Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 May 1.
Published in final edited form as: Addict Behav. 2021 Jan 8;116:106822. doi: 10.1016/j.addbeh.2021.106822

A Measure of Expectancies for Alcohol Analgesia: Preliminary Factor Analysis, Reliability, and Validity

Lisa R LaRowe 1, Stephen A Maisto 1, Joseph W Ditre 1
PMCID: PMC7887050  NIHMSID: NIHMS1661257  PMID: 33460990

Abstract

Rates of alcohol consumption are substantially higher among persons with pain, and recent research has focused on elucidating bidirectional pain-alcohol effects. Expectancies for alcohol analgesia could influence the degree to which alcohol confers acute pain-relieving effects, and may amplify the propensity to respond to pain with drinking behavior. However, no validated measures of expectancies for alcohol analgesia are available. Therefore, we developed a five-item measure of Expectancies for Alcohol Analgesia (EAA), which assesses the perceived likelihood that alcohol will reduce pain. The goal of this project was to examine psychometric properties of the EAA among a sample of 273 current alcohol users with chronic pain (Mage = 32.9; 34% female) who completed an online survey of pain and substance use. Confirmatory factor analysis (CFA) results indicated that the hypothesized single-factor structure of the EAA provided good model fit (Bollen-Stine bootstrap p = .13). The EAA also showed excellent internal consistency (α = .97), and scores were positively associated with quantity/frequency of alcohol use, alcohol outcome expectancies, coping-related drinking motives, and pain severity (ps < .01). These findings provide initial support regarding the single-factor structure, reliability, and validity of the EAA. Examination of predictive utility and further validation are important next steps.

Keywords: pain, alcohol, analgesia, expectancies

1. Introduction

Epidemiological estimates indicate that individuals who suffer from chronic musculoskeletal pain are twice as likely to meet diagnostic criteria for alcohol dependence, when compared to their pain-free counterparts (Von Korff et al., 2005). Similarly, the prevalence of pain appears to be substantially higher among problem drinkers (vs. non-problem drinkers; Brennan, Schutte, & Moos, 2005), and up to three-quarters of substance use treatment patients who identify alcohol as their drug of choice also report moderate-to-severe past-month pain (Larson et al., 2007). Given the high co-occurrence of pain and alcohol use, recent work has begun to elucidate bidirectional effects in pain-alcohol relations (e.g., Ditre, Zale, & LaRowe, 2019; Edwards, Vendruscolo, Gilpin, Wojnar, & Witkiewitz, 2020; Zale, Maisto, & Ditre, 2015).

An established reciprocal model posits that pain and alcohol use interact in the manner of a positive feedback loop, resulting in the exacerbation and maintenance of both conditions over time (Ditre et al., 2019; Zale et al., 2015). Although alcohol can confer acute analgesia (Thompson, Oram, Correll, Tsermentseli, & Stubbs, 2017), excessive alcohol consumption is associated with the onset and severity of numerous painful conditions. For example, heavy alcohol use is a causal factor in the development of alcohol-induced pancreatitis (Lerch et al., 2003) and alcohol-related neuropathy (Chopra & Tiwari, 2012), and may increase the risk of developing osteoarthritis (Cheng et al., 2000) and pain following musculoskeletal injury (Sá, Baptista, Matos, & Lessa, 2008). There is also converging research indicating that pain can increase motivation to drink alcohol. For example, laboratory pain induction increases self-reported urge to consume alcohol (Moskal, Maisto, De Vita, & Ditre, 2018), greater levels of pain unpleasantness have been associated with increased motivation to drink (Lawton & Simpson, 2009), nearly one-quarter of patients enrolled in both pain treatment and inpatient substance abuse programs have endorsed using alcohol to cope with pain (Goebel et al., 2011; Sheu et al., 2008), and pain intensity has been positively associated with alcohol coping motives (Rogers, Zegel, Tran, Zvolensky, & Vujanovic, 2020). Indeed, acute alcohol analgesia may negatively reinforce alcohol use and strengthen beliefs about the pain-relieving effects of alcohol (Ditre et al., 2019). Importantly, using alcohol to reduce pain can lead to increased consumption over time (Brennan et al., 2005).

Ditre and colleagues (2019) proposed that bidirectional pain-alcohol effects are likely influenced by outcome expectancies (i.e., estimates that a given behavior will lead to specific outcomes), which are considered to be important determinants of motivation and behavior (e.g., Bandura, 1989; Rotter, 1954). There is a vast literature documenting the role of outcome expectancies in the initiation, progression, and maintenance of alcohol use, and several measures have been developed to assess alcohol outcome expectancies (e.g., Brown, Christiansen, & Goldman, 1987; Fromme, Stroot, & Kaplan, 1993; Leigh & Stacy, 1993; Solomon & Annis, 1989). These measures assess both general and specific alcohol outcome expectancies, across a variety of domains (e.g., social facilitation, tension reduction, cognitive impairments). Higher scores on measures of positive alcohol outcome expectancies (i.e., estimates that alcohol use will result in desired consequences) have consistently been correlated with greater drinking motives and quantity/frequency of alcohol consumption (e.g., Jones, Corbin, & Fromme, 2001; Madden & Clapp, 2019; Monk & Heim, 2013). In contrast, negative alcohol outcome expectancies (i.e., estimates that alcohol use will result in undesired consequences) have been associated with reduced consumption and a greater desire to restrain from drinking (e.g., Jones et al., 2001; Monk & Heim, 2013). Given that positive (vs. negative) outcomes of alcohol use are often more immediate, it has been suggested that positive outcome expectancies are often more influential on drinking behavior (e.g., Stacy, Widaman, & Marlatt, 1990). Moreover, it has also been noted that the predictive utility of expectancy measures further improves with greater specificity of measurement (e.g., Fromme et al., 1993).

Previous work has demonstrated the importance of assessing specific expectancies that substance use will reduce pain. For example, a measure of pain and smoking expectancies (PSE; Ditre, 2006) was developed to assess beliefs that cigarette smoking will have acute analgesic effects. The PSE has demonstrated excellent internal consistency (Ditre, 2006; Ditre, Heckman, Butts, & Brandon, 2010), and has been shown to account for nearly one-third of the variance in pain-induced urge to smoke cigarettes (Parkerson & Asmundson, 2016). Alcohol users likely hold similar expectations for pain relief (Ditre et al., 2019; Zale et al., 2015), which may increase propensity to respond to actual or anticipated pain with drinking behavior. Consistent with evidence that positive alcohol outcome expectancies are correlated with greater quantity/ and frequency of alcohol consumption (Jones et al., 2001), it is also possible that expectancies for alcohol analgesia may lead to increased alcohol consumption and the development/maintenance of hazardous drinking patterns. Moreover, an accumulating literature demonstrates that analgesic outcome expectancies can influence the experience of pain (e.g., Atlas & Wager, 2014; Bingel et al., 2011; Butcher & Carmody, 2012; Ossipov, Dussor, & Porreca, 2010; Peerdeman, van Laarhoven, Peters, & Evers, 2016), and expectancies for alcohol analgesia may influence the degree to which alcohol use confers acute pain-relieving effects. However, we are not aware of any validated measure of alcohol outcome expectancies for pain relief.

We developed a measure of Expectancies for Alcohol Analgesia (EAA) to assess the perceived likelihood that alcohol consumption will reduce pain. The EAA consists of 5 items that were adapted from the PSE and are hypothesized to reflect a single-factor. All authors, who are experts in the domain of pain and substance use, reviewed and approved the adapted items for content validity. The primary goal of this study was to conduct an initial evaluation of the EAA factor structure, reliability, and validity among a sample of current alcohol users with chronic musculoskeletal pain. We hypothesized that the EAA would demonstrate (1) a single-factor structure, (2) acceptable internal consistency (α > .7), (3) initial evidence of concurrent validity via medium-to-large sized correlations with outcomes related to both alcohol consumption and clinical pain experience, and (4) initial evidence of discriminant validity via the absence of associations with cannabis use (a theoretically distinct construct). An exploratory aim of this study was to assess whether EAA scores differed as a function of sociodemographic characteristics (e.g., gender, race) and/or the presence of a high level of alcohol problems.

2. Method

2.1. Participants

Participants included 300 alcohol users who were recruited to complete an online survey of pain and alcohol use behaviors via Amazon Mechanical Turk. Participants were included if they were at least 21 years-old and a current resident of the United States, and endorsed any past-month alcohol use and current chronic musculoskeletal pain. Participants were excluded if they were younger than 21 years of age, resided outside of the United States, or did not endorse past-month alcohol use and current chronic musculoskeletal pain. We also included a response accuracy check (“To monitor quality, please respond with a two for this item”), and participants who responded incorrectly to this item were excluded from analyses (n = 27). Thus, the final sample consisted of N = 273 participants.

2.2. Procedure/Online Survey

All study procedures were completed using Amazon Mechanical Turk. A recruitment post was used to invite Mechanical Turk “workers” to participate in a survey on pain and substance use behaviors. Prospective participants completed a screening questionnaire to determine eligibility. Participants were asked to indicate their age and country of residence (a dropdown menu of all possible responses was provided). Past-month alcohol use and the presence of current chronic musculoskeletal pain were also assessed. Eligible participants were then invited to complete the brief (~40 minutes) online survey. Mechanical Turk has been shown to offer advantages that can reduce costs and increase recruitment feasibility (Ipeirotis, 2010). Prior work has also shown that the accuracy and representativeness of data collected from Mechanical Turk samples are similar to that of traditional participant pools (e.g., universities; Ipeirotis, 2010; Paolacci, Chandler, & Ipeirotis, 2010).

2.3. Measures

2.3.1. Demographic Variables.

Participants were asked to report sociodemographic information, including age, gender, race, ethnicity, education, employment status, and annual income.

2.3.2. Expectancies for Alcohol Analgesia.

Expectations for alcohol-related pain inhibition were assessed using a newly developed measure of Expectancies for Alcohol Analgesia (EAA; see Appendix). The EAA includes five items that assess the perceived likelihood that drinking alcohol will reduce or help one cope with pain, and is hypothesized to have a single-factor structure. Items are rated on a scale ranging from 0 (completely unlikely) to 9 (completely likely), and responses are summed to generate a total score (possible range: 0–45). The format (Likert-type likelihood scale) and item phrasing (i.e., first-person, hypothetical) of the EAA are comparable to widely-used measures of alcohol outcome expectancies (e.g., Fromme et al., 1993; Leigh & Stacy, 1993), and, consistent with recommendations (e.g., Morean, Corbin, & Treat, 2012), each item only assesses one anticipated effect of alcohol. Moreover, the items are face valid, do not include reverse-scored items, and have a Flesch-Kincaid Grade-Level of 4.1. All authors, who are experts in the domain of pain and substance use, reviewed and approved the adapted items for content validity.

2.3.3. Quantity and Frequency of Alcohol Consumption.

Patterns of alcohol consumption over the past 3 months were assessed using the Modified Daily Drinking Questionnaire (DDQ-M; Dimeff, 1999), and the Quantity-Frequency-Variability Questionnaire (QFV; Cahalan, Cisin, & Crossley, 1969). The DDQ-M allows for calculation of average number of drinks consumed each day, average number of hours spent drinking each day, and frequency of binge drinking (≥ 5 drinks within a couple of hours of each other). The QFV yields categorical classifications of alcohol use behavior (i.e., abstainers, infrequent, light, moderate, and heavy drinkers). Both the DDQ-M and the QFV are valid and reliable instruments that are commonly used in research examining patterns of drinking behavior (e.g., Carey, Henson, Carey, & Maisto, 2009; Simons, Maisto, Wray, & Emery, 2016).

2.3.4. Hazardous and Harmful Patterns of Alcohol Consumption.

The 10-item Alcohol Use Disorders Identification Test (AUDIT) was used to assess hazardous and harmful patterns of drinking (Babor, Higgins-Biddle, Saunders, & Monteiro, 1992). Items were summed to generate a total score, with scores ≥ 16 indicating a high level of alcohol problems (Babor, Higgins-Biddle, Saunders, & Monteiro, 2001). The AUDIT also includes three subscales: AUDIT-Consumption (assesses quantity/frequency of alcohol use), AUDIT- Harmful Use (assesses for drinking that results in consequences to physical and mental health), and AUDIT-Dependence (assesses for drinking that has resulted in dependence/addiction). Previous work has consistently demonstrated the reliability and validity of the AUDIT (e.g., Reinert & Allen, 2002).

2.3.5. Alcohol Outcome Expectancies.

Alcohol outcome expectancies were assessed using the 34-item Alcohol Outcome Expectancies Scale (AOES; Leigh & Stacy, 1993). This measure assesses two global factors (positive and negative outcome expectancies), as well as eight sub-factors, including four positive alcohol outcome expectancies (i.e., social facilitation, fun, sex, and tension reduction), and four negative outcome expectancies (social, emotional, physical, and cognitive performance). This valid and reliable measure has previously been shown to predict drinking behavior (Leigh & Stacy, 1993).

2.3.6. Drinking Motives.

Motives for drinking alcohol were assessed using the Revised Drinking Motives Questionnaire (DMQ-R; Cooper, 1994; Martin, Ferreira, Haase, Martins, & Coelho, 2016). The DMQ-R is 20-item measure that assesses various reasons for drinking using a 5-point Likert scale ranging from 1 (almost never/never) to 5 (almost always/always). Four different motives were assessed, including negative reinforcement/coping, social, enhancement, and conformity motives. The DMQ-R has previously been shown to predict greater alcohol use, risky drinking, and alcohol-related problems (Kuntsche, Stewart, & Cooper, 2008).

2.3.7. Clinical Pain Variables.

The Graded Chronic Pain Scale (GCPS; Von Korff, Ormel, Keefe, & Dworkin, 1992) provides a categorical classification of chronic pain by grade (severity) that ranges from Grade 1 (low intensity, low interference) to Grade 4 (severe interference). The GCPS also provides measures of characteristic pain intensity and pain-related disability. Consistent with scoring instructions (Von Korff et al., 1992), the characteristic pain intensity score was computed by summing ratings (0 = no pain to 10 = pain as bad as it could be) of pain “right now,” “on average,” and at its “worst” in the past three months. The pain-related disability score was computed by summing responses from three items assessing the extent to which pain has interfered with daily functioning over the past 3 months (0 = no interference to 10 = unable to carry on any activities) and one item reflecting the number of days that has interfered with usual activities (0 = none to 10 = 76–90 days). The GCPS is reliable and valid measure of chronic pain in both clinical and non-clinical samples (Von Korff, 2011). Finally, participants were asked to indicate whether or not they were currently taking prescription opioid medication.

2.3.8. Cannabis Use.

First, participants responded to a single yes/no item: “Have you used any cannabis (marijuana) over the past six months?” Participants who answered “yes” were then prompted to complete the Cannabis Use Disorder Identification Test – Revised (CUDIT-R; Adamson et al., 2010). The CUDIT-R is an eight item questionnaire that assesses the problematic cannabis use within the past 6 months. Higher scores are indicative of more hazardous cannabis use. If a participant denied past 6-month cannabis use, their CUDIT-R score was coded as 0. Both past 6-month cannabis use (yes/no) and CUDIT-R scores were used as indices of discriminant validity.

2.4. Data Analytic Plan

All statistical analyses were conducted using IBM SPSS Statistics and Amos 24. First, responses to each EAA item were examined for univariate and multivariate normality. Second, a confirmatory factor analysis (CFA) was fit to the data to confirm the hypothesized five-item, one-factor structure of the EAA. CFA (vs. exploratory factor analysis) was used given the empirical basis for specifying a single-factor model (e.g., high internal consistency and hypothesized unidimensionality of the PSE; Ditre, 2006; Ditre et al., 2010; Parkerson & Asmundson, 2016), and because CFA reduces the likelihood of benefitting from chance characteristics of the data (Fabrigar, Wegener, MacCallum, & Strahan, 1999). Consistent with recommendations (e.g., Hu & Bentler, 1999; Matsunaga, 2010; Schreiber, Nora, Stage, Barlow, & King, 2006), model fit was determined by examining: standardized root mean square residual (SRMR; < .08), root mean square error of approximation (RMSEA; < .10), comparative fit index (CFI; ≥ .90), and non-normed fit index (NNFI; ≥ .95). Third, internal consistency of the EAA was tested using Cronbach’s alpha (Santos, 1999). Fourth, we examined bivariate/point-biserial correlations between EAA scores and (1) past-month alcohol consumption patterns (DDQ-M and QFV scores), (2) problematic alcohol use (AUDIT scores), (3) alcohol outcome expectancies (AOES scores), (4) negative reinforcement/coping drinking motives (DMQ-R scores), (5) clinical pain variables (GCPS scores, current prescription opioid use), and (6) cannabis use (past 6-month cannabis use status [yes/no] and CUDIT-R scores). Finally, we examined associations between EAA total scores and sociodemographic factors (i.e., gender, age, race, ethnicity, marital status, income, education) and the presence of a high level of drinking problems (AUDIT score ≥ 16) using bivariate correlations (for continuous variables) and analysis of variance (ANOVA; for categorical/dichotomous variables).

3. Results

3.1. Participant Characteristics

Participants included 273 alcohol users with chronic musculoskeletal pain (34.4% female; 36.3% non-white; 18.7% Hispanic; Mage = 32.9, SD = 9.2, range: 22–66). The sample was generally well-educated (67.8% completed at least a 4-year college degree), and almost half (48%) reported a total household income greater than $50,000. Participants reported drinking approximately 1.6 alcoholic beverages each day (SD = 1.4), and nearly half (46.5%) scored above the AUDIT cut-off for high level of drinking problems (M = 15.5, SD = 11.1). The most commonly endorsed pain locations were back/neck (43.6%), head/face (18.7%), and lower extremities (14.7%), and nearly half of the sample (46.3%) reported that their current pain problem has lasted for over 1 year. The majority of participants (65.9%) endorsed either Grade 3 or Grade 4 chronic pain, indicating high levels of pain-related disability. Nearly one-quarter of the sample (23.8%) reported current prescription opioid use. Additional participant characteristics are presented in Table 1.

Table 1.

Sociodemographic, Alcohol, and Pain Characteristics (N = 273)

N (%)
Gender
 Male 179 (65.6%)
Race
 White 174 (63.7%)
 Black or African American 32 (11.7%)
 Asian 50 (18.3%)
 American Indian/Alaska Native 10 (3.7%)
 Other 6 (2.2%)
Ethnicity
 Hispanic 51 (18.7%)
Marital Status
 Single 136 (49.8%)
 Married 125 (45.8%)
 Divorced 12 (4.4%)
Education
 Did not graduate high school 1 (0.4%)
 High school graduate or GED 16 (5.9%)
 Some college/Technical school/Associates degree 71 (26.0%)
 4-year college degree 157 (57.5%)
 Some school beyond 4-year college degree 11 (4.0%)
 Professional degree (e.g., MD, JD, PhD) 17 (6.2%)
Household Income
 < $10,000 12 (4.4%)
 $10,000 – $49,999 130 (47.6%)
 $50,000 – $100,000 121 (44.3%)
 > $100,000 10 (3.7%)
Chronic Pain Grade a
 Grade 1 50 (18.3%)
 Grade 2 43 (15.8%)
 Grade 3 72 (26.4%)
 Grade 4 108 (39.6%)
Primary Pain Location
 Back/neck 119 (43.6%)
 Head/face 51 (18.7%)
 Upper extremities 29 (10.6%)
 Lower extremities 40 (14.7%)
 Chest/breast 12 (4.4%)
 Stomach/abdomen 22 (8.1%)
Prescription Opioid Use
 Yes 65 (23.8%)
M (SD)
Age 32.86 (9.24)
Average daily drinks 1.57 (1.43)
AUDIT – total score b 15.48 (11.10)
EAA score c 25.75 (12.81)
Open in a new tab

Note.

a

Graded Chronic Pain Scale

b

Alcohol Use Disorders Identification Test

c

Expectancies for Alcohol Analgesia.

3.2. Confirmatory Factor Analysis

The skewness and kurtosis values for each item fell within acceptable limits (< |2.0|; Table 2), and no univariate outliers were identified. However, even after excluding multivariate outliers (n = 9; identified via Mahalanobis distance; e.g., Blunch, 2012), the data remained multivariate non-normal (Mardia’s multivariate kurtosis coefficient = 12.59). Of note, multivariate outliers did not differ from non-outliers on any pain or alcohol variable (all ps > .2). Given multivariate non-normality, we utilized naïve bootstrapping with 2000 samples to obtain parameter estimates, adjusted standard errors, and confidence intervals, and the Bollen-Stine bootstrap χ2 test statistic to gauge model-fit (Bollen & Stine, 1992; Hoyle, 2012). Additional model-fit indices (i.e., SRMR, RMSEA, CFI, NNFI) were estimated using a maximum likelihood estimation procedure, which is remarkably robust even when there is departure from multivariate normality (e.g., Olsson, Foss, Troye, & Howell, 2000). There were no missing data.

Table 2.

Item Characteristics and Intercorrelations for the Expectancies for Alcohol Analgesia Scale (EAA)

Variable Min Max Skew Kurtosis 1 2 3 4 5
1 EAA_1 .000 9.000 −.633 −.805 1.000
2 EAA_2 .000 9.000 −.595 −.849 .868** 1.000
3 EAA_3 .000 9.000 −.514 −.920 .815** .832** 1.000
4 EAA_4 .000 9.000 −.680 −.688 .831** .869** .846** 1.000
5 EAA_5 .000 9.000 −.628 −.696 .771** .791** .824** .853** 1.000
Open in a new tab

Note.

**

p < .01.

Factor loadings are displayed in Table 3. Fit indices were as follows: Bollen-Stine bootstrap p = .006, CFI = .985, NNFI = .970, SRMR = .014, and RMSEA= .145 (90% CI: .100 – .194). Given that the Bollen-Stine bootstrap and RMSEA values indicated poor model fit and possible model misspecifications, standardized residual covariances and modification indices were evaluated (Byrne, 2010; Chau, 1997). Standardized residual covariances were all low (< 1.96). Modification indices suggested misfit resulting from correlated errors between item #4 (i.e., “When I feel pain, drinking alcohol can really help”) and item #5 (i.e., “I feel like drinking alcohol would help me cope with pain”; MI = 14.09), which could be explained by semantic overlap. Indeed, these items are the only two that do not directly assess expectancies that alcohol will reduce pain, but instead, assess whether alcohol can “help” more generally. Therefore, the error covariance between these items was freed up (Muthen & Muthen, 2010). Following model modification, fit indices were as follows: Bollen-Stine bootstrap p = .126, CFI = .995, NNFI = .988, SRMR = .009, and RMSEA = .092 (90% CI: .039 – .151). Factor loadings for the modified model are displayed in Table 3.

Table 3.

Factor Loadings with Bootstrap Standard Errors and Confidence Intervals

Unstandardized Factor Loadings
Hypothesized Model Modified Model
Variable Estimate SE 95% CI p Estimate SE 95% CI p
1 EAA_1 1.000 .000 1.000–1.000 - 1.000 .000 1.000–1.000 -
2 EAA_2 1.023 .024 .974–1.070 .001 1.021 .024 .971–1.067 .002
3 EAA_3 1.043 .028 .990–1.098 .001 1.035 .028 .983–1.090 .001
4 EAA_4 1.097 .030 1.044–1.158 .001 1.073 .028 1.021–1.129 .001
5 EAA_5 1.068 .036 1.003–1.146 .001 1.037 .035 .974–1.106 .001
Standardized Factor Loadings
Hypothesized Model Modified Model
Variable Estimate SE 95% CI p Estimate SE 95% CI p
1 EAA_1 .924 .013 .893–.946 .001 .933 .012 .905–.952 .001
2 EAA_2 .939 .011 .914–.958 .002 .946 .011 .922–.965 .002
3 EAA_3 .931 .011 .906–.951 .001 .932 .011 .908–.952 .001
4 EAA_4 .962 .007 .946–.974 .001 .950 .009 .929–.966 .030
5 EAA_5 .917 .015 .883–.942 .001 .899 .018 .859–.929 .001
e4 ↔ e5 .355 .110 .118–.549 .009
Open in a new tab

Note. The modified models allowed for correlated error terms for EAA items #4 and #5.

3.3. Internal Consistency

The EAA evinced excellent internal consistency (α = .97).

3.4. Correlates of EAA Scores

EAA scores were positively associated with average number of drinks consumed per day (r = .31, p < .001), average number of hours spent drinking each day (r = .22, p < .001), frequency of binge drinking (r = .47, p < .001), and QFV drinking classification (r = .45, p < .001). EAA scores were also positively associated with AUDIT total scores (r = .59, p < .001), AUDIT- Consumption scores (r = .53, p < .001), AUDIT- Dependence scores (r = .54, p < .001), and AUDIT- Harmful Use scores (r = .52, p < .001). In addition, EAA scores were positively associated with both positive (r = .45, p < .001) and negative (r = .49, p < .001) alcohol outcome expectancies, and correlations with individual AOES subscales ranged from r = .33 - .52 (all ps < .001). Similarly, EAA scores were positively associated with coping (r = .54, p < .001), social (r = .30, p < .001), enhancement (r = .46, p < .001), and conformity (r = .40, p < .001) motives for drinking.

To determine whether EAA scores were associated with alcohol use and alcohol-related problems above and beyond established predictors of alcohol use (e.g., other outcome expectancies, alcohol motives), we elected to run a series of post-hoc hierarchical linear regressions. In each model, predictors were entered in the following order: Step 1 (all AOES and DMQ subscale scores); Step 2 (EAA score). Results indicated that EAA scores remained a significant predictor of average number of daily drinks (β = .28, p < .001; ΔR2 = .04, p < .001), average number of hours spent drinking each day (β = .17, p = .029; ΔR2 = .02, p = .027), AUDIT total scores (β = .21, p < .001; ΔR2 = .03, p < .001), AUDIT- Consumption scores (β = .33, p < .001; ΔR2 = .06, p < .001), AUDIT- Dependence scores (β = .167, p < .001; ΔR2 = .02, p < .001), and AUDIT- Harmful Use scores (β = .15, p = .002; ΔR2 = .01, p = .002), even after accounting for other alcohol outcome expectancies and motives.

In terms of clinical pain variables, EAA scores were positively associated with chronic pain grade (r = .39, p < .001), characteristic pain intensity (r = .39, p < .001), pain-related disability (r = .38, p < .001), and current prescription opioid use (r = .19, p = .002). In terms of associations between clinical pain variables and more general alcohol expectancies/motives, chronic pain grade was positively associated with both positive (r = .30, p < .001) and negative (r = .55, p < .001) general alcohol outcome expectancies, and coping (r = .56, p < .001), social (r = .31, p < .001), enhancement (r = .37, p < .001), and conformity (r = .44, p < .001) motives for drinking. Similar patterns of positive associations were observed between general alcohol outcome expectancies/motives and both characteristic pain intensity and pain-related disability (ps < .05).

As expected, EAA scores were not associated with past 6-month cannabis use (r = .046, p = .459) or CUDIT-R scores (r = .075, p = .225).

3.5. EAA Scores as a Function of Sociodemographic Characteristics and High Level of Drinking Problems

Male participants scored higher (M = 26.87, SD = 12.55) on the EAA than female participants (M = 23.64, SD = 13.09; F(1, 271) = 3.96, p = .048), and EAA scores were negatively associated with age (r = −.18, p = .003). EAA scores also differed as a function of race (F(1, 267) = 3.13, p = .009), with Asian participants scoring significantly higher (M = 31.44, SD = 11.89) than Black/African American (M = 21.59, SD = 15.03) and White (M = 24.87, SD = 15.03) participants. Similarly, Hispanic participants scored higher on the EAA (M = 30.37, SD = 11.51) than non-Hispanic participants (M = 24.69, SD = 12.88; F(1, 271) = 8.38, p = .004). No differences in EAA scores were observed as a function of marital status, education, or income (ps > .05). Finally, EAA scores were higher among participants who scored above the AUDIT cut-off for a high level of drinking problems (M = 32.64, SD = 7.84), compared to those who scored below the cut-off (M = 19.77, SD = 13.29; F(1, 271) = 91.40, p < .001).

4. Discussion

This study represents the first examination of psychometric properties of the Expectancies for Alcohol Analgesia Scale (EAA), which is a novel, five-item measure designed to assess expectancies that drinking alcohol will reduce pain. The EAA was administered to 273 current alcohol users with chronic musculoskeletal pain, and results provided support for the single-factor structure, reliability, and validity. Although initial evaluation of the hypothesized single-factor structure of the EAA indicated good model-fit across several indices, the Bollen-Stine bootstrap χ2 test statistic and RMSEA suggested poor model fit. To improve model fit, we made one post-hoc adjustment in the factor structure by allowing correlated measurement errors for items 4 (“When I feel pain, drinking alcohol can really help”) and 5 (“I feel like drinking alcohol would help me cope with pain”). Overlap in the content of these items may give rise to covariation in measurement errors (i.e., both items address pain coping versus pain reduction), and modification indices suggested that substantial improvement in model-fit would be achieved by allowing the error covariance of these items to be estimated freely. This post-hoc modification resulted in good model-fit according to the Bollen-Stine bootstrap χ2 test statistic, CFI, NNFI, and SRMR (e.g., Bollen & Stine, 1992; Hu & Bentler, 1999), and acceptable fit according to the RMSEA (e.g., Lai & Green, 2016). Moreover, Cronbach’s α coefficients indicated that internal consistency of the EAA was excellent, which provides further evidence that the EAA items measure a single construct (Tavakol & Dennick, 2011).

Results also indicated that EAA scores were positively associated with quantity/frequency of alcohol consumption, and correlations tended to be medium-to-large in magnitude (rs ≥ .3; e.g., Pallant, 2013). Post-hoc analyses further indicated that these associations remained significant, even after covarying for other established predictors of alcohol consumption (i.e., other alcohol outcome expectancies, drinking motives), with EAA scores accounting for 6% of unique variance in AUDIT- Consumption scores. These findings are consistent with expectancy theory, which dictates that there should be lawful relationships between alcohol outcome expectancies and quantity/frequency of drinking (Jones et al., 2001), and with an established literature indicating that positive alcohol outcome expectancies are related to greater drinking behavior (e.g., Jones et al., 2001; Leigh & Stacy, 1993; Monk & Heim, 2013). EAA total scores were also positively associated with scores on a widely used measure of general alcohol outcome expectancies, and it was particularly notable that a relatively higher correlation (r = .47) was observed between EAA scores and scores on the Tension Reduction/Negative Reinforcement subscale of the AOES, which assesses expectations that alcohol will alleviate negative affect (e.g., “I feel less stressed”). Similarly, although EAA scores were positively associated with motives for drinking in general, expectancies for alcohol analgesia were most strongly associated with coping motives (r = .54; e.g., “To forget about your problems”). These findings suggest that EAA scores are most closely related to scores on measures that also assess negative reinforcement processes involved in drinking expectancies/motivation.

EAA scores were also positively associated with greater chronic pain grade, characteristic pain intensity, pain-related disability, and current prescription opioid use. Individuals with more severe and disabling pain have likely encountered a greater number of opportunities to learn about the effects of alcohol on pain, which, in turn, may strengthen expectancies for alcohol analgesia. Moreover, given associations between EAA scores and quantity/frequency of alcohol consumption, it is possible that chronic and heavy alcohol use has led to increased pain facilitation among those with higher EAA scores (e.g., Ditre et al., 2019; Egli, Koob, & Edwards, 2012).

In addition, results provided support for the discriminant validity of the EAA, which was tested by comparing EAA scores to a theoretically distinct construct (i.e., current cannabis use). This finding is consistent with recommendations that correlations with scores on discriminant measures should be lower than correlations with scores on measures that are theoretically-related to the construct of interest (Michalos, 2014). Taken together, results from this study indicated that the EAA demonstrated concurrent validity with theoretically-related constructs (e.g., frequency/quantity of alcohol consumption, alcohol outcome expectancies, clinical pain variables), as well as discriminant validity with a theoretically-distinct construct (i.e., cannabis use).

Interestingly, EAA scores were higher among males than females, and this pattern of findings is consistent with previous work demonstrating that males (vs. females) are more likely to hold positive alcohol outcome expectancies (e.g., Kirmani & Suman, 2010). We also observed a negative relationship between age and EAA scores, suggesting that younger (vs. older) adults with chronic pain may report greater expectancies that alcohol will provide pain relief. In addition, Hispanic (vs. non-Hispanic) participants scored higher on the EAA, which is consistent with previous findings that Hispanic drinkers may hold more positive expectancies (e.g., social extroversion) regarding the effects of alcohol use (Marin, Posner, & Kinyon, 1993), but inconsistent with evidence that non-Hispanic White participants may be most likely to self-medicate pain with alcohol (Riley III & King, 2009). Similarly, the current results suggest that Asian drinkers (vs. Black/African American and White drinkers) may hold stronger expectancies for alcohol analgesia, which is not consistent with prior work suggesting that positive alcohol outcome expectancies are generally lower among Asians (e.g., Meier, Slutske, Arndt, & Cadoret, 2007). In the current study, Hispanic and Asian (vs. Black or African American or White) participants reported greater pain intensity (ps < .05), and this may be one possible explanation for higher EAA scores among these groups. However, it is important to interpret these results with caution, and future work is needed to replicate these findings among larger, more diverse samples.

Several limitations and directions for future research are worth noting. First, although EAA items assessed the likelihood of experiencing pain-relief from drinking alcohol, the quantity of alcohol consumed and/or level of intoxication were not specified. Previous research has noted that expectancies may vary based on the amount of alcohol that a person imagines consuming and the duration of the drinking episode, and that assessment of dose-related expectancies may yield important information about the perceived effects of drinking (e.g., Fromme et al., 1993; Morean et al., 2012). Future work should consider anchoring EAA items to specific quantities of alcohol, and testing whether expectancies for alcohol analgesia increase as one imagines consuming a greater number of drinks over a specified time period. Second, we did not assess the frequency at which participants engage in pain self-management via alcohol use, and future work would benefit from testing whether EAA scores are associated with the quantity and frequency of alcohol use specifically for pain-coping. Third, only the concurrent and discriminant validity of the EAA were assessed, and future research is needed to assess the predictive validity of this measure. For example, it is important to test whether EAA scores are prospectively associated with the development of problematic patterns of alcohol consumption and poorer pain outcomes. Fourth, reliability was assessed using Cronbach’s alpha, and future research is needed to assess other forms of reliability (e.g., test-retest reliability). Fifth, although the EAA items were developed to be highly face valid, we did not assess participants’ understanding of each item and the response options, and it is possible that this could have impacted study findings. Future qualitative work could offer insight into the interpretation and conceptualization of EAA items (Krause, 2006). Finally, all participant characteristics, including chronic pain status, drinking behavior, and country of residence, were assessed via self-report, without additional verification (e.g., medical records, physician report).

It is also important to note several limitations related to the study sample. Although we included several measures of quantity/frequency of alcohol consumption, the presence of alcohol use disorder (AUD) was not assessed. Future research should extend these findings to treatment-seeking drinkers with AUD, and determine whether the EAA predicts treatment outcomes among this population. In the contrary, future work should also test the psychometric properties of the EAA among a sample that includes never and former drinkers. Another limitation is that the study was limited to participants who endorsed current chronic musculoskeletal pain, and future research should replicate these findings among participants with neuropathic pain conditions (e.g., fibromyalgia) and among treatment-seeking pain patients. Moreover, we did not assess the source of participants’ musculoskeletal pain (beyond pain location), or the presence/severity of other medical conditions, and future research is needed to determine whether these factors may influence EAA scores/psychometric properties. Given that individuals who do not experience persistent pain may still develop alcohol outcome expectancies for pain relief (e.g., expectancies can be influenced by social/cultural transmission; Asmundson, Gomez-Perez, Richter, & Carleton, 2014; Johnson, Nagoshi, Danko, Honbo, & Chau, 1990), it will also be important to examine the psychometric properties of the EAA among individuals without chronic pain (e.g., those with no pain, acute pain, and/or sub-acute pain). Finally, participants were recruited using Amazon Mechanical Turk, and additional work is needed to generalize these results across larger samples that are recruited via a variety of sampling methods (e.g., university participant pools, pain/substance treatment centers).

In addition to conducting supplemental validation studies of the EAA, future research is needed to clarify the role of expectancies for alcohol analgesia in bidirectional pain-alcohol effects. Alcohol can produce acute analgesia (Thompson et al., 2017), and expectations for pain relief have been shown to increase the magnitude of analgesic effects (e.g., Schenk, Sprenger, Geuter, & Büchel, 2014). Thus, it is possible that the experience of pain may be influenced by an interaction between alcohol consumption and expectancies for alcohol analgesia, and future work should test whether higher EAA scores are associated with greater reductions in pain following drinking. It is also possible that expectancies for alcohol analgesia may lead to greater drinking in response to pain, and future work should test whether EAA scores moderate the effects of pain on alcohol use behavior (i.e., pain as a motivator of drinking). Researchers could also consider utilizing a randomized experimental design to test the effects of a manipulation designed to challenge alcohol-related outcome expectancies for pain reduction on alcohol urge/consumption, as this may provide evidence for a causal pathway between expectancies for alcohol analgesia and drinking behavior in the context of pain (Ditre et al., 2010). It will also be important to elucidate the role of prescription opioid use in associations between expectancies for alcohol analgesia and alcohol use, particularly given recent empirical/clinical interest in alcohol/opioid co-use (e.g., LaRowe et al., 2020; Witkiewitz & Vowles, 2018; Zegel, Rogers, Vujanovic, & Zvolensky, 2020). Finally, research should continue to examine longitudinal associations/interrelations between pain-related and general alcohol outcome expectancies, drinking motives, and pain, given that this work will ultimately inform the development of tailored interventions for the millions of alcohol users with chronic pain.

Highlights.

  • Expectancies for alcohol analgesia may influence pain-alcohol interrelations.

  • The EAA is a novel 5-item measure of expectancies that alcohol will reduce pain.

  • The EAA evinced a single-factor structure and evidence of reliability/validity.

  • Future work should include further validation and tests of predictive utility.

Acknowledgments

Role of Funding Sources

This research was supported by NIH Grant No. R01AA024844 awarded to Joseph W. Ditre and Stephen A. Maisto, and by a Syracuse University dissertation fellowship awarded to Lisa R. LaRowe.

Appendix

Expectancies for Alcohol Analgesia

Throughout our lives, most of us have experienced pain from time to time (ranging from minor headaches and sprains, to more persistently painful conditions like neck, knee, or lower back pain). Below is a list of statements about how drinking alcohol may influence your experience of pain.

Please rate how LIKELY or UNLIKELY you believe each statement is for you when you drink alcohol. If the statement seems UNLIKELY to you, select a number from 0–4. If the statement seems LIKELY to you, select a number from 5–9. For example, if you believe the statement would never happen, select 0; if you believe the statement would happen every time you drink alcohol, select 9.

Please use the guide below to help you rate each statement:

0 1 2 3 4 5 6 7 8 9
Completely Extremely Very Somewhat A little A little Somewhat Very Extremely Completely
UNLIKELY LIKELY
Open in a new tab
1. Drinking alcohol would ease my pain. 0 1 2 3 4 5 6 7 8 9
2. If I were to experience pain, drinking would help me reduce it. 0 1 2 3 4 5 6 7 8 9
3. If I hurt myself, I would feel less pain if I could drink alcohol. 0 1 2 3 4 5 6 7 8 9
4. When I feel pain, drinking alcohol can really help. 0 1 2 3 4 5 6 7 8 9
5. I feel like drinking alcohol would help me cope with pain. 0 1 2 3 4 5 6 7 8 9
Open in a new tab

Footnotes

Conflict of Interest

All authors declare that they have no conflicts of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adamson SJ, Kay-Lambkin FJ, Baker AL, Lewin TJ, Thornton L, Kelly BJ, & Sellman JD (2010). An improved brief measure of cannabis misuse: the Cannabis Use Disorders Identification Test-Revised (CUDIT-R). Drug and Alcohol Dependence, 110(1–2), 137–143. [DOI] [PubMed] [Google Scholar]
  2. Asmundson GJ, Gomez-Perez L, Richter AA, & Carleton RN (2014). The psychology of pain: models and targets for comprehensive assessment In van Griensven H, Strong J, & Unruh A (Eds), Pain: A textbook for health professions (pp. 35–48). Churchill Livingstone. [Google Scholar]
  3. Atlas LY, & Wager TD (2014). A meta-analysis of brain mechanisms of placebo analgesia: consistent findings and unanswered questions Placebo (pp. 37–69): Springer. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babor TF, Higgins-Biddle JC, Saunders JB, & Monteiro MG (1992). The alcohol use disorders identification test Guidelines for use in primary health care. Geneva: World Health Organization. [Google Scholar]
  5. Babor TF, Higgins-Biddle JC, Saunders JB, & Monteiro MG (2001). The alcohol use disorders identification test (AUDIT): Guidelines for use in primary care. World Health Organization, Department of Mental Health and Substance Abuse. [Google Scholar]
  6. Bandura A (1989). Human agency in social cognitive theory. American Psychologist, 44(9), 1175–1184. [DOI] [PubMed] [Google Scholar]
  7. Bingel U, Wanigasekera V, Wiech K, Mhuircheartaigh RN, Lee MC, Ploner M, & Tracey I (2011). The effect of treatment expectation on drug efficacy: imaging the analgesic benefit of the opioid remifentanil. Science Translational Medicine, 3(70), 70ra14–70ra14. [DOI] [PubMed] [Google Scholar]
  8. Blunch N (2012). Introduction to structural equation modeling using IBM SPSS statistics and AMOS: Sage. [Google Scholar]
  9. Bollen KA, & Stine RA (1992). Bootstrapping goodness-of-fit measures in structural equation models. Sociological Methods & Research, 21(2), 205–229. [Google Scholar]
  10. Brennan PL, Schutte KK, & Moos RH (2005). Pain and use of alcohol to manage pain: prevalence and 3‐year outcomes among older problem and non‐problem drinkers. Addiction, 100(6), 777–786. [DOI] [PubMed] [Google Scholar]
  11. Brown SA, Christiansen BA, & Goldman MS (1987). The Alcohol Expectancy Questionnaire: an instrument for the assessment of adolescent and adult alcohol expectancies. Journal of Studies on Alcohol, 48(5), 483–491. [DOI] [PubMed] [Google Scholar]
  12. Butcher B, & Carmody J (2012). Sex differences in analgesic response to ibuprofen are influenced by expectancy: A randomized, crossover, balanced placebo‐designed study. European Journal of Pain, 16(7), 1005–1013. [DOI] [PubMed] [Google Scholar]
  13. Byrne BM (2010). Structural equation modeling with AMOS: Basic concepts, applications, and programming (2nd ed.). New York: Taylor and Francis Group, LLC. [Google Scholar]
  14. Cahalan D, Cisin IH, & Crossley HM (1969). American drinking practices: a national study of drinking behavior and attitudes: Publications Division, Rutgers Center of Alcohol Studies; distributed by College & University Press, New Haven, Conn. [Google Scholar]
  15. Carey KB, Henson JM, Carey MP, & Maisto SA (2009). Computer versus in-person intervention for students violating campus alcohol policy. Journal of Consulting and Clinical Psychology, 77(1), 74–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Chau PY (1997). Reexamining a model for evaluating information center success using a structural equation modeling approach. Decision Sciences, 28(2), 309–334. [Google Scholar]
  17. Cheng Y, Macera CA, Davis DR, Ainsworth BE, Troped PJ, & Blair SN (2000). Physical activity and self-reported, physician-diagnosed osteoarthritis: is physical activity a risk factor? Journal of Clinical Epidemiology, 53(3), 315–322. [DOI] [PubMed] [Google Scholar]
  18. Chopra K, & Tiwari V (2012). Alcoholic neuropathy: possible mechanisms and future treatment possibilities. British Journal of Clinical Pharmacology, 73(3), 348–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Cooper ML (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6(2), 117–128. [Google Scholar]
  20. Dimeff LA (1999). Brief alcohol screening and intervention for college students (BASICS): A harm reduction approach: Guilford Press. [Google Scholar]
  21. Ditre J (2006). Pain as a motivator to smoke: Effects of pain induction on smoking urge and behavior (Master’s Thesis, University of South Florida). Retrieved from Scholar Commons (http://scholarcommons.usf.edu/etd/2503) [Google Scholar]
  22. Ditre JW, Heckman BW, Butts EA, & Brandon TH (2010). Effects of expectancies and coping on pain-induced motivation to smoke. Journal of Abnormal Psychology, 119(3), 524–533. doi: 10.1037/a0019568 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ditre JW, Zale EL, & LaRowe LR (2019). A Reciprocal Model of Pain and Substance Use: Transdiagnostic Considerations, Clinical Implications, and Future Directions. Annual Review of Clinical Psychology, 15, 503–258. [DOI] [PubMed] [Google Scholar]
  24. Edwards S, Vendruscolo LF, Gilpin NW, Wojnar M, & Witkiewitz K (2020). Alcohol and pain: A translational review of preclinical and clinical findings to inform future treatment strategies. Alcoholism: Clinical and Experimental Research, 44(2), 368–383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Egli M, Koob GF, & Edwards S (2012). Alcohol dependence as a chronic pain disorder. Neuroscience & Biobehavioral Reviews, 36(10), 2179–2192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Fabrigar LR, Wegener DT, MacCallum RC, & Strahan EJ (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299. [Google Scholar]
  27. Fromme K, Stroot EA, & Kaplan D (1993). Comprehensive effects of alcohol: Development and psychometric assessment of a new expectancy questionnaire. Psychological Assessment, 5(1), 19–26. [Google Scholar]
  28. Goebel JR, Compton P, Zubkoff L, Lanto A, Asch SM, Sherbourne CD, . . . Lorenz KA (2011). Prescription sharing, alcohol use, and street drug use to manage pain among veterans. Journal of Pain and Symptom Management, 41(5), 848–858. [DOI] [PubMed] [Google Scholar]
  29. Hoyle RH (2012). Handbook of structural equation modeling: Guilford press. [Google Scholar]
  30. Hu L. t., & 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. [Google Scholar]
  31. Ipeirotis PG (2010). Demographics of Mechanical Turk. NYU Working Paper No. CEDER-10–01 Available at SSRN: https://ssrn.com/abstract=1585030 [Google Scholar]
  32. Johnson RC, Nagoshi CT, Danko GP, Honbo KAM, & Chau LL (1990). Familial transmission of alcohol use norms and expectancies and reported alcohol use. Alcoholism: Clinical and Experimental Research, 14(2), 216–220. [DOI] [PubMed] [Google Scholar]
  33. Jones BT, Corbin W, & Fromme K (2001). A review of expectancy theory and alcohol consumption. Addiction, 96(1), 57–72. [DOI] [PubMed] [Google Scholar]
  34. Kirmani MN, & Suman L (2010). Gender differences in alcohol related attitudes and expectancies among college students. Journal of the Indian Academy of Applied Psychology, 36(1), 19–24. [Google Scholar]
  35. Krause N (2006). The use of qualitative methods to improve quantitative measures of health-related constructs. Medical Care, 44(11), S34–S38. [DOI] [PubMed] [Google Scholar]
  36. Kuntsche E, Stewart SH, & Cooper ML (2008). How stable is the motive–alcohol use link? A cross-national validation of the Drinking Motives Questionnaire Revised among adolescents from Switzerland, Canada, and the United States. Journal of Studies on Alcohol and Drugs, 69(3), 388–396. [DOI] [PubMed] [Google Scholar]
  37. Lai K, & Green SB (2016). The problem with having two watches: Assessment of fit when RMSEA and CFI disagree. Multivariate Behavioral Research, 51(2–3), 220–239. [DOI] [PubMed] [Google Scholar]
  38. LaRowe LR, Powers JM, Garey L, Rogers AH, Zvolensky MJ, & Ditre JW (2020). Pain-related anxiety, sex, and co-use of alcohol and prescription opioids among adults with chronic low back pain. Drug and Alcohol Dependence, 214, 108171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Larson MJ, Paasche‐Orlow M, Cheng DM, Lloyd‐Travaglini C, Saitz R, & Samet JH (2007). Persistent pain is associated with substance use after detoxification: a prospective cohort analysis. Addiction, 102(5), 752–760. [DOI] [PubMed] [Google Scholar]
  40. Lawton J, & Simpson J (2009). Predictors of alcohol use among people experiencing chronic pain. Psychology, Health & Medicine, 14(4), 487–501. [DOI] [PubMed] [Google Scholar]
  41. Leigh BC, & Stacy AW (1993). Alcohol outcome expectancies: Scale construction and predictive utility in higher order confirmatory models. Psychological Assessment, 5(2), 216–229. [Google Scholar]
  42. Lerch MM, Albrecht E, Ruthenbürger M, Mayerle J, Halangk W, & Krüger B (2003). Pathophysiology of alcohol-induced pancreatitis. Pancreas, 27(4), 291–296. [DOI] [PubMed] [Google Scholar]
  43. Madden DR, & Clapp JD (2019). The event-level impact of one’s typical alcohol expectancies, drinking motivations, and use of protective behavioral strategies. Drug and Alcohol Dependence, 194, 112–120. [DOI] [PubMed] [Google Scholar]
  44. Marin G, Posner SF, & Kinyon JB (1993). Alcohol expectancies among Hispanics and non-Hispanic Whites: Role of drinking status and acculturation. Hispanic Journal of Behavioral Sciences, 15(3), 373–381. [Google Scholar]
  45. Martin JL, Ferreira JA, Haase RF, Martins J, & Coelho M (2016). Validation of the Drinking Motives Questionnaire-Revised across US and Portuguese college students. Addictive Behaviors, 60, 58–63. doi: 10.1016/j.addbeh.2016.03.030 [DOI] [PubMed] [Google Scholar]
  46. Matsunaga M (2010). How to factor-analyze your data right: do’s, don’ts, and how-to’s. International Journal of Psychological Research, 3(1), 97–110. [Google Scholar]
  47. Meier MH, Slutske WS, Arndt S, & Cadoret RJ (2007). Positive alcohol expectancies partially mediate the relation between delinquent behavior and alcohol use: Generalizability across age, sex, and race in a cohort of 85,000 Iowa schoolchildren. Psychology of Addictive Behaviors, 21(1), 25–34. [DOI] [PubMed] [Google Scholar]
  48. Michalos AC (2014). Encyclopedia of quality of life and well-being research: Springer; Netherlands Dordrecht. [Google Scholar]
  49. Monk RL, & Heim D (2013). A critical systematic review of alcohol-related outcome expectancies. Substance Use & Misuse, 48(7), 539–557. [DOI] [PubMed] [Google Scholar]
  50. Morean ME, Corbin WR, & Treat TA (2012). The Anticipated Effects of Alcohol Scale: Development and psychometric evaluation of a novel assessment tool for measuring alcohol expectancies. Psychological Assessment, 24(4), 1008–1023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Moskal D, Maisto SA, De Vita M, & Ditre JW (2018). Effects of experimental pain induction on alcohol urge, intention to consume alcohol, and alcohol demand. Experimental and Clinical Psychopharmacology, 26(1), 65–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Muthen LK, & Muthen BO (2010). Mplus user’s guide. 6th Los Angeles, CA: Muthén & Muthén, 19982006. [Google Scholar]
  53. Olsson UH, Foss T, Troye SV, & Howell RD (2000). The performance of ML, GLS, and WLS estimation in structural equation modeling under conditions of misspecification and nonnormality. Structural Equation Modeling, 7(4), 557–595. [Google Scholar]
  54. Ossipov MH, Dussor GO, & Porreca F (2010). Central modulation of pain. The Journal of Clinical Investigation, 120(11), 3779–3787. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Pallant J (2013). SPSS survival manual: McGraw-Hill Education (UK). [Google Scholar]
  56. Paolacci G, Chandler J, & Ipeirotis PG (2010). Running experiments on amazon mechanical turk. Judgment and Decision Making, 5(5), 411–419. [Google Scholar]
  57. Parkerson HA, & Asmundson GJ (2016). The role of pain intensity and smoking expectancies on smoking urge and behavior following experimental pain induction. Drug and Alcohol Dependence, 164, 166–171. [DOI] [PubMed] [Google Scholar]
  58. Peerdeman KJ, van Laarhoven AI, Peters ML, & Evers AW (2016). An integrative review of the influence of expectancies on pain. Frontiers in Psychology, 7, 1270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Reinert DF, & Allen JP (2002). The alcohol use disorders identification test (AUDIT): a review of recent research. Alcoholism: Clinical and Experimental Research, 26(2), 272–279. [PubMed] [Google Scholar]
  60. Riley III JL, & King C (2009). Self-report of alcohol use for pain in a multi-ethnic community sample. The Journal of Pain, 10(9), 944–952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Rogers AH, Zegel M, Tran JK, Zvolensky MJ, & Vujanovic AA (2020). Pain intensity, alcohol use motives, and alcohol use among firefighters: the moderating role of pain-related anxiety. Addictive Behaviors, 106415. [DOI] [PubMed] [Google Scholar]
  62. Rotter JB (1954). Social learning and clinical psychology. Englewood Cliffs, NJ: Prentice-Hall. [Google Scholar]
  63. Sá KN, Baptista AF, Matos MA, & Lessa Í (2008). Chronic pain and gender in Salvador population, Brazil. Pain, 139(3), 498–506. [DOI] [PubMed] [Google Scholar]
  64. Santos JRA (1999). Cronbach’s alpha: A tool for assessing the reliability of scales. Journal of Extension, 37(2), 1–5. [Google Scholar]
  65. Schenk LA, Sprenger C, Geuter S, & Büchel C (2014). Expectation requires treatment to boost pain relief: an fMRI study. PAIN®, 155(1), 150–157. [DOI] [PubMed] [Google Scholar]
  66. Schreiber JB, Nora A, Stage FK, Barlow EA, & King J (2006). Reporting structural equation modeling and confirmatory factor analysis results: A review. The Journal of Educational Research, 99(6), 323–338. [Google Scholar]
  67. Sheu R, Lussier D, Rosenblum A, Fong C, Portenoy J, Joseph H, & Portenoy RK (2008). Prevalence and characteristics of chronic pain in patients admitted to an outpatient drug and alcohol treatment program. Pain Medicine, 9(7), 911–917. [DOI] [PubMed] [Google Scholar]
  68. Simons JS, Maisto SA, Wray TB, & Emery NN (2016). Acute effects of intoxication and arousal on approach/avoidance biases toward sexual risk stimuli in heterosexual men. Archives of Sexual Behavior, 45(1), 43–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Solomon KE, & Annis HM (1989). Development of a scale to measure outcome expectancy in alcoholics. Cognitive Therapy and Research, 13(5), 409–421. [Google Scholar]
  70. Stacy AW, Widaman KF, & Marlatt GA (1990). Expectancy models of alcohol use. Journal of Personality and Social Psychology, 58(5), 918. [DOI] [PubMed] [Google Scholar]
  71. Tavakol M, & Dennick R (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2, 53–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Thompson T, Oram C, Correll CU, Tsermentseli S, & Stubbs B (2017). Analgesic effects of alcohol: a systematic review and meta-analysis of controlled experimental studies in healthy participants. The Journal of Pain, 18(5), 499–510. [DOI] [PubMed] [Google Scholar]
  73. Von Korff M (2011). Assessment of chronic pain in epidemiological and health services research In Turk DC & Melzack R (Eds.), Handbook of Pain Assessment: Third Edition New York, NY: The Guilford Press. [Google Scholar]
  74. Von Korff M, Crane P, Lane M, Miglioretti DL, Simon G, Saunders K, . . . Kessler R (2005). Chronic spinal pain and physical-mental comorbidity in the United States: results from the national comorbidity survey replication. Pain, 113(3), 331–339. doi: 10.1016/j.pain.2004.11.010 [DOI] [PubMed] [Google Scholar]
  75. Von Korff M, Ormel J, Keefe FJ, & Dworkin SF (1992). Grading the severity of chronic pain. Pain, 50(2), 133–149. [DOI] [PubMed] [Google Scholar]
  76. Witkiewitz K, & Vowles KE (2018). Alcohol and opioid use, co‐use, and chronic pain in the context of the opioid epidemic: A critical review. Alcoholism: Clinical and Experimental Research, 42(3), 478–488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Zale EL, Maisto SA, & Ditre JW (2015). Interrelations between pain and alcohol: an integrative review. Clinical Psychology Review, 37, 57–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Zegel M, Rogers AH, Vujanovic AA, & Zvolensky MJ (2020). Alcohol use problems and opioid misuse and dependence among adults with chronic pain: The role of distress tolerance. Psychology of Addictive Behaviors, doi: 10.1037/adb0000587 [DOI] [PubMed] [Google Scholar]

RESOURCES