Changes in non-opioid substitution treatment episodes for pharmaceutical opioids and heroin from 2002 to 2011

Highlights

• There were important differences between heroin and pharmaceutical opioid (PO)s.

• Those using codeine differed substantially from those using stronger opioids, such as oxycodone.

• Notable changes over time were observed with demographic characteristics for episodes.

• PO episodes were more likely to be short-term (e.g. detoxification, assessment).

• Dynamic changes over time highlight the importance of ongoing monitoring.

Abstract

Background

There has been a well-documented increase in the non-medical use of pharmaceutical opioids (PO) worldwide. However, there has been little detailed examination of treatment demand, or the characteristics of those presenting for treatment, particularly for treatments other than opioid substitution.

Methods

Data from closed drug and alcohol treatment episodes from the Alcohol and Other Drug Treatment Services National Minimum Data Set (AODTS-NMDS, representing non-opioid substitution treatment) in Australia for 2002–2003 to 2010–2011 were examined. In the four jurisdictions where detailed data were available, episodes where heroin was the principal drug of concern were compared to episodes for the four most frequently reported pharmaceutical opioids (morphine, codeine, fentanyl and oxycodone).

Results

In 2002–2003, most (93%) opioid treatment was related to heroin with seven percent of all opioid treatment episodes reporting a PO as the principal drug of concern. In 2010–2011, 20% of all opioid treatment episodes were attributed to POs. Distinct changes over time were observed for different opioids. There was an increase in the average age at the start of treatment for heroin and oxycodone episodes, and a reduction in the proportion of females for codeine episodes, with 67% in 2002–2003 compared with 44% in 2010–2011. Codeine and oxycodone episodes had the lowest current or past injection rates.

Conclusions

Clear differences were observed over time and between different opioids. Monitoring these emerging patterns will be important to inform treatment needs, particularly in light of different patterns of poly drug use, different routes of administration and changing demographic characteristics.

Introduction

Pharmaceutical opioid (PO) use and associated harms are an important international issue (Fischer et al., 2013c, Maxwell, 2011). Increases in mortality have been described as an ‘epidemic’ (Calcaterra et al., 2013, ONCDP, 2011). Over the past 10–15 years, substantial increases in prescribing and non-medical use of a range of opioids have been reported globally (Atluri et al., 2014, Degenhardt et al., 2007, Leong et al., 2009). In Australia, clear patterns of increased opioid prescribing (Leong et al., 2009) and increased non-medical use of analgesics and ‘other opiates’ (including morphine and oxycodone) have been seen in the general population (Australian Institute of Health and Welfare, 2014a). Likewise, increases in morphine and oxycodone use amongst sentinel samples of people who inject drugs have been reported (Stafford and Burns, 2012).

One important indicator of harm associated with PO use is seeking treatment for drug dependence. In the US, the proportion of all treatment admissions related to prescription opioid abuse increased from 2.2% to 9.8% between 1998 and 2008, with 26.5% of admissions for medication-assisted treatment being for the treatment of prescription opioid dependence (Substance Abuse and Mental Health Services Administration, 2010). Increased treatment admissions for codeine have been reported in South Africa (Myers et al., 2003). Australia has recorded an increase in hospital presentations for PO poisoning (Roxburgh and Burns, 2013), and increases in oxycodone- and fentanyl-related deaths have also been documented (Rintoul et al., 2010, Roxburgh et al., 2011, Roxburgh et al., 2013).

Studies of PO users indicate that some non-treatment seeking populations of pharmaceutical opioid dependent people, such as codeine users, may differ from heroin users in important ways such as employment and education (Nielsen et al., 2011). Further, a recent case series of drug treatment entrants identified differences between codeine and strong opioid users in terms of presenting characteristics and types of treatment received (Nielsen et al., 2014). To date, limited information is available to give a broader picture of PO treatment at a jurisdictional or national level, or to examine the patterns of drug and alcohol service utilization over the time period during which increased pharmaceutical opioid use has been observed. Recent indicators from opioid substitution pharmacotherapy treatment (OST) suggest a significant minority (around one in three) of people on OST report pharmaceutical opioids as the principal drug of concern at treatment entry (Australian Institute of Health and Welfare, 2014c).

Oxycodone has been the subject of much research interest, with new formulations of oxycodone being developed to counter concerns with misuse (Coplan et al., 2013, Sees et al., 2005). Less is known about opioids such as fentanyl and codeine. Fentanyl is a potent opioid with higher efficacy at the mu opioid receptor. Increased fentanyl prescribing has been reported in Australia and the US, with associated increases in mortality, and intentional misuse and injection described in the majority of deaths (Kuhlman et al., 2003, Roxburgh et al., 2013). Recent US studies have identified that between 9 and 20% of patients prescribed fentanyl display signs of non-adherent medication use (Layton et al., 2014, Passik et al., 2014). Reports are emerging of misuse and harms associated with codeine (Dutch, 2008, Frei et al., 2010, McDonough, 2011, Myers et al., 2003, Pilgrim et al., 2013, Sproule et al., 1999), particularly in those countries where access to codeine is less restricted. Estimates on rates of misuse and harms are less readily available.

The aim of this study was to use national drug treatment statistics to examine patterns of pharmaceutical opioid related presentations to services other than opioid substitution therapy (OST), including withdrawal, counselling, case management and support, information and education, and residential rehabilitation services. The two key areas we sought to examine were: (1) numbers of drug treatment episodes where a pharmaceutical opioid was reported as the principal drug of concern compared with episodes where heroin was reported as the principal drug of concern; and (2) to examine if there were differences in demographic and substance use characteristics reported with treatment episodes for ‘weaker’, less restricted opioids such as codeine and strong opioids such as morphine, oxycodone and fentanyl.