I recently had the opportunity to see John Kelly present some of his research on recovery. Kelly is doing some of the most important work in the addiction and recovery space. I say this because he’s one of the few researchers publishing on long-term outcomes and quality-of-life. Most contemporary research gives us little more than a narrow view of reduced harm, relapse, and stabilization. Addiction is a chronic condition, making long-term support and monitoring critical, yet we have very little empirical knowledge about recovery over months, years, and decades.
Kelly’s presentation highlighted two important studies. I recently posted on one of those studies. The other study is several years older, and I was surprised to find that we’d never posted on it, so it will be the focus of this post.
The paper we’re focused on today is: Kelly, J. F., Greene, M. C., & Bergman, B. G. (2018). Beyond Abstinence: Changes in Indices of Quality of Life with Time in Recovery in a Nationally Representative Sample of U.S. Adults. Alcoholism, clinical and experimental research, 42(4), 770–780. https://doi.org/10.1111/acer.13604
This study focuses on quality-of-life in recovery over a period of years. It uses data from the National Recovery Survey (NRS) and includes people who reported that they “believe they once had a problem with alcohol/drugs but no longer do.” This likely leaves us with a very broad range of problem severity, and the authors note that only 45.2% identified as being “in recovery” and that 48.6% reported current use of at least one substance. So, keep in mind that the paper uses very inclusive boundaries for recovery.
The paper’s findings are really well conveyed through its graphs.
It gets better

Respondents reported that their problem had been resolved for time frames ranging from less than a year to 40 years. The good news is that quality-of-life increases and psychological distress decreases over the years, with relatively rapid improvements in the first ten years and slower continuing improvements in the following decades. (Note the lines for quality-of-life, happiness, recovery capital, and self-esteem starting very low and rising over the years, while psychological distress starts very high and declines over the years.)
The less good news is that it took, on average, about 15 years for quality-of-life to match that of people who never had alcohol or other drug problems.
But it gets worse before it gets better

So… we know that the quality-of-life improves over the years and decades. However, the early months are difficult. The graph on the right side (0 to 2 years) shows quality-of-life, happiness, and self-esteem declining over the first 9 months and psychological distress increasing over that same period.
The authors observe that this makes sense. Distress is a rational response as their brains clear and they begin to face the damage their alcohol or drug problem may have caused in their work, finances, family, housing, legal issues, mental health, etc.
The message suggested by these data is that if people can find a way to navigate through this difficult early phase in the first year of recovery the picture looks brigher in terms of gains in happiness and self-esteem. It also behooves those clinically involved in recovery management to help those early in recovery to shift their inter-temporal horizon toward the 6 and 12-month milestones as times of potentially significant positive cognitive and affective change.
Kelly, J. F., Greene, M. C., & Bergman, B. G. (2018). Beyond Abstinence: Changes in Indices of Quality of Life with Time in Recovery in a Nationally Representative Sample of U.S. Adults. Alcoholism, clinical and experimental research, 42(4), 770–780. https://doi.org/10.1111/acer.13604
Life’s terms aren’t distributed equitably

If a significant part of recovery is “living life on life’s terms,” it would make sense that life’s terms may be harder for some and easier for others. The authors noted demographic differences in the trajectories of respondents’ quality of life, happiness, and psychological distress.
During the first five years of recovery we observed that the same patterns of improvements were not evident for all individuals, with men, for instance, showing consistently higher quality of life and self-esteem and lower psychological distress compared to women…. Similarly, individuals reporting mixed race or a racial group other than Black, White, or Hispanic (e.g., Native American) showed persistent challenges following AOD problem resolution in these indices of quality of life and psychological well-being. This group includes native/indigenous Americans and other mixed ethnicities who appear to be assiduously disadvantaged and show worsening, rather than improvements, in quality of life and psychological distress over time. More needs to be learned about the scope and nature of the recovery needs of this subgroup, in particular, to help them achieve the same benefits of recovery as those in other racial-ethnic groups.
Kelly, J. F., Greene, M. C., & Bergman, B. G. (2018). Beyond Abstinence: Changes in Indices of Quality of Life with Time in Recovery in a Nationally Representative Sample of U.S. Adults. Alcoholism, clinical and experimental research, 42(4), 770–780. https://doi.org/10.1111/acer.13604
Differences by substance

Similarly, people with primary opioid and stimulant problems were significantly disadvantaged in comparison to people with primary alcohol and cannabis problems.
Finally, those with opioid or other drug use problems (e.g., stimulants) appear to begin their recovery journey at a substantial disadvantage in terms recovery capital relative to those resolving alcohol or cannabis problems, taking several years to achieve similar rates to other primary drug use groups. Individuals with these primary substance-related problems, in particular, may be among the most marginalized and stigmatized (i.e., those with heroin, methamphetamine, or crack cocaine as their primary substance).
What it all means
This paper is important because it gives us a glimpse into the quality-of-life over months and years into recovery.
I believe it’s important for a few reasons:
- Most published research focuses on active addiction, reduced harms, stabilization, and the narrow effects of interventions or treatments over a period of days and weeks.
- This study focuses on the experience of recovery, as it’s lived in the world, over a period of years.
- Its focus on evaluating quality-of-life avoids arguments about the conceptual boundaries of recovery. After all, people pursue recovery to improve quality-of-life.
- It shows that quality-of-life improves over the years, but the first 9 months are very difficult and characterized by elevated distress and declining happiness. This is really important for a few reasons:
- Without this information, those early months of elevated distress and declining happiness might be interpreted as a sign that they are on the wrong path when it’s actually just part of the process.
- It can be motivating to know that the difficulties of early recovery do pass if we stay on the path.
- Patients and treatment professionals can develop approaches to provide extra support through this difficult period and strategies to reduce this suffering.
- The knowledge that there are differences in quality-of-life across gender, race, and primary drug can be investigated and used to develop strategies to give everyone access to the same quality-of-life improvements.
It’s also worth noting that the paper does not tell us much about variation in quality-of-life for mild, moderate, and severe alcohol and other drug problems. (Though the challenges identified for people with primary opioid and stimulant problems may provide a glimpse.)
I hope to see more similar research in the future, particularly research that focuses on people with addiction (the most severe and chronic form of alcohol and drug problems) and gives us a better understanding of their starting point, their trajectory, and their challenges.
Dr. Kelly leads the Recovery Research Institute. You can follow their work here.

That paper was incorporated in the 5-year model monograph/practice guideline, as was the finding of feeling worse during early improving.
https://recoveryreview.blog/2025/04/08/5-year-continuing-care-system-for-high-severity-complexity-and-chronicity-suds-clinical-targets-methods-and-increments-of-time/