Opioid maintenance treatment is the most common form of treatment for people with opioid dependence in Norway. The treatment is often lifelong and constitutes a major interference in everyday life, partly because of frequent supervised administration of the opioid agonist drug, counselling sessions and supervised urine tests. This systematic review compares effect of unobserved with observed administration of opioid agonist drug for peoples receiving opioid maintenance treatment.
After a systematic literature search in relevant databases we included five primary studies that investigated unobserved compared to observed administration of opioid agonist drug. The most common used opioid agonist drugs are represented in these studies (methadone, buprenorphine and buprenorphine-naloxone). There was a variation in how participants in the unobserved administration group received the drug and how many take-home doses they were given every time.
In summary, we found the quality of the evidence to be very low. For that reason we are uncertain if unobserved administration compared with observed administration for people in opioid maintenance treatment influence:
- retention in treatment after 1, 4, 12 and 26 weeks
- use of illicit/ not prescribed opioids after 12 weeks
- number of persons that committed crimes after 12 weeks
- patient satisfaction and self-reported use of illicit drugs
- adverse events
There is a lack of evidence concerning mortality.
Opioid maintenance treatment (OMT) is the most common form of treatment for people with opioid dependence in Norway. Persons who receives OMT show improved survival, reduced damage to health, and better quality of life. People in treatment should be involved in forming a plan for their own rehabilitation to reach or sustain desired levels of function. The treatment is often lifelong and constitutes a major interference in everyday life, partly because of frequent supervised administration of the opioid agonist, counselling sessions and supervised urine tests.
This systematic review summaries the effect of unobserved administration compared with observed administration of opioid agonist drugs (methadone, buprenorphine or buprenorphine-naloxone) for individuals receiving opioid maintenance treatment.
We searched for primary studies in MEDLINE, Embase, PsycINFO, CENTRAL (Wiley) and CINAHL until the 12th. April 2016. In addition, we searched the reference lists in the studies examined in full text. Two people examined 3778 titles and abstracts independently, and 21 papers in full text and included five studies. The population in this systematic review is individuals with opioid dependency undergoing opioid maintenance treatment. The intervention is unobserved administration compared with observed administration of opioid agonist drug. We looked for randomized controlled trials, controlled studies with before and after measurements, interrupted time series with at least three measurement points before and three after the intervention, and a clearly defined point in time when the intervention started. We considered the following outcomes: Retention in treatment, use of illicit drugs, crime, patient satisfaction, self-reported misuse of drugs, unwanted events and mortality. Two persons assessed independently the risk of bias. The project leader retrieved data from the studies, analysed and assessed our confidence in the documentation; and another investigator double-checked the information. We used GRADE (Grading of Recommendation Assessment, Development and evaluation) to assess our confidence to the effect estimate and documentation. The confidence can be rated as high, medium, low or very low.
We included five studies: Four randomized controlled trials and one randomized controlled trial with crossover design. In total 521 eligible persons were randomized. All the studies compared unobserved with observed administration of the opioid agonist drug, apart from one with partly unobserved administration. The opioid agonist drugs used were methadone, buprenorphine or buprenorphine with naloxone. There was also a variation in how participants in the unobserved administration group received the drug and how many take-home doses the were given every time.
Our confidence in the effect estimate of unobserved versus observed drug administration was considered to be very low for all outcomes reported. Retention in treatment was measured after 1, 4, 12 and 26 weeks. None of the measurement points showed any differences between the groups. After 26 weeks, the difference in the number of retained patients measured as adjusted OR was 1.01 (95% CI 0.62 to 1.64). The confidence intervals were wide at all measuring points, and we are uncertain if retention in treatment is different with unobserved versus observed drug administration. One study reported no clear differences in use of heroin as measured in urine after 12 weeks with considerable uncertainty around the effect estimate as RR 0.96 (95% CI 0.61 to 1.52). Another study reported use of illicit/not-prescribed opioids measured in urine after 12 weeks as adjusted OR of 1.25 (95% CI 0.55 to 2.82). Self-reported use of illicit drugs also resulted in estimated effects with wide confidence intervals. Only one relatively small study with few events reported on the number of participants who committed criminal acts during the third month of the trial in the two groups. They reported somewhat higher proportion involved in crime in the group receiving observed administration measured as an adjusted OR of 2.82 (95% CI 1.11 to 7.16), but this evidence of effect is of very low quality. Only one relatively small study reported on six adverse events, but none of these were considered to be related to the treatment. Three of the studies reported that the study participants were in favour of unobserved drug administration.
We did not find sufficient information to answer our research question adequately. Based on five relatively small studies, we had very low confidence in the evidence of effects. We can therefore only indicate the probable effects of unobserved drug administration. A danger with unobserved administration is higher leakage of substitution drugs to the illegal street market, and an increased risk of overdoses. None of the included studies had measured these outcomes.
We found sparse evidence to indicate whether unobserved or observed drug administration is most appropriate for opioid maintenance treatment for opioid dependence. The quality of the evidence of effect in the studies we found is very low. We are uncertain if unobserved comparted to observed drug administration give better, worse or equally effect on retention in treatment, use of illicit drugs, crime, patient satisfaction, self-reported misuse of drugs, unwanted events and mortality.