TOPLINE:

Substance-related intrusive memories in individuals with cocaine use disorder (CUD) frequently occurred independently of craving (42.4% of episodes); however, when these memories were more vivid and intrusive, they were associated with a greater intensity of craving, a Swiss study showed.

METHODOLOGY:

• Researchers conducted a 14-day ecological momentary assessment study in Switzerland using a smartphone app for the real-time data collection of intrusions, craving, and related experiences in naturalistic settings.
• The study involved 43 participants (age range, 18-59 years; 16.3% women) with a current diagnosis of CUD recruited from April 2022 to April 2023, with a median compliance rate of 82.8%.
• Participants reported a total of 360 event-based episodes, categorised as pure intrusions, intrusions with subsequent or simultaneous craving, or pure craving. Detailed data collection included intrusion characteristics (vividness, intrusiveness, and modalities); craving intensity; triggers; and cognitive-behavioural, emotional, and physiologic responses.
• Cocaine use severity (CUS) was computed as a standardised composite score of log-transformed weekly cocaine use, duration of use, and the Obsessive-Compulsive Cocaine Use Scale total score.
• Researchers employed linear regression models for participant-level outcomes and two multilevel models to account for repeated entries per participant, with variables adjusted for both individual and group averages.

TAKEAWAY:

• Intrusions rated as more vivid were significantly associated with a higher intensity of craving (unstandardised regression coefficient [b], 0.53; 95% CI, 0.21-0.86; P = .002) and a greater likelihood of craving (odds ratio [OR], 1.53; 95% CI, 1.27-1.85; P < .001).
• Intrusions rated as more intrusive were significantly associated with a higher intensity of craving (b, 0.48; 95% CI, 0.30-0.67; P < .001) and a greater likelihood of craving (OR, 1.67; 95% CI, 1.39-2.01; P < .001).
• Episodes involving craving were characterised by significantly greater distress (P < .001) and a greater loss of control (P < .001) than those involving pure intrusions and were associated with higher odds of reporting cocaine obtainment (OR, 19.90; P < .001).
• Higher CUS predicted more frequent intrusion episodes (b, 0.52; 95% CI, 0.26-0.79; P < .001); however, psychotherapy experience was associated with lower vividness (b, -1.45; P = .008), intrusiveness (b, -1.33; P = .004), and craving intensity (b, -1.56; P = .010).

IN PRACTICE:

"This study offers the first ecologically valid evidence of cocaine-related intrusions as frequent, distinct and impactful cognitive-affective events in CUD. Their dissociability from craving, yet strong predictive value for craving intensity when more vivid and intrusive, underscores their clinical relevance," the authors of the study wrote.

"Our findings position intrusions as key phenomena — transdiagnostic in nature and a key target in substance use disorders," they added.

SOURCE:

This study was led by Amelie Zacher, University of Zurich, Zürich, Switzerland. It was published online on April 27, 2026, in Addiction.

LIMITATIONS:

This study was limited by the moderate sample size (N = 43), the use of convenience sampling, and the low proportion of female participants. The reporting structure lacked an option to document craving followed by subsequent intrusions. Cocaine use measures relied on self-reports, which can be affected by recall and social desirability biases. Additionally, the event-based design introduced potential underreporting, and the study did not account for interoceptive awareness or comorbid conditions such as attention-deficit/hyperactivity disorder and depression.

DISCLOSURES:

The research was supported by the Clinical Research Priority Program Synapse, Trauma and Addiction of the University of Zurich. The authors declared having no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.