Mortality among clients of a state-wide opioid pharmacotherapy program over 20 years: Risk factors and lives saved

https://doi.org/10.1016/j.drugalcdep.2009.05.021Get rights and content

Abstract

Background

The small size of previous studies of mortality in opioid dependent people has prevented an assessment of the extent to which elevated mortality risks are consistent across time, clinical and/or patient groups. The current study examines reductions in mortality related to treatment in an entire treatment population.

Methods

Data from the New South Wales (NSW) Pharmaceutical Drugs of Addiction System, recording every “authority to dispense” methadone or buprenorphine as opioid replacement therapy, 1985–2006, was linked with data from the National Deaths Index, a record of all deaths in Australia. Crude mortality rates and standardized mortality ratios were calculated according to age, sex, calendar year, period in- or out-of-treatment, medication type, previous treatment exposure and cause of death.

Results

Mortality among 42,676 people entering opioid pharmacotherapy was elevated compared to age and sex peers. Drug overdose and trauma were the major contributors. Mortality was higher out of treatment, particularly during the first weeks, and it was elevated during induction onto methadone but not buprenorphine. Mortality during these risky periods changed across time and treatment episodes. Overall, mortality was similarly reduced (compared to out-of-treatment) whether patients were receiving methadone or buprenorphine. It was estimated that the program produced a 29% reduction in mortality across the entire cohort.

Conclusions

Mortality among treatment-seeking opioid-dependent persons is dynamic across time, patient and treatment variables. The comparative reduction in mortality during buprenorphine induction may be offset by the increased risk of longer out-of-treatment time periods. Despite periods of elevated risk, this large-scale provision of pharmacotherapy is estimated to have resulted in significant reductions in mortality.

Introduction

Illicit opioid use, especially heroin injection, has caused significant personal and public health problems in many countries across the globe (United Nations Office on Drugs and Crime, 2008). Apart from the burden to users, their families and the broader community, opioid dependence increases the risk of premature mortality (Darke et al., 2006). This elevated risk is concentrated across several causes of death: accidental drug overdose, suicide, trauma (e.g. motor vehicle accidents, homicide or other injuries), and HIV (in countries where HIV is prevalent among people who inject drugs) (Degenhardt et al., 2004, Degenhardt et al., 2006, Darke et al., 2006).

The mainstays of treatment for opioid dependence are pharmacological maintenance on methadone and buprenorphine, both of which are listed on the World Health Organization's (WHO) Model List of Essential Medicines (World Health Organization, 2005) for this indication. Methadone is an orally administered opioid agonist with a half-life of 24–36 h. Multiple randomized controlled trials have found that methadone treatment decreases illicit opioid use, improves social functioning, decreases offending behaviors and improves health (Ward et al., 1998, Mattick et al., 2003).

The need for supervised daily dosing of methadone in a defined treatment setting, and evidence of increase overdose death on induction into treatment prompted the search for alternative pharmacological treatment options (Mattick et al., 2001). As a partial agonist, buprenorphine produces less depression of respiration and consciousness than methadone, thereby reducing the overdose risk. Buprenorphine is longer acting than methadone, allowing for less than daily dosing.

Opioid pharmacotherapy is not without its own risks (Ward et al., 1998), nor does it completely remove the excess mortality risks that opioid dependent persons are known to face (Darke et al., 2006). Work has shown, for example, high mortality during the period of induction onto methadone (Caplehorn, 1998, Buster et al., 2002). More recent work has found that induction onto methadone, and cessation, carry elevated mortality risks (Caplehorn and Drummer, 1999, Buster et al., 2002, Brugal et al., 2005).

The small sample size of these studies has prevented an assessment of the extent to which these elevated risks are consistent across time and/or patient groups. Few existing examinations have had sufficient power to examine differences in risk across time and patient level variables. Further, these studies have typically focused on treatment groups rather than across entire treatment programs. No estimates exist of the size of reductions in mortality related to treatment for an entire treatment population while also considering other important predictors of mortality risk.

New South Wales (NSW) is the most populous State of Australia, with approximately six million residents. It has had an expanding and expansive opioid replacement program in place for almost thirty years. Over 40,000 people have entered treatment since 1985 (Burns et al., 2009). The size of this entire treatment population allows for an examination of important questions of clinical and population health interest. The aims of this study were to:

  • (i)

    Estimate overall mortality for all persons entering opioid pharmacotherapy between 1985 and 2006, by demographic and treatment variables;

  • (ii)

    Examine whether demographic or treatment variables were related to mortality levels during and following cessation of treatment;

  • (iii)

    Estimate mortality risk, according to specific causes of death, during time within treatment and following cessation of treatment;

  • (iv)

    Estimate the number of lives that may have been saved by the provision of methadone and buprenorphine in NSW over this period;

  • (v)

    Consider the estimated lives saved from improved clinical delivery of these treatments.

Section snippets

Sample

The NSW Pharmaceutical Drugs of Addiction System (PHDAS) is a database that records when an authority to dispense methadone or buprenorphine in NSW as an opioid replacement therapy to a particular person has been approved by the NSW Health Department. This study examined unit record data from the PHDAS database on all persons entering pharmacotherapy treatment between 1985 and 2006.

Exclusions from the analysis included: those who did not commence treatment; those in temporary programs, such

Overall results

Over the study period 42,676 clients entered treatment for a total of 425,998 person-years of follow-up (PY; median 9.2 years). The median episode length was 198 days, and participants entered into an average of 2.5 treatment episodes. Further details of treatment retention and re-entry are presented elsewhere (Burns et al., 2009) (see also Web Appendix 3).

During the follow-up period there were 3803 deaths, with an overall CMR of 8.9 deaths per 1000 PY (95% CI: 8.6–9.2; Table 1). CMRs were

Discussion

This is one of the largest and longest follow up studies of persons receiving opioid pharmacotherapy for illicit opioid dependence. Data were examined on over 40,000 treatment entrants across a large State-based program for whom patterns of entry and departure from treatment were tracked. Time in treatment was associated with lower mortality than time out of treatment, with an overall in-treatment SMR of 4.5 (95% CI 4.3, 4.8), compared to an out-of-treatment SMR of 8.0 (95% CI 7.7, 8.3) (RR

Conclusions

Mortality among opioid dependent people entering opioid pharmacotherapy is elevated compared to age and sex peers, with overdose, external causes and suicide the major contributors. This elevated mortality is higher when out of treatment (i.e. treatment reduces mortality), and it is particularly elevated during the first weeks out of treatment. The elevation in mortality varied in ways that probably reflect heroin availability and use. Mortality was highest during induction onto methadone. This

Role of funding source

The National Drug and Alcohol Research Centre and the National Centre in HIV Epidemiology and Clinical Research are funded by the Australian Government Department of Health and Aging, and are both affiliated with the Faculty of Medicine at the University of New South Wales. This study is funded through the National Health and Medical Research Centre project grant 455451. Louisa Degenhardt is supported by a NHMRC Senior Research Fellowship (ID #510279). Tony Butler is supported by a NHMRC Career

Contributors

L. Degenhardt conceived and supervised the study, and led the writing of the article. D.A. Randall undertook the statistical analysis. W.D. Hall contributed to the study design. M. Law provided advice on the statistical analysis of the study. All authors contributed to and have approved the final manuscript.

Conflict of interest

L. Degenhardt has been provided with funding by Reckitt Benckiser in the form of an untied educational grant to monitor the extent of injection of buprenorphine-naloxone injection after its introduction in Australia and to compare this with the injection of other OST forms. The design, conduct and interpretation of that study's findings were the work of the study investigators; Reckitt Benckiser had no role in these.

Acknowledgements

We would like to acknowledge the NSW Department of Health for providing the PHDAS data, and Ms Pia Salmelainen of the Pharmaceutical Services Branch, NSW Health, for her assistance with data extraction and interpretation.

References (21)

  • L. Degenhardt et al.

    The impact of the Australian heroin shortage on demand for and compliance with treatment for drug dependence

    Drug and Alcohol Dependence

    (2005)
  • Australian Bureau of Statistics, 2007. 2005 Causes of Death, Australia. Canberra, Australian Bureau of...
  • M. Brugal et al.

    Evaluating the impact of methadone maintenance programmes on mortality due to overdose and AIDS in a cohort of heroin users in Spain

    Addiction

    (2005)
  • Burns, L., Randall, D., Degenhardt, L., Hall, W., Law, M., Butler, T., Bell, J., 2009. Opioid agonist pharmacotherapy...
  • M.C. Buster et al.

    An increase in overdose mortality during the first 2 weeks after entering or re-entering methadone treatment in Amsterdam

    Addiction

    (2002)
  • J.R. Caplehorn

    Deaths in the first two weeks of maintenance treatment in NSW in 1994: identifying cases of iatrogenic methadone toxicity

    Drug and Alcohol Review

    (1998)
  • J.R. Caplehorn et al.

    Mortality associated with New South Wales methadone programs in 1994: lives lost and saved

    Medical Journal of Australia

    (1999)
  • C. Day et al.

    Documenting the heroin shortage in New South Wales

    Drug and Alcohol Review

    (2006)
There are more references available in the full text version of this article.

Cited by (284)

View all citing articles on Scopus

Additional background materials and data analyses are provided in six appendicies available with the online version of this article at doi:xxxxxxxx.

View full text