Abstract
Background:
Neural circuitry-based treatments, such as transcranial direct current stimulation (tDCS), have demonstrated efficacy in reducing craving in individuals with alcohol use and other addictive substances.
Aim:
The study aimed to investigate the effectiveness of tDCS on craving, time taken to first drink, and relapse to drinking over 3 months among individuals with severe alcohol use disorder.
Methods:
A randomized sham-controlled trial included adults aged 18–55 years with severe alcohol dependence. Participants (n = 149) were abstinent from alcohol for at least 3 days, underwent a benzodiazepine washout, and exhibited active craving. tDCS was administered twice daily for 5 consecutive days, with bilateral stimulation being given by placing the anode over F3 and the cathode over F4 to the ‘active’ (A) and ‘sham’ (S) intervention groups. Clinical parameters were assessed at baseline, 1 month (1 m), and 3 months (3 m).
Results:
At completion, out of the 149 randomized subjects (n (A) =75, n (S) =74), 107 participants (n (A) =51, n (S) =56) received the intended tDCS sessions. Baseline characteristics were comparable between the two groups. Intention-to-treat analysis showed significantly lower craving scores in group A than in group S at 1 month and 3 month follow-up time points in comparison to the baseline (baseline: A = 48.33 ± 1.94, S = 48.27 ± 2.45; 1 m: A = 30.37 ± 11.66, S = 33.55 ± 13.73; 3 m: A = 28.50 ± 13.23, S = 34.75 ± 14.07; F (2,294) = 5.52, P < 0.01). Intervention group A also exhibited fewer relapses [3 m A = 33 (44%), 3 m S = 47 (63.5%); χ2 (1) = 5.70, P = 0.01] and a longer time to first drink compared to S (A = 38.50 ± 27.0 days; S = 29.40 ± 23.83 days; t = 2.20, P = 0.03).
Conclusion:
Adjunctive tDCS demonstrated efficacy in reducing craving and preventing relapse in individuals with severe alcohol dependence. These findings suggest the potential of tDCS as a therapeutic intervention for severe alcohol dependence which is less intense in terms of resources and time and can further be tailored to monitor neurobiological correlates in recovery.
Keywords: Craving, neuromodulation, relapse, severe alcohol dependence, transcranial direct current stimulation, tDCS
INTRODUCTION
Alcohol use disorder (AUD) is a prevalent, chronic brain condition with a relapsing pattern, constituting about 5% of the global disease burden and displaying a high relapse rate.[1] In India, around 14.6% of individuals indulge in alcohol use, with 5.7 crore facing problematic alcohol consumption and 2.9 crore grappling with dependence.[2] Those dependent on alcohol experience a reduction in life expectancy by up to 20 years, and alcohol is implicated in over 200 diseases’ development and progression.[3] The disability-adjusted life years (DALYs) due to AUD in India are 180.26 per 100,000 people.[4]
Existing therapies for AUD, predominantly cognitive-behavioral interventions with or without medication, exhibit limited efficacy, with medications proving ineffective for many patients. A lack of comprehension regarding the causes and mechanisms of substance use disorders impedes the development of more effective treatments. Consequently, there is an urgent need to explore innovative and potentially more effective therapies and research approaches. Even medications with substantial support fail to yield significant improvements in outcomes for many patients in clinical trials.[5]
The most researched type of transcranial electrical stimulation (tES), known as transcranial direct current stimulation (tDCS), involves administering a mild electrical current directly to the scalp via anodal and cathodal electrodes. Anodal tDCS heightens cortical excitability, while cathodal tDCS diminishes it. Animal models have demonstrated its efficacy in decreasing craving and relapse.[6] This method enables the study of neuroelectric characteristics and cognitive processing in the human brain, with potential therapeutic applications due to its modulatory effects.[7]
tDCS stands out among noninvasive brain stimulation techniques due to its cost-effectiveness, portability, potential for home use, ease of application, and minimal adverse effects. Recent research suggests its promising role in curbing cravings in individuals with AUD and other addictive disorders, which is crucial in preventing relapse. Conventional craving treatments consisting of psychosocial and pharmacological approaches individually or in combination have demonstrated moderate success and sluggish reduction rates. Thus, there is a need for looking toward other treatment options that are novel, effective, safe, affordable, and targeted to dysfunctional neurocircuitry like tDCS.[8,9]
Earlier research has suggested that bilateral tDCS of the dorsolateral prefrontal cortex (DLPFC) is efficacious in reducing cravings and relapses.[10,11,12,13,14] There is, however, heterogeneity in the studies regarding active tDCS (cathodal or anodal) over F3/F4, the daily number of sessions, total duration of treatment, outcome measures, assessment methods, and follow-up assessments. Most studies have limited sample sizes and provide mixed results in reducing cravings. There have been a few studies to date in the Indian population to assess the effect of tDCS on craving in alcohol use disorder. Holla et al.[15] conducted a double-blind, randomized sham-controlled study, giving bilateral tDCS for once a day for consecutive 5 days, by placing the anode at the right dorsolateral prefrontal cortex (dlPFC) and the cathode at left dlPFC in 24 subjects over 90 days observing no significant difference in craving scores between the two groups. Astha et al.[16] studied the effects of 2 mA of anodal stimulation over right dlPFC and cathodal placement over left dlPFC in 76 alcohol-dependent subjects in a single blind sham-controlled trial, with a total of 10 sessions over 5 days to find a reduction in craving after 5 days of treatment, and did no longitudinal follow-ups. In a similar study, Aman and Sharma in 46 alcohol-dependent subjects receiving 10 sessions over 5 days of bilateral tDCS (right dlPFC anodal and left dlPFC cathodal stimulation) reported a significant reduction in craving scores at the end of 5 days, and they also did no longitudinal follow-ups.[17] Gairola et al.[18] studied 34 subjects with alcohol use disorder randomized to receive five sessions over 1 week of bifrontal tDCS sessions with the anode over right dlPFC and the cathode over left dlPFC observing a reduction in craving at the end of week 1 and week 4 but not at 8 weeks. Dayal et al.,[19] in a recent randomized sham-controlled trial among 44 detoxified inpatients, administered adjunctive transcranial direct current anodal stimulation at left dlPFC and the cathode at right dlPFC for 10 sessions over 5 days and reported no significant reduction in craving at the end of 3 months. At the time of conception and carrying out our study, only one Indian study by Holla et al.[15] was published. It had a longitudinal follow-up to deliver bilateral stimulation (right anodal and left cathodal dlPFC); however, it has a small sample size. There are no studies from India that included a large sample size (>100 subjects), from the usual mix of inpatients and outpatients, recruiting clinically diagnosed alcohol use disorder (DSM-5 criterion) with severe dependence (using Severity of Alcohol Dependence Questionnaire (SADQ)), including only right-handed subjects receiving 10 sessions of tDCS adjunctive to treatment as usual, recruited after at least 3 days of completing acute management of alcohol withdrawal/alcohol intake without withdrawal signs.
Hence, this study was aimed to assess if 10 sessions of adjunctive active bilateral tDCS with anodal left and cathodal right dlPFC stimulation delivered over 5 days, in a tertiary care general hospital psychiatric unit irrespective of outpatient or inpatient treatment setting at the initiation, in comparison to sham tDCS stimulation group delivered in randomized allocation, could reduce craving for alcohol as assessed by Obsessive Compulsive Drinking Scale (OCDS) at the end of 3 months. Another objective of the study was to assess the difference in clinical parameters between the two randomized groups receiving adjunctive active and sham tDCS intervention alongside treatment as usual, assessed by the abstinence and relapse rates.
MATERIALS AND METHODS
It was a longitudinal interventional study aimed at assessing changes in craving as assessed by Obsessive Compulsive Drinking Scale at 3 months, in subjects with severe alcohol use disorder using adjunctive tDCS, visiting the department of psychiatry at a tertiary care hospital providing specialist addiction treatment services, over the period of 1 year.
The sample consisted of 149 subjects in the age group of 18–55 years (range guided from previous 1-year chart review), diagnosed with alcohol use disorder as per DSM-5, Severity of Alcohol Dependence Questionnaire (SADQ) score >30, Obsessive Compulsive Drinking Scale (OCDS) Score >1, and right-handedness as per Edinburgh Handedness Inventory. They were included in the study after complete management of acute withdrawal (if any) and the washout period of at least 3 days.
Subjects with any other substance use disorder except nicotine, those with the presence of psychiatric morbidity in the past 6 months or at present, those with history of complicated withdrawal from alcohol in the past 6 months, those with history of Wernicke’s encephalopathy or Korsakoff syndrome, those who received electroconvulsive treatment or any other form of brain neuromodulation in the past 6 months, those with history suggestive of any significant neurological or neurosurgical illness, those with medical implants such as defibrillators or pacemakers or brain aneurysmal clips, and those with tDCS contraindications such as migraine and skin lesions involving scalp were excluded.
The permission for this study was obtained beforehand through the Institutional Ethics Committee via letter no. IEC/21/525 and was registered with the Clinical Trials Registry of India (CTRI) with registration number CTRI/2022/03/040823.
The subjects satisfying our inclusion and exclusion criteria were invited and informed of our procedure, following which well-informed written consent was obtained. The subjects who consented to participate were allocated to two groups, active treatment and sham treatment, based on an online random sequence generator.
Sampling
Each of the subjects was randomized into two groups based on their serial number and the group it was allotted to, as per the pregenerated sequence using an online computer-based random number generator, and the process went on till 3 months prior to the study completion.
The sample size was calculated using the statistical software G*Power 3.1 (freely available online). Assuming a moderate effect size of 0.5 derived from previous meta-analytical study,[20] a power of 80%, two-sided probability of Type 1 error at 5%, and principle aim as difference between two independent means (two groups) as a principal statistical test, we arrived at a number of 128 subjects being necessary. We aimed to recruit as many as possible for the duration of this study, halting 3 months prior to the time of concluding the study. Only the investigators knew about these two groups. The subjects were blinded via the tDCS sham setting in our instrument.
Tools used
The instruments applied at baseline were semistructured proforma, Edinburgh Handedness Inventory (EHI) to include only right-handed persons, Severity of Alcohol Dependence Questionnaire (SADQ) to assess the severity of alcohol dependence, and the Obsessive Compulsive Drinking Scale (OCDS) to measure craving for alcohol and are described subsequently.
The instrument administered after every session of tDCS for 5 days was the tDCS checklist, whereas the Timeline Follow Back tool was applied at each follow-up at 1 month and 3 months.
Semistructured proforma: The proforma recorded the sociodemographic and clinical details like diagnostic criteria fulfilled, age at initiation, age at dependence, usual dose, last intake, past and cumulative abstinence period, previous treatment seeking attempts and dispositions, various complications due to alcohol use, and other significant personal and family history.
Severity of Alcohol Dependence Questionnaire (SADQ)[21]: Each item is scored on a 4-point scale, giving a possible range of 0 to 60. A score of over 30 indicated severe alcohol dependence.
Obsessive Compulsive Drinking Scale (OCDS)[22]: It was used in English as well as Hindi after obtaining the required permissions to assess craving and its related behavior. OCDS is sensitive to and specific for the obsessive and compulsive characteristics of drinking-related thoughts, urges to drink, and the ability to resist those thoughts and urges in alcohol-abusing and alcohol-dependent populations. OCDS is sensitive as a monitoring tool and has predictive validity for relapse to alcohol use. It was administered before the first intervention, at 1 month and 3 months.
Edinburgh Handedness Inventory (EHI)[23]: It is a 10-item inventory to classify the handedness into right, left, or ambidextrous. The test–retest correlation coefficient is 0.75–0.86.
tDCS Side-Effect Checklist: It was developed to assess tDCS-related adverse effects from existing literature and administered for every session.
Timeline Follow-Back (TLFB)[24]: The instrument assessed the participant’s alcohol intake. It has been a widely used instrument to measure an accurate portrayal of one’s drinking timeline summary. It was applied at 1 month and 3 months post intervention.
For this study, we imbibed the definition of relapse as the return to drinking in previous pattern, as did Klauss et al.[25]
The intervention
Transcranial Direct Current Stimulation (tDCS): It was given as two sessions a day spaced by at least 2 hours. The current intensity was 2 mA, with the anode placed at F3 and the cathode at F4 (using the 10:20 EEG electrodes placement method) for a bilateral stimulation protocol. The active intervention consisted of ramping up to a current intensity of 2 mA in 30 seconds and maintaining it for 20 minutes. In contrast, the sham stimulation was given with a ramp-up time of 30 seconds at the beginning.
Procedure
Figure 1 illustrates the CONSORT diagram of the study. A convenience sample was used as all subjects included were a subset of the overall daily footfall of the addiction treatment facility. The sample was also purposive as the study used extensive exclusion criteria to homogenize the group. Allocation random numbers were generated in two batches (active and sham) in 1:1 ratio prior to the commencement of subject recruitment in the trial. The number sequence was kept with an investigator not involved in delivering the intervention. Upon satisfying the inclusion and exclusion criteria, participant’s enrolment number was tallied with the random number sequence and allocation into different treatment groups was done accordingly. Participants remained blinded of the group allocation as well as during the treatment phase. All participants were given the intervention, that is, ten sessions of tDCS twice a day over 5 consecutive days after they were abstinent from alcohol as well as benzodiazepines for at least 3 days. Two hundred and seventy-nine (n = 279) participants were screened, out of which 130 participants (n = 130) were excluded (114 did not meet the inclusion criteria and 26 declined to give consent to participate). The remaining 149 participants were randomized, with 75 (n = 75) participants allotted the active intervention group (A) and 74 (n = 74) participants allocated the sham intervention group (S). They were given the intervention as an adjunct to treatment as usual after they had completed detoxification and were off benzodiazepines for at least 3 days. Baseline SADQ and OCDS scores were obtained. A hundred and seven participants (51 participants in group A and 56 participants in group S) completed the intervention, that is, 10 sessions of tDCS, whereas the rest could not complete all sessions. All 149 participants were followed up in person or telephonically, wherever required. OCDS was reassessed, and TLFB was also applied, at 1 and 3 months.
Figure 1.
CONSORT flow diagram
Data analysis
All data were acquired and input for analysis using IBM Statistical Package for the Social Sciences (SPSS) version 25 for Windows (IBM Corp., Armonk, NY, United States). Descriptive statistics were frequency, percentages, means, and standard deviations. The primary objective of comparison of OCDS scores between the two groups was achieved using independent samples t-test, and the effect size was calculated using Cohen’s d with thresholds set at 0.20 for small, 0.50 for medium, and 0.80 for large effect.
Mixed model analysis of variance (ANOVA), also known as split-plot ANOVA, was performed to determine if the mean craving scores differed across the time points at baseline, 1 month, and 3 months. It was a two-way 2 × 3 [(intervention: active or sham) ×3 (time point: baseline, 1 month, and 3 months)] mixed model ANOVA with repeated measures on the time point. Partial η squared (ηp2) was used to indicate effect sizes, with thresholds set at 0.09 for small, 0.14 for medium, and 0.22 for large effects. Significance was attributed to P values below or at 0.05.
RESULTS
Sample characteristics
A hundred and forty-nine (n = 149) male participants, abstinent from alcohol and benzodiazepines for at least 3 days, with no withdrawal symptoms, were recruited. Table 1 illustrates the sample characteristics for the participants in each group. Both the groups (A and S) were comparable and had no statistically significant differences in their baseline characteristics. The participants’ age ranged from 22 to 53 years with the mean age being 36.22 ± 6.4 years (mean±s.d). The majority of them were married (78.5%), had matriculated (73.8%), and were employed (99.3%). About 3/5th of the population (60.4%) belonged to urban areas and belonged to at least the upper lower middle-class socioeconomic class as per the modified Kuppuswamy social class.[26]
Table 1.
Socio-demographic characteristics of the participants
| Variables | Active (n=75) (mean±sd)/n (%) | Sham (n=74) (mean±sd)/n (%) | Test of Significance | |
|---|---|---|---|---|
| Age (in years) | 36.83±6.3 | 35.62±6.5 | t (147)=1.15, P=0.3 | |
| Gender | Male | 52 (100%) | 55 (100%) | - |
| Education | Primary | 0 (0) | 2 (2.7) | χ2 (6)=4.30, P=0.6 |
| Middle | 16 (21.3) | 21 (28.4) | ||
| Matriculation | 21 (28.0) | 17 (23.0) | ||
| Intermediate | 20 (26.7) | 14 (18.9) | ||
| Graduate | 13 (17.3) | 15 (20.3) | ||
| Post-Graduate | 1 (1.3) | 1 (1.4) | ||
| Professional | 4 (5.3) | 4 (5.4) | ||
| Occupation | Professional | 6 (8.0) | 5 (6.8) | χ2 (5)=3.0, P=0.7 |
| Skilled worker | 39 (52.0) | 39 (52.7) | ||
| Unskilled worker | 1 (1.3) | 4 (5.4) | ||
| Unemployed | 1 (1.3) | 0 (0) | ||
| Business | 14 (18.7) | 13 (17.6) | ||
| Farmer | 14 (18.7) | 13 (17.6) | ||
| Income | <20,000/month | 30 (40.0) | 29 (39.2) | χ2 (1)=0.01, P=0.9 |
| ≥20,000/month | 45 (60.0) | 45 (60.8) | ||
| Marital Status | Never married | 10 (13.3) | 14 (18.9) | χ2 (3)=1.40, P=0.7 |
| Married | 60 (80.0) | 57 (77.0) | ||
| Separated | 4 (5.3) | 2 (2.7) | ||
| Divorced | 1 (1.3) | 1 (1.4) | ||
| Residence | Rural | 30 (40.0) | 29 (39.2) | χ2 (1)=0.01, P=0.9 |
| Urban | 45 (60.0) | 45 (60.8) | ||
*P value significant at <0.05; sd=standard deviation; t=student t-test value; χ2=Chi-square value
Table 2 depicts the baseline clinical characteristics of the participants in each group. The mean age at first alcohol use for all the participants was 19.16 ± 2.5 years (mean ± s.d). The mean duration of daily or almost daily consumption of alcohol was 7.41 ± 3.85 years, the mean duration of physical dependence was 4.21 ± 3.10 years, and the average daily consumption was found to be 20.65 ± 3.65 units. Almost every participant (1.86 ± 0.9) had had a significant attempt at abstinence as a formal treatment, that is, under the supervision of a psychiatrist or in a recognized inpatient facility, except four participants who never tried treatment but only had a self-motivated abstinence. ‘Significant attempt’ was operationalized as remaining completely abstinent from alcohol for at least 30 days. The participants attempted abstinence for a mean of 2.6 times in their lives after beginning to drink daily/almost daily. The participants had a median cumulative period of abstinence for 270 days (9 months) after the onset of daily/almost daily drinking. The mean severity of dependence score was found to be 45.61 ± 1.9. All our patients were receiving treatment as usual, which consisted of psychological as well as pharmacological interventions. All the subjects received anticraving drugs, but no one was on a deterrent drug. Naltrexone was received by 62.7% of subjects in the active tDCS group compared to 54.1% of subjects in the sham. Acamprosate was received by 37.3% and 45.9% of subjects from the active and sham tDCS groups, respectively. Naltrexone was supplied from the addiction treatment facility free of cost, and Acamprosate had to be purchased by the subjects. Medication adherence could not be commented upon. Both groups did not vary statistically in terms of the pharmacotherapy received [χ2(1) = 1.14; P = 0.3].
Table 2.
Baseline clinical parameters of the participants
| Variables | Active (n=75) (mean±sd)/n (%) | Sham (n=74) (mean±sd)/n (%) | Test of Significance | |
|---|---|---|---|---|
| Type of Alcohol Used | Country-made liquor | 18 (24.0) | 13 (17.6) | χ2 (1)=0.93; P=0.3 |
| Indian-made foreign liquor | 57 (76.0) | 61 (82.4) | ||
| Age at Onset of Consuming Alcohol (in years) | 19.40±2.6 | 18.93±2.4 | t (147)=1.10, P=0.3 | |
| Years Since Daily/Almost Daily Consumption of Alcohol (in years) | 19.40±2.6 | 18.93±2.4 | t (147)=1.10, P=0.3 | |
| Years Since Physical Dependence (in years) | 4.31±3.2 | 4.11±3.0 | t (147)=0.40, P=0.7 | |
| Usual Dose (units) | 20.60±3.7 | 20.70±3.6 | t (147)=-0.20, P=0.9 | |
| Maximum Dose (units) | 34.73±7.6 | 34.10±5.8 | t (147)=0.61, P=0.5 | |
| Number of Significant Abstinence Attempts | 2.72±1.1 | 2.43±1.1 | t (147)=1.60, P=0.1 | |
| Number of Significant Abstinence Attempts with Formal Treatment | 2.03±0.9 | 1.70±0.9 | t (147)=2.70, P=0.01* | |
| Cumulative Abstinence Period (in months) | 8.81±2.0 | 9.10±1.9 | t (147)=-0.84, P=0.4 | |
| History of Complicated Withdrawals | Yes | 15 (20.0) | 11 (14.9) | χ2 (1)=0.70; P=0.4 |
| No | 60 (80) | 63 (85.1) | ||
| Family History of Alcohol Use Disorder | Yes | 11 (14.7) | 5 (6.8) | χ2 (1)=0.70; P=0.4 |
| No | 64 (85.3) | 69 (93.2) | ||
| Ever Use (any other substance except alcohol or tobacco) | Yes | 20 (26.7) | 28 (37.8) | χ2 (1)=2.12; P=0.1 |
| No | 55 (73.3) | 46 (62.2) | ||
| Type of Substances Used (other than alcohol and/or tobacco) | Cannabinoids | 12 (16.0) | 14 (18.9) | χ2 (10)=6.90; P=0.7 |
| Inhalants | 0 (0) | 1 (1.4) | ||
| Opioids | 1 (1.3) | 4 (5.4) | ||
| Sedative/Hypnotics | 1 (1.3) | 1 (1.4) | ||
| Cannabinoids, Sedative/Hypnotics, Inhalants | 1 (1.3) | 1 (1.4) | ||
| Cannabis, Inhalants | 0 (0) | 1 (1.4) | ||
| Sedative/Hypnotics, Opioids | 3 (4) | 4 (5.4) | ||
| Cannabis, Opioids | 1 (1.3) | 1 (1.4) | ||
| Anabolic Steroids | 1 (1.3%) | 0 (0) | ||
| Cannabis, Opioids, Inhalants | 0 (0) | 1 (1.4) | ||
| Tobacco Use | No | 4 (5.3) | 8 (10.8) | χ2 (2)=3.43; P=0.2 |
| Smoking | 50 (66.7) | 39 (52.7) | ||
| Smokeless | 21 (28.0) | 27 (36.5) | ||
*P value significant at <0.05; sd=standard deviation; t=student t-test value; χ2=Chi-square value
The number of participants receiving the intended treatment (10 sessions of tDCS) was 107 (n = 107). Out of these, 51 (n = 51) participants received active intervention, and 56 (n = 56) received the sham intervention. However, we conducted an intention-to-treat analysis (n = 149, n (A) = 75 and n (S) =74) to conclude our results. After every tDCS session, we applied a tDCS side-effect checklist to assess and monitor the adverse effects. None of the participants from either of the treatment arm reported any side effects.
Craving
The primary objective of our study was to assess the effect of adjunctive tDCS on OCDS scores by the end of study. The independent samples t-test revealed a significant difference in test scores between the active tDCS intervention group [28.50 ± 13.23] and sham tDCS intervention group [34.75 ± 14.07], t(147) = -2.79. The effect size, as measured by Cohen’s d, was d = 0.45, indicating a small effect.
A mixed effects ANOVA to study the longitudinal effects of tDCS on craving over 3 months was done. The assumption of homogeneity of covariance was violated (Box’s M = 16.11; F (6,156485.7) =2.6; P = 0.02), but we proceeded with the analysis given that the sample sizes in both groups were similar.[14] The sphericity assumption was also violated, following which we report the Greenhouse-Geisser corrected differences for subjects to check if there was a fixed effect of time, which came out to be statistically significant (F (1.5) =201.2, p = <0.001). The effect size as estimated by partial η squared (ηp2) was 0.58, indicating a large effect size.
The equality of error variances was checked using Levene’s test, which found the assumption valid for scores compared at baseline (F = 3.31, P = 0.07) and 3 months (F = 0.73, P = 0.39) but violated for scores at 1-month follow-up (F = 4.87, P = 0.03).
We still carried forward the interpretation since both the groups had equal sample sizes.[27] Figure 2 illustrates that the mean craving scores, as assessed by OCDS in the active intervention group, are significantly lower than those of the sham group, at 1 month (A = 30.37 ± 11.66; S = 33.55 ± 13.73) and 3 month (A = 28.50 ± 13.23; S = 34.75 ± 14.07) follow-up points, as compared to the baseline (A = 48.33 ± 1.94, S = 48.27 ± 2.45); F(2,294) = 5.52, P < 0.01. The effect size as estimated by partial η squared (ηp2) was 0.04, indicating a small effect size.
Figure 2.
Mean craving scores over time
Follow-up parameters
Time to first drink, relapse, and abstinence
The mean time taken to drink again by the group that received active tDCS intervention was more than 5 weeks (38.46 ± 26.99 days), which was significantly longer than that of the sham tDCS intervention group that relapsed before a mean duration of 5 weeks post treatment initiation (29.39 ± 23.83), t (147) = 2.17, P = 0.03. The effect size measured by Cohen’s d was d = 0.35, 95%CI [0.03–0.67].
Our study found that the subjects in the active tDCS intervention group took a significantly longer time to relapse with a mean time of more than 2 months (67.77 ± 29.70 days) compared to the sham tDCS intervention group that took less than 2 months (55.02 ± 30.72), t (147) = 2.57, P = 0.01. The effect size, as measured by Cohen’s d for this, was d = 0.42, 95% CI [0.09–0.74].
The incidence of relapse in the active tDCS intervention group was 19.5% lower than that of the sham tDCS intervention group at the end of study, being statistically significant as well [χ2 (1) = 5.7; P = 0.02]. The number needed to treat (NNT) was 6, 95% CI [2.8–26.2].
We analyzed the subjects who were not completely abstinent but had not relapsed. This was done by using the Kaplan–Meier survival curve. The log-rank method was used to compare the probabilities statistically. The likelihood of not relapsing was significantly higher in the active tDCS intervention group, with the mean survival time being 55.4 days, 95% CI (47.2–63.7), compared to the sham tDCS intervention group having a mean survival time of 38.5 days, 95% CI (30.9–46.2); (χ2(1) = 8.1, P = 0.004).
Figure 3 illustrates the Kaplan–Meier survival curve to compare the probabilities for the active and sham intervention groups.
Figure 3.
Survival analysis curve
DISCUSSION
We conducted a randomized sham-controlled study in a set of 149 participants who were visiting a specialized addiction treatment facility to assess the effect of adjunctive active tDCS compared to sham tDCS in modifying craving for alcohol at the end of the study. Convenience and purposive samples were taken, which was randomized, bearing the caveats that this brings along and discussed by Andrade.[28]
The review and meta-analyses by Cavicchioli et al.[29] found craving as well as severity of dependence to have had associations with relapses in alcohol use disorders. The same was highlighted for the Indian population by Soundararajan and Murthy.[28,30] Hence, our focus was to target the craving as assessed by OCDS which was used in Hindi as well as English after obtaining permission from the copyright holders. It has been validated in the Indian population.
The subjects between the two groups were comparable in terms of all sociodemographic characteristics as well as clinical details with no statistically significant difference except in the number of abstinent attempts where abstinence periods were assisted by treatment providers. The participants in the active group had a significantly higher number of such treatments, and this could imply higher severity of related problems, requiring assistance with abstinence.[31]
Our study had one of the highest numbers of participants compared to all the previous studies with similar treatment protocols.[32] It was possible since it was the first dedicated addiction treatment facility, funded by the government in this area in a general hospital psychiatric setting which may have helped them overcome a lot of barriers to treatment including stigma, although comparative studies in the past have not focused on it.[33,34] Many other factors possibly contributing toward this level of participation could be sprawling unregulated treatment centers in the absence of trustworthy alternatives for substance use disorder.[35] As per a recent secondary analysis, the sociodemographic characteristics of all our participants were similar to those of the participants in the multicentric COMBINE trial.[36]
Several researchers have chosen either right or left anodal stimulation, with some reviews suggesting that right dlPFC anodal stimulation would provide better overall clinical outcomes in alcohol use disorder. However, there are meta-analyses that report that the site of stimulation of dlPFC (being either left or right) does not have a statistical difference on craving outcomes, yet a higher Hedges’ g for the right side has made it a preferable site for many studies.[37] A preference for the left dlPFC was made in view of a greater number of studies[10,11,12,13,19] assessing tDCS in addictive disorders stimulating the left dlPFC. Bilateral montages were advised so that the targeted dlPFC could be potentially activated in activity by the anode and reduce the interhemispheric inhibition from the opposite dlPFC by cathodal stimulation decreasing the potential cortical activity.[38]
OCDS is a valid craving scale; we used it in Hindi language for accuracy and reliability. Its predictive value for treatment outcomes has been demonstrated for outpatient and inpatient treatment.[39] The study has recruited subjects with severe dependence and demonstrated the efficacy of tDCS on craving as assessed on OCDS. Meta-analyses that reviewed the role of modulators in efficacy of this therapy mention that the severity of dependence influences the effect sizes.[37,38]
tDCS has been seen to improve outcomes by modulating and increasing the cognitive control that results in a decrease in craving. The dlPFC is a part of the ‘Executive Control Network’ hypothesized to control cravings.[37] This is central to the way tDCS may help us since lower levels of dlPFC functioning are reflective of prefrontal hypoactivity and the suboptimal functioning could be related to lower levels of self-control in subjects with alcohol use disorders. It is proven that the dlPFC must be optimally functioning to inhibit any behavioral or cognitive response, inefficiency of which is observed as hypoactivation. tDCS is therefore instrumental in activating the dlPFC through its anodal stimulation. Researchers are putting together consensus to translate the effects of tDCS into the clinical world guiding everyone.[8]
In our study, the tDCS sessions were given only for 5 days, but the effects were felt for months at a stretch. Explanations for effects outlasting the duration of stimulation are the long-term effects of tDCS, namely, synaptic plasticity, long-term potentiation and long-term depression, and even galvanotropism that have been discussed by Bikson et al.[40] The efficacy of our treatment on craving is evidently significant over 3 months on direct comparison of mean scores, with a small effect size of d = 0.45. The same when assessed using time x group interaction effects revealed that the effect size for reduction of craving with increasing time was significantly large in both groups at 1 and 3 months; there was significantly greater reduction in cravings in the active tDCS intervention group compared to sham tDCS intervention group with a small effect size. The study by Gairola et al.[18] did not find any significant reduction in craving beyond 1–4 weeks; they however differed from our study in terms of having opposite stimulation sites, fewer numbers of participants, different craving assessment scales, and administering disulfiram in addition to other anticraving agents as well that could have contributed to the initial positive results. Another study by Astha et al. showed a significant reduction in craving scores with a medium effect size of 0.58, though it was limited in its methodology by lack of randomization, no longitudinal follow-up, and sampling only inpatients while stimulating the dlPFC with electrode placement opposite to ours.[16] Another similar study reported positive outcomes without longitudinal data.[17] Dayal and colleagues published a similar study as ours including stimulation sites and the same follow-up period; however, they reported no significant change in craving and abstinence rates.[19]
Our study results are similar to those of the widely cited study done by Klauss et al.,[25] with a larger sample size than theirs but with bilaterally opposite stimulation parameters at dlPFC and half the duration of follow-up. Jitendriya and colleagues reported sample size being too small to draw a conclusion despite having achieved a reduction in craving.[41]
The focus on craving is warranted, given its pivotal role in the journey of addiction treatment and its association with the probability of use of the substance in the future as well as the completion of treatment. This leads to increased treatment costs and frequent relapses. Other clinical outcome parameters studied were maintenance of abstinence, the time taken to have their first drink, and relapse of drinking. By the end of our study, 10% of the participants remained abstinent, where 3 out of 5 belonged to the active intervention group. The rates of relapse are like that reported in the Indian population through a systematic review by Sarkar et al.[42]
The survival analysis remarked that as the study progressed, the probability of relapse in the sham tDCS intervention group was higher in comparison with active tDCS. Relapse is a complex human behavior where people who relapse do so in a shorter amount of time in the first few months post intervention (Project MATCH).[43] Furthermore, it is evident from research that it could be modeled as a nonlinear dynamic system, as demonstrated by the Cusp Catastrophe model of this process.[43,44,45] As our analysis did not factor in the other determinants of relapse but only craving, we hypothesize that a timeframe of 3 months in our study may not allow for an elaborate interpretation of this phenomenon yet. The craving scores were observed to be linearly decreasing in both the groups, hinting at the efficacy of treatment as usual (psychosocial interventions + pharmacotherapy). However, it was observed to further decrease in the active tDCS intervention group at the end of 3 months but not in the sham tDCS intervention group, possibly due to the effect of multiple sessions. The effect sizes were small at the end of our study for the changes in craving scores and time taken to first drink and incidence of abstinence.
Studying the real-world implications, we assessed the NNT to find that given our sample characteristics, at the end of 3 months, six subjects needed to be treated to benefit 1 subject from relapsing to alcohol use. tDCS is safe to use, as has been reported in literature.[25] Our subjects reported even no minor adverse effects during or after tDCS treatment.
Limitations
In our study, most of the measures employed were self-reported and these reports are dynamic in nature, so future studies may be planned to reduce the reporting bias or incorporate real time assessments. The study was single-blinded; double blinding could have lent more veracity to our results. The changes in subjects’ status like inpatient management, changes in medications, medication adherence, external factors, and comorbidities in between our follow-up points may have confounded our results. Future studies with a longer follow-up course are required to see the long-term effects of tDCS and to help develop guidelines for administering these modalities.
CONCLUSION
The study examined the efficacy of adjunctive tDCS in a randomized sham-controlled trial on craving in alcohol use disorder. The study highlighted that tDCS significantly leads to a decrease in craving for alcohol and overall relapses in the persons with severe alcohol use disorder. Furthermore, the study points out the effectiveness of tDCS in significantly enhancing the time taken to first drink. In the future, multicenter studies with similar populations and protocols can be planned to have better results and understandings, guiding the translation of this useful treatment into clinical practice.
Conflicts of interest
There are no conflicts of interest.
Funding Statement
Nil.
REFERENCES
- 1.World Health Organization . World Health Organization; 2014. Global status report on noncommunicable diseases 2014. Available from: https://apps.who.int/iris/bitstream/handle/10665/148114/9789241564854_eng.pdf . [Last accessed on 2024 May 20] [Google Scholar]
- 2.Ambekar A, Agrawal A, Rao R, Mishra AK, Khandelwal SK, Chadda RK. New Delhi: Ministry of Social Justice and Empowerment, Government of India; 2019. Magnitude of substance use in India; pp. 11–16. [Google Scholar]
- 3.Carvalho AF, Heilig M, Perez A, Probst C, Rehm J. Alcohol use disorders. Lancet. 2019;394:781–92. doi: 10.1016/S0140-6736(19)31775-1. [DOI] [PubMed] [Google Scholar]
- 4.Ritchie H, Roser M. Alcohol consumption. 2022. Available from: https://ourworldindata.org/alcohol-consumption . [Last accessed on 2024 May 25]
- 5.Lohoff FW. Targeting unmet clinical needs in the treatment of alcohol use disorder. Front Psychiatry. 2022;13:767506. doi: 10.3389/fpsyt.2022.767506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pedron S, Dumontoy S, González-Marín M del C, Coune F, Van Schuerbeek A, Haffen E, et al. Transcranial direct current stimulation (tDCS) reduces motivation to drink ethanol and reacquisition of ethanol self-administration in female mice. Sci Rep. 2022;12:198. doi: 10.1038/s41598-021-03940-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Reato D, Salvador R, Bikson M, Opitz A, Dmochowski J, Miranda PC. Principles of transcranial direct current stimulation (tDCS): Introduction to the biophysics of tDCS. In: Knotkova H, Nitsche MA, Bikson M, Woods AJ, editors. Practical Guide to Transcranial Direct Current Stimulation: Principles, Procedures and Applications. Cham: Springer International Publishing; 2019. pp. 45–80. [Google Scholar]
- 8.Bollen Z, Dormal V, Maurage P. How should transcranial direct current stimulation be used in populations with severe alcohol use disorder? A clinically oriented systematic review. Clin EEG Neurosci. 2022;53:367–83. doi: 10.1177/15500594211001212. [DOI] [PubMed] [Google Scholar]
- 9.Reus VI, Fochtmann LJ, Bukstein O, Eyler AE, Hilty DM, Horvitz-Lennon M, et al. The American Psychiatric Association practice guideline for the pharmacological treatment of patients with alcohol use disorder. Am J Psychiatry. 2018;175:86–90. doi: 10.1176/appi.ajp.2017.1750101. [DOI] [PubMed] [Google Scholar]
- 10.Boggio PS, Sultani N, Fecteau S, Merabet L, Mecca T, Pascual-Leone A, et al. Prefrontal cortex modulation using transcranial DC stimulation reduces alcohol craving: A double-blind, sham-controlled study. Drug Alcohol Depend. 2008;92:55–60. doi: 10.1016/j.drugalcdep.2007.06.011. [DOI] [PubMed] [Google Scholar]
- 11.den Uyl TE, Gladwin TE, Rinck M, Lindenmeyer J, Wiers RW. A clinical trial with combined transcranial direct current stimulation and alcohol approach bias retraining. Addict Biol. 2017;22:1632–40. doi: 10.1111/adb.12463. [DOI] [PubMed] [Google Scholar]
- 12.den Uyl TE, Gladwin TE, Lindenmeyer J, Wiers RW. A clinical trial with combined transcranial direct current stimulation and attentional bias modification in alcohol-dependent patients. Alcohol Clin Exp Res. 2018;42:1961–9. doi: 10.1111/acer.13841. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Klauss J, Anders QS, Felippe LV, Nitsche MA, Nakamura-Palacios EM. Multiple sessions of transcranial direct current stimulation (tDCS) reduced craving and relapses for alcohol use: A randomized placebo-controlled trial in alcohol use disorder. Front Pharmacol. 2018;9:716. doi: 10.3389/fphar.2018.00716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dubuson M, Kornreich C, Vanderhasselt MA, Baeken C, Wyckmans F, Dousset C, et al. Transcranial direct current stimulation combined with alcohol cue inhibitory control training reduces the risk of early alcohol relapse: A randomized placebo-controlled clinical trial. Brain Stimul. 2021;14:1531–43. doi: 10.1016/j.brs.2021.10.386. [DOI] [PubMed] [Google Scholar]
- 15.Holla B, Biswal J, Ramesh V, Shivakumar V, Bharath RD, Benegal V, et al. Effect of prefrontal tDCS on resting brain fMRI graph measures in alcohol use disorders: A randomized, double-blind, sham-controlled study. Prog Neuropsychopharmacol Biol Psychiatry. 2020;102:109950. doi: 10.1016/j.pnpbp.2020.109950. [DOI] [PubMed] [Google Scholar]
- 16.Astha, Patil S, Patil NM, Tekkalaki B, Chate SS. Efficacy of tDCS on craving in patients of alcohol dependence syndrome: A single-blind, sham-controlled trial. Indian J Psychiatry. 2024;66:98–105. doi: 10.4103/indianjpsychiatry.indianjpsychiatry_492_23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Aman S, Sharma G. Paper Presentation: Effectiveness of tDCS in reduction of craving in alcohol dependence syndrome. Indian J Psychiatry. 2022;64(Suppl 3):S599. [Google Scholar]
- 18.Gairola A, Nischal A, Kar SK, Arya A, Singh A. Randomized controlled trial of bifrontal transcranial direct current stimulation on craving in alcohol use disorder. Indian J Psychol Med. 2024:02537176231223314. doi: 10.1177/02537176231223314. doi: 10.1177/02537176231223314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dayal P, Kaloiya GS, Verma R, Kumar N. Need to rethink tDCS protocols for the treatment of alcohol use disorder: Insights from a randomized sham-controlled clinical trial among detoxified inpatients. J Addict Dis. 2023;42:544–50. doi: 10.1080/10550887.2023.2257106. [DOI] [PubMed] [Google Scholar]
- 20.Song S, Zilverstand A, Gui W, Li HJ, Zhou X. Effects of single-session versus multi-session non-invasive brain stimulation on craving and consumption in individuals with drug addiction, eating disorders or obesity: A meta-analysis. Brain Stimul. 2019;12:606–18. doi: 10.1016/j.brs.2018.12.975. [DOI] [PubMed] [Google Scholar]
- 21.Stockwell T, Murphy D, Hodgson R. The severity of alcohol dependence questionnaire: Its use, reliability and validity. Br J Addict. 1983;78:145–55. doi: 10.1111/j.1360-0443.1983.tb05502.x. [DOI] [PubMed] [Google Scholar]
- 22.Anton RF, Moak DH, Latham PK. The Obsessive compulsive drinking scale: A new method of assessing outcome in alcoholism treatment studies. Arch Gen Psychiatry. 1996;53:225–31. doi: 10.1001/archpsyc.1996.01830030047008. [DOI] [PubMed] [Google Scholar]
- 23.Oldfield RC. The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia. 1971;9:97–113. doi: 10.1016/0028-3932(71)90067-4. [DOI] [PubMed] [Google Scholar]
- 24.Sobell LC, Sobell MB. In: Measuring Alcohol Consumption. Springer; 1992. Timeline follow-back; pp. 41–72. [Google Scholar]
- 25.Klauss J, Penido Pinheiro LC, Silva Merlo BL, Correia Santos G de A, Fregni F, Nitsche MA, et al. A randomized controlled trial of targeted prefrontal cortex modulation with tDCS in patients with alcohol dependence. Int J Neuropsychopharmacol. 2014;17:1793–803. doi: 10.1017/S1461145714000984. [DOI] [PubMed] [Google Scholar]
- 26.Radhakrishnan M, Nagaraja SB. Modified Kuppuswamy socioeconomic scale 2023: Stratification and updates. Int J Community Med Public Health. 2023;10:4415–8. [Google Scholar]
- 27.Well AD, Myers JL, Lorch RF., Jr . 3rd. New York: Routledge; 2010. Research Design and Statistical Analysis: Third Edition; p. 832. [Google Scholar]
- 28.Andrade C. The inconvenient truth about convenience and purposive samples. Indian J Psychol Med. 2021;43:86–8. doi: 10.1177/0253717620977000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Cavicchioli M, Vassena G, Movalli M, Maffei C. Is craving a risk factor for substance use among treatment-seeking individuals with alcohol and other drugs use disorders? A meta-analytic review. Drug Alcohol Depend. 2020;212:108002. doi: 10.1016/j.drugalcdep.2020.108002. [DOI] [PubMed] [Google Scholar]
- 30.Soundararajan S, Murthy P. High craving is associated with fewer abstinent days and lesser time to relapse during treatment in severe alcohol use disorder. Indian J Psychiatry. 2023;65:319–26. doi: 10.4103/indianjpsychiatry.indianjpsychiatry_550_22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Tucker JA, Chandler SD, Witkiewitz K. Epidemiology of recovery from alcohol use disorder. Alcohol Res. 2020;40:02. doi: 10.35946/arcr.v40.3.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Coles AS, Kozak K, George TP. A review of brain stimulation methods to treat substance use disorders. Am J Addict. 2018;27:71–91. doi: 10.1111/ajad.12674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Schomerus G, Matschinger H, Angermeyer MC. Do psychiatric units at general hospitals attract less stigmatizing attitudes compared with psychiatric hospitals? Epidemiol Psychiatr Sci. 2013;22:163–8. doi: 10.1017/S2045796012000510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Verhaeghe M, Bracke P, Bruynooghe K. Stigmatization in different mental health services: A comparison of psychiatric and general hospitals. J Behav Health Serv Res. 2007;34:186–97. doi: 10.1007/s11414-007-9056-4. [DOI] [PubMed] [Google Scholar]
- 35.Joseph SA, Hemalatha K. Addiction treatment in India: Legal, ethical and professional concerns reported in the media. Indian J Med Ethics. 2021;VI:306–13. doi: 10.20529/IJME.2021.030. [DOI] [PubMed] [Google Scholar]
- 36.Haass-Koffler CL, Piacentino D, Li X, Long VM, Lee MR, Swift RM, et al. Differences in sociodemographic and alcohol-related clinical characteristics between treatment seekers and non-treatment seekers and their role in predicting outcomes in the COMBINE study for alcohol use disorder. Alcohol Clin Exp Res. 2020;44:2097–108. doi: 10.1111/acer.14428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Chen J, Qin J, He Q, Zou Z. A meta-analysis of transcranial direct current stimulation on substance and food craving: What effect do modulators have? Front Psychiatry. 2020;11:598. doi: 10.3389/fpsyt.2020.00598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Kim HJ, Kang N. Bilateral transcranial direct current stimulation attenuated symptoms of alcohol use disorder: A systematic review and meta-analysis. Prog Neuropsychopharmacol Biol Psychiatry. 2021;108:110160. doi: 10.1016/j.pnpbp.2020.110160. [DOI] [PubMed] [Google Scholar]
- 39.Schmidt P, Helten C, Soyka M. Predictive value of obsessive-compulsive drinking scale (OCDS) for outcome in alcohol-dependent inpatients: Results of a 24-month follow-up study. Subst Abuse Treat Prev Policy. 2011;6:14. doi: 10.1186/1747-597X-6-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Bikson M, Paulus W, Esmaeilpour Z, Kronberg G, Nitsche MA. Mechanisms of acute and after effects of transcranial direct current stimulation. In: Knotkova H, Nitsche M, Bikson M, Woods A, editors. Practical Guide to Transcranial Direct Current Stimulation. Cham: Springer; 2019. pp. 81–113. [Google Scholar]
- 41.Jitendriya B, Holla B, Shivkumar V, Chand PK, Murthy P, Venkatsubaramanian G, et al. Effect of transcranial direct current stimulation on relapse of alcohol. Indian J Psychiatry. 2022;64(Suppl 3):S629–30. [Google Scholar]
- 42.Sarkar S, Tom A, Das S, Bharadwaj B, Ghosh A. Systematic review: Rates and determinants of relapse to alcohol. J Ment Health Hum Behav. 2022;27:8–18. [Google Scholar]
- 43.Witkiewitz K, van der Maas HL, Hufford MR, Marlatt GA. Nonnormality and divergence in posttreatment alcohol use: Reexamining the Project MATCH data “another way”. J Abnorm Psychol. 2007;116:378–94. doi: 10.1037/0021-843X.116.2.378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Witkiewitz K, Marlatt GA. Modeling the complexity of post-treatment drinking: It’s a rocky road to relapse. Clin Psychol Rev. 2007;27:724–38. doi: 10.1016/j.cpr.2007.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Hufford MR, Witkiewitz K, Shields AL, Kodya S, Caruso JC. Relapse as a nonlinear dynamic system: Application to patients with alcohol use disorders. J Abnorm Psychol. 2003;112:219–27. doi: 10.1037/0021-843x.112.2.219. [DOI] [PubMed] [Google Scholar]