Families rarely ask this out loud, because asking it feels like admitting they expect to fail. It usually arrives instead as a statement, somewhere near the end of a phone call:
She’s already been twice.
The tone is always the same — that this fact is damning, that it means something is fundamentally different about this person, and that the odds have now run out.
The research does not support that reading.
The number is more than one
Kelly and colleagues (2019) approached the question directly, asking a national sample of people who had resolved a significant alcohol or drug problem how many serious attempts it had taken them.
The median answer was two. Half of everyone who resolved a serious alcohol or drug problem did it within two serious attempts, and the middle half of the sample fell between one and four.
The mean, though, was 5.35.
That gap is the most useful thing in the paper. A relatively small number of people who needed a great many attempts drags the average up to more than double the median. So “the average person needs five tries” is true and badly misleading — most people needed far fewer, and a minority needed very many.
Two consequences follow, and they point in opposite directions, which is why both are worth saying. If your person is on attempt two or three, they are squarely in the ordinary range and nothing unusual is happening. And if they are on attempt eight, they are in a real and recognised group — one the researchers specifically noted — for whom more attempts were needed. That is not the same as a group for whom it did not work.
The low median also held when the authors looked only at people in stable recovery of five years or more, who are much less likely to add to their tally later. It is not an artefact of counting people early.
And a great many people do get there
The second finding is the one worth telling families, because almost nobody knows it.
Kelly and colleagues (2017) produced the first national probability-based estimate of how many US adults have resolved a significant alcohol or drug problem. The figure is 9.1% — about 22.35 million people.
That is roughly one adult in eleven, walking around in ordinary life, with a resolved problem behind them. They are simply not visible, because resolved problems do not announce themselves. What is visible is the people still struggling — which produces a badly skewed impression of the odds.
That study also found people get there by different routes, and that which route they take varies by substance — use of formal treatment or mutual help was significantly higher for opioids and significantly lower for cannabis than for alcohol. There is not one path, and the absence of a single correct route is itself good news for a family that has tried one thing and found it did not fit.
What a previous attempt actually tells you
If a relapse were random, a previous attempt would tell you nothing. It isn’t, and it does.
Each attempt produces information that did not exist before. You now know something about when it came apart — which week, which circumstance, how long after discharge. You know what was in place and what wasn’t at that moment. You know which parts of the plan the person could actually sustain and which were never realistic. And you know something about the specific conditions under which their resolve fails, which is much more useful than a general belief that it does.
A third attempt designed around all of that is not a repeat of the second. It is a different intervention that happens to be the third one.
The version that genuinely does not work is the one where nothing changes except everyone trying harder.
The pattern in repeated attempts
Something worth noticing: when families describe a sequence of attempts, the failure point is frequently the same each time.
Often it is the transition — the first weeks after treatment ends, when every protective structure disappears at once. Sometimes it is a specific recurring event: a season, an anniversary, a work cycle, a particular relationship. Sometimes it is the point at which support was withdrawn because things appeared to be going well.
If the same stage keeps recurring, that stage is the thing to design around — not the treatment programme, which may have been perfectly adequate up to the point where it stopped.
This is a large part of why we work on the period after treatment rather than on treatment itself. For a great many people, the repeated failure point is not the thirty days. It is the thirty-first.
How to think about the number
If someone you love has been through this twice, three times, five times, the reasonable interpretation is not that they are beyond help.
It is that they are somewhere on a distribution that most people who eventually resolve this were also on — and that each attempt has produced information which, if anyone actually uses it, improves the design of the next one.
The question worth asking is not how many more times. It is what was different about the moment it came apart, and what would have to be in place for that moment to go otherwise.
References
- Kelly JF, Greene MC, Bergman BG, White WL, Hoeppner BB (2019). How Many Recovery Attempts Does it Take to Successfully Resolve an Alcohol or Drug Problem? Alcoholism: Clinical and Experimental Research. 10.1111/acer.14067
- Kelly JF, Bergman B, Hoeppner BB, Vilsaint C, White WL (2017). Prevalence and pathways of recovery from drug and alcohol problems in the United States population. Drug and Alcohol Dependence. 10.1016/j.drugalcdep.2017.09.028
Read more about Personal Recovery Assistant.
