Introduction
You've lost count, which is itself an answer of a kind. Each time you said it was the last, and each time something in you found another one. Now you're asking whether there's a rule — a defensible number, after which nobody could blame you for stopping.
There isn't. But the question can be improved, and the evidence is more encouraging than the exhaustion suggests.
The Number Is Probably Lower Than You Fear
Something worth knowing before you decide anything.
A large national study of American adults who once had a significant alcohol or drug problem and no longer do asked how many serious recovery attempts it took. The answer surprised the researchers.
The median was two.1 The middle 50% of people needed between one and four attempts. The single most common answer was one. Roughly 13% reported making no "serious" attempt at all.
The mean was 5.35 — but that average is dragged upward by a small number of people at the extreme end, where attempts ranged as high as 100. The researchers themselves concluded that the median is the more appropriate figure to communicate, and warned that the cultural image of addiction as an endlessly relapsing condition may reduce hope and discourage people from seeking help.
Notably, the number of attempts did not differ by primary substance. Opioids and alcohol looked the same.
What That Does and Doesn't Mean
It does not mean your person will recover on attempt two. Higher numbers of attempts were associated with greater severity, prior treatment, and psychiatric comorbidity — which may well describe your situation.
What it means is that the expectation you're carrying — that this will go on forever, that people like this never stop — is not supported by the evidence, and it may be shaping how you think about a decision you haven't made yet.
The Thing You're Actually Afraid Of
Underneath the question is usually a specific fear, and it should be said out loud rather than left to operate silently.
That if you stop, they will die, and it will have been because you stopped.
This is the belief that keeps people in place for decades, and it needs examining. Your proximity has not stopped them so far. People do not use because someone left and they do not recover because someone stayed. The idea that you are the load-bearing wall of another adult's survival is something addiction installs in the people who love someone with it, and it has ruined a great many lives on both sides of the arrangement.
If they die, it will not be because you protected yourself. That sentence is worth reading twice.
But "Chances" Is the Wrong Frame
Here is the more important thing.
"Chances" implies a fixed quantity you dispense, after which you have run out and may leave with your conscience clear. It converts a relationship into an account, and it makes the decision about them — their behavior, their sincerity, whether they have earned another.
That framing has a cruel feature: it means you can never leave in good conscience while they are still trying. You are permanently on the hook, waiting for a failure severe enough to authorize your departure.
The better question is not how many chances they deserve. It's what you can survive.
Ask What This Is Costing You
Specifically, and out loud.
What has it taken from your health, your sleep, your finances, your other relationships, your children if you have them? What has it cost the version of you that existed before this?
Those costs are real and they belong to you. You are permitted to weigh them without first proving that the other person is beyond hope. Leaving is not a verdict on their worth. It can simply be a recognition of what you have left.
Distinguish Chances From Conditions
Something more useful than a count.
A chance is unconditional and it teaches nothing. A condition is specific and observable: I will stay while you are engaged in treatment. I will not stay while you are using in this house.
The difference is that a condition can be met, and can be seen to be met, by both of you. It doesn't require you to assess sincerity — which is unknowable — only behavior, which is visible.
Conditions also protect you from the trap of the count. You are not deciding whether they deserve a fifth chance. You are stating what you require, and they are choosing.
Only Set What You Will Enforce
The most important operational rule.
A condition you don't enforce is worse than no condition. It teaches, unmistakably, that your stated limits are negotiable, and it means the next one carries less weight than the last.
Which means: do not announce a boundary you are not prepared to act on. Set a smaller one you will actually hold. A small enforced boundary is worth more than a large empty one.
You Are Allowed to Stop
Said plainly, because nobody says it to you.
There is no threshold of suffering you must reach before you are permitted to protect yourself. You do not have to wait for a catastrophe that authorizes your departure. You do not owe someone unlimited access to your life because you love them, and love does not obligate you to be destroyed alongside them.
People leave, and the person left sometimes recovers, and sometimes doesn't, and neither outcome tells you whether leaving was right.
Get Support of Your Own
Whatever you decide.
Family groups, therapy, and structured programs for people in your position exist because this position is a recognized and specific ordeal. You are not a supporting character in someone else's recovery. Your life is also happening, and it has been happening in a state of emergency for a long time.
The Bottom Line
The median number of serious attempts before people resolve a substance problem is two, and the cultural picture of endless relapse is not what the data shows. But "how many chances" is the wrong question, because it makes your survival contingent on their failure being severe enough. Replace chances with conditions you will actually enforce, weigh what this is costing you without apology, and know that you are permitted to stop.
Sources
- Median 2; endless-relapse picture unsupported — 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, 43(7):1533-1544. View source ↗