Optimizing the length and reliability of measures of mechanisms of change to support measurement-based care in alcohol use disorder treatment.

Abstract

Objective: Clients who receive alcohol use disorder (AUD) treatment experience variable outcomes. Measuring clinical progress during treatment using standardized measures (i.e., measurement-based care) can help indicate whether clinical improvements are occurring. Measures of mechanisms of behavioral change (MOBCs) may be particularly well-suited for measurement-based care; however, measuring MOBCs would be more feasible and informative if measures were briefer and if their ability to detect reliable change with individual clients was better articulated. Method: Three abbreviated measures of hypothesized MOBCs (abstinence self-efficacy, coping strategies, anxiety) and a fourth full-length measure (depression) were administered weekly during a 12-week randomized trial of cognitive-behavioral therapy (CBT) for women with AUD. Psychometric analyses estimated how reliably each measure distinguished within-person change from between-person differences and measurement error. Reliability coefficients were estimated for simulated briefer versions of each instrument (i.e., instruments with fewer items than the already-abbreviated instruments) and rates of reliable improvement and reliable worsening were estimated for each measure. Results: All four measures had good reliability (.86-.90) for detecting within-person change. Many participants (41.4%-62.5%) reliably improved on MOBCs from first to last treatment session. Reliable improvement on MOBCs was associated with reductions in percentage of drinking days (PDD) at 3, 9, and 15-month follow-ups. Simulated briefer versions of each instrument retained good reliability for detecting change with only 3 (self-efficacy), 11 (coping strategies), 5 (anxiety), or 10 items (depression). Conclusions: Brief MOBC measures can detect reliable change for individuals in AUD treatment. Routinely measuring MOBCs may help with monitoring clinical progress. Public Health SignificanceBrief measures of abstinence self-efficacy, coping strategies, depression, and anxiety reliably measure within-person clinical changes during alcohol use disorder (AUD) treatment. The 41%-63% of clients who showed reliable improvement in these measures during outpatient AUD treatment had the largest reductions in drinking days after treatment. Measuring these constructs during AUD treatment can be one way of delivering measurement-based care, an evidence-based practice that is rarely utilized in routine AUD treatment.

What do clinicians want? Understanding frontline addiction treatment clinicians’ preferences and priorities to improve the design of measurement-based care technology

“…Graphical displays of clinical progress over time (e.g., line graphs) may be one means of illustrating such progress. Additionally, MBC systems could potentially provide interpretations as to whether a trend reflects reliable improvement or decline, increased or decreased risk for adverse outcomes, and/or suggested interventions based on the measures collected [ 30 ].…”

The rise and fall of substance use during treatment: Applying recency and expectancy principles to daily substance use patterns.

“…Future studies may directly test this hypothesis by measuring the salience of treatment mechanisms during therapy on a daily basis, which would require the development of measures that can reliably detect daily changes in treatment mechanism salience. Relatively short measures of specific treatment mechanisms have been developed to be used at weekly, daily, and hourly intervals in SUD treatment (Hallgren et al, 2021; Roos et al, 2020), which may reflect the salience of treatment mechanisms in memory.…”

“…Although the salience of treatment mechanisms may be difficult to measure daily during a clinical trial, they may be indirectly inferred by assessing the patterns of daily substance use in temporal proximity to the occurrence of therapy sessions. Daily and hourly substance use patterns during outpatient treatment reliably covary with psychosocial variables that indicate increased substance use motivation (Epstein & Preston, 2010; Hallgren et al, 2021; Padovano & Miranda, 2018; Palij et al, 1996; Roos et al, 2020). Substance use behavior is a good proxy for substance use motivation, although there may be situations when a patient has strong motivation to use substances, but it is extremely impractical for them to do so (e.g., during inpatient treatment or incarceration).…”

Practical assessment of DSM‐5 alcohol use disorder criteria in routine care: High test‐retest reliability of an Alcohol Symptom Checklist

“…Our findings suggest that the Alcohol Symptom Checklist may be a reliable tool for assessing AUD criteria and for supporting clinicians in diagnosing AUD and determining its severity (mild, moderate, or severe) in routine care settings. In particular, good‐to‐excellent test‐retest reliability observed in the full sample and the primary care subsample indicate that the measure may be a viable tool for monitoring whether patients are experiencing changes in their AUD symptoms over time, for example, when checklists are used for clinical monitoring or measurement‐based care (Bradley et al, 2019; Hallgren et al, 2021a, 2021b). Reliable change analyses indicated that an individual primary care patient who increases or decreases the number of AUD criteria reported by at least 3 criteria may be concluded to have experienced statistically reliable change (i.e., change that is unlikely attributable to measurement error alone) at the p < 0.20 level, and that a primary care patient who increases or decreases the number of AUD criteria by at least 4 criteria may be concluded to have experienced reliable change at the p < 0.05 level.…”

“…Our findings suggest that the Alcohol Symptom Checklist may be a reliable tool for assessing AUD criteria and for supporting clinicians in diagnosing AUD and determining its severity (mild, moderate, or severe) in routine care settings. In particular, goodto-excellent test-retest reliability observed in the full sample and the primary care subsample indicate that the measure may be a viable tool for monitoring whether patients are experiencing changes in their AUD symptoms over time, for example, when checklists are used for clinical monitoring or measurement-based care (Bradley et al, 2019;Hallgren et al, 2021aHallgren et al, , 2021b…”