A J-shaped relationship between supervised methadone consumption and retention in methadone maintenance treatment in primary care represents a ‘double edge sword’, according to authors of a study recently published in Drug and Alcohol Dependence.1 Funded by the Heath Research Board (HRB) through the HRB Centre for Primary Care Research, the study is the first to examine the influence of supervised methadone consumption on retention in methadone treatment over multiple treatment episodes in primary care.

Supervised methadone consumption

Supervised methadone consumption entails the administration of methadone to patients by a pharmacist or clinician, thus ensuring patients take methadone as prescribed. Ensuring patient compliance can prevent diversion of methadone to illicit drug markets and can reduce relapse to heroin use. Research has found supervised methadone consumption to be associated with a reduction in drug-related deaths, including those attributed to methadone. However, long-term supervision is resource intensive and may promote dropout from treatment due to the disruption to patients’ lives. Conflicting findings have emerged from the few studies that have compared supervised and unsupervised consumption.

At the time of the study, and consistent with World Health Organization recommendations, Irish guidelines for methadone maintenance treatment in primary care advised a minimum of one dose per week administered under pharmacy supervision.2,3 A dose of 60—120 mg daily, with prescriptions issued to dispense methadone for up to seven days, was further recommended.

Method

The sample consisted of 6,393 patients who experienced at least one methadone treatment episode between 2004 and 2010, and 19,715 treatment episodes. Patients were mostly male (68.5%) and aged under 30 years (58.6%).

The sample was identified by linking data from the Central Treatment List (the national register for methadone maintenance treatment); records from the Health Service Executive’s Methadone Treatment Scheme; the General Medical Services (GMS) pharmacy claims; and the HRB’s National Drug-Related Deaths Index (NDRDI). Included were persons aged 16—65 years who had at least three methadone prescriptions prescribed and dispensed in primary care during the study period. GMS provided data on all other prescription medications dispensed to these patients, while data from the NDRDI enabled the identification of persons who had died during the timeframe.

Prescription refill data were used to assess the level of supervised methadone for each treatment episode, with those dispensing a single dose categorised as supervised. For each patient, the percentage of supervised prescriptions was calculated and classified as 20%, 20—39%, 40—59%, 60—79%, or 80% or more. Prescription data were also used to calculate the total number of prescriptions (comorbidity score) for other drugs issued to each patient across the timeframe.

Patients were deemed to be in continuous treatment if they had received a new prescription within seven days of the end of coverage of a prescription, and as ceased treatment if they had not. Retention in treatment was designated for treatment episodes that had no interruption in prescribed methadone lasting more than seven days. The length of treatment episodes was based on the date of the first prescription and coverage of the last. Only episodes that started within the timeframe of interest were included in the analyses.

Statistical analyses examined the relationship between supervised methadone consumption and time to discontinuation of treatment across multiple treatment episodes, accounting for recurrent methadone treatment episodes, and including age, gender, median daily methadone dose and comorbidities as potential confounders.

Results

• 36% of patients were supervised for less than 20% of prescriptions, 16% for 20—59%, and 48% for 60% or more during the initial treatment episode.
• Across episodes, treatment discontinuation was least among patients supervised for 20—59% of prescriptions, and was greatest among patients supervised for 60% or more (indicating a J-shaped relationship).
• 67% of patients experienced more than one treatment episode; the median episode length for the initial treatment episode was 224 days; and the overall median episode length was 104 days.
• Daily methadone doses ranging from 60 to 120 mg per day were more effective at retaining patients in treatment than doses of less than 60 mg, or greater than 120 mg per day.
• The minimum recommended daily dose (60 mg) was not received by one-third of patients during the initial treatment episode.
• Many patients received co-prescriptions, most commonly for benzodiazepines (72%) and antidepressants (49%). Those with 11 or more were significantly more likely to have shorter treatment episodes than those with five or fewer.

Discussion

The J-shaped relationship identified suggests that with too little or too much supervision, patients may drop out of methadone treatment. This finding is consistent with trials in the US and Scotland, and is supported by qualitative research suggesting that supervision can be acceptable to patients in the short-term, as they develop a routine and establish relationships with staff, but that patients prefer to be unsupervised in the longer-term. The authors propose that other studies that found no differences in retention based on whether consumption was supervised or unsupervised failed to account for the relapsing nature of opioid addiction and the recurrence of treatment episodes. Consistent with research from Canada, many patients in the current study experienced multiple treatment episodes and were retained longer in later treatment episodes. Findings from a Scottish cohort study also suggest that cumulative exposure to opiate substitution improves patient survival. Although one-third of patients in the current study did not receive the recommended dosage, this proportion is lower than in the UK (57%) and Canada (51%). A further key finding is that many patients have comorbid conditions, and these patients tend to have shorter treatment episodes.

Conclusion

The findings highlight a challenge for clinicians: reducing supervision risks increasing the availability of street methadone and hence the population level risk of methadone deaths, while increasing supervision risks dropout from treatment and greater patient mortality.

The authors emphasise that further research is needed to profile patients suitable for unsupervised dosing, with the aim of retention in treatment and reduced diversion. The authors caution that the study methodology did not capture patients transferring from primary care to specialised settings, may have underestimated retention, and did not consider the quality of treatment.

2   Irish College of General Practitioners (2003) Working with opiate users in community based primary care. Dublin: Irish College of General Practitioners. https://www.drugsandalcohol.ie/5450/

3   Health Service Executive (2016) Clinical guidelines for opioid substitution treatment. Dublin: Health Service Executive. Available online at https://www.drugsandalcohol.ie/26573/