A Randomized Controlled Trial of Video Psychoeducation for Electroconvulsive Therapy in the United States

Electroconvulsive therapy (ECT) is a robust, effective, and safe treatment for severe psychiatric conditions, including major depression, mania, and schizophrenia (1, 2). In contemporary clinical trials, ECT has demonstrated response rates in the range of 60% to 80% and remission rates of 50% to 60% among patients with treatment-resistant depression. These rates are substantially better than rates observed with psychopharmacology (3–5). Despite its demonstrated effectiveness and advancements in procedures, ECT is utilized for less than 0.5% of individuals with major depression in the United States, with population utilization rates decreasing over time (6–8).

There may be several reasons for the underutilization of ECT. For one, patients, family members, mental health professionals, and the public often have negative perceptions and inaccurate knowledge about the treatment (9–11). Stigma surrounding ECT may be a major barrier to its acceptance by the public (12); fear and lack of knowledge might be a barrier to ECT use among psychiatric patients (13, 14). The media have also often perpetuated stigma and fear related to ECT through inaccurate portrayals of the procedure in movies (15, 16). A comprehensive review found that American films have increasingly portrayed ECT as a brutal, harmful procedure with little or no therapeutic benefit, despite robust evidence to the contrary (17). One study found that support for ECT among medical students decreased further after the students watched movie scenes involving the procedure (18).

Thus it is important that the health care field take responsibility for educating patients and their families about ECT. Providers who have clinical experience and accurate knowledge of ECT tend to have more positive attitudes toward the treatment (19, 20). As such, various tools have been developed to better inform patients, caregivers, health care professionals, and the general public about the benefits of ECT through educational videos, audio recordings, printed pamphlets, workshops, and training programs. Collectively, these educational tools have been associated with improved perceptions and knowledge about ECT (21–26).

Only a few studies have compared the efficacy of different psychoeducation tools in promoting knowledge of ECT. One study conducted in the 1980s found that patients assigned to watch a videotape about ECT did not feel that the videotape provided more information than the usual consent process (27). A study conducted in the early 2000s found that students who watched an educational video or read a pamphlet about ECT had improved perceptions and knowledge compared with a control group, but neither the video nor the pamphlet was superior to the other (22). With major developments in the creation and dissemination of online videos in the past decade and research demonstrating the positive effects of visual aids in informing patients about medical treatments (28, 29), more research is needed.

In this study, we conducted a randomized controlled trial comparing the effectiveness of two forms of ECT psychoeducation (video and printed materials) in a national sample of U.S. adults who screened positive for depression. Our main outcomes of interest were changes in perceptions and knowledge about ECT and willingness to accept ECT. We hypothesized that both interventions would lead to increased positive perceptions and accurate knowledge about ECT and willingness to receive it. We further hypothesized that video psychoeducation, which offers audiovisual stimuli, would lead to greater positive changes than a printed brochure with no illustrations.

Methods

A national sample of participants was recruited via Amazon Mechanical Turk (MTurk) in July 2019. MTurk is an online platform designed to allow individuals (known as requesters) to pay other individuals (known as workers/participants) for completing small online tasks known as human intelligent tasks (HITs) (30). Requesters can assign many participants to work on the same HIT and can restrict each participant’s ability to complete the HIT more than once (31).

MTurk has been used widely by social science researchers (32) to recruit participants for experimental (33–36) and observational research (37). The quality of data obtained through MTurk has been found to be commensurate with that collected through traditional methods (e.g., samples of undergraduates) (37).

Participants were invited to participate in a study to help us “understand attitudes, opinions, and knowledge about treatment for mental disorders.” After participants provided informed consent, they completed a brief screener to determine study eligibility. Eligible participants were randomly assigned to receive either video psychoeducation or an informational brochure about ECT. Random assignment was conducted through Qualtrics by assigning every other participant to one of two interventions. After random assignment, participants completed a baseline survey before the intervention was delivered online, followed by a postsurvey. Participants were compensated $1 for completing the HIT, consistent with compensation for similar HITs on MTurk, and were restricted to completing the HIT only once. All study procedures were approved by the institutional review board at Yale University School of Medicine.

Participants

Participants were asked to complete a brief screener to determine study eligibility before they were enrolled in the study. The screener asked participants about their age and current country of residence and included the nine-item Patient Health Questionnaire (PHQ-9), a widely used screening instrument for depression. Eligibility criteria for the study were age 18 or over, English speaking, currently living in the United States and a positive screen for current depression (score of ≥8 on the PHQ-9) (38, 39). In addition, only participants who had an approval rate of greater than 90% on HITs and who had completed more than 100 approved HITs were allowed to participate. This safeguard was taken to ensure high-quality participation (35, 40).

A total of 865 participants initially agreed to participate in the survey, but 271 participants did not meet the eligibility criteria and an additional 38 participants either did not complete the intervention or the follow-up survey or failed data validity checks. The final analytic sample consisted of 556 participants. (A CONSORT diagram of the recruitment and randomization process in available in the online supplement for this article.)

Psychoeducation Video Versus Brochure

Participants were randomly assigned either to watch an 8-minute psychoeducation video about ECT or to read a four-page online brochure that presented identical information. The psychoeducation video was created by the first author and was made available to the public online on YouTube (41). The video was designed to provide psychoeducation about ECT and information about research on its effectiveness and to correct common misconceptions and myths about ECT. The video featured members of the public, subject matter experts, and testimonials of patients discussing their perceptions and knowledge about ECT. Closed captions were included in the video. The informational brochure was a direct transcript of the video, and no images were presented. Thus the psychoeducation video and brochure contained the same content and differed only in communication format.

Measures

Background characteristics of participants—including age, gender, sexual orientation, education level, marital status, employment status, annual income, geographic region of residence, and any active duty service in the U.S. military—were collected through self-report.

Participants were asked to complete the PHQ-9 (as a screener) and to self-report any mental or substance use disorders from a list of conditions. Participants were also asked whether they had ever been homeless (“e.g., living in a shelter, in a car, on the streets, in an abandoned building, couch surfing”) and whether they had ever attempted suicide in their lifetime.

The ECT-PK scale was developed on the basis of a systematic review (42) and through expert consensus and psychometric testing (43). The ECT-PK scale consists of two subscales: a six-item perception subscale and a 12-item knowledge subscale. For the perception subscale, participants are asked to rate their level of agreement with six statements on a 5-point Likert scale, from 1 (strongly disagree) to 5 (strongly agree). A total perception subscale score was calculated as the mean response to the items (after recoding certain items), with higher scores indicating more positive perceptions of ECT. On the knowledge subscale, participants were asked to respond dichotomously (yes/no) to 12 statements about ECT based on whether they thought the statement was true or false. A total knowledge subscale score was calculated as the total number of correct items, indicating greater knowledge about ECT. Both ECT-PK subscales have good construct validity, criterion validity, and internal consistency reliability (43).

In addition to completing the ECT-PK scale, participants were asked to respond (yes/no) to an additional question about their willingness to receive ECT (“If I were depressed, I’d be willing to try ECT”).

Data Analyses

The intervention and control groups were compared on background characteristics with bivariate tests, including independent t tests and chi-square tests. The groups were also compared on changes in ECT-PK subscale and item scores before and after the intervention by using a two-way mixed analysis of variance, with time as the within-subject factor and psychoeducation group as the between-subjects factor. Assumptions of normality and homogeneity of variance were assessed by Q-Q plots and Levene’s test of homogeneity of variance, and no violations of assumptions were found. For individual knowledge items that had categorical responses, generalized estimating equations were used to examine changes between groups in pre-post ECT knowledge. Effect sizes were calculated by using Cohen’s d to compare changes between the two groups in scores on the ECT-PK before and after the intervention

Multiple regression analyses were conducted to examine background characteristics associated with greater change in ECT-PK subscale scores. Change scores were calculated as pre-post differences in perception and knowledge subscale scores and were designated as dependent variables. The regression analyses controlled for group assignment, and standardized beta coefficients and total R² were calculated. Binary logistic regressions were conducted to identify background characteristics associated with a change in willingness to receive ECT after the intervention (video or brochure). Odds ratios with 95% confidence intervals were calculated as effect sizes, and a total Nagelkerke R² value was calculated reflecting the total variance explained. Finally, bivariate correlations were conducted to investigate associations between changes in perceptions, knowledge, and willingness to receive ECT. SPSS, version 23.0, was used for all analyses, and the significance level was set at 0.05.

Results

As shown in Table 1, there were no significant differences in background characteristics between participants in the video and brochure groups, which supports the conclusion that randomization was successful. A majority of participants were heterosexual, white females in their twenties and thirties who had at least a college degree and who were working full-time. Participants were from diverse income levels and regions of the country. All participants screened positive for depression, but many also reported that they had been diagnosed as having a range of other mental disorders, most prominently anxiety disorder, posttraumatic stress disorder, alcohol use disorder, and bipolar disorder.

Table 2 shows that there were significant increases among both groups in positive perceptions of ECT on all perception items and in the total perception subscale score. In both groups, changes in perception scores were generally associated with moderate to large effect sizes (d=0.21–1.19 among items in the video group and d=0.44–1.15 among items in the brochure group). There were no significant differences between groups in change in perception scores. Group × time interaction effects were significant for two items (“concerns about the possibility of memory loss” and “frightened by ECT”). The brochure group had a significantly greater reduction than the video group in concerns about memory loss related to ECT, whereas the video group had a significantly greater reduction than the brochure group in fear about ECT.

Table 3 shows that there were also significant improvements in knowledge of ECT among both groups on nearly all knowledge items and on the total knowledge subscale score. There were exceptions on two items. On one item, both groups showed no significant change in knowledge that muscle relaxants are administered during ECT. On the other item, both groups showed a significant decline in knowledge about whether ECT can induce a seizure. For all other items, there were moderate or large improvements in knowledge in both the video group (14% to 56% change) and the brochure group (12% to 56% change). There was a large effect size increase in total knowledge subscale scores in both groups (d=1.47 and 1.25 for the video group and brochure group, respectively). Comparing the two psychoeducation groups, the video group showed a greater increase in knowledge than the brochure group about whether ECT is an outdated treatment and whether ECT is used for control or punishment. There were no other significant group × time interaction effects on knowledge items.

Both the video and brochure groups showed significant and large increases in willingness to be treated with ECT (Wald χ²=114.67, df=1, p<0.001). In the video group, 85 (31%) participants reported being willing to receive ECT at preintervention compared with 175 (63%) participants at postintervention. In the brochure group, 80 (29%) reported being willing to receive ECT at preintervention compared with 155 (56%) participants at postintervention. However, there was no significant difference between groups in change in willingness to use ECT, and the group × time interaction effect was not significant, suggesting that change in willingness to receive ECT was similar for both groups.

As shown in Table 4, regression analyses with the total sample identified a few background characteristics associated with changes in ECT-PK subscale scores and willingness to receive ECT. Being female, not having bipolar disorder, having a drug use disorder, and higher PHQ-9 scores were significantly associated with greater increases in perception subscale scores. Being male and having a history of suicide attempts were significantly associated with a greater likelihood of being more willing to receive ECT. No background characteristics were significantly associated with changes in knowledge subscale scores.

Correlational analyses revealed that changes in perception subscale scores were significantly associated with changes in knowledge subscale scores (r=0.33, p<0.001) and changes in willingness to receive ECT (r=0.33, p<0.001). Changes in knowledge subscale scores were also significantly associated with changes in willingness to receive ECT (r=0.19, p<0.001).

Discussion

In this randomized trial comparing participants who received video psychoeducation or an informational brochure about ECT, there were several noteworthy findings. First, many participants had negative perceptions about ECT at baseline, particularly fears that ECT causes memory loss and brain damage, and that it is painful. The majority of participants also showed limited knowledge and understanding that ECT is a modern, safe, and beneficial treatment for several psychiatric disorders (1, 2). More specifically, many participants did not understand that ECT is provided with anesthesia. These findings are perhaps not surprising in the context of 4 decades of stigma, negative perceptions, inaccurate knowledge, and actual abuses with ECT use (12–14, 17). However, since then, there have been advances in ECT procedures, treatment of mental disorders has become more humane in general, and patients have enjoyed greater autonomy to select their own treatment (7, 44, 45).

Second, both the video and brochure groups showed moderate to large increases in positive perceptions and knowledge about ECT. That is important because it suggests that adults who screen positive for depression are receptive to brief psychoeducation about ECT and that both video and written information are effective in improving ECT perceptions and knowledge. These findings are consistent with a small study (22) that compared effects of a video and pamphlet intervention on ECT perceptions and knowledge among psychology students. The study found greater improvements in perception and knowledge about ECT among participants in both interventions compared with a control group, but neither active intervention was superior to the other.

Although we did not find that the video was more effective than the informational brochure in changing overall perceptions and knowledge about ECT, there were significant group differences on some specific domains. Compared with the brochure group, the video group showed significantly greater reductions in fears about ECT and greater knowledge that ECT is a modern procedure used for treatment and not for control or punishment. At the same time, the brochure group showed significantly greater reductions than the video group in concerns about memory loss related to ECT. Thus, whereas both formats were effective, video psychoeducation may have some advantages in allaying fears about ECT and demonstrating the modernity of ECT through presentation of images, patient testimonials, and credible commentary from subject matter experts. However, these findings need to be replicated with samples of real patients to ensure that the findings are not a statistical artifact. It is also worth considering that some information, such as ECT’s effects on memory loss, may be best communicated to patients through written materials, much as prescription labels report potential side effects of medications.

Third, there was a dramatic increase in reported willingness to receive ECT regardless of type of psychoeducation. In fact, the number of participants who expressed willingness to receive ECT more than doubled in both groups (from 31% to 63% in the video group compared with from 29% to 56% in the brochure group). This finding was quite remarkable and suggested that brief informational interventions such as those offered in this study can effectively increase engagement and acceptability of ECT. In addition, we found that changes in perceptions and knowledge about ECT and willingness to receive ECT were all correlated, so improving one domain may affect others. It is reasonable to deduce that improving perceptions and knowledge may increase willingness to receive ECT. Therefore, providing brief psychoeducation for a broad, targeted group who may benefit from ECT may be an inexpensive and effective population-based approach to increasing use of this powerful treatment option.

There were several study limitations. This study, including both assessments and delivery of the interventions, was conducted entirely online. The generalizability of the results to real-world, in-person conditions is thus unknown, although psychoeducation about ECT is increasingly offered online (46–48). However, the participants may not be representative of patients, and further study is needed in clinical settings. Perceptions and knowledge about ECT and willingness to receive ECT were assessed, but whether they translate to actual behaviors and utilization of ECT remains unknown and also needs to be studied. These limitations notwithstanding, the study had several strengths, including its randomized controlled design, large sample size, and inclusion of only patients who screened positive for depression.

Conclusions

ECT remains a maligned, underutilized treatment, and psychoeducation may be important in improving access to those who may benefit from it. Brief psychoeducation, in both video or written format, can markedly improve perception and knowledge about ECT and willingness to receive ECT. Further research is needed on when and how best to engage individuals in ECT psychoeducation and whether changes after ECT psychoeducation are long-lasting and influence treatment decisions.