This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.
Obsessive-compulsive disorder (OCD) is a common and chronic mental illness with a high rate of disability. Internet-based cognitive behavioral therapy (ICBT) makes online treatment available to patients and has been shown to be effective. However, 3-arm trials on ICBT, face-to-face cognitive behavioral group therapy (CBGT), and only medication are still lacking.
This study is a randomized, controlled, assessor-blinded trial of 3 groups for OCD: ICBT combined with medication, CBGT combined with medication, and conventional medical treatment (ie, treatment as usual [TAU]). The study aims to investigate the efficacy and cost-effectiveness of ICBT related to CBGT and TAU for adults with OCD in China.
In total, 99 patients with OCD were selected and randomly assigned to the ICBT, CBGT, and TAU groups for treatment for 6 weeks. The primary outcomes were the Yale-Brown Obsessive-Compulsive Scale (YBOCS) and the self-rating Florida Obsessive-Compulsive Inventory (FOCI), compared at baseline, during treatment (3 weeks), and after treatment (6 weeks), to analyze efficacy. The secondary outcome was the EuroQol Visual Analogue Scale (EQ-VAS) scores of the EuroQol 5D Questionnaire (EQ-5D). The cost questionnaires were recorded to analyze cost-effectiveness.
Repeated-measures ANOVA was used for data analysis, and the final effective sample size was 93 (ICBT: n=32, 34.4%; CBGT: n=28, 30.1%; TAU: n=33, 35.5%). After 6-week treatment, the YBOCS scores of the 3 groups significantly decreased (
Therapist-guided ICBT combined with medication is as effective as face-to-face CBGT combined with medication for OCD. ICBT combined with medication is more cost-effective than CBGT combined with medication and conventional medical treatment. It is expected to become an efficacious and economic alternative for adults with OCD when face-to-face CBGT is not available.
Chinese Clinical Trial Registry ChiCTR1900023840; https://www.chictr.org.cn/showproj.html?proj=39294
Obsessive-compulsive disorder (OCD) is a common and chronic mental illness with a high rate of disability [
Internet-based cognitive behavioral therapy (ICBT) makes online treatment available to patients. ICBT is divided into therapist-guided ICBT and self-guided ICBT. ICBT has the advantages of convenience and time saving, which can solve the problem of uneven distribution of medical resources and make up for the limitations of conventional CBT. The results of meta-analyses have shown that therapist-guided ICBT is as effective as face-to-face CBT (ffCBT) for various mental health conditions [
ICBT for OCD has been generally proven to be effective. A randomized controlled trial (RCT) in 2013 showed that ICBT and the iCBT program in book format (bCBT) were more effective than the waiting treatment group (waitlist) in improving the obsessive-compulsive and depressive symptoms of patients with OCD [
At present, although the number of international studies on the treatment of OCD with ICBT is gradually increasing, most RCTs conducted only set a waitlist or traditional medicine group to compare with ICBT and do not use ffCBT as the control group. In most cases, these waitlist controls are not restricted to take part in any psychotherapy or physiotherapy other than ffCBT, which introduces irrelevant variables. In addition, many current studies on ICBT for OCD are affected by comorbidities and heterogeneous medications. The latest ICBT research on OCD mainly focuses on children and adolescents, and there is a lack of research on adults. Up to now, there is no research on ICBT for OCD and its health economics analysis of OCD in China. Given these deficiencies, the efficacy and costs of ICBT for OCD in China deserve further research. Considering feasibility and cost-effectiveness, cognitive behavioral group therapy (CBGT) was set as the ffCBT active control group in this study. In a meta-analysis of CBT for OCD, individual CBT showed small and nonsignificant effect sizes (0.17) compared to CBGT and showed a noninferior efficacy of CBGT to individual CBT [
The study conducted an RCT to compare the feasibility, safety, efficacy, and cost-effectiveness of ICBT, CBGT, and TAU in patients with OCD. TAU involved the administration of psychotropic medications as usual. Both ICBT and CBGT were combined with medical treatment, and patients in the ICBT or the CBGT group also took psychotropic medications as usual. The study used therapist-guided ICBT because a meta-analysis showed that therapist-guided ICBT has better effects than self-guided ICBT [
The study aimed to investigate the efficacy and cost-effectiveness of ICBT related to CBGT and TAU for adults with OCD in China. The study was a noninferiority trial and hypothesized that the efficacy of therapist-guided ICBT is no less than that of face-to-face CBGT and medical treatment for OCD. In addition, ICBT was expected to be more cost-effective due to lower costs of treatment.
The study was a 6-week, assessor-blinded, clinical RCT with patients with OCD allocated 1:1:1 to the ICBT, CBGT, and TAU groups. Primary and secondary outcomes were measured 3 times: at baseline (0 weeks), during treatment (3 weeks), and after treatment (6 weeks). The study was carried out at the Shanghai Mental Health Center in Shanghai, China, and was registered in the Clinical Trial Registry (registration no. ChiCTR1900023840). Recruitment and intervention for the trial started in 2018.
The study adhered to the Consolidated Standards of Reporting Trials (CONSORT) and was approved by the Shanghai Ethical Review Committee (2018-57).
The study subjects were 99 patients with OCD who were consecutively enrolled from a public psychiatric hospital. Information about the trial was sent to clinicians and advertised on posters in the clinic lobby and the online WeChat public account. Participation was referred by psychiatrists or via self-referral, and both were diagnosed as OCD by psychiatrists at the Shanghai Mental Health Center, China. After that, participants were asked to complete an online screening and a brief phone interview. Suitable participants were invited to take part in a face-to-face psychiatric assessment conducted by trained evaluators on duty to determine inclusion or exclusion. The evaluator used the Mini-International Neuropsychiatric Interview (MINI) [
The inclusion criteria were(1) age between 18 and 54 years, (2) being satisfied with the diagnostic criteria for OCD in the
All patients in the 3 groups had taken medication before the start of the experiment, 84.9% (n=79) of the patients had taken medication stably for more than 8 weeks, most of the remaining patients had taken medication stably for less than 8 weeks, and a few patients had discontinued medication before enrollment. The types of medications were SSRIs, and the types/doses were not changed during the study.
Patients in the ICBT group used the Cognitive Behavioral Therapy China (CBTC) platform linked to the CBTC website [
CBTCa platform intervention modules.
Module | Training program | Homework | Number of training sessions (N=12), n (%) |
1. Understanding OCDb and subjective discomfort units | Psychological education of OCD | To understand the Subjective Units of Distress Scale (SUDS), symptom monitoring | 2 (16.7) |
2. Training methods and exposure checklist | Psychological education on ERPc | To create an exposure registration form | 2 (16.7) |
3. Reward list and exposure design | Introduction to exposure design | To create a reward list, establish a code of practice for ritual prevention, design for exposure | 2 (16.7) |
4. Exposure practice | Exposure content adjustment, use of exposure sheets for instruction | Exposure practice | 5 (41.7) |
5. Relapse prevention | Psychological education | To rebuild new rules | 1 (8.3) |
aCBTC: Cognitive Behavioral Therapy China.
bOCD: obsessive-compulsive disorder.
cERP: exposure and response prevention.
The therapist conducted a pretreatment interview with each patient before and trained them all on how to use the CBTC platform for practice. A senior domestic OCD and ICBT therapist supervised the treatment. The patients used their accounts and passwords to log on to the platform and receive treatment for the module of facing OCD. There were in total 12 treatments, which lasted 6 weeks, about 120 minutes per treatment, twice a week. After each module treatment was completed, the platform arranged corresponding tasks. Before the next treatment, the patients filled in the completion status of the homework so that the therapist could obtain the severity of their symptoms in real time. The therapist provided feedback to the patients through the platform during the entire treatment process and answered questions on time, but the therapist allocated no more than 15 minutes to each patient per week.
Patients participating in ICBT needed to log on to the platform at least twice a week to complete their studies and the homework after each module. If a patient failed to log on to the platform for study or to complete the homework on time, the online therapist would send a short message to remind the patient. If the patient failed to log on to the platform or failed to complete the homework 3 times, the platform would automatically cancel the account and the patient would not be able to log on to the platform for treatment. For patients who missed a certain treatment for some reason or faced difficulties, the researcher would remind them to complete this treatment in time via phone and email.
The CBGT group was headed by a nationally registered therapist who had received systematic training. The group was closed and homogeneous, and all face-to-face sessions were audio-recorded to ensure that therapists adhered to the treatment manual. Each subgroup included 6-8 patients, who were treated twice a week for a total of 12 treatments, each 120 minutes long, which lasted 6 weeks. A professional group cognitive therapist supervised this group during treatment. The treatment location was the group psychotherapy room of the Shanghai Mental Health Center. The treatment room was not changed during the treatment period.
The structure, content, and number of treatments followed the
For handling other special situations, patients who were absent from a certain treatment for some reason were informed of the homework via telephone and email, and those who had difficulty completing the homework could listen to the recording with the researcher for the retrospective study. Patients were deemed to have dropped out of the study if they were absent 3 times in total.
The TAU group was treated with only medication. The medication interventions of the 3 groups were under the charge of an associate chief psychiatrist, who was unaware of the grouping of patients. The medicines used for the treatment of OCD in this study were SSRIs approved by the State Food and Drug Administration (SFDA). The medication treatment lasted for 6 weeks. Patients with sleep disorders could use benzodiazepines in combination, but benzodiazepines should not be taken continuously for more than 2 weeks, and other psychotropic medicines were not used in combination. The medicines used in this study were commonly used in the clinic, which had good safety and fewer side effects. The evaluations of patients’ side effects and adverse information by an outpatient doctor were collected regularly. If patients failed to comply with the agreement and stopped the medicines by themselves, they were deemed to have withdrawn from the study.
The demographic data of the 3 groups were collected during the enrollment assessment. Obsessive-compulsive symptoms, the quality of life, and costs were collected 3 times: at baseline, during treatment (3 weeks), and after treatment (6 weeks). All assessments at baseline were offline, and subsequent assessments in the ICBT and TAU groups were conducted online. In addition, the study collected process indexes and recorded adverse events to assess feasibility and safety. The process indexes included the attendance rate, dropout rate, homework completion, and subjective satisfaction of patients.
Obsessive-compulsive symptoms were measured using a masked assessor–rated YBOCS [
The quality of life was measured using the EQ-VAS score of the EQ-5D [
The study designed the Sinicized Cost Questionnaire based on Trimbos and the Institute of Medical Technology Assessment Cost Questionnaire for Psychiatry [
Direct costs included registration, psychological treatment, medication, and rehabilitation. Indirect costs included transportation expenses, accommodation expenses, food expenses, medical and health care expenses due to illness, and the loss of productivity for patients and their families. To avoid confusion regarding the inability to work/study due to illness with unemployment/dropout, only the time spent in the hospital for consultation and treatment was counted as the number of days the patient was unable to work/study.
The cost-effectiveness analysis (CEA) method was used to measure the spending of the 3 groups. The results of the total cost divided by the reduction in the YBOCS total score reflected the cost required to obtain each unit of curative effect.
The statistical method suitable for this study was 2-way (1 within-subject and 1 between-subject factor) repeated-measures ANOVA. According to the theoretical considerations [
and the minimum sample size was nl=n2=n3=23. Considering the dropout rate and clinical operability, the TAU, ICBT, and CBGT groups should each have had 33 patients, so finally, 99 patients meeting the criteria were included.
Random sequences were generated in Microsoft Excel with the RAND function, and all eligible patients were randomized 1:1:1 to the ICBT, CBGT, and TAU groups. The evaluators were blind to the patients’ treatment. At the same time, patients were asked not to mention the psychotherapeutic conditions to the evaluator during assessment. The arrangement of the time and location of the assessments for the 99 patients was the responsibility of the coordinator, who did not disclose the grouping of patients.
Statistical data were analyzed with SPSS 25.0 (IBM Corporation). All outcome analyses were conducted according to the intention-to-treat principle, and the missing data were dealt with using conditional mean imputation [
The study repeated the analysis for the primary outcome using 2 different approaches as sensitivity analyses to assess the robustness of our conclusions. The first approach was complete case analysis. In this situation, only cases that completed all the follow-up measures (n=80, 86.0%) were analyzed, while cases with missing values were removed. The second approach was analysis with the addition of covariates. In this case, demographic variables, such as age, years of education, age at first onset, and total duration of the disease, were included as covariates in the repeated-measures ANOVA.
In cost analyses, the total cost was the sum of direct and indirect costs. The total direct/indirect cost of each group was obtained by adding the direct/indirect cost provided by each group. To quantify the cost of lost productivity in the indirect cost, the study averaged China's 2018 annual per capita income (renminbi [RMB] 27,996 [US $4235.40]) for 365 days (RMB 76.70 [US $11.60]) and then multiplied it by the number of patient working days lost. The mean (95% CI) and SD of costs were described. The cost-effectiveness ratio (CER) for each group was the result of the total cost divided by the total reduction in the YBOCS score, which was the amount of money spent for each reduction in the YBOCS score. The incremental cost-effectiveness ratio (ICER) took the cost of the group with the least expenditure as the standard to compare the additional expenditure of the other groups per reduction in the YBOCS score. The yuan (RMB) is the national currency of the People’s Republic of China, and the average exchange rate was US $1.00=RMB 6.61 for 2018.
A total of 93 subjects (n=56, 60.2%, men; n=37, 39.8%, women) were included in the analysis. Participant flow and reasons for dropout throughout the trial are shown in
Participant flow in a study of the effect of ICBT vs CBGT vs TAU on OCD symptoms in adults. CBGT: cognitive behavioral group therapy; ICBT: internet-based cognitive behavioral therapy; TAU: treatment as usual; OCD: obsessive-compulsive disorder.
Demographic data of the 3 groups at baseline (N=93).
Characteristics | CBGTa group (n=28) | ICBTb group (n=32) | TAUc group (n=33) | |
|
||||
|
Male | 15 (53.6) | 18 (56.3) | 23 (69.7) |
|
Female | 13 (46.4) | 14 (43.7) | 10 (30.3) |
|
||||
|
Unmarried | 14 (50.0) | 20 (62.5) | 22 (66.7) |
|
Married | 14 (50.0) | 12 (37.5) | 11 (33.0) |
|
||||
|
Mean (SD) | 30.29 (8.09) | 29.16 (6.35) | 27.52 (7.73) |
|
Min-max (years) | 18-48 | 19-45 | 18-45 |
Years of education, mean (SD) | 13.82 (2.79) | 13.81 (3.31) | 14.55 (2.41) | |
|
||||
|
Psychiatric medications | 28 (100) | 32 (100) | 33 (100) |
|
Traditional Chinese medicine | 0 | 0 | 1 (3.0) |
|
Psychological treatment | 0 | 0 | 1 (3.0) |
Duration of SSRId treatment>8 weeks, n (%) | 26 (92.9) | 24 (75.0) | 29 (87.9) | |
|
||||
|
Acute (within a month) | 1 (3.6) | 2 (6.3) | 6 (18.2) |
|
Subacute | 1 (3.6) | 5 (15.6) | 5 (15.2) |
|
Chronic (>3 months) | 26 (92.9) | 25 (78.1) | 22 (66.7) |
Total duration of the disease, mean (SD) | 7.71 (6.71) | 6.06 (6.18) | 8.96 (6.65) | |
Age at first onset, mean (SD) | 22.54 (6.61) | 23.78 (7.25) | 18.85 (6.53) | |
|
||||
|
Persistent | 23 (82.1) | 28 (87.5) | 16 (48.5) |
|
Intermittent | 5 (17.9) | 4 (12.5) | 17 (51.5) |
Family history, n (%) | 3 (10.7) | 1 (3.1) | 8 (24) |
aCBGT: cognitive behavior group therapy.
bICBT: internet-based cognitive behavior therapy.
cTAU: treatment as usual (conventional medical treatment).
dSSRI: selective serotonin reuptake inhibitor.
In the ICBT group, 7 (21.2%) of 33 patients dropped out (n=3, 42.9%, men): 1 (14.3%) patient lost contact before the start of treatment, 3 (42.9%) patients discontinued the medication because their condition improved, and 3 (42.9%) patients withdrew because they could not receive online treatment. In the CBGT group, 7 (21.2%) of 33 patients dropped out (n=4, 57.1%, men): 5 (71.4%) patients dropped out because they could not wait until the treatment group was assembled, 1 (14.3%) patient withdrew because they discontinued medication themselves, and 1 (14.3%) female patient dropped out due to the lack of female patients in the group. In the TAU group, 5 (15.2%) patients dropped out (n=3, 60.0%, men) because they discontinued medication themselves. There was no statistically significant difference in the dropout rate among the 3 groups (
In conclusion, a total of 19 (19.2%) of 99 patients dropped out from the 3 groups during treatment. Of them, 6 (31.6%) patients did not participate in the treatment (n=5, 83.3%, in the CBGT group; n=1, 16.7%, in the ICBT group) and had no baseline data. The remaining 13 (68.4%) patients dropped out in the middle of treatment (n=2, 15.4%, in the CBGT group; n=6, 46.2%, in the ICBT group; and n=5, 38.4%, in the TAU group). According to the intention-to-treat principle, we replaced the of these 13 (68.4%) patients with the mean imputation, and the final effective data of 93 patients (ICBT: n=32, 34.4%; CBGT: n=28, 30.1%; TAU: n=33, 35.5%) were considered.
The results of the 7-level satisfaction scale evaluation of the 3 groups were as follows:
ICBT group: 3-week mean 5.73 (SD 0.96); 6-week mean 5.77 (SD 0.82)
CBGT group: 3-week mean 5.19 (SD 0.94); 6-week mean 5.30 (SD 0.97)
TAU group: 3-week mean 4.86 (SD 0.76); 6-week mean 4.67 (SD 0.81)
During the treatment (3 weeks), the satisfaction degree of the TAU (
Among the 12 treatments, there was no significant difference between the ICBT (18.59%) and CBGT (13.14%) groups in the absence rate (
This study recorded 2 adverse events: 1 (3.1%) patient in the ICBT group expressed anxiety and vertigo symptoms during the reading text of module training and finally terminated the treatment and withdrew from the study, while 1 (3.6%) female patient in the CBGT group felt uncomfortable and withdrew from the study due to the lack of female members in the group. No serious adverse events occurred in this study.
At baseline, there were no statistically significant differences in the YBOCS score (
There was a significant group×time interaction effect on the YBOCS score (
Mean and SD of the YOBCSa and FOCIb scores at baseline, during treatment (3 weeks), and after treatment (6 weeks) for the 3 groups.
Measure | Baseline, mean (SD) | During treatment, mean (SD) | After treatment, mean (SD) | ||
|
|||||
|
ICBTc group | 22.06 (4.48) | 16.50 (3.77) | 13.28 (5.33) | |
|
CBGTd group | 23.00 (4.31) | 19.14 (4.52) | 13.18 (4.99) | |
|
TAUe group | 21.82 (3.06) | 18.06 (4.61) | 16.00 (4.96) | |
|
|||||
|
ICBT group | 11.13 (3.11) | 9.00 (2.92) | 8.50 (2.77) | |
|
CBGT group | 11.64 (3.31) | 10.29 (2.00) | 9.25 (2.91) | |
|
TAU group | 11.55 (3.24) | 11.45 (2.28) | 11.03 (2.39) |
aYBOCS: Yale-Brown Obsessive-Compulsive Scale.
bFOCI: Florida Obsessive-Compulsive Inventory.
cICBT: internet-based cognitive behavioral therapy.
dCBGT: cognitive behavioral group therapy.
eTAU: treatment as usual.
Results of repeated-measures ANOVAs and effect sizes for the primary efficacy measure at baseline, during treatment (3 weeks), and after treatment (6 weeks) for the 3 groups.
Measure | Difference at 3 weeks, mean (95% CI), |
Difference at 6 weeks, mean (95% CI), |
||||
|
|
Within group | Between group | Within group | Between group | Group×time interaction |
|
||||||
|
ICBTb group | –5.56 (–7.33 to –3.79), <.001 | N/Ac | –8.78 (–10.95 to –6.61), <.001 | N/A | .001 |
|
ICBT group vs CBGTd group | N/A | –2.64 (–5.36 to 0.08), .06 | N/A | 0.10 (–3.12 to 3.32), |
N/A |
|
ICBT group vs TAUe group | N/A | –1.56 (–4.17 to 1.05), .444 | N/A | –2.72 (–5.80 to 0.37), .103 | N/A |
|
CBGT group | –3.86 (–5.75 to –1.97), <.001 | N/A | –9.82 (–12.14 to –7.50), <.001 | N/A | N/A |
|
CBGT group vs TAU | N/A | 1.08 (–1.62 to 3.78), .993 | N/A | –2.82 (–6.02 to 0.37), .102 | N/A |
|
TAU group | –3.76 (–5.50 to –2.02), <.001 | N/A | –5.82 (–7.96 to –3.68), <.001 | N/A | N/A |
|
||||||
|
ICBT group | –2.13 (–3.39 to –0.86), <.001 | N/A | –2.63 (–3.99 to –1.26), <.001 | N/A | .02 |
|
ICBT group vs CBGT group | N/A | –1.29 (–2.83 to 0.26), .135 | N/A | –0.75 (–2.45 to 0.95), .851 | N/A |
|
ICBT group vs TAU group | N/A | –2.46 (–3.93 to –0.98), <.001 | N/A | –2.53 (–4.16 to –0.90), .001 | N/A |
|
CBGT group | –1.36 (–2.71 to –0.01), <.001 | N/A | –2.39 (–3.85 to –0.93), <.001 | N/A | N/A |
|
CBGT group vs TAU group | N/A | –1.17 (–2.70 to 0.36), .198 | N/A | –1.78 (–3.47 to –0.10), .04 | N/A |
|
TAU group | –0.09 (–1.34 to 1.16), .99 | N/A | –0.52 (–1.86 to 0.83), .99 | N/A | N/A |
aYBOCS: Yale-Brown Obsessive-Compulsive Scale.
bICBT: internet-based cognitive behavioral therapy.
cN/A: not applicable.
dCBGT: cognitive behavioral group therapy
eTAU: treatment as usual.
fFOCI: Florida Obsessive-Compulsive Inventory.
Primary outcome of YBOCS in a study of the effect of ICBT vs CBGT vs TAU on OCD symptoms in adults. CBGT: cognitive behavioral group therapy; ICBT: internet-based cognitive behavioral therapy; OCD: obsessive-compulsive disorder; TAU: treatment as usual; YBOCS: Yale-Brown obsessive-compulsive scale.
Primary outcome of FOCI in a study of the effect of ICBT vs CBGT vs TAU on OCD symptoms in adults. CBGT: cognitive behavioral group therapy; FOCI: Florida obsessive-compulsive inventory; ICBT: internet-based cognitive behavioral therapy; OCD: obsessive-compulsive disorder; TAU: treatment as usual.
Treatment response and remission under 3 interventions after 6 weeks of treatment.
Time | ICBTa group (n=32), n (%) | CBGTb group (n=28), n (%) | TAUc group (n=33), n (%) | ||||||
|
|||||||||
|
3 weeks | 10 (31) | 3 (11) | 4 (12) | .06 | ||||
|
6 weeks | 20 (63) | 22 (79) | 11 (33) | .001 | ||||
|
|||||||||
|
3 weeks | 0 | 0 | 1 (3) | .40 | ||||
|
6 weeks | 4 (13) | 5 (18) | 1 (3) | .16 |
aICBT: internet-based cognitive behavioral therapy.
bCBGT: cognitive behavioral group therapy.
cTAU: treatment as usual.
d
eYBOCS score reduction rate≥35.0% after treatment was defined as the treatment response.
fAfter treatment, YBOCS score<8 was classified as remission.
The VAS scores in the EQ-5D scale for the 3 groups had no significant differences at baseline (
Secondary outcome of VAS in a study of the effect of ICBT vs CBGT vs TAU on OCD symptoms in adults. CBGT: cognitive behavioral group therapy; ICBT: internet-based cognitive behavioral therapy; OCD: obsessive-compulsive disorder; TAU: treatment as usual; VAS: visual analogue scale.
After 6-week treatment, the total cost per person in the CBGT group was RMB 6678.45, 95% CI 4460.88-8896.01 [US $1010.36, 95% CI 678.87-1345.84]), that in the ICBT group was RMB 3308.81, 95% CI 2476.89-4140.73 [US $500.58, 95% CI 374.72-626.43]), and that in the TAU group was RMB 2259.61, 95% CI 2074.16-2445.05 [US $341.85, 95% CI 313.79-369.90]). There was a significant difference in the total cost between the 3 groups (
Costs (RMBa/US $b) of the 3 groups after 6-week treatment.
Type of cost | CBGTc group (n=28) | ICBTd group (n=32) | TAUe group (n=33) | Between-group comparisonf | |||||||
|
Total cost | Mean (95% CI) | Total cost | Mean (95% CI) | Total cost | Mean (95% CI) | |||||
Total cost | 186,997.00/28,290.02 | 6678.45 (4460.88-8896.01)/1010.36 (678.87-1345.84) | 105,882.00/16,018.46 | 3308.81 (2476.89-4140.73)/500.58 (374.72-626.43) | 74,567.00/11,280.94 | 2259.61 (2074.16-2445.05)/341.85 (313.79-369.90) | 13.45 | <.001 | |||
Direct cost | 124,844.00/18,887.14 | 4458.70 (3639.37-5278.03)/674.54 (550.59-798.49) | 73,701.00/11,149.92 | 2303.16 (1717.54-2888.77)/348.44 (259.84-437.03) | 50,246.00/7601.51 | 1522.61 (1382.19-1663.03)/230.35 (209.11-251.59) | 29.80 | <.001 | |||
Indirect cost | 62,153.00/9,402.87 | 2219.75 (673.77-3765.83)/335.82 (101.93-569.72) | 32,181.00/4868.53 | 1005.66 (635.49-1375.73)/152.14 (96.14-208.13) | 24,321.00/3679.43 | 737.00 (632.77-841.06)/111.50 (95.73-127.24) | 3.57 | .03 |
aRMB: renminbi.
bThe average exchange rate was US $1.00=RMB 6.61 for 2018.
cCBGT: cognitive behavioral group therapy.
dICBT: internet-based cognitive behavioral therapy.
eTAU: treatment as usual.
fThe
After 6-week treatment, it cost RMB 376.80 (US $57.00) for the ICBT group to reduce the YBOCS score by 1, RMB 388.37 (US $58.75) for the TAU group, and RMB 679.99 (US $102.87) for the CBGT group. The ICBT group spent RMB 303.19 (US $45.97) less than the CBGT group and RMB 11.57 (US $1.75) less than the TAU group. See
Cost-effectiveness analysis (RMBa/US $b) of patients after 6-week treatment.
Group | Total cost (RMB/US $) | YBOCSc average score | YBOCS total reduction score | CERd (RMB/US $) | ICERe (RMB/US $) | |
|
Baseline | After treatment |
|
|||
CBGTf | 186,996.50/28,289.94 | 23.00 | 13.18 | 275.00 | 679.99/102.87 | 303.19/45.97 |
TAUg | 74,567.00/11,280.94 | 21.82 | 16.00 | 192.00 | 388.37/58.75 | 11.57/1.75 |
ICBTh | 105,882.00/16,018.46 | 22.06 | 13.28 | 281.00 | 376.80/57.00 | 0.00/0.00 |
aRMB: renminbi.
bThe average exchange rate was US $1.00=RMB 6.61 for 2018.
cYBOCS: Yale-Brown Obsessive-Compulsive Scale.
dCER: cost-effectiveness ratio.
eICER: incremental cost-effectiveness ratio.
fCBGT: cognitive behavioral group therapy.
gTAU: treatment as usual.
hICBT: internet-based cognitive behavioral therapy.
This study did not include the time of therapists for cost calculation but collected the total time of the therapists occupied by the 3 groups of interventions. The results showed that patients in the ICBT group consumed a total of 1078 minutes of the therapist’s time during the 6-week treatment, with an average of 32.67 (SD 13.37) minutes per person; the CBGT group consumed a total of 2880 minutes of the therapist’s time during the 6-week treatment, with an average of 102.86 (SD 0) minutes per person. Compared to the CBGT group, the ICBT group saved a total of 1802 minutes of the therapist’s time, with an average of 70.19 (SD 13.37) minutes saved per person.
The study performed complete case analysis (n=80, 86.0%) and analysis with the addition of covariates, and the results of the 2 sensitivity analyses were consistent with those of the primary analyses, suggesting the robustness of our conclusions.
This study was an RCT of ICBT, CBGT, and conventional medical treatment for OCD in China, which aimed to evaluate the efficacy and cost-effectiveness of therapist-guided ICBT for Chinese patients with OCD. All 3 groups had curative effects and improved quality of life after treatment for 6 weeks, and the ICBT group was more cost-effective than the other groups. The satisfaction of patients was high, and there were no serious adverse events.
In terms of efficacy, the significant efficacy of SSRIs was consistent with the results shown in earlier studies [
A difference between the results of the YBOCS and FOCI was found in this study. After treatment (6 weeks), the FOCI score of the ICBT and CBGT groups was significantly lower than that of the TAU group, while this difference was not detected in the YBOCS. It was probably because the YBOCS is an assessor-rated scale, while FOCI is a self-rating scale. Patients may have more acute observations of the changes in their own obsessive-compulsive symptoms and experience. This might be more sensitively reflected in the FOCI score after ICBT/CBGT rather than the blind assessor–rated YBOCS. In addition, the satisfaction degree of treatment in the ICBT/CBGT group was higher than that in the TAU group, which might also affect the self-rating scores.
With regard to feasibility, the results showed that the TAU group had the lowest dropout rate, and the ICBT group had the same dropout rate as the CBGT group. Half of the patients dropped out in the ICBT group due to discomfort with reading text on an electronic screen. Completing therapy and homework by reading and typing on the keyboard might be unfriendly to some people, especially older patients. A patient in the ICBT group withdrew from the study because of anxiety and vertigo symptoms during training and reading. It was difficult for ICBT when the patient was not accustomed to using the computer, lacked motivation for treatment, or had too severe symptoms. Patient compliance did correlate with treatment modality, and some patients who dropped out reported that they were not used to online therapy. A previous paper indicated that a single human-computer interaction might lack face-to-face treatment, resulting in decreased compliance of patients during treatment [
Health economics results showed that ICBT is significantly less expensive than CBGT and was the most cost-effective option of the 3. The cost savings of the ICBT group might be achieved by less transportation expenses, accommodation expenses, food expenses, and productivity loss because patients in the ICBT group could choose their own ERP practice time at home without extra transportation, accommodation, and work absence compared to the CBGT group. From health organizational perspectives (ie, therapist time) due to the online technology and the therapist-guided self-help mode, compared to the CBGT group, the ICBT group saved time for therapists, with an average of 76.43 minutes per person, which may help solve the human resource shortage in mental health services in China. Nowadays, the burden on mental health in various countries is increasing. For example, the United States spent more than US $300 billion on mental health each year [
This study has a few limitations. First, the dose and duration of patients’ SSRI treatment before enrollment were not clearly recorded, which were variables that could be included in the analysis. In addition, 8 weeks should be used as a standard for stable medication to screen subjects to reduce the impact of medication duration on the trial. Furthermore, the mean imputation may underestimate the variability of the data, although the sensitivity analysis of complete cases (n=80) showed consistent results with the primary analyses. In terms of health economics, it would be more comprehensive to include the cost of treatment space and therapists’ time in the cost analyses in the future. For example, CBGT requires renting treatment rooms and takes more time of therapists, which may lead to more costs for CBGT.
Based on the results of the equivalent efficacy of ICBT and CBGT, researchers and physicians can provide ICBT as a treatment option for patients in clinical practice, try to enhance the interactivity of ICBT courses or develop a combination of online and offline treatment methods, and gradually carry out some in-depth research in the future. For example, researchers should develop and validate the feasibility and efficacy of combined online and offline treatments. In addition, researchers can try to explore ICBT based on individualized assessment and make better use of the advantages of the online form through a step care model. Multicenter studies need to be carried out in the future. In addition, some studies have focused on the long-term efficacy of ICBT [
Second, this study only focused on therapist-guided ICBT and did not explore self-guided ICBT. The latest research has verified the feasibility of self-guided ICBT (unguided ICBT) [
Therapist-guided ICBT combined with medication is as effective as face-to-face CBGT combined with medication for OCD. ICBT combined with medication is more cost-effective than CBGT combined with medication and conventional medical treatment. It is expected to become an efficacious and economic alternative for adults with OCD when face-to-face CBGT is not available.
CONSORT-eHEALTH checklist (V 1.6.1).
cognitive behavioral group therapy
cognitive behavioral therapy
Cognitive Behavioral Therapy China
cost-effectiveness ratio
EuroQol 5D Questionnaire
EuroQol Visual Analogue Scale
exposure and response prevention
face-to-face CBT
Florida Obsessive-Compulsive Inventory
internet-based cognitive behavioral therapy
incremental cost-effectiveness ratio
least significant difference
obsessive-compulsive disorder
renminbi
randomized controlled trial
selective serotonin reuptake inhibitor
treatment as usual
Visual Analogue Scale
Yale-Brown Obsessive-Compulsive Scale
This work was supported by the National Natural Science Foundation of China (82230045, 81771460), the General project of Shanghai Municipal Health Commission (202140054), the Academic Leader of Health Discipline of Shanghai Municipal Health Commission (2022XD025), the Shanghai Jiaotong University "star of Jiaotong University" medical engineering cross research fund key project (YG2021ZD28), the Shanghai Science and Technology Committee Project (20DZ2253800), and the Key Laboratory of Psychotic Disorders (13dz2260500).
None declared.