Drug and Alcohol Findings home page in a new window Background notes

Methadone maintenance therapy versus no opioid replacement therapy for opioid dependence


One of the four post-2000 studies used to demonstrate greater retention was initiated in prison The analysis was based on the 1 month post-release follow-up study. and in another patients were placed on a waiting list. The remaining two studies compared community methadone maintenance with either a short (21 days followed by aftercare) or a long (stabilisation for four months then methadone reduction over the next two followed by aftercare) detoxification regimen. These studies were a strong but limited indication that even fairly basic methadone maintenance retains patients far better than detoxification-based approaches.

Conducted in Sweden in the 1970s when methadone maintenance was rare and experimental, another study recruited young adult patients who had repeatedly relapsed after completing detoxifications. It was not included in the featured analysis because all the patients randomly allocated to drug-free treatments "refused to accept the ... alternatives offered and asked for discharge from the clinic as soon as they learned that methadone would not be given". In effect this amounted to zero retention compared to 11 of the 17 patients randomised to methadone remaining in treatment after typically four years.

Opiate use

In respect of opiate use (whether biologically confirmed or based on the patients' accounts), just two of the studies included in the relevant analyses implemented an active alternative treatment, though another offered this but it was turned down by all the patients. One was the previously cited comparison with a short (21 days followed by aftercare) detoxification option conducted in the USA. Conducted in Thailand, the second compared methadone maintenance against a 45-day methadone detoxification. It selected patients who had repeatedly relapsed after at least six previous detoxifications, possibly poor candidates for a further attempt and prime candidates for a maintenance option. In this study maintenance led to a statistically significant and substantial reduction in urine tests indicative of illegal opiate use. In the US study the reduction (indicated both by tests and the patients' accounts) was less but consistent and statistically significant. The featured review combined the two methadone maintenance groups (standard and minimal counselling) and found no statistically significant difference between these and detoxification, though it is unclear what stage in the study this was based on. The original paper reports that across the follow-up period, the difference between detoxification and minimal counselling methadone maintenance was statistically significant, and that that the two methadone regimens did not significantly differ. The analysis based on the patients' accounts included the Swedish study previously referred to, during which drug-free treatment was on offer but was refused by all the patients allocated to it. Two years later (the time when patients denied this could reapply for methadone maintenance), 12 of 17 methadone patients were no longer using illicit drugs compared to just five offered drug-free treatments.

A third study conducted in San Francisco was not included in the analyses, possibly because it combined urine test results and the patients' accounts in its assessment of whether they had used illegal opiates over the past month. It compared methadone maintenance against four months of stabilisation on methadone followed by two months' gradual withdrawal and continuing aftercare. During and after the withdrawal phase, significantly more detoxification patients used illegal opiates and on average they used on many more days than patients continuing to be maintained on methadone.


Across the three studies included in the analysis of criminality there was a substantial relative reduction in patients allocated to methadone which just failed to reach statistical significance. One of the studies was the Swedish study previously referred to, where two years later two of the patients denied methadone were in prison but none of those offered it. These seem to have been the figures incorporated in the featured analysis, but probably under-represent the reduction in crime in the methadone patients associated with reduced illegal heroin use. The same is true of a study which allocated patients to a waiting list or to a stripped down methadone programme. On the basis that two of the 149 methadone patients were in prison at the end of the study and just one of the 152 waiting list patients, the featured analysis records this study as indicating a non-significantly higher rate of criminality in the methadone patients. This is almost certainly the reverse of the truth. Waiting list patients had access to methadone after a month, so any reductions in crime could have been due to methadone treatment. During that month, twice as many waiting list patients gave urine tests indicative of continued heroin use; along with this probably went higher levels of criminality. The remaining study involved a programme initiated in prison which was compared to placing patients on a waiting list.

Set in Hong Kong, a further study not included in the analysis found that while still in the study, patients on methadone maintenance were convicted at half the rate per month of patients unknowingly detoxified from methadone and then prescribed a placebo. Data was available only while patients were retained in the study, and far fewer of the detoxified patients were retained. It seems fair to assume that the conviction rate of patients after drop-out from treatment and from the study would have been even higher. If so, the relative advantage of methadone would have been underestimated.


The most dramatic indication of the lifesaving potential of methadone was provided by the Swedish study previously referred to, where typically four years later, four of the 17 patients at first denied methadone were dead but none of the 17 offered it. In the featured analysis the previously cited study set in Hong Kong is recorded has having a higher death rate among the methadone maintenance patients. However, data was available only while patients were retained in the study, and far fewer of the detoxified patients were retained; by the end of the three-year study just one of the 50 detoxified patients was still in treatment compared to 28 of the 50 maintained patients. The fact that three deaths were recorded among maintained patients but one among the detoxified patients could simply be because the former were observed for on average 23 months, the latter for just five months. Of the remaining two studies included in the analysis, one was of a programme initiated in prison and the other involved patients allocated to methadone maintenance or to a waiting list.

Comment on these background notes Return to main entry
L10 Web Stats Reporter 3.15 LevelTen Hit Counter - Free PHP Web Analytics Script
LevelTen dallas web development firm - website design, flash, graphics & marketing