Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 1.
Published in final edited form as: J Consult Clin Psychol. 2022 Aug;90(8):601–612. doi: 10.1037/ccp0000750

Mechanisms of quality-of-life improvement in treatment for alcohol use disorder

Hannah A Carlon 1, Margo C Hurlocker 1,*, Katie Witkiewitz 1
PMCID: PMC9899433  NIHMSID: NIHMS1864922  PMID: 36066862

Abstract

Objective:

For individuals in alcohol use disorder (AUD) treatment, many argue that holistic indicators such as quality of life (QoL) should be more consistently used in addition to drinking-related indicators. QoL increases from pre- to post-AUD treatment, but the mechanisms are unclear. The present study examined the roles of positive and negative affect in QoL change during AUD treatment, and additionally explored the relationship between QoL change and medication adherence.

Methods:

We examined the mediating roles of end-of-treatment positive affect (i.e., vigor) and negative affect (i.e., stress and tension) in the relationship between baseline and 26-week QoL among participants in the COMBINE study randomized to medication management [MM; n=468] or medication management plus combined behavioral intervention [MMCBI; n=479] for AUD. We also explored whether changes in QoL were associated with medication adherence.

Results:

Change in psychological health QoL was mediated by increased vigor (i.e., positive affect) and decreased stress, and change in environmental QoL was mediated by decreased stress. There were also differences by treatment group, with stress mediating changes in environmental QoL among participants in MM, and vigor mediating changes in psychological health QoL among participants in MMCBI. Medication adherence was not associated with greater QoL after controlling for post-treatment alcohol use.

Conclusions:

The present study identified potential mechanisms of QoL change in AUD treatment, thus contributing to the growing knowledge surrounding alternative indicators of treatment success for AUD treatment and recovery. Targeting affective states and stress during treatment may improve QoL and recovery outcomes for persons with AUD.

Keywords: alcohol use disorder, quality of life, positive affect, medication adherence

Introduction

Alcohol use disorder (AUD) is highly prevalent (i.e., approximately 14% of the U.S. population received a past-year AUD diagnosis in 2012–2013; McCabe et al., 2017) and a major cause of preventable death worldwide (Whiteford et al., 2013). Despite the existence of several empirically supported treatments for AUD, rates of returning to hazardous alcohol consumption (i.e., “relapse”) after completing AUD treatment remain high: around 67% of those who complete treatment will return to hazardous alcohol consumption within six months (Durazzo & Meyerhoff, 2017; Kirshenbaum et al., 2009; Maisto et al., 2006; Meyerhoff & Durazzo, 2010). Further, relapse to heavy drinking post-treatment is linked to longer periods of risky alcohol consumption and unfavorable physical and psychosocial consequences (Durazzo et al., 2008; Estruch et al., 2005; Maisto et al., 2006; Wood et al., 2018).

The focus on relapse has long dominated treatment and research on AUD, however over the last two decades there has been increasing focus on a chronic versus acute model of care, integration of behavioral and medical services, and a greater focus on resilience and well-being in treatment to address the chronically relapsing nature of substance use disorders (SUDs; Dennis & Scott, 2007; White, 2000). Among the many ways that recent research has sought to address the high rates of relapse and related impacts, investigations have grown on the effects of AUD treatment on psychosocial factors beyond just drinking reductions (e.g., percent days abstinent, drinks per drinking day) (e.g., Donovan et al., 2012; Kirouac & Witkiewitz, 2019). One such salient factor in the treatment of AUD that has gained traction is a greater focus on quality of life (QoL) (e.g., Kirouac et al., 2017).

Quality of Life and Alcohol Treatment

Considered to be a holistic indicator of wellbeing, QoL is defined as an individual’s “perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” (WHOQOL Group, 1997, p. 1). Individuals who meet criteria for an AUD typically score lower on measures of QoL (particularly health-related QoL) than those in the general population (Donovan et al., 2005; Volk et al., 1997), especially before treatment initiation (Srivastava & Bhatia, 2013). Additionally, low QoL has been linked to more alcohol-related problems and consequences (Patience et al., 1997). However, QoL is not consistently used as an indicator of treatment success in clinical research and practice surrounding AUD (Foster et al., 1999; Longabaugh et al., 1994). Currently, abstinence and no heavy drinking days are the only endpoints for AUD medication trials that are currently recommended by the Food and Drug Administration (Food and Drug Administration, 2015). And yet, QoL undoubtedly influences AUD treatment success, and vice-versa such that worse drinking outcomes tend to be associated with worse QoL (LoCastro et al., 2009), and decreased risky drinking yields better QoL scores in AUD treatment (Bold et al., 2017; Daeppen et al., 2014; Kraemer et al., 2002; Witkiewitz et al., 2018). Additional research has suggested that behavioral interventions for AUD, including those combined with anti-craving medication maintenance, have favorable impacts on drinking outcomes through QoL increases (Morgan et al., 2004; Prisciandaro et al., 2012). QoL also impacts individuals with AUD long into post-treatment recovery: higher QoL is associated with greater achievements during post-treatment recovery (Eddie et al., 2021), and relapse to heavy drinking post-treatment is associated with decreased QoL (Foster et al., 2000a). Yet, there are also mixed findings with respect to AUD outcomes and QoL. For example, one study found that those who relapsed post-treatment did not show any reductions in QoL (Picci et al., 2014), whereas another study found medium effect size improvements in QoL following alcohol treatment (Kirouac et al., 2017). In summary, the preponderance of evidence suggests that drinking and QoL are related throughout and following treatment (with the exception of Picci et al., 2014), however other psychosocial factors may help explain changes in QoL for individuals during and following AUD treatment.

Quality of Life: Mechanisms of Change

Prior work has identified QoL itself as a mechanism of change in alcohol use during treatment (Prisciandaro et al., 2012), but the mechanisms by which QoL confers AUD treatment success are somewhat unexplored in the literature. The scarce research that has examined mediators of QoL in AUD treatment has largely focused on more traditional outcomes, such as reductions in alcohol consumption during treatment (Foster et al., 2000b), baseline alcohol consumption, and other drug use (Lahmek et al., 2009). Exploring mechanisms of QoL change during and after AUD treatment could help identify modifiable factors to target in treatment, and could help explain why some individuals do, and some do not, show changes in QoL from pre- to post-treatment, and after relapsing post-treatment (Picci et al., 2014).

Positive and Negative Affect

Extant work surrounding the role of affect (defined as the experience of acute emotional states, often divided into positive and negative; Watson et al., 1988) in AUD treatment has largely focused on the effects of negative affect (e.g., tension, stress, anxiety, anger) on drinking related outcomes. Changes in negative affect has predicted both heavy drinking during treatment (Witkiewitz et al., 2011) and relapses after treatment completion (Foster et al., 2000a; Witkiewitz & Villarroel, 2009). Evidence shows that in particular, tension and stress are associated with AUD relapses via alterations of the hypothalamic and prefrontal-limbic-striatal pathways (Blaine & Sinha, 2017), and negative affect has been found to be inversely associated with QoL (He et al., 2019; Pokrajac-Bulian et al., 2015). Yet, to our knowledge, no studies have examined negative affect as a mediator of QoL change in the context of AUD treatment.

There also is a dearth of research examining the role of positive affect (e.g., happiness, optimism, excitement, contentment) in AUD treatment. The broaden-and-build theory of positive emotions posits that experiencing positive emotions allows one’s behavioral repertoire to broaden, making room for a new variety of thoughts, actions, and problem-solving ideas, thus increasing QoL (Fredrickson, 2004). Specifically, long-term QoL and well-being is achieved in this way through a theorized “upward spiral” effect: increased positive emotions leads to a broadened thought-action repertoire, leading to further positive emotions, and the cycle repeats (Fredrickson & Joiner, 2002, 2018). Randomized, controlled longitudinal studies have supported this upward spiral effect: individuals randomized to skills-based interventions targeting positive emotions have shown increases in personal resources 2.5 months post-treatment (Fredrickson et al., 2008), as well as improved physical health indices over time (Kok et al., 2013). Such effects have been associated with increased life satisfaction and decreased depression (Fredrickson et al., 2008), and the effects of positive emotion-targeting interventions tend to increase over time, further supporting this upward spiral nature (Fredrickson et al., 2008; Moskowitz et al., 2017). Based on this theory, negative affect serves only the evolutionary purpose of narrowing this same behavioral repertoire, similar to the way fear elicits a fight-or-flight response (Fredrickson & Levenson, 1998). Given the broaden-and-build theory and the lasting upward spiral effect of positive emotions on overall QoL, examining the role of positive and negative affect as mechanisms of QoL change in AUD treatment is pertinent. And, taking into consideration the role of QoL in preventing relapse (Foster et al., 2000a), better characterizing “why” QoL changes in AUD treatment can help us to identify treatment targets for those more at-risk of relapse after treatment.

Medication Adherence in Alcohol Treatment

Several medications designed to prevent craving and relapse (i.e., acamprosate and naltrexone) have proven vital to AUD treatment success (Zweben et al., 2008), yet poor adherence remains a central concern (Oslin et al., 2008; Swift et al., 2011). Negative affect, positive affect, and QoL have all been shown to impact medication adherence behaviors in a variety of populations, including individuals with HIV (Kalichman & Kalichman, 2016; Mugavero et al., 2009), post-traumatic stress disorder (Salas et al., 2020), and hypertension (Dyussenova et al., 2018; Kretchy et al., 2014). The relationship between QoL and medication adherence in AUD samples is unclear. Among participants in the Combined Pharmacotherapies and Behavioral Interventions for Alcohol Dependence (COMBINE) study, those who experienced adverse events during AUD treatment yielded worse medication adherence (Gueorguieva et al., 2013). Additionally, depression has been associated with naltrexone non-adherence in outpatient AUD treatment (Feeney et al., 2001). Taking into consideration the broaden-and-build theory, increased QoL would provide the cognitive flexibility surrounding problem-solving that is perhaps favorable to greater medication adherence in AUD treatment. On the counterbalance, decreased QoL would provide a lack of flexibility surrounding problem-solving, and is perhaps unfavorable for medication adherence. The association between QoL and medication adherence in AUD treatment remains relatively unexplored, but the broaden-and-build theory might provide a framework through which to study the associations between QoL and adherence.

The Present Study

The present study was a secondary analysis of the COMBINE study (Anton et al., 2006). The primary aim was to examine the mediating roles of end-of-treatment positive affect (i.e., vigor) and negative affect (i.e., stress and tension) in the relationship between baseline and 26-week (10 weeks post-treatment) QoL in a sample of individuals with AUD who received medication management (MM; n=468) or medication management plus combined behavioral intervention (MMCBI; n=479). Given the importance of medication adherence to pharmacological treatment success, we also explored whether medication adherence during AUD treatment predicted changes in QoL from baseline to post-treatment (26 weeks).

Methods

Participants

The COMBINE study, the largest pharmacotherapy study to date for the treatment of AUD, took place between 2001 and 2004. Participants were recruited from 11 outpatient treatment centers across the United States (See COMBINE Study Research Group, 2003 for a more detailed description of study methods). In short, eligible participants were adults who (1) met diagnostic criteria for AUD, and (2) were abstinent from alcohol for at least four days before being assigned to a treatment modality. Participants were assessed on drinking behavior and other mental health constructs at baseline, during-treatment, and at three follow-up time points (10 weeks, 36 weeks, and 52 weeks following treatment). The COMBINE study involved randomly assigning participants to one of nine different 16-week treatment conditions, but the current study consolidated treatment conditions into two groups: those who received medication only (MM; n=468) and those who received medication plus combined behavioral intervention (MMCBI; n=479). We excluded participants who only received CBI, given our exploratory focus on medication adherence in the present study. The sample for the current study was 69% male, 61% were employed full-time, and ~40% of the sample had a four-year college degree or higher. The racial and ethnic composition included 77% non-Hispanic White, 11% Hispanic, 8% Black or African American, 1% bi- or multi-racial, 1% American Indian or Alaska Native, <1% Asian American or Pacific Islander, and 1% other race/ethnicity. The hypotheses of the present secondary data analysis were not preregistered.

Measures

Quality of Life

QoL was assessed using the World Health Organization Quality of Life Brief Version (WHOQOL-BREF; WHOQOL Group, 1998). The WHOQOL-BREF is a self-report measure that includes 25 items which measure QoL across four domains: physical health (e.g., “To what extent do you feel that physical pain prevents you from doing what you need to do?”), psychological health (e.g., “How often do you have negative feelings such as blue mood, despair, anxiety, depression?”), social relationships (e.g., “How satisfied are you with the support you get from your friends?”), and environmental (e.g., “How satisfied are you with the conditions of your living place?”) quality of life. Responses are rated on a 5-item Likert scale (1 = “not at all,” 5 = “an extreme amount”), where higher scales indicate better QoL. The WHOQOL-BREF was administered to participants at baseline, 26 weeks (10 weeks post-treatment), and 52 weeks (36 weeks post-treatment). For the purpose of the current study, we focused on QoL scores at baseline and 26 weeks. The WHOQOL-BREF has demonstrated good to excellent reliability and validity (Skevington et al., 2004) and was further validated and found to produce reliable scores in the COMBINE study (Kirouac et al., 2017; LoCastro et al., 2009).

Positive Affect and Tension

The Profile of Mood States (McNair, 1992) was used to assess positive affect and tension. The POMS is a 30-item, self-report measure of fluctuating mood states and participants indicate the extent they have experienced each mood “in the past week including today” using a 5-point Likert scale (0 = “not at all,” 4 = “extremely”). Positive affect was captured using items within the vigor-activity subscale and included: “lively,” “active,” “energetic,” “full of pep,” and “vigorous.” Tension was captured using the tension-anxiety subscale and included: “tense,” “shaky,” “uneasy,” “nervous,” and “anxious.” The POMS was administered to participants at baseline and weeks 1, 2, 4, 8, 12, and 16 (end) of treatment. For the purpose of the current study, we focused on positive affect and tension scores at week 16 (i.e., end of treatment). In a previous psychometric validation study, this scale has demonstrated internal consistency across its subscales (.84–.95), test-retest reliability (.65–.74), and concurrent validity (McNair, 1992). Internal consistency was strong among our sample for the vigor and tension subscales at 16 weeks, with Cronbach’s alpha values of .93 and .88, respectively.

Stress

Stress was measured with the Perceived Stress Scale 4 (PSS-4; Cohen, 1988), a 4-item self-report measure that captures symptoms of stress within the last week (e.g., “In the last week, how often have you felt that you were unable to control the important things in your life?”). Items are rated on a 5-point Likert scale (0 = “never,” 4 = “very often”). The PSS-4 was administered at baseline and weeks 1, 2, 4, 6, 8, 10, 12, and 16 (end of treatment), as well as at 52 weeks (36 weeks post-treatment). For the purpose of the current study, we focused on stress scores at week 16 (i.e., end of treatment). While the 4-item version of the PSS has demonstrated less favorable reliability than its two longer counterparts (i.e., the original 14-item version (Cohen et al., 1983) and the 10-item version (Cohen, 1988), the PSS-4 was ascertained to be useful when perceived stress is to be measured briefly (Cohen, 1988), as is in the COMBINE study. The PSS-4 showed adequate internal consistency among our sample at 16 weeks (Cronbach’s alpha = .78).

Post-Treatment Alcohol Use

Post-treatment alcohol use was captured using the Form 90 interview (Miller & Del Boca, 1994) and consisted of a sum-score of number of drinks participants consumed between week 16 (end of treatment) and week 26 (10 weeks post-treatment). This study controlled for post-treatment alcohol use, given that prior research has demonstrated relationships between alcohol use and QoL and negative affect among those in the COMBINE study (Kirouac et al., 2017; LoCastro et al., 2009; Witkiewitz et al., 2011).

Medication Adherence

Medication adherence was captured with the Medication Non-Compliance checklist (MNC checklist), a 21-item staff administered form developed for the COMBINE study to identify reasons for non-adherence among those participates who did not take all pills as prescribed at a given visit. First, an item on the checklist reads “Patient took all pills as prescribed?”; if this item was answered “yes,” the rest of the checklist was then administered. The remaining checklist items were divided into two types of non-adherence: (1) intentional (e.g., “wanted to drink or use illicit drugs and not mix pills”), and (2) unintentional (e.g., “lost pills”). Total scores of each non-adherence type were created by summing all items, where a higher number on a subscale would indicate more reasons for medication non-adherence. The MNC checklist was administered to all participants at eight in-treatment visits: weeks 1, 2, 4, 6, 8, 10, 12, and 16 (end of treatment). A sum score of the preliminary general adherence item (“Patient took all pills as prescribed?”) was used as a marker of general medication adherence (i.e., 0 = no and 1 = yes, such that a score of 8 would indicate total adherence throughout all eight in-treatment visits). We excluded cases in analyses that were missing across any of the eight treatment time points for this item, with the understanding that a lower sum score due to missingness might not necessarily indicate lower medication adherence (e.g., a participant’s score might reflect a 7 due to study staff error during one time point, when the participant might have actually been adherent across all eight time points). This resulted in a final sample of 570 participants for exploratory medication adherence analyses. In addition to the general adherence item sum score, intentional and unintentional MNC sum scores were used in analyses.

Statistical Analyses

All analyses were conducted using Mplus version 8.1 (Muthén & Muthén, 2017). We conducted a multiple mediation model to test the direct and indirect associations among baseline QoL domains, end-of-treatment vigor, end-of-treatment stress, end of treatment tension, and 26-week QoL domains. Specifically, we modeled each of the four domains of QoL (physical health, psychological health, social relationships, and environmental QoL) at baseline as predictors, end-of-treatment positive affect, stress, and tension as the mediating variables, and the four QoL domains at 26 weeks as the outcomes. We controlled for post-treatment alcohol use, as well as baseline levels of vigor, tension, and stress. We additionally conducted a multi-group, multiple mediation model using the same predictors, covariates, mediating variables, and outcomes to examine differences in the structural parameters across treatment groups (i.e., MM versus MMCBI). Specifically, we tested for cross-group invariance by comparing two nested models – a baseline model with no specified constraints and a second model with all paths constrained to be invariant between groups. We used a chi-square difference test to compare nested models and determine if the multiple mediation model significantly differed between treatment groups. Several indicators were used to evaluate model fit across the proposed mediation models, including the chi-square statistic, comparative fit index (CFI), Tucker Lewis Index (TLI), and root mean square error of approximation (RMSEA). A good (adequate) fit was indicated with scores of .94 (.90–.94) or higher for CFI and TLI and .06 (.07–.08) or lower for RMSEA (Chen et al., 2008; Hu & Bentler, 1999). Significant mediation is indicated if the bootstrap confidence intervals of the indirect effect does not contain zero. We used two methods for estimating the effect size: (1) we calculated the amount of variance explained by the mediator(s) by dividing the products of the indirect paths by the total effect (Preacher & Hayes, 2004), and (2) additionally calculated the difference of coefficients (i.e., direct – total paths), given the former method can inflate effect size estimates with lower beta values (Preacher & Kelley, 2011). Given the importance of medication adherence to pharmacological treatment success, we also conducted regression analyses to explore the relationship between medication adherence (both general and intentional versus unintentional) and changes in QoL domains from baseline to 26 weeks. Medication adherence analyses comprised all participants who completed the MNC checklist, regardless of treatment group. Information regarding accessing data from the COMBINE study can be found at https://www.niaaa.nih.gov/combine-study.

Results

Descriptive analyses

Means, standard deviations, and intercorrelations of study variables in each treatment condition are displayed in Table 1. For main outcome mediation analyses, the final analytic sample consisted of 947 participants (MM=468; MMCBI=479). For exploratory medication adherence regression analyses, the final analytic sample was further tapered to 570 participants who were not missing QoL data at baseline or 26 weeks, nor medication adherence data at any of the eight treatment time points.

Table 1.

Means, standard deviations, and intercorrelations of study variables

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 MM MM
M SD
1. BL physical 1.00 .69** .42** .56** .34** −.39** −.49** .49** .40** .29** .42** .31** −.27** −.37** −.06 27.47 4.28
2. BL psychological .64** 1.00 .58** .59** .44** −.41** −.58** .41** .53** .38** .43** .36** −.32** −.41** −.03 21.13 4.01
3. BL social .45** .54** 1.00 .48** .26** −.24** −.39** .28** .34** .53** .33** .25** −.19** −.31** −.04 9.88 2.65
4. BL environment .55** .57** .43** 1.00 .26** −.31** −.51** .36** .34** .30** .63** .25** −.23** −.42** −.05 29.66 5.60
5. BL vigor .35** .45** .24** .30** 1.00 −.30** −.43** .31** .34** .21** .21** .47** −.11* −.18** −.02 8.51 4.44
6. BL tension −.35** −.34** −.26** −.28** −.29** 1.00 .52** −.28** −.31** −.19** −.22** −.10* .35** .22** .07 5.37 4.16
7. BL stress −.45** −.50** −.39** −.47** −.37** .55** 1.00 −.40** −.45** −.36** −.42** −.23** .31** .45** .08 5.83 3.00
8. 26-wk physical .57** .42** .29** .44** .29** −.34** −.44** 1.00 .74** .56** .66** .39** −.39** −.48** −.32** 29.26 4.24
9. 26-wk psychological .42** .58** .36** .43** .39** −.35** −.47** .73** 1.00 .67** .71** .47** −.43** −.54** −.24** 22.78 4.26
10. 26-wk social .33** .42** .58** .30** .21** −.23** −.35** .53** .63** 1.00 .58** .29** −.32** −.41** −.19** 10.93 2.63
11. 26-wk environment .42** .37** .28** .65** .26** −.32** −.46** .66** .66** .49** 1.00 .34** −.34** −.57** −.26** 31.09 5.64
12. 16-wk vigor .34** .37** .22** .29** .49** −.20** −.23** .41** .45** .31** .33** 1.00 −.31** −.42** −.16** 10.18 4.79
13. 16-wk tension −.28** −.25** −.16** −.25** −.11* .38** .27** −.48** −.44** −.35** −.44** −.33** 1.00 .61** .34** 2.90 3.37
14. 16-wk stress −.37** −.39** −.26** −.40** −.17** .25** .44** −.54** −.52** −.40** −.49** −.43** .63** 1.00 .33** 4.08 3.14
15. Alcohol use post-tx −.11* −.07 −.04 −.06 −.07 .03 −.02 −.25** −.31** −.20** −.20** −.17** .27** .22** 1.00 222.88 358.63
MMCBI M 27.06 20.92 9.76 29.76 8.19 5.40 5.78 29.04 22.69 10.74 31.55 10.11 3.01 4.14 186.32
MMCBI SD 4.22 3.92 2.58 5.21 4.44 3.98 2.83 4.37 4.25 2.54 5.42 4.77 3.46 3.11 298.90
Open in a new tab

Note. Correlations for the MMCBI condition are presented below the diagonal, whereas those for the MM condition are above the diagonal.

M=mean; SD=standard deviation

MM=medication management only condition; MMCBI=medication management plus combined behavioral therapy condition

*

p < .05

**

p < .01.

Multiple mediation model

The mediation model examined the mediating effects of positive affect (vigor) and negative affect (tension and stress) on the relationship between baseline and 26-week QoL domains, while accounting for post-treatment alcohol use and baseline positive and negative affect indicators (see model in Figure 1). Global fit statistics revealed a reasonable fit with the data (χ2 (79, N=947) = 22.432, p=.033, CFI=.998, TLI=.986, RMSEA=0.030 [0.008, 0.049]). Total, direct, and indirect parameter estimates are shown in Table 2. All four QoL domains at baseline were related to QoL at 26 weeks, such that QoL significantly increased from baseline to post-treatment (physical health QoL c=0.38, p<.01; psychological health QoL c=0.42, p<.01; social relationships QoL c=0.47, p<.01; environmental QoL c=0.55, p<.01). After accounting for the mediators in the model, the direct relationships of QoL weakened but remained significant across domains (physical health QoL c′=0.359, p<.01; psychological health QoL c′=0.355, p<.01; social relationships QoL c′=0.473, p<.01; environmental QoL c′=0.513, p<.01). Three significant partial mediation effects were identified. Vigor (β=.026, p=.004, 95% CI [.01, .04]) and stress (β=.021, p=.023, 95% CI [.01, .04]) partially mediated the direct association between baseline and post-treatment psychological health QoL, accounting for 6% of the variance, indicating a small effect size (Preacher & Kelley, 2011). Specifically, baseline psychological health QoL was positively related to end-of-treatment vigor (a=.16, p<.01), which in turn, predicted higher post-treatment psychological health QoL (b=.16, p<.01). In addition, baseline psychological health QoL was inversely related to end-of-treatment stress (a=−.14, p=.004), which, in turn, predicted lower post-treatment psychological health QOL (b=−.15, p<.01). Stress also partially mediated the relationship between baseline and post-treatment environmental QoL (β=.029, p=.002, 95% CI [.02, .05]), accounting for 4% of the variance, indicating a small effect size (Preacher & Kelley, 2011). Specifically, baseline environmental QoL was inversely related to stress (a=−.17, p<.01), which, in turn, predicted lower post-treatment environmental QoL (b=−.17, p<.01).

Figure 1.

Figure 1.

Open in a new tab

Observed multiple mediation model among entire sample. For parsimony, only significant, standardized parameter estimates are reported.

Note. ***p < .001, **p < .01, *p < .05

Table 2.

Total, direct, and indirect effects of baseline QoL on 26-week QoL (four domains)

Physical Health Psychological Health Social Relationships Environmental
β SE 95% CI β SE 95% CI β SE 95% CI β SE 95% CI
Total effect .39*** .03 .33, .43 .42*** .03 .37, .46 .47*** .03 .43, .52 .55*** .03 .51, .59
Direct effect .36*** .03 .31, .40 .36*** .03 .31, .40 .47*** .03 .43, .52 .51*** .03 .47, .55
Total indirect effect .03 .01 .00, .05 .06*** .02 .03, .09 −.00 .01 −.02, .02 .04** .01 .02, .06
Specific indirect effect
 Vigor .01 .01 .00, .02 .03** .01 .01, .04 .00 .00 −.01, .01 .00 .00 .00, .01
 Stress .02 .01 .00, .02 .02* .01 .01, .04 .00 .01 −.01, .01 .03** .01 .02, .05
 Tension .01 .01 .00, .02 .01 .01 .00, .03 −.00 .01 −.01, .01 .00 .00 −.00, .01
Open in a new tab

Note. SE=standard error; CI=confidence interval

***

p < .001

**

p < .01

*

p < .05

Multi-group, multiple mediation model

We next tested the same multiple mediation model as a function of treatment group (i.e., MM or MMCBI). When constraining the parameter estimates in the multiple mediation model to be equal across treatment groups, the overall model fit significantly worsened (Δχ2 = 986.07, Δdf = 82, p < .001), suggesting the path estimates (as a whole) are significantly different between MM and MMCBI. Global fit statistics of the baseline model revealed a good fit with the data (χ2 (158, N=947)=32.594, p=.1129, CFI=.998, TLI=.988, RMSEA=.03 [.00, .05]). Significant direct and indirect effects are shown in Figure 2. Similar to the multiple mediation model, total effects were significant across QoL domains and, after adding the mediators, direct effects weakened yet remained significant across domains for both groups. Two significant mediation effects were observed. Among those in the MM group, end-of-treatment stress significantly mediated the relationship between baseline and post-treatment environmental QoL (β=.051, p=.004, 95% CI [.03, .09]), accounting for 6% of the variance, indicating a small effect size (Preacher & Kelley, 2011). Baseline environmental QoL was negatively related to stress (a=−.174, p=.002), which, in turn, predicted lower post-treatment environmental QoL (b=−.293, p<.001). Among the MMCBI group, end-of-treatment vigor significantly mediated the association between baseline and post-treatment psychological health QoL (β=.022, p=.052, 95% CI [.01, .05]), accounting for 5% of the variance, indicating a small effect size (Preacher & Kelley, 2011). Baseline psychological health QoL was positively related to vigor (a=.18, p=.005), which, in turn, predicted higher post-treatment psychological health QoL (b=.12, p=.012).

Figure 2.

Figure 2.

Open in a new tab

Observed multi-group, multiple mediation model

Note. ***p < .001, **p < .01, *p < .05

Medication Adherence

We conducted regression analyses to examine the relationship between changes in QoL from baseline to 26 weeks and adherence to medication during treatment. The general adherence sum score (i.e., a sum of “Patient took pill as prescribed?” across eight treatment points) was calculated for 570 participants across both treatment groups (M=5.04, SD=2.37). Change in the four QoL domains (baseline to 26 weeks) were then entered as the predictor variables in regression analyses, with the medication adherence sum variable as the outcome. In a model without covariates, the change in environmental QoL from baseline to 26 weeks was significantly related to general medication adherence (β=.15, p=.048), such that greater environmental quality of life at 26 weeks was associated with more general medication adherence during treatment. However, when the models additionally controlled for the effects of alcohol use at post-treatment, the association between adherence and QoL was no longer significant. Finally, we fit the same model with intentional and unintentional medication adherence sum scores as the outcomes, but no significant relationships emerged when adherence was divided into these two categories.

Discussion

Predominant research support improvements in QoL during and following AUD treatment, yet the mechanisms of change in QoL are relatively unknown. The current study examined the mediating role of end-of-treatment positive and negative affect on changes in QoL domains from baseline to 26-week follow-up among individuals in the COMBINE study who underwent 16 weeks of either medication management only or medication management plus combined behavioral intervention for AUD. Results indicated that, even while controlling for post-treatment alcohol use, all four QoL domains improved from baseline to follow-up for the total sample. The multiple mediation model indicated that change in psychological health QoL was partially explained by increased vigor (i.e., positive affect) and decreased stress, and change in environmental QoL was partially explained by decreased stress. An examination of mediation effects across the two treatment groups revealed that stress explained changes in environmental QoL among participants who received only medication, and vigor explained changes in psychological health QoL among participants who received medication plus behavioral therapy. Additionally, the current study explored the associations between medication adherence and changes in QoL from baseline to follow-up. Medication adherence was not associated with QoL domains when controlling for post-treatment alcohol use.

The current findings that QoL increased across all four domains is in line with previous research using similar samples; QoL is known to increase during AUD treatment (Anton et al., 2006; LoCastro et al., 2009; Morgan et al., 2004; Prisciandaro et al., 2012). Additionally, these findings are consistent with those of LoCastro et al. (2009) in that QoL increases did not significantly differ between groups (i.e., MM versus MMCBI). Significant increases across QoL domains further evidences the point that treatment outcomes cannot be uniquely captured through only drinking-related indices.

Interestingly, the mediating effects of positive and negative affect only were identified among two QoL domains (i.e., psychological health and environmental). The finding that psychological health QoL was mediated by both positive and negative affect is perhaps expected, given that greater positive emotions and less negative emotions might naturally yield greater self-reported psychological functioning (e.g., Mauss et al., 2011). However, our finding that environmental QoL was mediated by stress is noteworthy, given that health-related QoL typically gains more attention in substance use research (despite literature that argues for non-health-related QoL to have a chance in the spotlight for these populations; e.g., Laudet et al., 2009). To uniquely focus on health-related QoL among addiction populations is to ignore other important areas of functioning (e.g., financial, work, or other environmental stressors), and it stands to reason that physical health QoL changes are better explained by physical health and behavioral variables (e.g., substance use, nutrition) rather than the affective variables examined in this study. Regarding social relationships QoL and its lack of significant mediation, previous research has found evidence that AUD treatment might not be effective in improving one’s quality of and satisfaction with social relationships. For example, one study found satisfaction with work, children, marriage, and sexual relationships did not change for individuals in treatment, suggesting that AUD treatment might not adequately address these aspects of wellbeing (Chodkiewicz, 2012). While the present study did find significant increases in the social relationships QoL domain, these increases were not mediated by affective variables. Perhaps there are additional non-treatment factors (e.g., social recovery capital) that contributed to the social relationships QoL increases observed in this study, given that typical AUD treatment might neglect this domain of wellbeing, and given its lack of mediation by affective variables. Non-health-related QoL has garnered the least attention of all domains, and thus is worthy of future investigation in studies of AUD treatment outcomes, including their mechanisms of change.

Prior research has posited that positive and negative affect are not in fact polar ends on a spectrum, but rather separate phenomena that are influenced by different factors (Sirgy, 2002). This is to say that factors that cause an increase in positive affect might not actually result in negative affect when there is an absence of the factors. In the case of the present study’s findings, there are perhaps additional psychosocial factors simultaneously associated with positive or negative affect and QoL domains. For example, vigor is only one specific aspect of positive affect, while psychological health QoL is only one aspect of QoL. It might be that a factor such as mindful awareness (i.e., awareness of the present moment) influences both the aforementioned variables, and thus helps to explain why vigor specifically mediated change in the psychological health QoL domain (Veneziani & Voci, 2015). Relatedly, multiple studies have found that increases in QoL are not stable over time, but rather vary during drinking stages and recovery (Amodeo et al., 1992; Chaturvedi et al., 1997), suggesting that QoL changes may have different mechanisms depending on one’s stage of recovery. The examination of additional psychosocial mechanisms of QoL changes is needed.

We found that positive affect only mediated psychological health QoL in the MMCBI group and not in the MM group. This finding is perhaps foreseen, given that participants in the MMCBI group underwent the behavioral intervention component of treatment. Previous research has shown that positive affect can be increased through more positive psychology-inspired behavioral interventions for AUD (Akhtar & Boniwell, 2010; Krentzman et al., 2015). While scarce research has examined positive affect in more typical behavioral treatment for AUD, the combined behavioral intervention (CBI) in the present study contained elements similar to strengths-based and positive psychology-inspired interventions. For example, phase two of the CBI involved identifying the participant’s personal strengths and resources that would aid in facilitating change in drinking in an effort to increase the participant’s optimism about reaching a change goal (COMBINE Study Research Group, 2003). This identification of personal strengths is a central element to many positive psychology interventions which have shown effectiveness in increasing positive affect (e.g., Kahler et al., 2015; Proyer et al., 2015; Seligman et al., 2005). Perhaps this element of the CBI assisted to confer positive affect-related benefits, resulting in the observed mediation effect. The finding that environmental QoL was mediated by stress in the MM group only is reasonable, as this group lacked the CBI component of treatment, and thus perhaps did not develop skills to cope with environmental stress during treatment that the behavioral intervention may have offered. Phase three of the CBI included nine possible modules that could be administered based on the therapist’s judgment and the client’s needs (COMBINE Study Research Group, 2003). One such module is titled “Coping with Cravings and Urges,” and incorporates techniques like urge surfing (Bowen et al., 2014; Marlatt & Donovan, 2005) to build coping strategies. A previous secondary analysis using COMBINE Study data found that those who received the “Coping with Cravings and Urges” module had weaker correlations between negative affect (e.g., stress) and alcohol dependence (Witkiewitz et al., 2011). Another module in phase three which possibly mitigated stress for those receiving the CBI was titled “Mood Management Training.” This module utilized cognitive behavioral techniques to manage negative mood states and responses to them (COMBINE Study Research Group, 2003). Participants who received these two modules as part of the CBI might have learned more skills to handle stress compared to those in the MM only group. Because partial mediation effects were modest, and different mediators were significant for different domains (e.g., positive affect only mediated psychological health QoL), it is likely that there are different psychosocial mechanisms of change for each QoL domain for individuals in AUD treatment. Further exploring the mechanisms of QoL change in AUD treatment is a worthy future investigation, given that QoL increases are significant and may support individuals in recovery post-treatment.

In exploratory regression analyses examining the relationship between medication adherence and QoL change, greater adherence was associated with greater environmental QoL, but not other QoL domains. These findings are in line with previous research: a prior analysis of the COMBINE study found that those who experienced more adverse life events showed lower adherence (Gueorguieva et al., 2013). Experiencing adverse or chaotic life events could conceivably hinder an increase in the environmental QoL domain, especially if such events are caused by unfavorable elements of one’s environment; our finding that environmental QoL was the only domain associated with medication adherence fits within this picture. However, this effect did not remain once post-treatment alcohol use was accounted for in the model. Research among non-AUD samples have found that poor medication adherence predicts relapse (e.g., Schoeler et al., 2017); it is likely that this relationship was more potent than the association between adherence and environmental QoL. Lastly, previous research with the COMBINE study examining the relationship between drinking risk reduction and health/QoL outcomes found that medication adherence did not affect the association between the aforementioned variables (Witkiewitz et al., 2018). These findings are comparable to those of the current study, given that we did not find significant relationships between medication adherence and health-related QoL (neither physical nor psychological), and given that the one significant regression path between environmental QoL change and adherence became non-significant after the model controlled for alcohol use.

Broadly, the present findings underscore the need to move beyond alcohol reduction when considering indices of AUD treatment success, and further supports a shift toward a recovery model that embraces holistic outcomes of what it means to “recover” from an AUD. In the present study, even controlling for post-treatment alcohol use, QoL significantly increased from pre- to post-treatment, and affective variables appeared to help explain some of this increase. Prior research shows mixed associations between alcohol relapse post-treatment and QoL (Foster et al., 2000; Picci et al., 2014), suggesting that alcohol-related variables cannot paint a full picture of recovery—the present study suggests that there is value in examining change in QoL to better capture treatment success.

Strengths and Limitations

The current study presents several strengths. Notably, this study adds to the growing body of work examining the role of QoL in alcohol treatment and recovery. Additionally, this study adds to the limited research on positive affect’s role in alcohol treatment outcomes, given the predominant focus on negative affect in research with AUD samples. This study also examined non-health-related QoL domains (in addition to two health-related domains) among an AUD sample, while much of the current literature gives complete attention to health-related QoL among those with chronic disorders such as AUD. Lastly, this study is the first to our knowledge to explore the relationship between QoL change and medication adherence in an AUD treatment sample.

The current study’s findings should be interpreted within the context of several limitations. Primarily, the COMBINE study’s inclusion criteria limits generalizability of our findings (e.g., participants were included if they were between 4 and 21 days abstinent from alcohol and excluded if they had a comorbid substance use disorder or psychiatric disorder requiring medication; Anton et al., 2006). Additionally, treatment only lasted 16 weeks with no relapse prevention aftercare, and participants were tasked with completing eight in-person assessments throughout the duration of treatment. Importantly, the COMBINE study was not designed to assess QoL or mechanisms of change in QoL. QoL data were not collected during the duration of treatment, either, which limits how fine-tuned we were able to characterize change in QoL from pre- to post-treatment. Primary outcomes in the COMBINE study were drinking-related (e.g., percent days abstinent, percent heavy drinking days), and the present study did not include drinking variables as outcomes, but rather controlled for the effects of drinking after treatment. While this was purposeful, given the present study’s focus on more holistic indicators of treatment success (i.e., QoL), omitting alcohol use as an outcome could be considered a limitation, given the nature of the COMBINE study. Finally, there was a fair amount of missing data. Such missingness on measures of QoL and mediators could have indicated much lower scores than what was included in the final analytic sample; for instance, missingness due to a skipped appointment could actually be indicative of an adverse life event that was not captured in the data by missingness.

Moreover, we have used the terms “mediator” and “mechanism of change” synonymously, yet it is imperative to note that statistical mediation and causal mechanisms of behavior change are not one in the same (Karno, 2007). Statistical mediators do not denote causality, while causal mechanisms of behavior change are thought to meet a more rigorous set of conditions to be deemed as such (i.e., Kazdin and Nock [2003; 2007] posit that for a variable to be suggested as causing a change in behavior, it must satisfy tests of strong association, specificity, gradient, experimental design, temporal relation, consistency, and plausible coherence). While vigor and stress did show evidence of being statistical mediators of QoL in the present study, more thorough tests would need to be conducted to identify vigor and stress as mechanisms of QoL change. Perhaps future studies might aim to fulfill the specificity requirement of identifying mechanisms of QoL change by conducting sensitivity analyses to elucidate biasing effects of possible omitted confounders (MacKinnon & Pirlott, 2015; Tofighi & Kelley, 2016). Fulfilment of the temporal relation criterion was also not achieved in this study, as we did not identify whether change in the mediators (i.e., positive and negative affect) in fact preceded change in the outcome (26-week QoL). Also, as this is the first study to our knowledge to identify mechanisms of QoL change, this research would need to be replicated to fulfill the consistency requirement.

Conclusion

AUD is a chronic disease and relapse following treatment is common. QoL is an indicator that can effectively quantify AUD’s effects on an individual’s psychosocial wellbeing and should be more consistently considered as an outcome variable in AUD treatment trials. The present study found evidence that positive affect and stress are possible mechanisms of QoL change in AUD treatment. Though effects were modest and should be interpreted in the context of the aforementioned limitations, the present study did identify promising potential mechanisms of QoL change in AUD treatment, and thus contributes to the growing area of knowledge surrounding alternative indicators of treatment success for this population. With further research, the field can better elucidate the processes by which QoL increases during AUD treatment.

Public health significance statement:

This study suggests that positive and negative affect, as well as decreased experiences of stress, help to explain why quality of life significantly increases for individuals after treatment for alcohol use disorder. Greater medication adherence, reductions in negative affect and stress, and increases in positive affect during treatment were all associated with better quality of life among persons with alcohol use disorder.

NIH Funding:

Time for manuscript preparation was provided by National Institutes of Health grant numbers: R01AA025539-05 (PI: K. Witkiewitz) and K23DA052646 (PI: M. Hurlocker).

References

  1. Akhtar M, & Boniwell I (2010). Applying positive psychology to alcohol-misusing adolescents. Groupwork, 20(3), 6–31. 10.1921/gpwk.v20i3.707 [DOI] [Google Scholar]
  2. Amodeo M, Kurtz N, & Cutter HSG (1992). Abstinence, reasons for not drinking, and life satisfaction. International Journal of the Addictions, 27(6), 707–716. 10.3109/10826089209068762 [DOI] [PubMed] [Google Scholar]
  3. Anton RF, O’Malley SS, Ciraulo DA, Cisler RA, Couper D, Donovan DM, Gastfriend DR, Hosking JD, Johnson BA, & LoCastro JS (2006). Combined pharmacotherapies and behavioral interventions for alcohol dependence: The COMBINE study: A randomized controlled trial. Journal of the American Medical Association, 295(17), 2003–2017. 10.1001/jama.295.17.2003 [DOI] [PubMed] [Google Scholar]
  4. Blaine SK, & Sinha R (2017). Alcohol, stress, and glucocorticoids: From risk to dependence and relapse in alcohol use disorders. Neuropharmacology, 122, 136–147. 10.1016/j.neuropharm.2017.01.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bold KW, Epstein EE, & McCrady BS (2017). Baseline health status and quality of life after alcohol treatment for women with alcohol dependence. Addictive Behaviors, 64, 35–41. 10.1016/j.addbeh.2016.08.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bowen S, Chawla N, & Witkiewitz K (2014). Mindfulness-based relapse prevention for addictive behaviors. In Mindfulness-Based Treatment Approaches (pp. 141–157). Academic Press. 10.1016/B978-0-12-416031-6.00007-4 [DOI] [Google Scholar]
  7. Chaturvedi SK, Kirthana R, & Desai M (1997). Quality of life and alcoholism: What aspects are perceived as important? Quality of Life Research, 632. [Google Scholar]
  8. Chen F, Curran PJ, Bollen KA, Kirby J, & Paxton P (2008). An empirical evaluation of the use of fixed cutoff points in RMSEA test statistic in structural equation models. Sociological Methods & Research, 36(4), 462–494. 10.1177/0049124108314720 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chodkiewicz J (2012). Quality of life and therapy effects among alcoholics. Polish Journal of Applied Psychology, 10(2), 49–65. [Google Scholar]
  10. Cohen S (1988). Perceived stress in a probability sample of the United States. In Spacapan S & Oskamp S (Eds.), The social psychology of health (pp. 31–67). Sage Publications, Inc. [Google Scholar]
  11. Cohen S, Kamarck T, & Mermelstein R (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396. 10.2307/2136404 [DOI] [PubMed] [Google Scholar]
  12. COMBINE Study Research Group. (2003). Testing combined pharmacotherapies and behavioral interventions in alcohol dependence: Rationale and methods. Alcoholism: Clinical and Experimental Research, 27(7), 1107–1122. 10.1111/j.1530-0277.2003.tb02873.x [DOI] [PubMed] [Google Scholar]
  13. Daeppen J-B, Faouzi M, Sanchez N, Rahhali N, Bineau S, & Bertholet N (2014). Quality of life depends on the drinking pattern in alcohol-dependent patients. Alcohol and Alcoholism, 49(4), 457–465. 10.1093/alcalc/agu027 [DOI] [PubMed] [Google Scholar]
  14. Dennis M, & Scott CK (2007). Managing addiction as a chronic condition. Addiction Science & Clinical Practice, 4(1), 45. 10.1151/ascp074145 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Donovan DM, Bigelow GE, Brigham GS, Carroll KM, Cohen AJ, Gardin JG, Hamilton JA, Huestis MA, Hughes JR, & Lindblad R (2012). Primary outcome indices in illicit drug dependence treatment research: Systematic approach to selection and measurement of drug use endpoints in clinical trials. Addiction, 107(4), 694–708. 10.1111/j.1360-0443.2011.03473.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Donovan D, Mattson ME, Cisler RA, Longabaugh R, & Zweben A (2005). Quality of life as an outcome measure in alcoholism treatment research. Journal of Studies on Alcohol, Supplement, s15, 119–139. 10.15288/jsas.2005.s15.119 [DOI] [PubMed] [Google Scholar]
  17. Durazzo TC, Gazdzinski S, Yeh P-H, & Meyerhoff DJ (2008). Combined neuroimaging, neurocognitive and psychiatric factors to predict alcohol consumption following treatment for alcohol dependence. Alcohol & Alcoholism, 43(6), 683–691. 10.1093/alcalc/agn078 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Durazzo TC, & Meyerhoff DJ (2017). Psychiatric, demographic, and brain morphological predictors of relapse after treatment for an alcohol use disorder. Alcoholism: Clinical and Experimental Research, 41(1), 107–116. 10.1111/acer.13267 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Dyussenova L, Pivina L, Semenova Y, Bjørklund G, Glushkova N, Chirumbolo S, & Belikhina T (2018). Associations between depression, anxiety and medication adherence among patients with arterial hypertension: Comparison between persons exposed and non-exposed to radiation from the Semipalatinsk Nuclear Test Site. Journal of Environmental Radioactivity, 195, 33–39. 10.1016/j.jenvrad.2018.09.016 [DOI] [PubMed] [Google Scholar]
  20. Eddie D, White WL, Vilsaint CL, Bergman BG, & Kelly JF (2021). Reasons to be cheerful: Personal, civic, and economic achievements after resolving an alcohol or drug problem in the United States population. Psychology of Addictive Behaviors, 35(4), 402–414. 10.1037/adb0000689 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Estruch R, Coca A, & Rodicio JL (2005). High blood pressure, alcohol and cardiovascular risk. Journal of Hypertension, 23(1), 226–229. 10.1097/00004872-200501000-00039 [DOI] [PubMed] [Google Scholar]
  22. Feeney GFX, Connor JP, Young RM, Tucker J, & Czajkowski F (2001). Adherence with naltrexone prescription advice in hospital outpatient alcohol rehabilitation programme. Journal of Clinical Pharmacy and Therapeutics, 26(1), 73–79. 10.1111/j.1365-2710.2001.00326.x [DOI] [PubMed] [Google Scholar]
  23. Food and Drug Administration. (2015). Alcoholism: Developing Drugs for Treatment. In Center for Drug Evaluation and Research (CDER). [Google Scholar]
  24. Foster JH, Marshall EJ, & Peters TJ (n.d.). Application of a quality of life measure, the life situation survey (LSS), to alcohol-dependent subjects in relapse and remission. Alcoholism: Clinical and Experimental Research, 24(11), 1687–1692. 10.1111/j.1530-0277.2000.tb01970.x [DOI] [PubMed] [Google Scholar]
  25. Foster JH, Peters TJ, & Marshall EJ (2000). Quality of life measures and outcome in alcohol-dependent men and women. Alcohol, 22(1), 45–52. 10.1016/S0741-8329(00)00102-6 [DOI] [PubMed] [Google Scholar]
  26. Foster JH, Powell JE, Marshall EJ, & Peters TJ (1999). Quality of life in alcohol-dependent subjects–a review. Quality of Life Research, 8(3), 255–261. 10.1023/A:1008802711478 [DOI] [PubMed] [Google Scholar]
  27. Fredrickson BL (2004). The broaden–and–build theory of positive emotions. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 359(1449), 1367–1377. 10.1098/rstb.2004.1512 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Fredrickson BL, Cohn MA, Coffey KA, Pek J, & Finkel SM (2008). Open hearts build lives: positive emotions, induced through loving-kindness meditation, build consequential personal resources. Journal of Personality and Social Psychology, 95(5), 1045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Fredrickson BL, & Joiner T (2002). Positive emotions trigger upward spirals toward emotional well-being. Psychological Science, 13(2), 172–175. [DOI] [PubMed] [Google Scholar]
  30. Fredrickson BL, & Joiner T (2018). Reflections on positive emotions and upward spirals. Perspectives on Psychological Science, 13(2), 194–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Fredrickson BL, & Levenson RW (1998). Positive emotions speed recovery from the cardiovascular sequelae of negative emotions. Cognition & Emotion, 12(2), 191–220. 10.1080/026999398379718 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Gueorguieva R, Wu R, Krystal JH, Donovan D, & O’Malley SS (2013). Temporal patterns of adherence to medications and behavioral treatment and their relationship to patient characteristics and treatment response. Addictive Behaviors, 38(5), 2119–2127. 10.1016/j.addbeh.2013.01.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. He S, Zhu J, Jiang W, Ma J, Li G, & He Y (2019). Sleep disturbance, negative affect and health-related quality of life in patients with maintenance hemodialysis. Psychology, Health & Medicine, 24(3), 294–304. 10.1080/13548506.2018.1515493 [DOI] [PubMed] [Google Scholar]
  34. Hu L, & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. 10.1080/10705519909540118 [DOI] [Google Scholar]
  35. Kahler CW, Spillane NS, Day AM, Cioe PA, Parks A, Leventhal AM, & Brown RA (2015). Positive psychotherapy for smoking cessation: A pilot randomized controlled trial. Nicotine & Tobacco Research, 17(11), 1385–1392. 10.1093/ntr/ntv011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Kalichman SC, & Kalichman MO (2016). HIV-related stress and life chaos mediate the association between poverty and medication adherence among people living with HIV/AIDS. Journal of Clinical Psychology in Medical Settings, 23(4), 420–430. 10.1007/s10880-016-9481-8 [DOI] [PubMed] [Google Scholar]
  37. Karno MP (2007). A case study of mediators of treatment effectiveness. Alcoholism: Clinical and Experimental Research, 31, 33s–39s. 10.1111/j.1530-0277.2007.00491.x [DOI] [PubMed] [Google Scholar]
  38. Kazdin AE, & Nock MK (2003). Delineating mechanisms of change in child and adolescent therapy: Methodological issues and research recommendations. Journal of Child Psychology and Psychiatry, 44(8), 1116–1129. 10.1111/1469-7610.00195 [DOI] [PubMed] [Google Scholar]
  39. Kirouac M, Stein ER, Pearson MR, & Witkiewitz K (2017). Viability of the World Health Organization quality of life measure to assess changes in quality of life following treatment for alcohol use disorder. Quality of Life Research, 26(11), 2987–2997. 10.1007/s11136-017-1631-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kirouac M, & Witkiewitz K (2019). Predictive value of non-consumption outcome measures in alcohol use disorder treatment. Addiction, 114(6), 1086–1092. 10.1111/add.14553 [DOI] [PubMed] [Google Scholar]
  41. Kirshenbaum AP, Olsen DM, & Bickel WK (2009). A quantitative review of the ubiquitous relapse curve. Journal of Substance Abuse Treatment, 36(1), 8–17. 10.1016/j.jsat.2008.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Kok BE, Coffey KA, Cohn MA, Catalino LI, Vacharkulksemsuk T, Algoe SB, Brantley M, & Fredrickson BL (2013). How positive emotions build physical health: Perceived positive social connections account for the upward spiral between positive emotions and vagal tone. Psychological Science, 24(7), 1123–1132. [DOI] [PubMed] [Google Scholar]
  43. Kraemer KL, Maisto SA, Conigliaro J, McNeil M, Gordon AJ, & Kelley ME (2002). Decreased alcohol consumption in outpatient drinkers is associated with improved quality of life and fewer alcohol-related consequences. Journal of General Internal Medicine, 17(5), 382–386. 10.1046/j.1525-1497.2002.10613.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Krentzman AR, Mannella KA, Hassett AL, Barnett NP, Cranford JA, Brower KJ, Higgins MM, & Meyer PS (2015). Feasibility, acceptability, and impact of a web-based gratitude exercise among individuals in outpatient treatment for alcohol use disorder. The Journal of Positive Psychology, 10(6), 477–488. 10.1080/17439760.2015.1015158 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Kretchy IA, Owusu-Daaku FT, & Danquah SA (2014). Mental health in hypertension: assessing symptoms of anxiety, depression and stress on anti-hypertensive medication adherence. International Journal of Mental Health Systems, 8(1), 1–6. 10.1186/1752-4458-8-25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lahmek P, Berlin I, Michel L, Berghout C, Meunier N, & Aubin H-J (2009). Determinants of improvement in quality of life of alcohol-dependent patients during an inpatient withdrawal programme. International Journal of Medical Sciences, 6(4), 160–167. 10.7150/ijms.6.160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Laudet AB, Becker JB, & White WL (2009). Don’t wanna go through that madness no more: Quality of life satisfaction as predictor of sustained remission from illicit drug misuse. Substance Use & Misuse, 44(2), 227–252. 10.1080/10826080802714462 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. LoCastro JS, Youngblood M, Cisler RA, Mattson ME, Zweben A, Anton RF, & Donovan DM (2009). Alcohol treatment effects on secondary nondrinking outcomes and quality of life: The COMBINE Study. Journal of Studies on Alcohol and Drugs, 70(2), 186–197. 10.15288/jsad.2009.70.186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Longabaugh R, Mattson ME, Connors GJ, & Cooney NL (1994). Quality of life as an outcome variable in alcoholism treatment research. Journal of Studies on Alcohol, Supplement, s12, 119–129. 10.15288/jsas.1994.s12.119 [DOI] [PubMed] [Google Scholar]
  50. MacKinnon DP, & Pirlott AG (2015). Statistical approaches for enhancing causal interpretation of the M to Y relation in mediation analysis. Personality and Social Psychology Review, 19(1), 30–43. 10.1177/1088868314542878 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Maisto SA, Clifford PR, Stout RL, & Davis CM (2006). Drinking in the year after treatment as a predictor of three-year drinking outcomes. Journal of Studies on Alcohol, 67(6), 823–832. 10.15288/jsa.2006.67.823 [DOI] [PubMed] [Google Scholar]
  52. Marlatt GA, & Donovan DM (2005). Relapse prevention: Maintenance strategies in the treatment of addictive behaviors. Guilford press. [Google Scholar]
  53. Mauss IB, Shallcross AJ, Troy AS, John OP, Ferrer E, Wilhelm FH, & Gross JJ (2011). Don’t hide your happiness! Positive emotion dissociation, social connectedness, and psychological functioning. Journal of Personality and Social Psychology, 100(4), 738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. McCabe SE, West BT, Jutkiewicz EM, & Boyd CJ (2017). Multiple DSM-5 substance use disorders: A national study of US adults. Human Psychopharmacology: Clinical and Experimental, 32(5), e2625. 10.1002/hup.2625 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. McNair DM (1992). Profile of mood states. Educational and Industrial Testing Service. [Google Scholar]
  56. Meyerhoff DJ, & Durazzo TC (2010). Predicting relapse and treatment outcome in alcohol use disorders: Time for an integrative biopsychosocial approach. Alcoholism: Clinical and Experimental Research, 34, 289A–290A. [Google Scholar]
  57. Miller WR, & Del Boca FK (1994). Measurement of drinking behavior using the Form 90 family of instruments. Journal of Studies on Alcohol, Supplement, 12, 112–118. [DOI] [PubMed] [Google Scholar]
  58. Morgan MY, Landron F, Lehert P, & Group NEATS (2004). Improvement in quality of life after treatment for alcohol dependence with acamprosate and psychosocial support. Alcoholism: Clinical and Experimental Research, 28(1), 64–77. 10.1097/01.ALC.0000108652.73143.4B [DOI] [PubMed] [Google Scholar]
  59. Moskowitz JT, Carrico AW, Duncan LG, Cohn MA, Cheung EO, Batchelder A, Martinez L, Segawa E, Acree M, & Folkman S (2017). Randomized controlled trial of a positive affect intervention for people newly diagnosed with HIV. Journal of Consulting and Clinical Psychology, 85(5), 409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Mugavero MJ, Raper JL, Reif S, Whetten K, Leserman J, Thielman NM, & Pence BW (2009). Overload: The impact of incident stressful events on antiretroviral medication adherence and virologic failure in a longitudinal, multi-site HIV cohort study. Psychosomatic Medicine, 71(9), 920. 10.1097/PSY.0b013e3181bfe8d2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Muthén B, & Muthén L (2017). Mplus. Chapman and Hall/CRC. [Google Scholar]
  62. Nock MK (2007). Conceptual and design essentials for evaluating mechanisms of change. Alcoholism: Clinical and Experimental Research, 31, 4s–12s. 10.1111/j.1530-0277.2007.00488.x [DOI] [PubMed] [Google Scholar]
  63. Oslin DW, Lynch KG, Pettinati HM, Kampman KM, Gariti P, Gelfand L, Ten Have T, Wortman S, Dundon W, & Dackis C (2008). A placebo-controlled randomized clinical trial of naltrexone in the context of different levels of psychosocial intervention. Alcoholism: Clinical and Experimental Research, 32(7), 1299–1308. 10.1111/j.1530-0277.2008.00698.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Patience D, Buxton M, Chick J, Howlett H, McKENNA M, & Ritson B (1997). The Seccat Survey: II. The alcohol related problem questionnaire as a proxy for resource costs and quality of life in alcoholism treatment. Alcohol and Alcoholism, 32(1), 79–84. 10.1093/oxfordjournals.alcalc.a008237 [DOI] [PubMed] [Google Scholar]
  65. Picci RL, Oliva F, Zuffranieri M, Vizzuso P, Ostacoli L, Sodano AJ, & Furlan PM (2014). Quality of life, alcohol detoxification and relapse: Is quality of life a predictor of relapse or only a secondary outcome measure? Quality of Life Research, 23(10), 2757–2767. 10.1007/s11136-014-0735-3 [DOI] [PubMed] [Google Scholar]
  66. Pokrajac-Bulian A, Kukić M, & Bašić-Marković N (2015). Quality of life as a mediator in the association between body mass index and negative emotionality in overweight and obese non-clinical sample. Eating and Weight Disorders-Studies on Anorexia, Bulimia and Obesity, 20(4), 473–481. 10.1007/s40519-015-0208-x [DOI] [PubMed] [Google Scholar]
  67. Preacher KJ, & Hayes AF (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36(4), 717–731. 10.3758/BF03206553 [DOI] [PubMed] [Google Scholar]
  68. Preacher KJ, & Kelley K (2011). Effect size measures for mediation models: quantitative strategies for communicating indirect effects. Psychological Methods, 16(2), 93. 10.1037/a0022658 [DOI] [PubMed] [Google Scholar]
  69. Prisciandaro JJ, DeSantis SM, & Bandyopadhyay D (2012). Simultaneous modeling of the impact of treatments on alcohol consumption and quality of life in the COMBINE study: A coupled hidden Markov analysis. Alcoholism: Clinical and Experimental Research, 36(12), 2141–2149. 10.1111/j.1530-0277.2012.01823.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Proyer RT, Gander F, Wellenzohn S, & Ruch W (2015). Strengths-based positive psychology interventions: A randomized placebo-controlled online trial on long-term effects for a signature strengths-vs. a lesser strengths-intervention. Frontiers in Psychology, 6, 456. 10.3389/fpsyg.2015.00456 [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Salas J, Scherrer JF, Tuerk P, van den Berk-Clark C, Chard KM, Schneider FD, Schnurr PP, Friedman MJ, Norman SB, & Cohen BE (2020). Large posttraumatic stress disorder improvement and antidepressant medication adherence. Journal of Affective Disorders, 260, 119–123. 10.1016/j.jad.2019.08.095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Schoeler T, Petros N, Di Forti M, Klamerus E, Foglia E, Murray R, & Bhattacharyya S (2017). Poor medication adherence and risk of relapse associated with continued cannabis use in patients with first-episode psychosis: a prospective analysis. The Lancet Psychiatry, 4(8), 627–633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Seligman MEP, Steen TA, Park N, & Peterson C (2005). Positive psychology progress: Empirical validation of interventions. American Psychologist, 60(5), 410. 10.1037/0003-066X.60.5.410 [DOI] [PubMed] [Google Scholar]
  74. Sirgy MJ (2002). The psychology of quality of life (Vol. 12). Dordrecht: Kluwer Academic Publishers. [Google Scholar]
  75. Skevington SM, Lotfy M, & O’Connell KA (2004). The World Health Organization’s WHOQOL-BREF quality of life assessment: Psychometric properties and results of the international field trial. A report from the WHOQOL group. Quality of Life Research, 13(2), 299–310. 10.1023/B:QURE.0000018486.91360.00 [DOI] [PubMed] [Google Scholar]
  76. Srivastava S, & Bhatia MS (2013). Quality of life as an outcome measure in the treatment of alcohol dependence. Industrial Psychiatry Journal, 22(1), 41. 10.4103/0972-6748.123617 [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Swift R, Oslin DW, Alexander M, & Forman R (2011). Adherence monitoring in naltrexone pharmacotherapy trials: A systematic review. Journal of Studies on Alcohol and Drugs, 72(6), 1012–1018. 10.15288/jsad.2011.72.1012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Tofighi D, & Kelley K (2016). Assessing omitted confounder bias in multilevel mediation models. Multivariate Behavioral Research, 51(1), 86–105. 10.1080/00273171.2015.1105736 [DOI] [PubMed] [Google Scholar]
  79. Veneziani CA, & Voci A (2015). The Italian adaptation of the Mindful Awareness Attention Scale and its relation with individual differences and quality of life indexes. Mindfulness, 6(2), 373–381. 10.1007/s12671-013-0270-2 [DOI] [Google Scholar]
  80. Volk RJ, Cantor SB, Steinbauer JR, & Cass AR (1997). Alcohol use disorders, consumption patterns, and health-related quality of life of primary care patients. Alcoholism: Clinical and Experimental Research, 21(5), 899–905. 10.1111/j.1530-0277.1997.tb03855.x [DOI] [PubMed] [Google Scholar]
  81. Watson D, Clark LA, & Tellegen A (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. 10.1037/0022-3514.54.6.1063 [DOI] [PubMed] [Google Scholar]
  82. White W (2000). Toward a new recovery movement: Historical reflections on recovery, treatment and advocacy. Recovery Community Support Program (RCSP) Conference. [Google Scholar]
  83. Whiteford HA, Degenhardt L, Rehm J, Baxter AJ, Ferrari AJ, Erskine HE, Charlson FJ, Norman RE, Flaxman AD, & Johns N (2013). Global burden of disease attributable to mental and substance use disorders: Findings from the Global Burden of Disease Study 2010. The Lancet, 382(9904), 1575–1586. 10.1016/S0140-6736(13)61611-6 [DOI] [PubMed] [Google Scholar]
  84. Group WHOQOL. (1997). Measuring quality of life. Geneva: The World Health Organization, 1–13. [Google Scholar]
  85. Group WHOQOL. (1998). Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychological Medicine, 28(3), 551–558. 10.1017/S0033291798006667 [DOI] [PubMed] [Google Scholar]
  86. Witkiewitz K, Bowen S, & Donovan DM (2011). Moderating effects of a craving intervention on the relation between negative mood and heavy drinking following treatment for alcohol dependence. Journal of Consulting and Clinical Psychology, 79(1), 54. 10.1037/a0022282 [DOI] [PMC free article] [PubMed] [Google Scholar]
  87. Witkiewitz K, Kranzler HR, Hallgren KA, O’Malley SS, Falk DE, Litten RZ, Hasin DS, Mann KF, & Anton RF (2018). Drinking risk level reductions associated with improvements in physical health and quality of life among individuals with alcohol use disorder. Alcoholism: Clinical and Experimental Research, 42(12), 2453–2465. 10.1111/acer.13897 [DOI] [PMC free article] [PubMed] [Google Scholar]
  88. Witkiewitz K, & Villarroel NA (2009). Dynamic association between negative affect and alcohol lapses following alcohol treatment. Journal of Consulting and Clinical Psychology, 77(4), 633. 10.1037/a0015647 [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Wood AM, Kaptoge S, Butterworth AS, Willeit P, Warnakula S, Bolton T, Paige E, Paul DS, Sweeting M, & Burgess S (2018). Risk thresholds for alcohol consumption: Combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. The Lancet, 391(10129), 1513–1523. 10.1016/S0140-6736(18)30134-X [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Zweben A, Pettinati HM, Weiss RD, Youngblood M, Cox CE, Mattson ME, Gorroochurn P, & Ciraulo D (2008). Relationship between medication adherence and treatment outcomes: The COMBINE study. Alcoholism: Clinical and Experimental Research, 32(9), 1661–1669. 10.1111/j.1530-0277.2008.00743.x [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES