The Language of Substance Use and Recovery: Novel Use of the Go/No-Go Association Task to Measure Implicit Bias.

Previous research has found initial evidence that word choice impacts the perception and treatment of those with behavioral health disorders through explicit bias (i.e., stigma). A more robust picture of behavioral health disorder stigma should incorporate both explicit and implicit bias, rather than relying on only one form. The current study uses the Go/No-Go Association Task to calculate a d' (sensitivity) indexed score of automatic attitudes (i.e., implicit associations) to two terms, "addict" and "person with substance use disorder." Participants have significantly more negative automatic attitudes (i.e., implicit bias) toward the term "addict" in isolation as well as when compared to "person with a substance use disorder." Consistent with previous research on explicit bias, implicit bias does exist for terms commonly used in the behavioral health field. "Addict" should not be used in professional or lay settings. Additionally, these results constitute the second pilot study employed the Go/No-Go Association Task in this manner, suggesting it is a viable option for continued linguistic stigma related research.

Comparison of Smartphone Ownership, Social Media Use, and Willingness to Use Digital Interventions Between Generation Z and Millennials in the Treatment of Substance Use: Cross-Sectional Questionnaire Study.

BACKGROUND: Problematic substance use in adolescence and emerging adulthood is a significant public health concern in the United States due to high recurrence of use rates and unmet treatment needs coupled with increased use. Consequently, there is a need for both improved service utilization and availability of recovery supports. Given the ubiquitous use of the internet and social media via smartphones, a viable option is to design digital treatments and recovery support services to include internet and social media platforms. OBJECTIVE: Although digital treatments delivered through social media and the internet are a possibility, it is unclear how interventions using these tools should be tailored for groups with problematic substance use. There is limited research comparing consumer trends of use of social media platforms, use of platform features, and vulnerability of exposure to drug cues online. The goal of this study was to compare digital platforms used among adolescents (Generation Zs, age 13-17) and emerging adults (Millennials, age 18-35) attending outpatient substance use treatment and to examine receptiveness toward these platforms in order to support substance use treatment and recovery. METHODS: Generation Zs and Millennials enrolled in outpatient substance use treatment (n=164) completed a survey examining social media use, digital intervention acceptability, frequency of substance exposure, and substance use experiences. Generation Zs (n=53) completed the survey in July 2018. Millennials (n=111) completed the survey in May 2016. RESULTS: Generation Zs had an average age of 15.66 (SD 1.18) years and primarily identified as male (50.9%). Millennials had an average age of 27.66 (SD 5.12) years and also primarily identified as male (75.7%). Most participants owned a social media account (Millennials: 82.0%, Generation Zs: 94.3%) and used it daily (Millennials: 67.6%, Generation Zs: 79.2%); however, Generation Zs were more likely to use Instagram and Snapchat, whereas Millennials were more likely to use Facebook. Further, Generation Zs were more likely to use the features within social media platforms (eg, instant messaging: Millennials: 55.0%, Generation Zs: 79.2%; watching videos: Millennials: 56.8%, Generation Zs: 81.1%). Many participants observed drug cues on social media (Millennials: 67.5%, Generation Zs: 71.7%). However, fewer observed recovery information on social media (Millennials: 30.6%, Generation Zs: 34.0%). Participants felt that social media (Millennials: 55.0%, Generation Zs: 49.1%), a mobile phone app (Millennials: 36.9%, Generation Zs: 45.3%), texting (Millennials: 28.8%, Generation Zs: 45.3%), or a website (Millennials: 39.6%, Generation Zs: 32.1%) would be useful in delivering recovery support. CONCLUSIONS: Given the high rates of exposure to drug cues on social media, disseminating recovery support within a social media platform may be the ideal just-in-time intervention needed to decrease the rates of recurrent drug use. However, our results suggest that cross-platform solutions capable of transcending generational preferences are necessary and one-size-fits-all digital interventions should be avoided.

Biased labels: An experimental study of language and stigma among individuals in recovery and health professionals.

BACKGROUND: Labels such as "addict" and "substance abuser" have been found to elicit implicit and explicit stigma among the general public previously. The difference in the levels of this bias among individuals in recovery and those employed in the health profession has not yet been identified, however. The current study seeks to answer this question using measures of implicit bias. METHODS: A subset sample (n = 299) from a previously completed study (n = 1288) was selected for analysis. Mixed-model ANOVA tests were completed to identify variance between d-prime automatic association scores with the terms "addict" and "substance abuser" among individuals in recovery and those identified as working in the health professions. RESULTS: Individuals in recovery did not have lower negative associations with either term, whereas individuals employed as health professionals had greater negative associations with the term "substance abuser" but did not have greater negative associations with the term "addict." CONCLUSIONS: Results provide further evidence that previously identified stigmatizing labels have the potential to influence medical care and medical practitioner perceptions of individuals with substance use disorders and should be avoided. Further exploration into the role negative associations derived from commonly used labels have in the individual recovery process is needed to draw appropriate recommendations.

Technology and Social Media Use Among Patients Enrolled in Outpatient Addiction Treatment Programs: Cross-Sectional Survey Study.

BACKGROUND: Substance use disorder research and practice have not yet taken advantage of emerging changes in communication patterns. While internet and social media use is widespread in the general population, little is known about how these mediums are used in substance use disorder treatment. OBJECTIVE: The aims of this paper were to provide data on patients' with substance use disorders mobile phone ownership rates, usage patterns on multiple digital platforms (social media, internet, computer, and mobile apps), and their interest in the use of these platforms to monitor personal recovery. METHODS: We conducted a cross-sectional survey of patients in 4 intensive outpatient substance use disorder treatment facilities in Philadelphia, PA, USA. Logistic regressions were used to examine associations among variables. RESULTS: Survey participants (N=259) were mostly male (72.9%, 188/259), African American (62.9%, 163/259), with annual incomes less than US $10,000 (62.5%, 161/259), and averaged 39 (SD 12.24) years of age. The vast majority of participants (93.8%, 243/259) owned a mobile phone and about 64.1% (166/259) owned a mobile phone with app capabilities, of which 85.1% (207/243) accessed the internet mainly through their mobile phone. There were no significant differences in age, gender, ethnicity, or socio-economic status by computer usage, internet usage, number of times participants changed their phone, type of mobile phone contract, or whether participants had unlimited calling plans. The sample was grouped into 3 age groups (Millennials, Generation Xers, and Baby Boomers). The rates of having a social media account differed across these 3 age groups with significant differences between Baby Boomers and both Generation Xers and Millennials (P<.001 in each case). Among participants with a social media account (73.6%, 190/259), most (76.1%, 144/190) reported using it daily and nearly all (98.2%, 186/190) used Facebook. Nearly half of participants (47.4%, 90/190) reported viewing content on social media that triggered substance cravings and an equal percentage reported being exposed to recovery information on social media. There was a significant difference in rates of reporting viewing recovery information on social media across the 3 age groups with Baby Boomers reporting higher rates than Millennials (P<.001). The majority of respondents (70.1%, 181/259) said they would prefer to use a relapse prevention app on their phone or receive SMS (short message service) relapse prevention text messages (72.3%, 186/259), and nearly half (49.1%, 127/259) expressed an interest in receiving support by allowing social media accounts to be monitored as a relapse prevention technique. CONCLUSIONS: To our knowledge, this is the first and largest study examining the online behavior and preferences regarding technology-based substance use disorder treatment interventions in a population of patients enrolled in community outpatient treatment programs. Patients were generally receptive to using relapse prevention apps and text messaging interventions and a substantial proportion supported social media surveillance tools. However, the design of technology-based interventions remains as many participants have monthly telephone plans which may limit continuity.

Peer-delivered harm reduction and recovery support services: initial evaluation from a hybrid recovery community drop-in center and syringe exchange program.

BACKGROUND: Recovery from substance use disorder (SUD) is often considered at odds with harm reduction strategies. More recently, harm reduction has been categorized as both a pathway to recovery and a series of services to reduce the harmful consequences of substance use. Peer recovery support services (PRSS) are effective in improving SUD outcomes, as well as improving the engagement and effectiveness of harm reduction programs. METHODS: This study provides an initial evaluation of a hybrid recovery community organization providing PRSS as well as peer-based harm reduction services via a syringe exchange program. Administrative data collected during normal operations of the Missouri Network for Opiate Reform and Recovery were analyzed using Pearson chi-square tests and Monte Carlo chi-square tests. RESULTS: Intravenous substance-using participants (N = 417) had an average of 2.14 engagements (SD = 2.59) with the program. Over the evaluation period, a range of 5345-8995 sterile syringes were provided, with a range of 600-1530 used syringes collected. Participant housing status, criminal justice status, and previous health diagnosis were all significantly related to whether they had multiple engagements. CONCLUSIONS: Results suggest that recovery community organizations are well situated and staffed to also provide harm reduction services, such as syringe exchange programs. Given the relationship between engagement and participant housing, criminal justice status, and previous health diagnosis, recommendations for service delivery include additional education and outreach for homeless, justice-involved, LatinX, and LGBTQ+ identifying individuals.

Peer-based recovery support services delivered at recovery community organizations: Predictors of improvements in individual recovery capital.

BACKGROUND: While clinical interventions used to support the recovery process of U.S. adults are well understood, community-based solutions such as peer-based recovery support services delivered by a recovery community organization are not. METHODS: Previously collected administrative data of 3459 participants at 20 recovery community organizations in the U.S. were analyzed using a paired samples t-test to examine intake and current recovery capital differences, and multiple linear regression models to examine the association between peer-based recovery support engagement on changes in recovery capital. RESULTS: Participants were mostly male (52.1%), non-Hispanic (80.2%), White (75.5%), with an average age of 39.38 years (SD = 12.57). Participants' average engagement was 130.68 days (SD = 166.6) with a total of 4290 engagement sessions (M = 4.75, SD = 4.74) and 8913 brief check-ins (M = 5.0, SD = 5.03) facilitated. Reported health events were 0.09 recurrences of substance use (SD = 0.61) and 0.02 emergency room visits (SD = 0.26) on average. Paired sample t-test results showed a statistically significant increase in recovery capital of 1.33 points (95% CI: 0.97-1.69). Multiple linear regression models for predicting improvements in recovery capital (adjusted r2 = 0.61) found number of follow-up engagements and completed recovery plan goals were statistically significant predictors. CONCLUSIONS: Peer-based recovery support services delivered by recovery community organizations assist in significantly improving individual recovery capital, as well as helping to facilitate involvement with an array of recovery support services that may contribute to other functional social determinant domain improvements and lower negative health events.

Attitudes toward opioid use disorder medications: Results from a U.S. national study of individuals who resolved a substance use problem.

The attitudes of individuals who receive, provide, or influence opioid use disorder (OUD) medication services, also called stakeholders, may enhance or hinder their dissemination and adoption. Individuals who have resolved a significant alcohol or other drug (AOD) problem are a group of key stakeholders whose OUD medication attitudes are not well understood empirically. This group subsumes, but is not limited to, individuals who identify as being "in recovery." Analyses leveraged the National Recovery Study, a geo-demographically representative survey of U.S. adults who resolved a significant AOD problem (N = 1,946). We examined the prevalence of positive, neutral, and negative attitudes toward agonists, such as buprenorphine/naloxone and methadone, and antagonists, such as oral and extended-release depot injection naltrexone. Single-predictor logistic regression models tested for demographic, clinical, and recovery-related correlates of these attitudes and, for those significant at the .1 level, multivariable-predictor logistic regression models tested unique associations between these correlates and attitudes. Results showed that participants were equally likely to hold positive (21.4 [18.9-24.0]%) and negative agonist (23.8 [21.2-26.7]%) attitudes but significantly more likely to hold negative (30.3 [27.4-33.3]%) than positive antagonist attitudes (18.0 [15.9-20.4]%). Neutral attitudes were most commonly endorsed for both agonists (54.8 [51.6-57.9]%) and antagonists (51.7 [48.5-54.8]%). For agonists, more recent AOD problem resolution was a unique predictor of positive attitude, whereas Black and Hispanic races/ethnicities, compared with White, were unique predictors of negative attitude. For antagonists, older age group (45-59 and 60 + vs. 18-29 years), lifetime opioid antagonist medication prescription, and past 90-day non-12-step mutual-help attendance were unique predictors of positive attitude, whereas greater spirituality was a unique predictor of negative attitude. This population-level study of U.S. adults who resolved an AOD problem showed that agonist attitudes may be more positive than anecdotal evidence suggests. Certain characteristics and experiences, however, highlight a greater likelihood of negative attitudes, suggesting these factors may be potential barriers to OUD medication adoption. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Digital recovery networks: Characterizing user participation, engagement, and outcomes of a novel recovery social network smartphone application.

BACKGROUND: Recovery support services, including in vivo (i.e., face to face) peer-based supports and social networks, are associated with positive effects on substance use disorder recovery outcomes. The translation of in vivo supports to digital platforms is a recent development that is mostly unexamined. The types of users and their engagement patterns of digital recovery support services (D-RSS), and the utility of objective and self-report data in predicting future recovery outcomes require further study to move the recovery support field forward. METHODS: De-identified individual user data from Sober Grid, a recovery social network site (R-SNS) smartphone application, for the years 2015-2018 was analyzed to identify the demographics, engagement patterns, and recovery outcomes of active users. Analysis of variance (ANOVA) tests were used to examine between generational group differences on activity variables and recovery outcomes. Logistic and linear regressions were used to identify significant predictors of sobriety length and relapse among users. RESULTS: The most active tercile of users (n = 1273; mAge = 39 years; 62% male) had average sobriety lengths of 195.5 days and had experienced 4.4 relapses on average since sign-up. Users have over 33,000 unilateral and nearly 14,000 bilateral connections. Users generated over 120,000 unique posts, 507,000 comments, 1617,000 likes, 12,900 check-ins, and 593,000 chats during the period of analysis. Recovery outcomes did not vary between generations, though user activity was significantly different between Generations (Millennials, Generation X, and Baby Boomers), with baby boomers and generation X having higher levels of engagement and connection among all activity markers. Logistic regression results revealed gender (female) was associated with a lower likelihood of reporting loneliness or sexual feelings as an emotional trigger. Linear regressions revealed generation, number of unilateral connections, and number of check-ins was associated with sobriety length, while generation and number of check-ins was associated with number of relapses. CONCLUSIONS: Active users of Sober Grid engage in several platform features that provide objective data that can supplement self-report data for analysis of recovery outcomes. Most commonly uses features are those similar to features readily available in open-ecosystem social network sites (e.g., Facebook). Prediction model results suggest that demographic factors (e.g., age, gender) and activity factors (e.g., number of check-ins) may be useful in deploying just-in-time interventions to prevent relapse or offer additional social support. Further empirical examination is needed to identify the utility of such interventions, as well as the mechanisms of support that accompany feature use or engagement with the D-RSS.

Cultural Differences in Tweeting about Drinking Across the US.

Excessive alcohol use in the US contributes to over 88,000 deaths per year and costs over $250 billion annually. While previous studies have shown that excessive alcohol use can be detected from general patterns of social media engagement, we characterized how drinking-specific language varies across regions and cultures in the US. From a database of 38 billion public tweets, we selected those mentioning "drunk", found the words and phrases distinctive of drinking posts, and then clustered these into topics and sets of semantically related words. We identified geolocated "drunk" tweets and correlated their language with the prevalence of self-reported excessive alcohol consumption (Behavioral Risk Factor Surveillance System; BRFSS). We then identified linguistic markers associated with excessive drinking in different regions and cultural communities as identified by the American Community Project. "Drunk" tweet frequency (of the 3.3 million geolocated "drunk" tweets) correlated with excessive alcohol consumption at both the county and state levels (r = 0.26 and 0.45, respectively, p < 0.01). Topic analyses revealed that excessive alcohol consumption was most correlated with references to drinking with friends (r = 0.20), family (r = 0.15), and driving under the influence (r = 0.14). Using the American Community Project classification, we found a number of cultural markers of drinking: religious communities had a high frequency of anti-drunk driving tweets, Hispanic centers discussed family members drinking, and college towns discussed sexual behavior. This study shows that Twitter can be used to explore the specific sociocultural contexts in which excessive alcohol use occurs within particular regions and communities. These findings can inform more targeted public health messaging and help to better understand cultural determinants of substance abuse.

"Abusing Addiction": Our Language Still Isn't Good Enough.

BACKGROUND: Previous research has found initial evidence that word choice impacts the perception and treatment of those with behavioral health disorders. These previous studies have relied on vignette-based methodologies, however, and a more quantifiable index of the stigma words can produce is needed. METHOD: The current study uses the Go/No-Go Association Task to calculate a d-prime (sensitivity) indexed score of automatic attitudes to two terms, "substance abuser" and "person with substance use disorder". RESULTS: Participants have significantly more negative automatic attitudes towards the term "substance abuser", as compared to "person with a substance use disorder". CONCLUSION: Consistent with previous research, implicit bias does exist for terms commonly used in the behavioral health field. "Substance Abuser" and its derivatives should not be used in professional or lay settings.

Recovery dialects: A pilot study of stigmatizing and nonstigmatizing label use by individuals in recovery from substance use disorders.

Previous research has found language used to describe individuals with a substance use disorder (SUD; e.g., "addict," "substance abuser") contributes to and elicits negative bias among the general public and health care professionals. However, the prevalence in which recovering individuals use these labels to self-identify and the impact of such labels are unknown. The current pilot study, a cross-sectional design, examined the usage of two labels ("addict," "person with a SUD") as well as the differences in recovery outcomes among individuals in recovery. Participants (n = 54) used both labels at high rates ("addict": 66.67%; "person with a SUD": 38.89%), though mutually exclusive use was lower ("addict" only: 35.19%, "person with a SUD" only: 7.5%). Common label use settings included mutual-aid recovery meetings, with friends and family, and on social media. Analysis of variance tests found no statistically significant differences between label groups for recovery capital, self-esteem, internalized stigma and shame, flourishing, or length in recovery. Descriptively, participants using only "person with a SUD" had more positive outcomes, although these individuals also had higher levels of internalized shame. Results suggest that language may have only a marginal impact on individuals in recovery, although professionals and the general public should continue to avoid using stigmatizing labels. Additionally, many individuals in recovery have the ability to discern context and setting, switching between positive and negative labels as appropriate. Future research is warranted given these pilot findings and should focus on long-term impacts of self-labeling and internalized stereotypes among individuals in recovery. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Reducing harm and promoting recovery through community-based mutual aid: Characterizing those who engage in a hybrid peer recovery community organization.

BACKGROUND: Peer-based support services are often used within harm reduction organizations, and more recently within recovery community organizations (RCO). Identifying the characteristics of individuals who engage with these novel RCOs is needed. Additionally, conducting collaborative research with communities of people who use drugs (PWUD) or are in recovery is an effective and rewarding approach that allows individuals to take ownership and play a critical role in the study. METHODS: This exploratory study employs a community-based participatory research (CBPR) framework in partnership with a peer-led hybrid recovery community organization, Rebel Recovery, in Florida. Peer staff participated in all phases of the study, helping to inform the study protocol, data collection, analysis, interpretation, and results write-up. A cross-sectional survey instrument was used to collect consumer intake data. Pearson Chi-square tests and multivariate binomial logistic regressions were used to examine relationships between consumer characteristics and service utilization. RESULTS: Consumers (n = 396) of Rebel Recovery peer support services had a mean age of 35.60 years (SD = 9.74). Many were experiencing homelessness (35.4%), unemployed (69.7%), high school graduates or GED holders (68.2%) and had a last year income of less than $10,000 (58.3%). The majority were users of heroin primarily (70.7%), with intravenous use being the preferred route of administration (63.9%). Exploratory analysis found that gender, marital status, and involvement in the child welfare system were significantly related to primary substance of use. Past 30-day engagement in recovery meetings had several statistically significant predictors including primary substance of use, age, housing status, annual income level, past-30-day arrests, tobacco use, and alcohol harm perception. Process findings from the CBPR methods used reconfirm the value of including peers in research involving PWUD and individuals in recovery. CONCLUSIONS: Results suggest that peer-based support services at a hybrid recovery community organization can successfully engage populations that are often underserved (i.e., experiencing homelessness, involved in drug court, intravenous users, etc.). Significant relationships identified in the exploratory analysis suggest that additional education concerning overdose and the potential benefits of recovery meetings may be useful for specific consumers. Additionally, several recommendations and benefits of engaging in community-based participatory research with peer-led organizations are made for future research.

Responding to the opioid and overdose crisis with innovative services: The recovery community center office-based opioid treatment (RCC-OBOT) model.

Opioid use disorder (OUD) and opioid-related overdose mortality are major public health concerns in the United States. Recently, several community-based and professional innovations - including hybrid recovery community organizations, peer-based emergency department warm handoff programs, emergency department buprenorphine induction, and low-threshold OUD treatment programs - have emerged or expanded in an effort to address significant obstacles to providing patients the care needed for OUD and to reduce the risk of overdose. Additional innovations are needed to address the crisis. Building upon the foundational frameworks of each of these recent innovations, a new model of OUD pharmacotherapy is proposed and discussed: the Recovery Community Center Office-Based Opioid Treatment model. Additionally, two potential implementation scenarios, the overdose and non-overdose event protocols, are detailed for communities, peers, and practitioners interested in implementing the model. Potential barriers to implementation of the model include service reimbursement, licensing regulations, and organizational concerns. Future research should seek to validate the model and to identify actual implementation and sustainability barriers and best practices.

Systemic barriers in substance use disorder treatment: A prospective qualitative study of professionals in the field.

BACKGROUND: The US is in the midst of one of the largest public health crises in recent history with over 63,000 drug poisoning deaths in 2016 and a projected annual economic cost of over $420 billion. With the rise of deaths and economic burden related to substance use, it is paramount that systemic barriers within the treatment industry be identified and resolved. METHODS: Data were collected from US substance use treatment professionals (N = 182) in the fall of 2016. Thematic analysis with axial coding was used on anonymized responses to an online open-ended survey. Additional ad hoc testing for variance (education, generation, regional location, and employment) was completed using Monte Carlo chi-square analyses. RESULTS: 7 major themes emerged: 1) additional training, education, and use of evidence-based practices, 2) expansion of treatment services, 3) increased resources, 4) stigma reduction, 5) increased collaboration and leadership, 6) reductions in regulations, requirements, and incentives, and 7) expansion of recovery support services. Participant response yielded a significant relationship between employment type (p = 0.002) and regional location (p = 0.046). CONCLUSIONS: Systemic barriers in the treatment field are prevalent from the perspective of professionals engaged in the field. While previously identified barriers are still present, newly reported barriers include: 1) lack of treatment services (e.g., capacity), 2) lack of technological resources (e.g., technological support tools), 3) lack of recovery support services (e.g., recovery housing), 4) lack of collaboration and leadership (e.g., communication and partnership), and 5) increasing unethical practices in the field (e.g., incentive-based patient brokering).

Substance use, recovery, and linguistics: The impact of word choice on explicit and implicit bias.

BACKGROUND: The general public, treatment professionals, and healthcare professionals have been found to exhibit an explicit negative bias towards substance use and individuals with a substance use disorder (SUD). Terms such as "substance abuser" and "opioid addict" have shown to elicit greater negative explicit bias. However, other common terms have yet to be empirically studied. METHODS: 1,288 participants were recruited from ResearchMatch. Participants were assigned into one of seven groups with different hypothesized stigmatizing and non-stigmatizing terms. Participants completed a Go/No Association Task (GNAT) and vignette-based social distance scale. Repeated-measures ANOVAs were used to analyze the GNAT results, and one-way ANOVAs were used to analyze vignette results. RESULTS: The terms "substance abuser", "addict", "alcoholic", and "opioid addict", were strongly associated with the negative and significantly different from the positive counterterms. "Relapse" and "Recurrence of Use" were strongly associated with the negative; however, the strength of the "recurrence of use" positive association was higher and significantly different from the "relapse" positive association. "Pharmacotherapy" was strongly associated with the positive and significantly different than "medication-assisted treatment". Both "medication-assisted recovery" and "long-term recovery" were strongly associated with the positive, and significantly different from the negative association. CONCLUSIONS: Results support calls to cease use of the terms "addict", "alcoholic", "opioid addict", and "substance abuser". Additionally, it is suggested that "recurrence of use" and "pharmacotherapy" be used for their overall positive benefits. Both "medication-assisted recovery" and "long-term recovery" are positive terms and can be used when applicable without promoting stigma.

Can Twitter be used to predict county excessive alcohol consumption rates?

OBJECTIVES: The current study analyzes a large set of Twitter data from 1,384 US counties to determine whether excessive alcohol consumption rates can be predicted by the words being posted from each county. METHODS: Data from over 138 million county-level tweets were analyzed using predictive modeling, differential language analysis, and mediating language analysis. RESULTS: Twitter language data captures cross-sectional patterns of excessive alcohol consumption beyond that of sociodemographic factors (e.g. age, gender, race, income, education), and can be used to accurately predict rates of excessive alcohol consumption. Additionally, mediation analysis found that Twitter topics (e.g. 'ready gettin leave') can explain much of the variance associated between socioeconomics and excessive alcohol consumption. CONCLUSIONS: Twitter data can be used to predict public health concerns such as excessive drinking. Using mediation analysis in conjunction with predictive modeling allows for a high portion of the variance associated with socioeconomic status to be explained.