Reliability of Differential Item Functioning in Alcohol Use Disorder: Bayesian Meta-Analysis of Criteria Discrimination Estimates

Understanding alcohol use and alcohol use disorders from a developmental psychopathology perspective: Research advances, challenges, and future directions

As part of the special issue of Development and Psychopathology honoring the remarkable contributions of Dr Dante Cicchetti, the current paper attempts to describe the recent contributions that a developmental psychopathology perspective has made in understanding the development of alcohol use and alcohol-related problems over the lifespan. The paper also identifies some of the future challenges and research directions. Because the scope of this task far exceeds the confines of a journal length article this paper does not attempt a comprehensive review. Rather, it builds on an earlier review and commentary that was published in Development and Psychopathology in 2013, with a similar goal.)Building on that work and updating its conclusions and suggestions for future directions, the current paper emphasizes findings from the research areas that were identified for further study in 2013 and the findings that have been published since that time.

Diagnostic Criteria for Identifying Individuals at High Risk of Progression From Mild or Moderate to Severe Alcohol Use Disorder

Importance

Current Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-5) diagnoses of substance use disorders rely on criterion count-based approaches, disregarding severity grading indexed by individual criteria.

Objective

To examine correlates of alcohol use disorder (AUD) across count-based severity groups (ie, mild, moderate, mild-to-moderate, severe), identify specific diagnostic criteria indicative of greater severity, and evaluate whether specific criteria within mild-to-moderate AUD differentiate across relevant correlates and manifest in greater hazards of severe AUD development.

Design, Setting, and Participants

This cohort study involved 2 cohorts from the family-based Collaborative Study on the Genetics of Alcoholism (COGA) with 7 sites across the United States: cross-sectional (assessed 1991-2005) and longitudinal (assessed 2004-2019). Statistical analyses were conducted from December 2022 to June 2023.

Main Outcomes and Measures

Sociodemographic, alcohol-related, psychiatric comorbidity, brain electroencephalography (EEG), and AUD polygenic score measures as correlates of DSM-5 AUD levels (ie, mild, moderate, severe) and criterion severity-defined mild-to-moderate AUD diagnostic groups (ie, low-risk vs high-risk mild-to-moderate).

Results

A total of 13 110 individuals from the cross-sectional COGA cohort (mean [SD] age, 37.8 [14.2] years) and 2818 individuals from the longitudinal COGA cohort (mean baseline [SD] age, 16.1 [3.2] years) were included. Associations with alcohol-related, psychiatric, EEG, and AUD polygenic score measures reinforced the role of increasing criterion counts as indexing severity. Yet within mild-to-moderate AUD (2-5 criteria), the presence of specific high-risk criteria (eg, withdrawal) identified a group reporting heavier drinking and greater psychiatric comorbidity even after accounting for criterion count differences. In longitudinal analyses, prior mild-to-moderate AUD characterized by endorsement of at least 1 high-risk criterion was associated with more accelerated progression to severe AUD (adjusted hazard ratio [aHR], 11.62; 95% CI, 7.54-17.92) compared with prior mild-to-moderate AUD without endorsement of high-risk criteria (aHR, 5.64; 95% CI, 3.28-9.70), independent of criterion count.

Conclusions and Relevance

In this cohort study of a combined 15 928 individuals, findings suggested that simple count-based AUD diagnostic approaches to estimating severe AUD vulnerability, which ignore heterogeneity among criteria, may be improved by emphasizing specific high-risk criteria. Such emphasis may allow better focus on individuals at the greatest risk and improve understanding of the development of AUD.

Temporal dynamics of emotional processing: Parsing trial-wise variance of the late positive potential using Generalizability Theory

The reliability of individual trial event-related potential (ERP) components extracted from electroencephalogram has been consistently questioned since ERP research began. This ambivalence is based on misunderstood assumptions stemming from Cronbach and Classical Test Theory. Contemporary methods allow for the reliability of individual ERP trials to be estimated and for analyses of these trial-level ERP components to be meaningfully parsed. We illustrate the use of Generalizability Theory procedures in estimating the reliability of trial-level ERPs using the late positive potential (LPP), a neural measure of motivated attention toward emotionally evocative stimuli. Individuals (N = 88) completed a passive viewing task while continuous EEG was recorded. Variability in trial-level LPP responses was decomposed into facets corresponding to individual differences, chronological trial within block, stimulus type, their two-way interactions, and specific stimuli. We estimated various reliability coefficients and found that both overall and category-specific person-level LPP estimates have good-to-excellent reliability, while the reliability of within-person differences (i.e., change) between arousal categories was fair for the early LPP. These results were generally consistent across time windows, but were highest early in the LPP time course. We argue that investigating reliability using trial-level data allows researchers to pursue hypotheses focused on neurophysiological dynamics that unfold over the course of an experiment and not risk false inferences (i.e., ecological fallacy) when using person-level aggregates to deduce such processes. Moreover, such analyses provide information that allows researchers to optimize their protocols by potentially reducing the number of individual trials, burden on participants, and cost, while retaining sufficient reliability.

An application of moderated nonlinear factor analysis to develop a commensurate measure of alcohol problems across four alcohol treatment studies

BACKGROUND

Self-report measures of alcohol problems are commonly included in studies evaluating treatment and recovery from alcohol use disorder (AUD), but no prior study has examined the replicability of the measurement of alcohol problems across studies with various measures and diverse samples. Further, it is unclear which items may be better indicators of alcohol problems for patient subgroups. In the present study, we integrated data from four large alcohol treatment studies to develop a commensurate measure of alcohol problems using moderated nonlinear factor analysis (MNLFA).

METHODS

Data were from the COMBINE study, Project MATCH, the Relapse Replication and Extension Project (RREP), and the United Kingdom Alcohol Treatment Trial (UKATT), yielding a total sample size of 4414. MNLFA was carried out on the Drinker Inventory of Consequences (COMBINE, MATCH, RREP) and Alcohol Problems Questionnaire (UKATT).

RESULTS

We successfully created a 78-item commensurate measure of alcohol problems and examined differential item functioning (DIF) by study membership, time, and socio-demographic characteristics. Sixty-two items demonstrated intercept DIF, suggesting differences in rates of item endorsement for clients with the same underlying levels of alcohol problems across patient subgroups. Six items demonstrated loading DIF, suggesting differences in the extent to which the items were indicative of alcohol problems across patient subgroups.

CONCLUSIONS

The self-reported measurement of alcohol problems replicates across measures and diverse samples. Items with DIF have clinical implications for the treatment of AUD. Finally, MNLFA scores can be used to test substantive research questions across these studies.

Endorsement of specific alcohol use disorder criterion items changes with age in individuals with persistent alcohol use disorders in 2 generations of the San Diego Prospective Study

BACKGROUND

Diagnostic and Statistical Manual (DSM) alcohol use disorder (AUD) criteria are written in broad enough terms to apply to diverse populations. The current analyses evaluate whether the endorsement of criteria changes with increasing age in individuals with persistent AUDs.

METHODS

Data regarding AUDs persisting across 3 timepoints between average ages of 31 and 43 were gathered about every 5 years from 318 interviews for 106 San Diego Prospective Study (SDPS) AUD male probands. Similar data regarding persistent AUDs across 2 timepoints were obtained from 136 interviews with 68 SDPS AUD offspring between average ages of 21 and 27. Changes in the endorsement of each AUD criterion were evaluated using Cochran's Q test.

RESULTS

For AUD probands across time, significant decreases were observed in the proportions endorsing 4 criteria (tolerance, withdrawal, failure to fulfill obligations, and using alcohol in hazardous situations). Increased rates of endorsement were documented for 3 criteria (drinking alcohol in higher amounts or for longer periods of time, spending a great deal of time regarding alcohol, and continued use despite social or interpersonal problems). Significant increases in rates of endorsements for offspring were seen for spending a great deal of time regarding alcohol and giving up or reducing important activities in order to drink.

CONCLUSIONS

These data indicate that the salience of many DSM AUD criterion items changed significantly with age in both SDPS generations among individuals with persistent AUDs. The current results support the need for additional systematic research to determine whether specific criterion items might need to be weighted differently in evaluating older and younger individuals with persistent AUDs.