Kelly J. F., Greene M. C., Bergman B. G., White W. L. & Hoeppner B. B. (2019) Alcoholism: Clinical and Experimental Research, 43(7): 1533-1544. DOI: 10.1111/acer.14067. Click here to read.

Are we overestimating the number of recovery attempts required to achieve stable recovery from alcohol and other drugs (AOD)?

This study presents the results of a cross-sectional survey of US adults (n=39, 809) who report resolving a significant AOD problem (n=2002). The authors estimated the mean number of serious recovery attempts before entering stable recovery to be 5.35. The median number of recovery attempts was 2. Black race, prior use of treatment and mutual help groups, and psychiatric comorbidity were associated with a higher number of attempts. The number of recovery attempts did not differ between primary substances.

The authors conclude that the average number of recovery attempts may be substantially lower than previously believed as outliers and extreme values influence the mean value. They suggest that the median number – 2 – be used in clinical and policy communications. The skewed distribution is indicative of the presence of subgroups requiring more attempts to resolve their AOD problem, and suggests that substance use disorders (SUDs) are spectrum disorders characterised by heterogeneous aetiological pathways, diverse clinical profiles, and highly variable courses. AOD services should therefore be flexible in tailoring the intensity of interventions to the specific needs of the individual.


The language we use to communicate with our patients is likely to elicit varying degrees of hope and motivation. Whilst there is ample evidence that SUDs are chronically relapsing disorders, and of course we must be open with patients about the nature of addiction, we risk demotivating patients by emphasising the higher number of recovery attempts needed by patients with higher severity SUDs. The results of this study support the argument that we should be using the median estimate of recovery attempts in clinical and policy communications.

Working in substance misuse services we often see more complex patients with comorbidity and more severe SUDs, so anecdotally at least it might appear that most patients go through many recovery attempts before achieving stability. However, for those of us who work in primary care, and non-specialist services, we may see many more patients with mild or moderate substance use disorders, who require lower-intensity treatment and fewer recovery attempts to achieve stability. It is important that all professionals working with patients with an AOD problem have an overview of the variable trajectories of SUDs, and are able to appropriately communicate this to patients.

The authors point out that a treatment system designed around the mean clinical profile would have the unintended consequences of over-treating those with lower severity SUDs and high recovery capital, and under-treating those with high problem severity and minimal recovery capital. If patients perceive treatment as overbearing they may drop out of treatment; conversely, if services are perceived as inadequate and not able to meet the individual needs of a patient, they may disengage with treatment. Both of these scenarios have the potential to feed into the revolving door of acute treatment. The authors state that these findings underscore that one size does not fit all and that highly individualised approaches to addressing AOD problems are required. Accurate assessment, including assessment of available recovery capital, is vital to ensure that treatment is tailored to the needs of the individual.

There are several limitations to this study. Importantly, the term “serious recovery attempt” is not clearly defined and was open to interpretation by survey respondents. The data is also based on self-report and retrospective recall so may be open to bias. Furthermore, resolution of an AOD problem is not clearly defined. Does resolution refer to complete abstinence from a problem substance, or simply a reduction in use so that use no longer causes subjective distress? Future research should aim to replicate these findings with clearer definitions of recovery and recovery attempt.

To view the full clinical update by Tom Jones, please register to become a Premium Member.

Papers included in the full clinical update:

  • Alcohol-induced blackouts at age 20 predict the incidence, maintenance and severity of alcohol dependence at age 25: a prospective study in a sample of young Swiss men.

  • Uptake into a bedside needle and syringe program for acute care inpatients who inject drugs.

  • Emergency department treatment of cannabinoid hyperemesis syndrome: a review.

  • Life after opioid-involved overdose: survivor narratives and their implications for ER/ED interventions.