# Long-term relapse: markers, mechanisms, and implications for disease management in alcohol use disorder

**Authors:** John F. Kelly, Morgan Klein, Katherine Zeng, Sydney Manske, Alexandra Abry

PMC · DOI: 10.3389/fpubh.2025.1706192 · Frontiers in Public Health · 2026-01-07

## TL;DR

The study explores factors leading to long-term relapse in alcohol use disorder, focusing on psychological and social changes that may predict recurrence.

## Contribution

The paper introduces new insights into long-term relapse mechanisms, emphasizing self-regulation and coping factors rather than neurophysiological triggers.

## Key findings

- Changes in recovery vigilance were both prevalent and potent contributors to long-term relapse.
- Psychological and social factors were identified as the most influential in predicting relapse.
- The accumulation of risk factors tended to intensify as relapse approached.

## Abstract

Much has been theorized and documented about factors involved in alcohol use disorder (AUD) relapse during the early months following a recovery attempt where biobehavioral classical conditioning (“cues/triggers”) and neurophysiological explanatory theories predominate. Little has been documented, however, about long-term relapse (LTR) factors following sustained AUD remission where self-regulation and stress and coping theories may predominate because LTR precursors are centered less around neurophysiological dysregulation and cue reactivity and more around factors such as lowered recovery vigilance, avoidant coping, or changes in recovery-support services (RSS) usage. Greater knowledge of factors involved in LTR could sensitize and empower clinicians to deliver more effective disease management protocols to monitor and intervene upon such risks prior to AUD recurrence.

Cross-sectional, retrospective study of individuals in recovery from primary AUD (N = 50; 44% Female; 50% White) who had experienced LTR within the past 5 years following at least 1 year of remission (M years remitted prior to relapse = 3.6; range = 1–23) and assessed for any change in bio-psycho-social domains or RSS usage during the year prior to LTR, along with their attributions of factors' contribution to relapse (risk “potency”; i.e., didn't contribute, possibly, probably, or definitely, contributed). Research questions focused on the year preceding the LTR assessing: (1) prevalence and nature of the bio-psycho-social and RSS use changes and degree of attributed LTR risk potency; (2) number and type of definitely-contributing relapse factors within participants; (3) dynamic temporal onset and nature of high-risk LTR precipitants; (4) single most influential LTR risk factor.

Several bio-psycho-social and RSS changes occurred during the year preceding LTR varying in prevalence and potency. Some were prevalent, but not potent, in terms of definitely contributing to LTR (e.g., sleep, energy); others occurred infrequently, but were potent (e.g., physical pain, recreational drug use); others were both highly prevalent and highly potent (e.g., change in recovery vigilance). Within participants, median number of definitely contributing LTR factors = 4, covering 2 different domains, on average. Temporal accumulation of LTR risks tended to intensify toward the relapse horizon over the preceding year. The single most important relapse factor tended to cluster in psychological (e.g., recovery vigilance, mental health) and social domains.

Findings have implications for long-term disease management during AUD recovery providing a set of potential preliminary markers and mechanisms that might be assessed, monitored, and, when necessary, intervened upon prior to the onset of heavy symptomatic alcohol use to prevent AUD recurrence.

## Full-text entities

- **Diseases:** pain (MESH:D010146), AUD (MESH:D000437)
- **Chemicals:** alcohol (MESH:D000438)

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12819679/full.md

## Figures

7 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12819679/full.md

## References

32 references — full list in the complete paper: https://tomesphere.com/paper/PMC12819679/full.md

---
Source: https://tomesphere.com/paper/PMC12819679