Regression tree applications to studying alcohol-related problems among college students.

Machine learning algorithms hold promise for developing precision medicine approaches to addiction treatment yet have been used sparingly to identify predictors of alcohol-related problems. Recursive partitioning, a machine learning algorithm, can identify salient predictors and clinical cut points that can guide treatment. This study aimed to identify predictors and cut points of alcohol-related problems and to examine result stability in two separate, large data sets of college student drinkers (n = 5,090 and 2,808). Four regression trees were grown using the "rpart" package in R. Seventy-one predictors were classified as demographics (e.g., age), alcohol use indicators (e.g., typical quantity/frequency), or psychosocial indicators (e.g., anxiety). Predictors and cut points were extracted and used to manually recreate the tree in the other data set to test result stability. Outcome variables were alcohol-related problems as measured by the Alcohol Use Disorder Identification Test and Brief Young Adult Alcohol Consequences Questionnaire. Coping with depression, conformity motives, binge drinking frequency, typical/heaviest quantity, drunk frequency, serious harm reduction protective behavioral strategies, substance use, and psychosis symptoms best predicted alcohol-related problems across the four trees; coping with depression (cut point range: 1.83-2.17) and binge drinking frequency (cut point range: 1.5-2.5) were the most common splitting variables. Model fit indices suggest relatively stable results accounting for 17%-30% of the variance. Results suggest the nine salient predictors, particularly coping with depression motives scores around 2 and binge drinking frequency around two to three times per month, are important targets to consider when treating alcohol-related problems for college students. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Internalized Stigma Measurement in Substance Use Treatment Settings: A Narrative Review.

Stigma relating to substance use disorders is one of the many barriers to enrolling in substance use treatment. Stigma is also related to poorer substance use treatment outcomes, yet few studies of substance use and substance use treatment outcomes include measures of stigma. Stigma is a multi-level experience occurring as a result of discrimination within a systematic power structure promoting inequities among marginalized populations. Several domains of stigma are manifested among individuals seeking treatment for a substance use disorder, with internalized stigma being the most commonly measured. The current paper is a narrative review of measures that have been developed to measure internalized stigma related to substance use in treatment settings. Measures of stigma (n=8) in substance use treatment settings were identified using PubMed and PsycINFO databases. The review identified various strengths of existing measures, including a broad range of measures with mostly excellent internal consistency. The review also identified limitations including the general lack of consideration for multiple domains and intersecting forms of stigma, samples with limited racial and ethnic diversity, and the lack of assessments of polysubstance use. The development of measures of stigma that assess multiple domains of stigma and that are tested in a wide range of substance use treatment settings with racially and ethnically diverse participants is needed. This is of particular importance because stigma remains a crucial barrier to successful initiation and completion of substance use treatment.

Early Birds and Night Owls: Distinguishing Profiles of Cannabis Use Habits by Use Times with Latent Class Analysis.

Background: Understanding, predicting, and reducing the harms associated with cannabis use is an important field of study. Timing (i.e., hour of day and day of week) of substance use is an established risk factor of severity of dependence. However, there has been little attention paid to morning use of cannabis and its associations with negative consequences. Objectives: The goal of the present study was to examine whether distinct classifications of cannabis use habits exist based on timing, and whether these classifications differ on cannabis use indicators, motives for using cannabis, use of protective behavioral strategies, and cannabis-related negative outcomes. Methods: Latent class analyses were conducted on four independent samples of college student cannabis users (Project MOST 1, N=2,056; Project MOST 2, N=1846; Project PSST, N=1,971; Project CABS, N=1,122). Results: Results determined that a 5-class solution best fit the data within each independent sample consisting of the classes: (1) "Daily-morning use",(2) "Daily-non-morning use", (3) "Weekend-morning use", (4) "Weekend-night use", and (5) "Weekend-evening use." Classes endorsing daily and/or morning use reported greater use, negative consequences and motives, while those endorsing weekend and/or non-morning use reported the most adaptive outcomes (i.e., reduced frequency/quantity of use, fewer consequences experienced, and fewer cannabis use disorder symptoms endorsed). Conclusions: Recreational daily use as well as morning use may be associated with greater negative consequences, and there is evidence that most college students who use cannabis do avoid these types of use. The results of the present study offer evidence that timing of cannabis use may be a pertinent factor in determining harms associated with use.

A comprehensive examination of motivational profiles for alcohol-related behaviors among college students.

Recent research demonstrates unique relations of types of motivation for drinking responsibly based on self-determination theory and drinking motives with alcohol-related outcomes among college students. In the present study, we sought to extend prior research by using a person-centered approach to simultaneously consider types of motivation within and across these motivational constructs as well as their synergistic relations with alcohol-related outcomes. We used cross-sectional survey data from 2,808 college students at 10 universities in eight states across the United States who reported past-month alcohol use (Mage = 20.59, SD = 4.18; 72.9% female; 58.2% non-Hispanic White). A series of latent profile analyses were conducted using types of motivation for drinking responsibly and drinking motives as indicators. A five-profile solution was selected as optimal. Mean comparisons indicated that profiles defined by high endorsement of higher quality motivations for drinking responsibly (i.e., more self-determined) and low endorsement of drinking motives in combination were related to the most frequent protective behavioral strategies use, least alcohol use, and fewest negative alcohol-related consequences. Additionally, these profiles were higher on dispositional autonomy and psychological need satisfaction and lower on psychological need frustration. These findings provide initial insight into simultaneously considering motivational profiles for the interrelated behaviors of drinking responsibly and drinking that can be leveraged in college drinking interventions. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Using Decision Trees to Identify Salient Predictors of Cannabis-Related Outcomes.

Cannabis use continues to escalate among emerging adults and college attendance may be a risk factor for use. Severe cases of cannabis use can escalate to a cannabis use disorder, which is associated with worse psychosocial functioning. Predictors of cannabis use consequences and cannabis use disorder symptom severity have been identified; however, they typically employ a narrow set of predictors and rely on linear models. Machine learning is well suited for exploratory data analyses of high-dimensional data. This study applied decision tree learning to identify predictors of cannabis user status, negative cannabis-related consequences, and cannabis use disorder symptoms. Undergraduate college students (N = 7000) were recruited from nine universities in nine states across the U.S. Among the 7 trees, 24 splits created by 15 distinct predictors were identified. Consistent with prior research, one's beliefs about cannabis were strong predictors of user status. Negative reinforcement cannabis use motives were the most consistent predictors of cannabis use disorder symptoms, and past month cannabis use was the most consistent predictor of probable cannabis use disorder. Typical frequency of cannabis use was the only predictor of negative cannabis-related consequences. Our results demonstrate that decision trees are a useful methodological tool for identifying targets for future clinical research.

A risk score assessment tool for peripheral arterial disease in women: From the National Health and Nutrition Examination Survey.

BACKGROUND: Peripheral arterial disease (PAD) carries a significant morbidity and mortality. Women are more commonly affected with this condition and are mostly asymptomatic, and undertreated. The objective of the study was to develop and validate a simple risk score to identify women with PAD. HYPOTHESIS: Identifying those at early stage of the disease could help reduce the risk of complications. METHODS: Using data from the National Health and Nutrition Examination Survey 1999-2004, we identified women who had data on ankle brachial index. The cohort was divided into development (70%) and validation (30%) groups. Using variables that are self-reported or measured without laboratory data, we developed a multivariable logistic regression to predict PAD, which was evaluated in the validation cohort. RESULTS: A total of 150.6 million women were included. A diagnosis of PAD was reported in 13.7%. Age, body mass index, hypertension, diabetes mellitus, smoking, non-oral contraceptive pill usage, and parity were all independently associated with PAD. The C-statistics was 0.74, with good calibration. The model showed good stability in the validation cohort (C-statistics 0.73). CONCLUSION: This parsimonious risk model is a valid tool for risk prediction of PAD in women, and could be easily applied in routine clinical practice.