and calculation
       Hospitalization
        Average cost per hospitalization$14,021Gamma$13,106$20,484AR-DRG v.6
       Adverse events
        Antidepressant$80.95Gamma$64.76$97.14Calculation ± 20%
        rTMS$81.79Gamma$65.43$98.15Calculation ± 20%
        ECT$72.53Gamma$58.03$87.04Calculation ± 20%
      AR-DRG, Australian refined diagnosis-related groups; AUD, Australian dollar; ECT, electroconvulsive therapy; HAM-D, Hamilton Depression Rating Scale; REM, remission; RESP, response; rTMS, repetitive transcranial magnetic stimulation.
      low asterisk The expert was a university professor in psychiatry and a practicing physician, and had no conflicts of interest to declare.

       Probabilities

      The treatment effects (probabilities of gaining response or remission) for rTMS are extensively reported in the literature. The range is wide, however, due to trial design and sample size variations. Meta-analyses also report varying response and remission rates, depending on the comparators, treatment frequency (low vs. high), and frontal side (left, right, or bifrontal) [
      • Berlim M.T.
      • Van den Eynde F.
      • Daskalakis Z.J.
      Efficacy and acceptability of high frequency repetitive transcranial magnetic stimulation (rTMS) versus electroconvulsive therapy (ECT) for major depression: a systematic review and meta-analysis of randomized trials.
      ,
      • Berlim M.T.
      • Van den Eynde F.
      • Jeff Daskalakis Z.
      Clinically meaningful efficacy and acceptability of low-frequency repetitive transcranial magnetic stimulation (rTMS) for treating primary major depression: a meta-analysis of randomized, double-blind and sham-controlled trials.
      ,
      • Berlim M.T.
      • van den Eynde F.
      • Tovar-Perdomo S.
      • et al.
      Response, remission and drop-out rates following high-frequency repetitive transcranial magnetic stimulation (rTMS) for treating major depression: a systematic review and meta-analysis of randomized, double-blind and sham-controlled trials.
      ,
      • Ren J.
      • Li H.
      • Palaniyappan L.
      • et al.
      Repetitive transcranial magnetic stimulation versus electroconvulsive therapy for major depression: a systematic review and meta-analysis.
      ]. To calculate the pooled estimate of the treatment effect for all relevant studies, meta-analyses were performed using the random-effect inverse variance–weighted method for binary outcomes (see File 1 in Supplemental Materials found at doi: 10.1016/j.jval.2015.04.004). The response and remission rates for rTMS were estimated as 37.5% and 21.5%, respectively. The efficacy outcome for antidepressant medication was derived from the STAR*D trial [
      • Rush A.J.
      STAR*D: what have we learned?.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
      ]. This study reported major outcomes (remission, response, and adverse effects) for different patient groups including patients who failed two adequate antidepressant courses in their current illness episode. The reported response and remission rates for antidepressant medications were 16.8% and 13.6%, respectively.
      The probabilities of regressing after a period of remission for both rTMS and antidepressant medications are not reported in the literature. Studies, however, have reported information on worsening and relapse rates for rTMS [
      • Janicak P.G.
      • Nahas Z.
      • Lisanby S.H.
      • et al.
      Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month, multisite, open-label study.
      ,
      • Mantovani A.
      • Pavlicova M.
      • Avery D.
      • et al.
      Long-term efficacy of repeated daily prefrontal transcranial magnetic stimulation (TMS) in treatment-resistant depression.
      ], antidepressants [
      • Rush A.J.
      STAR*D: what have we learned?.
      ], and ECT treatment [
      • Sackeim H.A.
      • Haskett R.F.
      • Mulsant B.H.
      • et al.
      Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial.
      ]. These rates were converted to probabilities of losing remission (regressing). Treatment efficacy tends to decrease if patients develop resistance [
      • Simpson K.N.
      • Welch M.J.
      • Kozel F.A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
      ,
      • Sinyor M.
      • Schaffer A.
      • Levitt A.
      The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
      ,
      • O’Reardon J.P.
      • Solvason H.B.
      • Janicak P.G.
      • et al.
      Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial.
      ]; however, the decrement rate is not reported quantitatively in the literature. The rates of efficacy decrement (for each subsequent treatment) were therefore assumed to be 20% and 15% of remission and response rates, respectively. It was also assumed that 75% of the patients who failed either rTMS or (acute) antidepressants would start augmentation medication. In Australia, lithium is indicated (and approved by the Pharmaceutical Benefit Schedule) to augment antidepressants for patients with MDD. Off-label use, however, includes atypical antipsychotic drugs such as aripiprazole, quetiapine, and olanzapine. The model used lithium augmentation as the base case, and tested a mix of augmentation therapies in the sensitivity analysis. Last, mortality risk was assumed to be higher for patients in acute depression or in mild/moderate depression than in the general population.

       Resources and Costs

      Costs related to all health care resource items for the economic model are summarized in Table 1. All costs were converted to monthly values to accommodate the 2-monthly cycle calculation. For simplicity, cost per treatment course (for both rTMS and antidepressant) was assumed to occur within one cycle. Psychiatric consultation for treatment and a management plan incurred a cost for each treatment course. Subsequent psychiatric consultations and short visits were part of regular MDD monitoring.
      Each rTMS session was estimated to cost approximately AU $150 covering the professional component and practice components. Each acute rTMS course consists of an average number of 28 sessions [
      • Carpenter L.L.
      • Janicak P.G.
      • Aaronson S.T.
      • et al.
      Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice.
      ] and lasts for a period of 4 to 6 weeks. Patients who respond positively to treatment (i.e., observed reduction in HAM-D17 score) will proceed to rTMS maintenance until potential relapse or dropout. For rTMS maintenance, the average number of sessions was estimated to be 26 per annum, equivalent to an average of two rTMS sessions per month.
      Each antidepressant (acute) treatment course was recommended for at least 3 months. The monthly cost was calculated as the weighted average of most commonly used antidepressant drugs prescribed by Australian doctors (Table 1) [
      • Stephenson C.P.
      • Karanges E.
      • McGregor I.S.
      Trends in the utilisation of psychotropic medications in Australia from 2000 to 2011.
      ]. This resulted in a monthly cost of AU $17.30, equivalent to AU $52 per 3-monthly (acute) treatment course. If a patient’s condition improved to full remission, further medication was not required unless the patient had a relapse later.
      For each treatment option, patients who failed two consecutive courses would move to either augmentation or ECT. Augmentation agents included lithium and atypical antipsychotic agents (e.g., quetiapine, aripiprazole, and olanzapine). The total cost for augmentation included at least 2-month supply of the medication, regular monitoring tests, and one psychiatrist consultation in addition to standard antidepressant medications. The ECT treatment cost covered 10 sessions with one psychiatric visit. The costs for hospitalizations and individual adverse events were identified by Sullivan et al. [
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ] and valued using Australian national cost schedules (Table 1).

       Health State Utility Values

      To calculate QALYs, patient utility values were assigned to each health state in the model. Fourteen published studies were identified from a systematic review of clinical trials and economic evaluations of MDD treatment (limited to publications after January 2000). Hawthorne et al.’s [
      • Hawthorne G.
      • Cheok F.
      • Goldney R.
      • et al.
      The excess cost of depression in South Australia: a population-based study.
      ] estimates were used for the model base case because they reported the utility weights for Australians but still aligned with values reported in the wider literature. There is limited information on the disutility from adverse events associated with each treatment. The most relevant study for this topic is Sullivan et al. [
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ] on the cost-effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions in the United States. For the economic model, the weighted averages of the disutilities of adverse events relevant to each treatment were calculated, with the utility values taken directly from Sullivan et al. [
      • Sullivan P.W.
      • Valuck R.
      • Saseen J.
      • et al.
      A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
      ].

       Analyses

      The main outcomes of the model were costs and quality-adjusted life-years (QALYs), both discounted at 5% to reflect time preferences. For the base case, microsimulations were performed with 50,000 trials to achieve stable results. Mean costs and QALYs for each treatment arm were produced to calculate the incremental cost-effectiveness ratio (ICER). Univariate and multivariate probabilistic sensitivity analyses were undertaken for key variables. The 95% confidence intervals or the high and low values, where available, were used to reflect wide variation in the base values. Beta distributions were assigned for utilities and probabilities, whereas gamma distributions were assigned for costs. All sensitivity analyses were performed with 50,000 trials for result stability and consistency with the base case. An upper threshold of AU $50,000 per QALY was used to indicate cost-effective results.

      Results

      Estimates of cost, effect, and ICER for the base case (3-year horizon) are presented in Table 2 and on the cost-effectiveness plane in Figure 1. The model predicted that QALYs gained with rTMS were higher than QALYs gained with pharmacotherapy (1.25 vs. 1.18 QALYs) while costs were slightly less (AU $31,003 vs. AU $31,190). Therefore, in the base case, the rTMS option was considered the superior alternative compared with pharmacotherapy for the treatment of treatment-resistant patients with MDD. At a threshold of AU $50,000 per QALY gain, the probability that rTMS was dominant was 32% and the probability that rTMS was cost-effective compared with antidepressants was 41%.
      Table 2Costs, effects, cost-effectiveness ratios, and net monetary benefit (2013–2014 AUD).
      Mean values3 y (base case)5 y (sensitivity analysis)
      AntidepressantrTMSAntidepressantrTMS
      Total cost$31,190$31,003$41,009$39,693
       Incremental total cost–$187–$1,316
      Total QALYs1.181.251.531.63
       Incremental total QALYS0.070.10
      Cost/QALY$26,432$24,803$26,803$24,352
      Incremental cost per QALYDominantDominant
      AUD, Australian dollar; QALY, quality-adjusted life-year; rTMS, repetitive transcranial magnetic stimulation.
      Figure thumbnail gr1
      Fig. 1Microsimulation results in the Base case (3 year horizon with 50,000 trials.) QALYs, quality-adjusted life-years.
      Table 3 presents the effect of parameter changes on the ICER. The analytical model was very robust with respect to all parameters. Univariate sensitivity analyses identified the most influential variables in the model: the probabilities of gaining and losing remission after antidepressant treatment; the probability of losing remission without rTMS maintenance; the probability of losing remission after treatment; and the doses and costs per session of rTMS and ECT. The model was not sensitive to the utility values, and for most probability and cost variables rTMS was a dominant alternative to antidepressant medication. The multivariate analyses showed that the model results were stable to variations in model values, with the likelihood of rTMS being dominant or cost-effective compared with antidepressants exceeding 70%.
      Table 3Summary of sensitivity analysis results (2013–2014 AUD).
      VariablesAntidepressantsrTMSCost-effectiveness results of rTMS vs. antidepressants
      Cost ($)QALYCost ($)QALY
      Utilities
       Univariate28,921–28,9471.19–1.2328,434–28,4661.23–1.30Dominant = 100%
       Multivariate (all utility values)28,9111.1628,4311.24Dominant = 97%; ICER < 50,000 = 3%
      Transition probabilities
       Gaining remission after treated with antidepressant29,1451.1928,5181.27ICER > 50,000 = 10%; Dominated = 1%
       Losing remission without rTMS maintenance28,9231.1928,4831.27ICER > 50,000 = 7%
       Losing remission after treated with antidepressant28,5771.2028,4411.27ICER > 50,000 = 13%; Dominated = 7%
       Other transition probabilities (univariate)28,833–29,4001.1928,388–28,9511.26–1.27Dominant = 60%–100%; ICER < 50,000 = 0%–40%
       Multivariate (all probabilities)31,1031.1830,4231.25Dominant = 57%; ICER < 50,000 = 34%; Dominated = 9%
      Costs
       rTMS dose for acute treatment28,9171.1928,4181.27Dominant = 98%; ICER < 50,000 = 2%
       rTMS cost per session28,9371.1928,4911.27Dominant = 67%; ICER < 50,000 = 33%
       ECT dose28,8761.1928,4001.27Dominant = 97%; ICER < 50,000 = 3%
       ECT cost per session28,9451.1928,4571.27Dominant = 89%; ICER < 50,000 11%
       Augmentation (including lithium, atypical antipsychotic drugs)28,9761.1928,4891.27Dominant 100%
       Other cost variables28,919–28,9451.1928,437–28,4591.27Dominant = 100%
       Multivariate (all cost variables)28,9151.1928,4521.27Dominant = 71%; ICER < 50,000 = 29%
      All variables (1,000 simulations with 50,000 trials each)30,5281.2030,0711.27Dominant = 56%; ICER < 50,000 = 15%; ICER > 50,000 = 17%; Dominated = 12%
      AUD, Australian dollar; ECT, electroconvulsive therapy; ICER, incremental cost-effectiveness ratio; rTMS, repetitive transcranial magnetic stimulation.
      Additional sensitivity analyses were also performed on the discount rates and model duration. When a longer time horizon was applied (5 instead of 3 years), the average cost saving increased to $1316 and the average QALY gain per patient was 0.10. This implies increasing cost-effectiveness in the medium term for rTMS treatment versus standard pharmacotherapies. The use of different discount rates had little impact on this overall conclusion. At a 3% discount rate, rTMS was a superior strategy, less costly, and more effective, compared with the antidepressant medication. At a 7% discount rate, the ICER was AU $127 per QALY gained, well below a willingness-to-pay threshold of AU $50,000.

      Discussion

      A cost-effective treatment for treatment-resistant patients with depression is an important challenge today because MDD is a chronic and debilitating disease that significantly decreases one’s quality of life, and is a leading cause of disability worldwide [
      • Mrazek D.A.
      • Hornberger J.C.
      • Altar C.A.
      • et al.
      A review of the clinical, economic, and societal burden of treatment-resistant depression: 1996–2013.
      ,
      • Greden J.F.
      The burden of recurrent depression: causes, consequences, and future prospects.
      ]. rTMS has been received as a clinically effective and safe option for treatment-resistant patients with MDD [
      • Carpenter L.L.
      • Janicak P.G.
      • Aaronson S.T.
      • et al.
      Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice.
      ,
      • Daskalakis Z.J.
      • Levinson A.J.
      • Fitzgerald P.B.
      Repetitive transcranial magnetic stimulation for major depressive disorder: a review.
      ,
      • Janicak P.G.
      • Nahas Z.
      • Lisanby S.H.
      • et al.
      Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month, multisite, open-label study.
      ,
      • Wall C.A.
      • Croarkin P.E.
      • McClintock S.M.
      • et al.
      Neurocognitive effects of repetitive transcranial magnetic stimulation in adolescents with major depressive disorder.
      ,
      • Janicak P.G.
      • Dunner D.L.
      • Aaronson S.T.
      • et al.
      Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of quality of life outcome measures in clinical practice.
      ]. In this economic evaluation, we have shown that rTMS is a cost-effective alternative to pharmacotherapy over 3 years. In general, patients with treatment-resistant MDD gain slightly more quality of life at a lower cost when treated with rTMS than with pharmacotherapies. The results of the analyses suggest that there is a low probability of antidepressant medications being cost-effective compared with rTMS at a willingness-to-pay level of AU $50,000 per QALY gain.
      To date, rTMS therapy appears to have a high degree of safety and acceptability among patients and clinicians [
      • Janicak P.G.
      • O’Reardon J.P.
      • Sampson S.M.
      • et al.
      Transcranial magnetic stimulation in the treatment of major depressive disorder: a comprehensive summary of safety experience from acute exposure, extended exposure, and during reintroduction treatment.
      ]. The magnetic pulse produces an audible high-frequency clicking sound, and ear protection (earplugs) is used during rTMS treatments. Common adverse events observed with rTMS are mild to moderate posttreatment headache and mild pain or discomfort at the treatment area. The most significant medical risk associated with the use of rTMS therapy is the inadvertent induction of a seizure. No seizures, however, were reported in the clinical research trials of the NeuroStar rTMS Therapy System [
      • Janicak P.G.
      • O’Reardon J.P.
      • Sampson S.M.
      • et al.
      Transcranial magnetic stimulation in the treatment of major depressive disorder: a comprehensive summary of safety experience from acute exposure, extended exposure, and during reintroduction treatment.
      ,
      • George M.S.
      • Lisanby S.H.
      • Avery D.
      • et al.
      Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial.
      ]. In postmarket use, the prevalence of seizure with the NeuroStar rTMS Therapy System, under recommended operating conditions, is estimated to be less than 0.1% per patient and lower than what is typically seen with routine antidepressant medications. There has also been no evidence of emergent suicidal ideation during acute treatment with the NeuroStar rTMS Therapy System.
      In the time of fiscal challenge and rising burden of mental illness, this cost-effectiveness evidence is timely and useful for both policymakers and service providers in resource allocation. Incorporating rTMS into the standard treatment algorithm for MDD expands the choice set available for patients, especially for those who develop intolerance to and/or fail pharmacotherapies. rTMS can also replace ECT in a subgroup of patients without psychotic symptoms or acute risks who are traditionally referred for ECT. As a substitution therapy, rTMS will be a cost-saving device for government budgets if more treatment-resistant patients with MDD switch from antidepressant medication and/or ECT to rTMS. That is, subsidizing rTMS is potentially an efficiency-improvement strategy for the health system.
      The findings from our model are consistent with conclusions from the Simpson et al. [
      • Simpson K.N.
      • Welch M.J.
      • Kozel F.A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
      ] study despite significant differences in modeling approaches. Simpson et al. used a Markov cohort model to compare rTMS with sham and pharmacotherapies under open-label conditions and for patients who were exposed to at least one but no more than four antidepressant medications. Based on the data derived from the published STAR*D study [
      • Rush A.J.
      STAR*D: what have we learned?.
      ] and on a multicentre randomized controlled trial [
      • O’Reardon J.P.
      • Solvason H.B.
      • Janicak P.G.
      • et al.
      Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial.
      ], the study found that rTMS provided a net cost saving of US $1123 per QALY gained compared with the current antidepressant medication therapies [
      • Simpson K.N.
      • Welch M.J.
      • Kozel F.A.
      • et al.
      Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
      ]. Although our model relies on the same STAR*D study for the clinical efficacy of pharmacotherapies (in patients with TRD), the efficacy for rTMS was derived from a large meta-analysis rather than one single study. This might better reflect the effectiveness of rTMS in practice. In addition, Simpson et al.’s model did not take into account the subsequent and rescue treatments (i.e., ECT, augmentation, and hospitalization) when patients failed either rTMS or standard pharmacotherapies. Our model closely mimics this pathway and thus better reflects the treatment algorithm for MDD in “real” practice.
      We did not consider psychotherapies in this model because of a lack of clinical evidence of rTMS versus psychotherapy (both efficacy and safety). Psychotherapies for patients with TRD currently do not show definitive benefits to warrant their use as a comparator for these difficult-to-treat patients. In addition, no clinical studies have directly compared the efficacy of rTMS treatment to any form of psychotherapy. There are also no high-quality studies comparing psychotherapies to pharmacotherapy or rTMS or placebo that are relevant for this article, that is, those who failed to respond to two previous medication treatments. Although an assessment of effectiveness or safety between psychotherapy and rTMS is not possible within the scope of this research, this option cannot be ruled out as being potentially beneficial to some patients.
      Despite using the best available evidence to use in the model, a number of assumptions were necessary. First, we have made various assumptions in the decrements of efficacy for subsequent courses of the same treatment. We also calculated the probabilities of losing remission from relapse rates (for the respective treatments) under an assumption that the rate of losing remission is constant over time. Second, utility weights were all sourced from the literature on patients with depression and may be different to the modeled population (i.e. treatment-resistant patients). Third, some cost estimates relied on expert advice, which might not fully represent those of all clinicians in current practice. The sensitivity analyses, however, showed that the effect of the above assumptions was small, and, if present, there was approximately 10% chance that antidepressants were a cost-effective treatment compared with rTMS. Finally, our analyses took an Australian perspective with the use of Australian-specific costs, utilities, and background mortality. Generalizability to other countries may be in question, but we believe that the relativities of the unit costs for the different treatments would likely be similar across jurisdictions.

      Conclusions

      The study shows that rTMS is a cost-effective treatment alternative for patients with MDD who have failed at least two adequate courses of antidepressant medications. This result supports providers in deciding to subsidize rTMS to increase the diversity of treatment options. This finding also has wider implications in relation of improving cost efficiency within the health system.

      Acknowledgments

      The study arises from a health technology assessment performed by authors independently for the Australian Government for advice to the Medical Services Advisory Committee. The authors are part of the Assessment Group at Griffith University contracted to the Australian Government. The publication of study results is not contingent on any third-party approval or censorship of the manuscript.

      Supplementary Materials

      References

        • Lepine J.P.
        • Briley M.
        The increasing burden of depression.
        Neuropsychiatr Dis Treat. 2011; 7: 3-7
        • Mrazek D.A.
        • Hornberger J.C.
        • Altar C.A.
        • et al.
        A review of the clinical, economic, and societal burden of treatment-resistant depression: 1996–2013.
        Psychiatr Serv. 2014; 65: 977-987
        • Greden J.F.
        The burden of recurrent depression: causes, consequences, and future prospects.
        J Clin Psychiatry. 2001; 62: 5-9
        • Carpenter L.L.
        • Janicak P.G.
        • Aaronson S.T.
        • et al.
        Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice.
        Depress Anxiety. 29. 2012: 587-596
        • Daskalakis Z.J.
        • Levinson A.J.
        • Fitzgerald P.B.
        Repetitive transcranial magnetic stimulation for major depressive disorder: a review.
        Can J Psychiatry. 2008; 53: 555-566
        • Janicak P.G.
        • Nahas Z.
        • Lisanby S.H.
        • et al.
        Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month, multisite, open-label study.
        Brain Stimul. 2010; 3: 187-199
        • Wall C.A.
        • Croarkin P.E.
        • McClintock S.M.
        • et al.
        Neurocognitive effects of repetitive transcranial magnetic stimulation in adolescents with major depressive disorder.
        Front Psychiatry. 2013; 4: 165
        • Fitzgerald P.
        Repetitive transcranial magnetic stimulation and electroconvulsive therapy: complementary or competitive therapeutic options in depression?.
        Australas Psychiatry. 2004; 12: 234-238
        • Simpson K.N.
        • Welch M.J.
        • Kozel F.A.
        • et al.
        Cost-effectiveness of transcranial magnetic stimulation in the treatment of major depression: a health economics analysis.
        Adv Ther. 2009; 26: 346-368
        • Knapp M.
        • Romeo R.
        • Mogg A.
        • et al.
        Cost-effectiveness of transcranial magnetic stimulation vs. electroconvulsive therapy for severe depression: a multi-centre randomised controlled trial.
        J Affect Disord. 2008; 109: 273-285
        • Kozel F.A.
        • George M.S.
        • Simpson K.N.
        Decision analysis of the cost-effectiveness of repetitive transcranial magnetic stimulation versus electroconvulsive therapy for treatment of nonpsychotic severe depression.
        CNS Spectr. 2004; 9: 476-482
        • Stephenson C.P.
        • Karanges E.
        • McGregor I.S.
        Trends in the utilisation of psychotropic medications in Australia from 2000 to 2011.
        Aust N Z J Psychiatry. 2013; 47: 74-87
        • Berlim M.T.
        • Van den Eynde F.
        • Daskalakis Z.J.
        Efficacy and acceptability of high frequency repetitive transcranial magnetic stimulation (rTMS) versus electroconvulsive therapy (ECT) for major depression: a systematic review and meta-analysis of randomized trials.
        Depress Anxiety. 2013; 30: 614-623
        • Berlim M.T.
        • Van den Eynde F.
        • Jeff Daskalakis Z.
        Clinically meaningful efficacy and acceptability of low-frequency repetitive transcranial magnetic stimulation (rTMS) for treating primary major depression: a meta-analysis of randomized, double-blind and sham-controlled trials.
        Neuropsychopharmacology. 2013; 38: 543-551
        • Berlim M.T.
        • van den Eynde F.
        • Tovar-Perdomo S.
        • et al.
        Response, remission and drop-out rates following high-frequency repetitive transcranial magnetic stimulation (rTMS) for treating major depression: a systematic review and meta-analysis of randomized, double-blind and sham-controlled trials.
        Psychol Med. 2014; 44: 225-239
        • Ren J.
        • Li H.
        • Palaniyappan L.
        • et al.
        Repetitive transcranial magnetic stimulation versus electroconvulsive therapy for major depression: a systematic review and meta-analysis.
        Prog Neuropsychopharmacol Biol Psychiatry. 2014; 51: 181-189
        • Rush A.J.
        STAR*D: what have we learned?.
        Am J Psychiatry. 2007; 164: 201-204
        • Sinyor M.
        • Schaffer A.
        • Levitt A.
        The sequenced treatment alternatives to relieve depression (STAR*D) trial: a review.
        Can J Psychiatry. 2010; 55: 126-135
        • Mantovani A.
        • Pavlicova M.
        • Avery D.
        • et al.
        Long-term efficacy of repeated daily prefrontal transcranial magnetic stimulation (TMS) in treatment-resistant depression.
        Depress Anxiety. 2012; 29: 883-890
        • Sackeim H.A.
        • Haskett R.F.
        • Mulsant B.H.
        • et al.
        Continuation pharmacotherapy in the prevention of relapse following electroconvulsive therapy: a randomized controlled trial.
        JAMA. 2001; 285: 1299-1307
        • O’Reardon J.P.
        • Solvason H.B.
        • Janicak P.G.
        • et al.
        Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial.
        Biol Psychiatry. 2007; 62: 1208-1216
        • Sullivan P.W.
        • Valuck R.
        • Saseen J.
        • et al.
        A comparison of the direct costs and cost effectiveness of serotonin reuptake inhibitors and associated adverse drug reactions.
        CNS Drugs. 2004; 18: 911-932
        • Hawthorne G.
        • Cheok F.
        • Goldney R.
        • et al.
        The excess cost of depression in South Australia: a population-based study.
        Aust N Z J Psychiatry. 2003; 37: 362-373
        • Janicak P.G.
        • Dunner D.L.
        • Aaronson S.T.
        • et al.
        Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of quality of life outcome measures in clinical practice.
        CNS Spectr. 2013; 18: 322-332
        • Janicak P.G.
        • O’Reardon J.P.
        • Sampson S.M.
        • et al.
        Transcranial magnetic stimulation in the treatment of major depressive disorder: a comprehensive summary of safety experience from acute exposure, extended exposure, and during reintroduction treatment.
        J Clin Psychiatry. 2008; 69: 222-232
        • George M.S.
        • Lisanby S.H.
        • Avery D.
        • et al.
        Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial.
        Arch Gen Psychiatry. 2010; 67: 507-516