Baseline characteristics of the sample (N = 19) are presented in Table 2. The sample consisted of middle-aged subjects: 53% women and 84% white. Most had at least a high school diploma (74%) and were employed at least half-time (68%). At enrollment, they reported receiving IOP for 23.7 (SD = 11.2) days and abstinence for 30.9 (SD = 22.2) days. Before IOP (4–20 weeks before enrollment), subjects reported abstinence (PDA) during 48% (SD = 30.2) of days; on drinking days (52% [SD = 9.1] of days), they usually drank heavily (HDD = 12.7 [SD = 8.9] of 28 days), with average consumption of 115 (SD = 89.9) drinks per 28 days. Almost half of the subjects had participated in another outpatient therapy, and more than one-third had undergone at least 1 alcohol detoxification and/or residential treatment in the past. There were no significant gender differences in the prestudy alcohol consumption and baseline clinical characteristics.

Four subjects (2 men) dropped out early in the study, after the first or second meditation session, and were lost to follow-up. Overall, the drop-outs did not substantially differ from the rest of the sample on main baseline clinical and sociodemographic characteristics.

Table 3 describes clinical characteristics of the study completers at baseline and during the study, and Figure 1 shows individual drinking trajectories over time. Baseline values reflect subjects’ scores at the end and/or immediately after the IOP therapy; 4- and 8-week values reflect scores during the meditation course; and 12- and 16-week values refer to “postcourse” follow-up. At 16 weeks, a majority of the subjects reported improved or unchanged PDA, HDD, and Total Drinks values (80%, 87%, and 80% of the subjects, respectively) compared with baseline (Fig. 1A–C). Statistical analysis confirmed that some aspects of subjects’ alcohol consumption improved (Table 3). PDA showed a tendency to improve during the meditation course (8 weeks: P = 0.074, Wilcoxon test), but by 16 weeks PDA values returned to baseline. The number of HDD showed a trend to improve during the study (P = 0.056, repeated measures analysis), with the main changes seen during the meditation course (weeks 4, 8: P < 0.05, Wilcoxon test). Compared with baseline, Total Drinks decreased significantly at 4 and 8 weeks (P < 0.05, Wilcoxon test), but the overall change during the study period did not reach statistical significance (repeated measures analysis; Table 3). During the 16-week study, subjects were abstinent on 94.5% (SD = 7.4) of study days, with 47% (N = 7) reporting complete abstinence, and 47% reporting 1 or more HDD.

Subjects’ severity of relapse triggers: depression, anxiety, stress (P < 0.05, repeated measures analysis), and craving (4, 8, 16 weeks: P ≤ 0.08, t test) decreased during the study (repeated measures analysis), with prepost effect sizes ranging from “moderate” to “large” (Table 3).

Compared with baseline, IL-6 level decreased at 16 weeks (P = 0.052, t test), which corresponded to a moderate effect size (Table 3). Although cortisol levels (N = 10) also were lower, neither those changes nor changes in liver enzymes (N = 12) were significant (P > 0.05, Wilcoxon test, Table 3).

Adherence to the meditation protocol was good. Study completers attended 82% of meditation course sessions and meditated on average 4.6 (SD = 1.1) days per week during the study, with maximum “meditating” time at 4 and 8 weeks (Table 3). Compared with the initial 4-week values, subjects decreased their at-home meditation practice at 12 and 16 weeks (P < 0.05, Wilcoxon test); this decrease mostly affected length of daily meditation practice (minutes per day, P < 0.05; minutes per meditating day, P = 0.066; Friedman test) rather than number of days per week spent on meditation (Table 3). At 16 weeks, all subjects reported meditating at home, with 53% meditating 4 or more days per week. During the study, subjects’ degree of mindfulness increased (P < 0.05, repeated measures analysis, Table 3).

Participation in “standard of care” sessions for alcohol dependence was reported by 67% of subjects through 8 weeks and by 20% of subjects at 16 weeks. This therapy was supplemented by Alcoholics Anonymous meetings in approximately half of the subjects (Table 3).

Using the Likert scale, subjects rated the meditation course as very important to them in general (8.7/10, SD = 1.8), useful relapse prevention tool (8.5/10, SD = 2.1), and reported being very likely to continue meditating (9/10, SD = 1.5). “Gaining skills to reduce stress,” “real-life skills for coping with craving,” and “good group support” were the most common qualitative themes reported as the most valuable aspects of meditation training. All subjects reported that meditation-related skills were “brand-new” to them and different from those learned during previous addiction treatments. In the 16-week exit survey, subjects continued to report high satisfaction with and usefulness of meditation as a therapy for their alcohol addiction (Table 4). In qualitative exit responses, Mini-Meditation²⁴ was the most commonly identified form of “the most useful” meditative technique (Table 4). No subject reported dissatisfaction, adverse events, or side effects associated with meditation therapy.

Bivariate correlations between drinking and meditation-related outcomes and variables of interest were assessed. At 16 weeks, PDA, HDD, and Total Drinks correlated to severity of craving (r = −0.76, 0.77, 0.76, P < 0.01), anxiety (r = −0.73, 0.74, 0.74, P < 0.01), and depression (r = −0.47, 0.46, 0.47, P = 0.08). Intensity of daily meditation practice (minutes/day) showed a negative association with severity of craving (r = −0.49, P = 0.06), anxiety (r = −0.7, P < 0.01), depression (r = −0.65, P < 0.01), and stress (r = −0.62, P < 0.05). Degree of mindfulness negatively correlated to the prelunch cortisol level (r = −0.78, P < 0.01). Change in outcome measures (16 weeks versus baseline): Change in PDA and Total Drinks correlated to change in craving (r = −0.55, P < 0.05; and r = 0.49, P = 0.06) and stress severity (r = −0.46, P = 0.08; and r = 0.47, P = 0.08). Change in degree of mindfulness was associated with change in stress severity (r = −0.72, P < 0.001). Change in biomarker levels did not correlate to change in drinking or meditation-related outcomes.

Subjects with at least 1 HDD during the study (N = 7), compared with those with no HDD (N = 8), reported more anxiety (all time points, P < 0.05), depression (12 and 16 weeks, P ≤ 0.05), craving (16 weeks, P < 0.01), lower degree of mindfulness (P < 0.05), and fewer meditation minutes per day (12.6 [SD = 12.3] versus 27.8 [SD = 19] minutes per day at 12 weeks, P < 0.05).

In this study, alcohol-dependent adults in early recovery who used meditation as adjunctive therapy reported 1) continued reduction of some aspects of their drinking, 2) improved mental health and stress-related outcomes during the study, 3) high level of satisfaction with meditation intervention, and 4) meditation intervention was a helpful therapeutic tool during their recovery. This study is the first to assess meditation as an adjunctive treatment for alcohol relapse prevention among alcoholics in a community setting, using both behavioral outcomes and biologic markers. The methods proved to be feasible and would likely be appropriate for use in a larger study. Recruitment, retention, compliance, and subject treatment satisfaction suggest high interest in meditation and feasibility of the intervention protocol in this population. These findings are comparable to those reported by other meditation studies of other types of patients.⁸

Several findings warrant additional comments. Subjects reported clinical improvement. Alcohol consumption decreased during the study, particularly during the meditation course, as assessed by PDA, Total Drinks, and HDD. These positive changes were achieved despite the fact that subjects were largely abstinent at baseline (PDA = 89%) after successful graduation from IOP. Our findings of subjects being abstinent on 94.5% of study days, with 47% reporting complete abstinence or at least 1 HDD, are consistent with those of large randomized trials of behavioral and/or pharmacological therapies for alcohol dependence at 12 or 16 week follow-up.^39,40

A decrease in severity of alcohol relapse triggers also represents a positive change. The SCL-90R anxiety and depression scores can be predictive of alcohol relapse.⁴¹ Although stress, anxiety, and depression scores were elevated at baseline, they decreased during the study and trended toward normative values.^30,32 A decrease in craving severity is consistent with improved drinking outcomes, suggesting progressive subject recovery; the OCDS have been shown to be responsive to treatment and may reflect subject drinking status and relapse risk.^33,34 This pilot study is the first study to evaluate craving severity among subjects with addiction treated with meditation. The results of our study are consistent with reports of positive meditation effects for stress, anxiety, depression,^8–11 and limited evidence for SUDs,^13–18 and suggest the possibility of efficacy of a meditation intervention as an adjunctive therapy for alcohol dependence. These results, combined with results of bivariate correlations and examination of differences between subjects who reported at least 1 HDD during the study and those who did not, also support the hypothesis that meditation may contribute to positive treatment outcomes in alcohol dependence through positive modification of relapse trigger severities.

This is the first study to examine IL-6 among meditation-treated subjects with SUDs. Interleukin-6, a stress-responsive and illness-sensitive cytokine, decreased during the study (P = 0.05), suggesting a positive change, because elevated IL-6 level is a predictor of poor health outcomes and mortality in adults,⁴² as well as disease severity and patient mortality in alcoholic hepatitis.⁴³ The decrease in IL-6 level may represent natural progression of the recovery process; early in the recovery, IL-6 level is elevated, but with abstinence, it decreases.⁴⁴ The lack of correlation between the change in drinking outcomes (or mental health outcomes) and change in IL-6 level suggests that decreased IL-6 levels were not a simple effect of a drinking pattern change. Small sample size and/or duration of follow-up may explain the lack of significance in the changes in cortisol and liver enzyme concentrations. Strong links between alcohol dependence and stress, a known relapse trigger,²⁰ make stress management behavioral therapies particularly attractive. Individuals in recovery can benefit from stress management skills. Meditation has been associated with stress reduction, improved cortisol regulation (including in subjects with SUDs¹⁴), and immunomodulation.¹⁰ These findings, along with our pilot results suggest that meditation may have contributed to positive effects on IL-6 in alcohol-dependent adults during recovery.

Degree of mindfulness reflects the extent to which a subject’s way of thinking is consistent with “mindfulness.” It has been shown to correlate with greater perceived well-being.³⁷ As expected, the degree of mindfulness, lower at baseline than normative values, increased during the study.³⁷ Our results suggest that subjects’ ways of thinking could have become more mindful as a result of meditation practice. This hypothesis is supported by correlations between the degree of mindfulness and stress-responsive biomarker, lunch cortisol concentration, and between change in the degree of mindfulness and change in stress severity. It also is supported by the observation that subjects with ≥1 HDD had a significantly lower degree of mindfulness and fewer meditation minutes per day at 12 weeks than other subjects.

Our subjects, well familiar with the strategies of “best practice” addiction treatment, appreciated meditation therapy. They reported that skills learned during the meditation course were “brand-new” to them in addition to being “important” and “helpful.” Patient satisfaction may contribute to treatment retention, which has been shown to increase likelihood of abstinence.^3,4 Meditation-related skills may expand the existing treatment strategies for alcohol dependence. Training in specific relapse prevention strategies and meditation training and practice can teach subjects to form an association between meditation and relapse prevention skills, and enhance skills acquired through cognitive behavioral therapy,^7,22 ideally improving the outcomes of interest. Meditation therapy has been documented to have long-lasting positive effects that can be maintained.⁸ It also seems to be safe; recent studies have not reported side effects or adverse events and a majority of individuals who completed a meditation course reported that the program “had a lasting value” and was highly important,⁸ results replicated in this pilot study.

This study is limited by the lack of a control group and its small sample size. Without an appropriate control group, we cannot know to what extent, or whether, meditation, participation in an intervention trial, effects of concomitant “standard of care” therapy and/or “natural course” of recovery over time contributed to subjects’ improvement. The strengths of this pilot include verification of the appropriateness of the outcome measures, strong subject involvement in the follow-up measures, and high adherence to the meditation protocol in this study population.

This is the first study to assess meditation as an adjunctive treatment for relapse prevention among alcohol dependent adults and, thus, also the first using behavioral outcomes and biologic markers. Although the results of this pilot study suggest that meditation may be efficacious in reducing drinking and decreasing severity of relapse triggers in recovering alcoholics, a more definitive, randomized trial that compares meditation to “best practice” is needed. The methods used in the current study were demonstrated to be feasible and appropriate for use in that type of a larger trial.