• Diamorphine;
  • heroin;
  • heroin-assisted treatment;
  • long-term outcome;
  • opioid dependence


Aims  To describe 4-year treatment retention and treatment response among chronic, treatment-resistant heroin-dependent patients offered long-term heroin-assisted treatment (HAT) in the Netherlands.

Design  Observational cohort study.

Setting and intervention  Out-patient treatment in specialized heroin treatment centres in six cities in the Netherlands, with methadone plus injectable or inhalable heroin offered 7 days per week, three times per day. Prescription of methadone plus heroin was supplemented with individually tailored psychosocial and medical support.

Participants  Heroin-dependent patients who had responded positively to HAT in two randomized controlled trials and were eligible for long-term heroin-assisted treatment (n = 149).

Measurements  Primary outcome measures were treatment retention after 4 years and treatment response on a dichotomous, multi-domain response index, comprising physical, mental and social health and illicit substance use.

Findings  Four-year retention was 55.7% [95% confidence interval (CI): 47.6–63.8%].

Treatment  Response was significantly better for patients continuing 4 years of HAT compared to patients who discontinued treatment: 90.4% versus 21.2% [difference 69.2%; odds ratio (OR) = 48.4, 95% CI: 17.6–159.1]. Continued HAT treatment was also associated with an increasing proportion of patients without health problems and who had stopped illicit drug and excessive alcohol use: from 12% after the first year to 25% after 4 years of HAT.

Conclusions  Long-term HAT is an effective treatment for chronic heroin addicts who have failed to benefit from methadone maintenance treatment. Four years of HAT is associated with stable physical, mental and social health and with absence of illicit heroin use and substantial reductions in cocaine use. HAT should be continued as long as there is no compelling reason to stop treatment.