Medication-assisted treatment (MAT) with buprenorphine, methadone, or naltrexone is now the evidence consensus for opioid use disorder, yet a substantial fraction of people with OUD still pursue abstinence-only recovery — through 12-step programs, residential detox, or counselling without pharmacotherapy — out of preference, stigma, or limited access. A 2023 NLP analysis of 4,048 patient reviews of buprenorphine/naloxone and methadone found that 82% of reviewers reported satisfaction with their medications and only 18% were dissatisfied. That 18% dissatisfaction figure is the closest available proxy for action-side regret, though it is a ceiling estimate: a dissatisfied patient may still, in retrospect, endorse the choice of medication over abstinence given the alternative outcomes.
The inaction side carries a harder clinical penalty. Post-detoxification literature consistently reports relapse rates of 72–88% within 12–36 months among patients who complete opioid detoxification without ongoing pharmacotherapy (PMC5046044). Dunn, Sigmon et al.’s 2011 systematic review found that only 20% of patients leaving residential detoxification remained abstinent from illicit opioid use at 30 days; the median opioid- negative rate at first post-taper follow-up across studies was 23%. A 70% midpoint of the 12–36 month relapse range is used as the inaction-side proxy, on the reasoning that patients who relapse after abstinence-only treatment — particularly those who subsequently accept MAT — frequently report wishing they had started medication earlier. No large-scale retrospective survey directly asks abstinence-only patients whether they regret not using medication, so this remains a proxy framing.
Barriers to starting MAT are well-documented and culturally entrenched. Bremer et al. (2023), comparing OUD-treatment barriers self-reported on social media against those found in published literature, found that stigma — including viewing buprenorphine as “a crutch” or “trading one addiction for another” — was the single most prominently reported barrier, appearing in 78.9% of reviewed publications and 47.6% of social-media barrier mentions. This stigma layer is what makes the temporal dynamics relevant: patients who initially resist MAT due to self-stigma and later relapse into active use are precisely the group expected, under Gilovich and Medvec’s framework, to experience inaction regret once the counterfactual (medication- supported stability) becomes visible. The 52-point gap between proxy rates should be read cautiously — it compares dissatisfaction with clinical failure, not two symmetrical retrospective measures — but the directional signal is clear: relapse after abstinence-only treatment is a categorically harder clinical outcome than dissatisfaction with a medication, which is why the inaction side of this comparison carries the heavier long-term cost.