Abstract
Introduction To assess the feasibility, acceptability and preliminary effectiveness of synchronous internet-mediated relapse prevention therapy (i-RPT) in alcohol use disorder. Methods This was a pilot, quasi-experimental study. Thirty-two adult men with alcohol use disorder were recruited through purposive sampling from an outpatient setting. We assessed patterns of drinking, craving, motivation and coping. Patients received five twice-weekly sessions of i-RPT. They were reassessed 12 weeks post-intervention (CTRI Trial REF/2020/09/036392). Results Thirty-two (48%) of the 67 patients fulfilled the eligibility criteria and all consented to the study. All participants completed at least two sessions and 23 (71.9%) completed all five sessions. Two-thirds of participants reported high satisfaction in the Telehealth Satisfaction Questionnaire. We observed modest intervention effects on days of abstinence in both per-protocol (P <0.001; r = 0.6) and worst-case (P <0.001, r = 0.5) analyses. There were also reductions in the amount of alcohol use, frequency of drinking and heavy drinking, craving and maladaptive coping behaviours. Per-protocol analysis revealed a positive post-intervention change in the motivational level to change alcohol use. Conclusion iRPT appears to be feasible, acceptable and possibly effective in alcohol use disorder.
Introduction
Alcohol use disorder (AUD) is a chronic relapsing brain disease. Individuals with AUD have a 75–95% one-year relapse rate following cessation attempts [1, 2]. Hence, prevention of relapse or minimisation of the severity and consequences of relapse are major goals in the treatment of AUD [3].
Relapse prevention therapy (RPT) is a cognitive-behaviour- based treatment with a goal of skill-building and improved coping to deal with high-risk drinking situations, cravings and ‘lapses’ [3]. Existing reviews show that RPT is generally effective in reducing substance use, improving psychosocial functioning, reducing the severity of relapse and motivation for behaviour change [4, 5]. RPT has been classified as an empirically validated treatment for AUD [6].
Although RPT has been shown to be effective across settings and modes of delivery, the feasibility and effectiveness of synchronous internet-delivered RPT (i-RPT) have not received adequate research attention. This is despite the fact that videoconferencing-based counselling for substance use disorder does not differ from in-person treatment with regard to patient satisfaction and outcome [7]. Telemedicine can be a cost-effective method to reduce the treatment gap and improve retention in substance use disorder [8]. The feasibility or efficacy of i-RPT was examined in three published studies; all were from high-income countries; one of these was on adolescents and the two others used text-based therapist guidance [9-11]. The initial results were encouraging.
The application of telemedicine for psychiatric disorders in India is relatively new [12]. The aim of our pilot study was to examine the acceptability, feasibility and preliminary effectiveness of i-RPT in patients with AUD.
Methods
This was a quasi-experimental pilot study intended to evaluate the feasibility, acceptability and preliminary effects of online nurse-delivered RPT for AUD. We have used a within-group design. Between 1 October and 31 December 2020, all patients registered in the tele-addiction clinic of our tertiary care addiction treatment center in North India were evaluated for eligibility. The inclusion criteria were: completed detoxification; not willing to take maintenance medications; had optimal technical support to participate in videoconferencing sessions; and had experienced at least one episode of relapse in the past. We excluded patients with comorbid substance use disorder (except for tobacco use disorder) and severe mental illness.
Although this was a pilot study, we performed a power calculation, intending to recruit as many study participants as possible to achieve that number within the 3-month study period. The estimated sample size was 35, assuming an effect of 0.5 on the primary outcome, α error probability 0.05 and the desired power of 0.80.
Feasibility and acceptability outcomes were measured by proportions of eligible patients, their readiness to accept i-RPT, attrition during and after the intervention, and participants' level of satisfaction with telemedicine. The primary outcome was the difference in the cumulative days of abstinence from alcohol. The secondary outcomes were divided into two groups: (i) alcohol-related, that is, reduction of alcohol amount on drinking days, frequency of drinking and heavy drinking; (ii) potential mediators of relapse, that is, improved motivation to quit alcohol, coping behaviour and reduction of craving. We obtained patterns of drinking by the Timeline Follow back method, craving by Alcohol Craving Questionnaire, motivation by Readiness to Change Questionnaire and coping by Coping Behaviours Inventory [13-17]. Patients were reassessed after 12 ± 2 weeks of intervention. We assessed patients' level of satisfaction and comfort with the online medium by Telehealth Satisfaction Scale [18].
A trained nurse delivered five twice weekly manual-based sessions of i-RPT through online Zoom meetings or WhatsApp. The training was performed by three of the authors, KD, AG and SS. AG prepared and adapted the manual to the need for telemedicine-based delivery. Further details are provided in the Supporting Information.
Analysis was performed using Statistical Package for Social Sciences (SPSS), version 20 [19]. The pre, post-intervention assessment scores were compared using McNemer's test and the Wilcoxon rank-sum test. We performed both per-protocol (PP) and worst-case (WC) analyses. Cohen's d was used to calculate within-group effect sizes.
All study participants provided written informed consent. The study was approved by the Institute Ethics Committee and registered with the Clinical Trial Registry-India.
Results
A total of 155 patients were registered in the tele-addiction clinic between 1 October and 31 December 2020. Among them, 67 (43.2%) patients were clinically diagnosed with alcohol dependence by a trained psychiatrist as per the International Classification of Diseases, 10th revision. Thirty-two (47.8%) patients met the eligibility criteria for our study; all of them consented to participate. We failed to achieve a sample size of 35 in the 3-month study period.
All participants were men and the mean age was 41 ± 9.4 years. The mean duration of alcohol dependence was 4.3 ± 3.7 years (see Table S1, Supporting Information). Twenty-three (71.9%) participants completed all five sessions. We contacted the nine participants who dropped out; six of them relapsed and three could not participate due to ‘time constraints’.
We performed a per-protocol analysis with 23 participants who completed all sessions of i-RPT and did a worst-case analysis including all study participants (N = 32). Wilcoxon signed-rank test revealed a statistically significant increase in number of days of abstinence [PP (P <0.001; r = 0.6) and WC (P <0.001, r = 0.5)], reduction of drinking on a drinking-day [PP (P = 0.027, r = 0.3) and WC (P = 0.23, r = 0.2)], frequency of drinking [PP (P <0.001, r = 0.6) and WC (P <0.001, r = 0.5)], and heavy drinking frequency [PP (P <0.001, r = 0.6) and WC (P <0.001, r = 0.4)]. There was a significant reduction in the craving scores [PP (P ≤0.001, r = 0.5) and WC (P <0.05, r = 0.4)]. The scores on the Coping Behavior Inventory also reduced significantly 3-months post-intervention [PP (P <0.001, r = 0.5) and WC (P <0.05, r = 0.4)]. Finally, only PP-analysis showed a higher proportion of participants [P = 0.006] in the action-stage of motivation following the intervention (Table 1). Fifteen (65.2%) patients were highly satisfied (TSS 32–40) with telemedicine (Table S1). Four (8.7%) participants relapsed. Stress and craving for alcohol were the major factors for relapse.
Table 1. Comparison of pattern of drinking and the median average score of coping behaviour and median general craving index (pre and post-intervention)
Table 1. Comparison of pattern of drinking and the median average score of coping behaviour and median general craving index (pre and post-intervention)Discussion
Our study showed that i-RPT as a stand-alone treatment (without medication support), is an acceptable, feasible and possibly effective treatment to reduce alcohol use, craving, improve motivation to quit and coping behaviour in individuals with alcohol dependence. All participants eligible for our study had consented to receive i-RPT, more than 70% completed all sessions, and a majority reported high satisfaction with telemedicine- suggesting high acceptability of our intervention. It was also possible to integrate telemedicine into routine clinical care. To our knowledge, this was the first study from a low-middle-income country and one of the handfuls of research examining the efficacy of telemedicine-based RPT [9-11].
Since previous studies on internet-based RPT tested a therapist-guided module rather than therapist-delivered i-RPT, the results of those studies are not entirely comparable with ours. Nevertheless, high-intensity therapist-guided i-RPT, in general, had better efficacy than unguided or self-guided RPT [20, 21]. The effectiveness of nurse-delivered i-RPT in our study supports the idea that higher involvement of the therapist may have better treatment outcome. The effect size of i-RPT was modest for alcohol use-related outcomes. Estimated effects were similar to previous internet-based RPT and with face-to-face RPT [5, 10].
Although we did not perform any mediation analysis (due to inadequate power to carry out such analysis), the improvement in coping and motivation, and reduction in alcohol craving may suggest the potential process of relapse prevention. Previous studies indicated the role of coping in predicting the risk of lapse and relapse, and improvement of coping skills to reduce the risk of relapse to alcohol [22, 23]. Higher craving also predicted a higher risk of relapse on alcohol [24]. Finally, all three factors, craving, coping and level of motivation, may interact with each other and influence the propensity of relapse- as per the reformulated cognitive-behaviour model or dynamic model of relapse [25].
Our study may encourage the clinicians and other professionals dealing with individuals with alcohol misuse to try internet-based RPT, especially when the patient is not willing to take a medication for maintaining abstinence and there are barriers to face-to-face treatment (e.g. travel restrictions and limited in-person contacts during the pandemic, distance from the clinic). The relatively low drop-out rate and high level of satisfaction with i-RPT should also guide the clinical decision.
Our study results should be considered preliminary because of their quasi-experimental design. We did not have a matched control group; hence, we are conservative in attributing the improved outcome to the effect of the intervention. Although we intended to reach a final sample of 35, we fell short of that goal; moreover, nine participants dropped out. Therefore, the study power may not have been adequate. Nevertheless, almost similar results in the per-protocol and worst-case analyses supported the preliminary effectiveness of i-RPT. All our research participants were men. This reflects the very low prevalence of AUD among women (men: women 17: 1) in India and treatment-seeking pattern [26, 27]. We can only comment on the efficacy of i-RPT to maintain short-term abstinence; however, 20–80% of treated samples who achieve short-term abstinence are estimated to relapse in 5 years [28]. Readers must exercise caution before generalizing our results to young adults and early-onset alcohol dependence as the large majority of our sample had late-onset alcohol dependence.
In sum, our study showed that a trained nurse can deliver internet-based RPT in a resource-limited setting; participants find it acceptable and it may reduce alcohol use in the short term. Future randomised clinical trials should examine the efficacy of i-RPT in a larger sample drawn from multiple centres, and with longer follow-up duration. Researchers may also like to explore several potential factors that could mediate the effect of i-RPT.
Conflict of Interest
The authors have no conflicts of interest.