Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Addict Med. Author manuscript; available in PMC 2012 September 1.
Published in final edited form as:
PMCID: PMC3156987

The Source of Methadone in Overdose Deaths in Western Virginia in 2004



Methadone-related overdose deaths increased in the United States by 468% from 1999 to 2005. Current studies associate the nonmedical use of methadone with methadone-related deaths. This study describes medical examiner cases in rural Virginia in 2004 with methadone identified by toxicology and compares cases according to source of methadone.


In 2004, all intentional and unintentional poisoning deaths from the Office of The Chief Medical Examiner, Western District of Virginia, were reviewed to identify cases in which methadone was a direct or contributing cause of death. The Virginia Prescription Monitoring Program was reviewed for prescription opioids in the name of these identified decedents. Decedent participation in local opioid treatment programs (OTP) was also assessed.


The source of methadone in the 61 methadone-related overdose deaths was mostly nonprescribed (67%), although 28% of decedents were prescribed methadone for analgesia. Only 5% of decedents were actively enrolled in an OTP. The majority of deaths were attributed to polysubstance overdose.


The majority of methadone overdose deaths in this study were related to illicit methadone use, rather than prescribed or OTP uses. Interventions to decrease methadone-related deaths should focus on reduction of nonprescription use of methadone.

Keywords: chronic pain, illicit drug use, methadone overdose, prescription monitoring

Methadone exhibits a long and variable half-life of 15 to 55 hours together with the potential for multiple drug interactions (Wolff, 2002). Pharmacokinetic studies of methadone indicate that it has high oral bioavailability, a rapid and extensive distribution phase into the tissues of the brain, liver, kidneys, muscle, and lungs, and a very slow elimination phase as these extravascular sites slowly release methadone into the plasma (Intrussi and Verebely, 1972; Inturrisi et al., 1987). The elimination half-life of methadone has considerable interindividual variability due to genetic differences and environmental factors (Eap et al., 2002). In addition, sustained respiratory depression lasting 24 to 48 hours occurs despite the analgesic effect of methadone lasting only 4 to 6 hours (Wolff, 2002). Effects are intensified in methadone-naive patients, patients nonadherent to dosing regimens, those using multiple substances to attain euphoria, and patients prescribed CYP450-inducing medications. Liver dysfunction further prolongs elimination of methadone. These aforementioned characteristics increase methadone's potential for delayed toxicity and poisoning relative to other opioids (Wolff, 2002).

Methadone has been used for the outpatient treatment of opioid addiction since 1966 (Joseph et al., 2000). Methadone maintenance treatment decreases illicit opioid use and decreases mortality among opioid-dependent patients (Gronbladh et al., 1990; Caplehorn et al., 1996; National Institutes of Health, 1999). Outside opioid treatment programs (OTPs), methadone prescribing surged in the last decade because of its reasonable cost and efficacious analgesic profile (Wheeler, 2000; Dickerson, 2001; Oregon Department of Human Services, 2003; Substance Abuse and Mental Health Services Administration [SAMHSA], 2004). Methadone prescriptions increased by 5-fold in New Hampshire from 2000 to 2003 (Andrew, Duval, abstract presented at the Joint Meeting of SOFT and TIAFT), 5-fold in Oregon from 1997 to 2001 (Oregon Department of Human Services, 2003), and 4-fold in North Carolina from 1997 to 2001 (Ballesteros et al., 2003; US Department of Justice [USDOJ], 2004). Similar pattern increases were noted for oxycodone and hydrocodone prescriptions, though to a lesser degree (USDOJ, 2004). From 1999 to 2005 methadone poisoning increased by 468% nationwide, while poisoning deaths due to heroin increased by 2.4% and other synthetic opioids increased by 138% (Fingerhut, 2008).

Given the concomitant rise in methadone prescribing and methadone-related deaths, some have hypothesized that physician inexperience prescribing methadone for chronic pain is a potential contributor, and have taken steps to better educate physicians about appropriate prescribing practices (Oregon Department of Human Services, 2009; SAMHSA, 2009). In 2006, the Food and Drug Administration also lowered the starting dosage recommendation of methadone to 2.5 to 10 mg every 8 to 12 hours from 2.5 to 10 mg every 3 to 4 hours as needed for initiation of analgesic therapy in opioid-naive patients (Roxane Laboratories, 2000, 2006). The new recommended initial dose for treatment of opioid dependence is no more than 40 mg/d (Roxane Laboratories, 2006). Alternative reasons for the concomitant rise in methadone prescribing and mortality include increased opportunities for the nonmedical use and diversion of methadone (Cairns et al., 1996; Lehder et al., 2002; Sorg and Greenwald, 2002; Ballesteros et al., 2003; USDOJ, 2004; Paulozzi et al., 2009). Empiric data regarding the source of methadone in overdose deaths are limited. Estimates of the contribution of physician methadone prescriptions to opioid overdose deaths vary widely from 15% to 75% of medical examiner cases (Barrett et al., 1996; Ballesteros et al., 2003; Gagajewski and Apple, 2003; Shah et al., 2005; Paulozzi et al., 2009). Only 2 studies, under the same Centers for Disease Control and Prevention investigation, have used a prescription-monitoring program to link methadone prescriptions to overdose deaths (Hall et al., 2008; Paulozzi et al., 2009). Both studies found evidence of significant prescription opioid nonmedical use and a high prevalence of a history of substance abuse among overdose decedents in West Virginia, regardless of whether they were prescribed methadone from a physician or not (Hall et al., 2008; McLellan and Turner, 2008; Paulozzi et al., 2009).

During the last decade there has been a 300% increase in medical examiner--investigated deaths in western Virginia identifying prescription opioids as a direct or contributing cause of overdose death (Wunsch et al., 2009). A recent study of opioid-related deaths in western Virginia from 1997 through 2003 reported that methadone was the most commonly identified opioid associated with accidental overdose (Wunsch et al., 2009). In that study, the source was not determined primarily because the Prescription Monitoring Program (PMP) was not established in Virginia until 2003.

The objective of this study was (a) to describe methadone-related deaths secondary to poisoning in western Virginia from January 1, 2004, to December 31, 2004, and (b) to determine the source of the medication by linking medical examiner and OTP records with Virginia's PMP. We hypothesized that in most cases the source of methadone was from physician prescriptions for analgesia rather than for opioid addiction treatment or an illicit source.


Study Design and Setting

This is a retrospective descriptive case series of persons who died from methadone poisoning in western Virginia from January 1, 2004, to December 31, 2004. Poisoning in this study is defined as death associated with intoxication by licit and/or illicit drug substances as determined by the Office of The Chief Medical Examiner (OCME), Western District, Va, and will be referred to by the term “overdose.”

Cases were drawn from the OCME, Western District, Va. The office's jurisdiction services 34 counties with a population of 1.6 million people. The region is predominantly rural with 7 metropolitan statistical areas, each with fewer than 100,000 people. Three private OTPs serve this region with a combined 2004 census of 1505 patients. All OTPs in the region participated in the study.

The institutional review board at the Virginia College of Osteopathic Medicine approved this study. Authorization to review medical examiner files and prescription monitoring data was obtained from the OCME who had access to the Virginia PMP for OCME cases.

Case Identification

Medical examiner cases include suspicious, violent, or unnatural deaths, or deaths related to the public interest such as incarcerated prisoners or patients in mental institutions. In the Western District, the suspicion of licit or illicit drug poisoning in a death confers jurisdiction to the OCME. Therefore, referrals of all suspected poisoning deaths in the region are made to the OCME for further evaluation. The cause of death in cases when a decedent has been hospitalized for greater than 24 hours is determined by the attending hospital physician and therefore is exempt from OCME referral. Cases when a decedent dies within 24 hours of hospitalization are referred to the OCME for cause of death determination. All decedents referred to the OCME undergo toxicology testing to screen for illicit drugs, ethanol, and prescribed pharmaceuticals including opioids, stimulants, depressants, and other prescribed medications (eg, antidepressants and antiemetics). All methadone deaths reviewed for this study were subject to a complete autopsy with comprehensive toxicological analysis including quantitative determinations of drug and drug metabolite concentrations.

Determination of cause and manner of death is the statutory responsibility of the medical examiner. The causes of death in this analysis were determined by medical examiners on the basis of the information gathered from the toxicological data (including postmortem methadone blood concentrations), autopsy findings, police reports, scene photographs, accident reconstruction, and witness statements.

For this analysis, all Western District medical examiner cases occurring between January 1, 2004, and December 31, 2004, where the cause of death was poisoning, were identified for possible inclusion in the study. Poisoning deaths where illicit drugs or medications were present on toxicological analysis and were a direct or contributing cause of death were selected for further review. Methadone-related cases were defined as those deaths where the manner of death was intentional and unintentional, methadone was present in the toxicological analyses, and methadone was deemed a direct or contributing cause of death by the medical examiner. [medium shade]Each decedent was given a unique identifier for tracking purposes.

Data Sources

We collected data from 3 sources: (1) medical examiner records, (2) OTP medical records, and (3) Virginia PMP database records.

Medical examiner records included autopsy reports; toxicological reports; hospital records including information about previous overdoses or suicide attempts; physician notes and records including information about a history of substance abuse, pain syndromes, history of mental illness, and medication records; death investigation and death scene reports including photographs and documentation of medications or drug paraphernalia found at the death scene; police reports including witness statements and information regarding previous criminal histories; and death certificates that include demographic information including the date of birth, sex, race, and home address. These records were extensively reviewed and pertinent data extracted for each decedent case.

Executive directors of the 3 regional OTPs were presented a list of decedents including name and date of birth and asked to verify decedents who were patients in their clinic at any time in the past. After confirming OTP enrollment, the principal investigator (MBW) extracted information from the patients' OTP medical charts including admission and discharge information, dosing orders, take-home allowance, and urine drug screen result history. A decedent could also be classified as OTP source if there was compelling evidence in the medical examiner files for active methadone maintenance therapy at an OTP other than the 3 that were contacted.

The Virginia PMP is an online database maintained by the Department of Health Professions and requires pharmacies, nonresident pharmacies, permitted physicians, and physicians holding permits to sell controlled substances to report dispensation of schedule II-, III-, and IV-controlled substances. The PMP was initiated in August 2003 as a pilot program in the region of western Virginia served by the Western District OCME. During the pilot program, only schedule II-controlled substances were tracked by the PMP. Methadone prescribed by an OTP is not listed in the PMP. Using each decedent's name and date of birth, the PMP was queried for any dispensed schedule II-controlled substances from August 2003 to date of decedent death, but no later than December 31, 2004. Reports generated from the PMP database included name of medication, the time when date prescription was filled, medication dose, quantity of medication dispensed, prescribing physician, and pharmacy name and address.

Decedent Characteristics

Decedent sociodemographic characteristics were abstracted from medical examiner files and included age at death (years), sex (male, female), and race (white, other). Decedent's place of residence was classified as rural or nonrural (eg, metropolitan or micropolitan) according to the Government Office of Management and Budget definitions (Federal Register, 2000). Decedents' history of substance abuse was defined as any autopsy evidence of nonmedical route of drug administration, documentation in OCME and medical records of a history of addiction or substance abuse, history of unintentional drug overdose requiring hospitalization, history of treatment in a substance abuse treatment program, or a positive toxicology screen for cocaine, marijuana, or heroin on postmortem toxicological examination. Decedents were classified as having questionable prescription activity if there was PMP evidence of greater than or equal to 4 physicians prescribing them opioids at one time or greater than or equal to 4 pharmacies used at one time (Katz et al., 2010).

Medical records within the medical examiner file were also searched for evidence of previous methadone overdose history, suicide attempt, or previous or current incarceration. Decedents were classified as having had a previous methadone overdose history or a previous suicide attempt if there was documentation within either a hospital record or other medical record. Decedents were classified as having a previous or current history of incarceration as documented in their police investigations.

The primary cause of death (ie, methadone only vs polysubstance overdose) and manner of death were extracted from OCME autopsy report diagnoses as found in the medical examiner file.

Source of Methadone

For each decedent, we attempted to identify all possible sources of methadone. The possible sources of methadone were (a) prescribed by a non-OTP physician, (b) prescribed by an OTP-physician, or (c) illicit. Decedents were classified as having received a prescription from a non-OTP physician (a) when their name and a corresponding methadone prescription were found in the PMP or (b) when a current prescription for methadone in the decedent's name was found at the death scene. Decedents were classified as having received methadone from an OTP if they were enrolled in an OTP within the 2 weeks before death. Decedents were classified as obtaining methadone from an illicit source when the decedent did not have a prescription for methadone in the PMP, from an OTP, or at the death scene. Cases in which there was suspicion for drug sharing or stealing were also classified as illicit source.

Toxicological Analyses

Postmortem blood concentrations of methadone were used along with all available objective information in the cases to formulate a cause of death statement. Postmortem toxicology data were available for all decedents in this study and included mean methadone, other drugs, and drug metabolites concentrations. The mean methadone concentrations reported were blood or body fluid measurements representing both central and peripheral collection sites, although, whenever available, quantitative analyses of methadone were performed in a peripheral specimen of blood collected at autopsy (eg, femoral or iliac blood). Quantitative analyses of drugs and drug metabolites in postmortem blood, body fluids (urine, gastric fluid, and bile), and tissues (liver) were performed by a number of validated analytical methods employing gas chromatography-mass spectrometry, gas chromatography-mass spectrometry/selected ion monitoring, high-performance liquid chromatography with diode array or fluorescence detection, and gas chromatography with electron capture detection.

Statistical Analyses

Descriptive statistics were utilized to report decedent characteristics and assess overall associations between decedent characteristics and source of methadone. Chi-square tests were employed for categorical data and simple regression with Wald test for continuous variables. Bivariate associations between decedent characteristics and source of methadone were identified using simple logistic regression.


Decedent Characteristics

In 2004, there were 203 overdose cases examined by the OCME. Methadone was detected by toxicology and deemed a direct or contributing cause of death in 68 of those cases (34%). Seven cases were excluded because decedents resided outside of the Virginia PMP pilot study area. Table 1 describes the characteristics of the remaining 61 decedents who exhibited an average age of 36 years (age range, 19-59 years), were predominantly white men, and were nearly as likely to be from rural areas as nonrural areas. Forty-six percent did not meet study criteria for history of substance abuse, yet autopsy and final toxicological analysis reported the cause of death in the majority of cases (61%) to be polysubstance overdose. The average number of substances contributing to decedents' deaths was 3, with a range of 1 to 6 substances. Of note, 6-acetylmorphine, the analytical toxicological marker for heroin, was identified in only 1 of the 61 decedents (#55; Table 2).

Methadone-Overdose Decedent Characteristics, Western Virginia 2004
Individual description of methadone-overdose decedents, western Virginia 2004

Table 2 provides a full qualitative report about each decedent. Among cases with a history of incarceration (n = 3), 1 decedent (#13) was on house arrest at the time of death and 1 decedent (#1) was released from jail within 30 days of death. One decedent (#67) had a history of previous nonintentional methadone overdose occurring 1 week before her final overdose. Six decedents (10%; #6, 16, 27, 28, 34, and 59) had a medical history of prior suicide attempts documented in their medical examiner files. One manner of death (#27) was classified suicide. A history of substance abuse was found in 33 decedents. Eight decedents (#2, 4, 9, 14, 17, 18, 25, and 27) had a history of ever being enrolled in an OTP, although only 3 (#2, 17, and 18) of these decedents were actively enrolled at the time of death. Nine decedents (#26, 33, 35, 37, 42, 55, 56, 67, and 68) exhibited concomitant cocaine use by toxicology.

Source of Methadone

The source of methadone was illicit in 41 (67%) decedents, obtained from an OTP in 3 (5%) decedents, and prescribed by a non-OTP physician in 17 (28%) decedents. Two of the 17 decedents (#6 and 13) with prescriptions by a non-OTP physician had prescriptions found at the death scene, but not in the PMP. There were no cases in which more than 1 source was suggested on the basis of a combination of information obtained from the death investigation and from the PMP and OTP records. Table 1 compares decedents' characteristics by source of methadone.

Illicit Source of Methadone

Forty-one decedents neither had a prescription recorded in the PMP nor were enrolled in an OTP at any time before death and nor had a prescription in their name at the death scene; therefore, the source of methadone was classified as illicit. Decedents with illicit methadone source were younger than those with other sources of methadone (Table 1). For every 1-year increase in age, the odds of illicit source of methadone decreased by 8% (OR = 0.92; 95% confidence interval [CI], 0.86-0.97). Decedents with illicit methadone source were also less likely to have antidepressants found by toxicology (OR = 0.17; 95% CI, 0.05-0.61). Toxicological analysis of decedents with illicit methadone frequently revealed other opioids (OR = 3.27; 95% CI, 0.82-13.0).

OTP Source of Methadone

Three decedents' source of methadone was from an OTP. Two decedents (#2 and 17) with OTP source died within 1 day of their administered dose of methadone. One decedent (#18) died within 2 days of his or her administered dose of methadone. Two of the 3 (#17 and 18) had been prescribed methadone for longer than 1 month. The decedent in the third case (#2) had rejoined the OTP after a short hiatus for a positive urine drug screen for marijuana. The decedent died on the second day of dose reinduction with 30 mg of methadone and the toxicology screen revealed only methadone. Decedent #17 was prescribed methadone 105 mg daily with 2 take-home doses per week and died with methadone and benzodiazepines in her toxicology screen. Decedent #18's dose was not known because the patient was enrolled in an OTP in West Virginia.

Five decedents with a history of OTP enrollment were not actively enrolled at the time of death. Four of the 5 inactive OTP decedents (#4, 9, 14, and 25) died less than 4 months after discharge from an OTP and the longest interval between discharge and a decedent's death was 8 months (#27). Two of the inactive OTP decedents (#4 and 9) were prescribed 6 take-home methadone doses before discharge from the OTP. One of the inactive OTP decedents (#25) had received a prescription for methadone from a non-OTP physician while enrolled in the OTP based on PMP information; however, this prescription was obtained 9 months before death.

Prescription for Methadone in PMP

Individuals prescribed methadone for analgesia were older (OR = 1.16; 95% CI, 1.06-1.26) and more likely to have antidepressants identified by postmortem toxicology (OR = 8.78; 95% CI, 2.3-33.2) than other decedents. None of the decedents with physician prescriptions in the PMP had 6 or more physicians prescribing opioids at one time. Two decedents (#39 and 67) obtained their medications at 4 or more pharmacies. Four decedents (24%) had cocaine in their toxicological analyses (OR = 3.08; 95% CI, 0.67-14.1). Among the 17 decedents with non-OTP physician prescriptions, 14 (82.4%) had a documented chronic pain condition in their medical examiner files.

When decedent prescription dosage was available either at the death scene or from PMP data (n = 9), the range of doses for patients was 20 to 160 mg/d. Four of these decedents were on dosages higher than 40 mg/d and 1 (#8) was simultaneously receiving a prescription for methadone from 2 physicians (1 for 20 mg/d and 1 for 60 mg/d). Five of the methadone prescriptions were initial prescriptions and 2 decedents had sporadically prescribed prescriptions as documented in the PMP.

The majority of deaths in the PMP population occurred around the time of their prescriptions refills. The average number of days from the time of a prescription to a decedent's death was 7 days. In 13 cases (77%), a prescription for methadone was obtained within 7 days of death, and in 8 cases (47%), a prescription for methadone was filled within 4 days of death. Many of these decedents had death scene evidence of large quantities of missing pills from their pill bottles. Medical examiner file information, together with PMP, data provided 1 scenario of the circumstances leading to a decedent's death (#28): the decedent was discharged from a physician's practice because of a positive urine drug screen and secured and filled a prescription for methadone from another physician and died 2 days later.


In the current study, 1 of 3 overdose deaths was methadone-related in western Virginia during 2004. Among methadone-related deaths, methadone was more commonly obtained from illicit sources, rather than from physician prescriptions for analgesia or from OTPs. Among those with an analgesic prescription for methadone noted in the PMP, the majority did not exhibit evidence of questionable prescription activity before their deaths. Polysubstance use was common, occurring in two-thirds of methadone-associated deaths.

The methadone deaths described in this article were more numerous than those in prior studies, which had larger base populations and longer follow-up periods (Barrett et al., 1996; Mikolaenko et al., 2002; Ballesteros et al., 2003; Gagajewski and Apple, 2003; Center for Substance Abuse Research, University of Maryland, 2004; Distefano, 2004; Wolf et al., 2004; Shields et al., 2007; Baker and Jenkins, 2008; Paulozzi et al., 2009). The large number of deaths occurring in this small rural region in just 1 year underscores the significance of the problem of methadone-related nonmedical drug use in rural Virginia. This study is also unique in that it provides specific data from 3 OTPs with 1500 patients.

The majority of decedents had no record of methadone prescription or participation in an OTP. This finding is consistent with other studies that reported high rates of illicitly obtained methadone in methadone-related overdose deaths, ranging from 39% to 67% (Barrett et al., 1996; Gagajewski and Apple, 2003; Paulozzi et al., 2009). All of the aforementioned studies collaborated with medical examiner offices and OTPs to help determine circumstances of overdose deaths and participation in methadone maintenance therapy. The more recent study utilized a PMP to determine source of methadone (Paulozzi et al., 2009). The current study extends the results of previous studies by further exploring the source of diverted methadone and characteristics of decedents.

Evidence from this study suggests that the source of diverted medication is varied. Police reports from decedents' medical examiner files corroborated that several patients illicitly bought methadone or obtained methadone from family members. In the 2008 National Survey on Drug Use and Health, 70% of people aged 12 years and older who used pain relievers for nonmedicinal purposes obtained those pain relievers from friends or relatives and only 4% bought them from a drug dealer (SAMHSA, 2009).

For nearly two-thirds of decedents, the cause of death was polysubstance toxicity from an average of 3 substances. The high prevalence of polysubstance-related overdose deaths in this study is similar to that in previous studies (Barrett et al., 1996; Mikolaenko et al., 2002; Oregon Department of Human Services, 2003; Wolf et al., 2004; Hall et al., 2008; Paulozzi et al., 2009; Wunsch et al., 2009) and underscores the need to educate providers, patients, and illicit drug users regarding the increased risk of overdose from enhanced respiratory depression and fatal arrhythmias when combining opioids, sedative hypnotics, and alcohol (Leavitt and Krantz, 2003).

Like other studies, one of the most commonly observed drug combinations was benzodiazepines and methadone (Oregon Department of Human Services, 2003; Wolf et al., 2004; Wunsch et al., 2009). Benzodiazepines, particularly alprazolam, may “boost” the euphoric effects of methadone when taken concomitantly (Stitzer et al., 1981). Methadone and other opioids were also frequently observed in the illicit category. Although the study was underpowered to detect significant differences for this variable, the high proportion of decedents with polyopioid overdose suggests that decedents had access to several illicit opioids. Decedents likely did not appreciate the important pharmacokinetic differences between methadone and other opioids and may have expected it to produce euphoric effects similar to other opioids. We also may hypothesize that the lower cost of both licit and illicit methadone, relative to other opioids (eg, oxycodone or hydromorphone), makes methadone more affordable and easily accessible. Unfortunately, we do not know when or for how long decedents may have been illicitly using methadone, but in many cases the medical examiner information would suggest that decedents were methadone-naive, thus making its delayed toxicity even greater.

Another drug combination observed was cocaine and methadone in 4 of the PMP decedents. These decedents did not have known recent histories of substance abuse treatment that could have placed them at increased risk for overdose death due to lack of tolerance. If cocaine abuse is known at the time of methadone prescription, close physician monitoring is necessary in assessing potential toxic cardiac effects and other forms of substance use. In addition, cocaine is reported to accelerate methadone elimination. Decedents may have therefore taken more methadone than prescribed to achieve adequate pain relief or euphoria (Moolhan et al., 2001).

Our study highlights the importance of previously reported concurrent antidepressants and methadone in overdose deaths (Wunsch et al., 2009). This finding may be particularly salient for older patients who are also at increased risk for methadone overdose deaths from prescribed, instead of illicitly procured, methadone. It has been hypothesized that certain SSRIs and SNRIs may increase the pharmacological effect of methadone by inhibiting its biotransformation in the body (Leavitt, 2005). This drug combination is particularly concerning due to lack of awareness of its potential to increase methadone overdose risk.

Together, the characteristics of the methadone-overdose deaths described in this study are similar to the risk factors reported in heroin overdose (eg, male sex, young age [30s], use of heroin for 10 years, use of heroin alone, lack of enrollment in drug treatment, recent reduction in heroin use, reduced tolerance due to recent release from prison or drug detoxification treatment, injection of heroin, and the use of benzodiazepines, alcohol, or antidepressants (especially tricyclic antidepressants) (Darke and Hall, 2003). Like fatal heroin overdoses, many decedents in this study had a history of substance abuse, most decedents were male, most were not enrolled in drug treatment, and there was a high frequency of polysubstance use.

Finally, this study expands on previous studies that used a PMP to determine source of controlled substances (Hall et al., 2008; Paulozzi et al., 2009). Currently, PMPs exist in 40 states with variable objectives by state; however, the consensus goal of most PMPs is to minimize prescription drug abuse and diversion while supporting legitimate medical practice (Katz et al., 2008; Oregon Department of Human Services, 2009). The few studies to investigate PMP use reported helping providers make informed decisions about opioid prescribing (Katz et al., 2008; Sinha et al., 2009). A key limitation to PMPs is practitioners' infrequent use. Also, patients will cross state lines to avoid monitoring because many PMPs do not report data to other states. Finally, PMPs do not have access to controlled substances obtained from the Internet, OTPs, Department of Veteran's Affairs, or Indian Health Service facilities (US Government Accountability Office, 2009). Interestingly, in this study, there was little PMP evidence of questionable prescription activity in the decedents prescribed methadone, thus raising concern that the PMP may not be an effective method to forewarn prescribers of their patients' potential risk for fatal overdose. Or, it may indicate that the previously described definition of questionable prescription activity is too lenient. Further research is needed to better establish the clinical usefulness of PMP data to aid prescribing practices and predict risk for overdose.


This study should be interpreted in the context of several potential limitations. The 61 cases are reviewed from a specified region of the country; they may not be representative of deaths elsewhere. Although rich in detail, the overall small sample size provides limited power to detect statistical differences in trends observed in Tables 1 and and2.2. Third, because this study depended on toxicology results to identify decedents, it may not comprise a total census of all methadone occurrences in 2004 and may, in fact, underestimate the problem. Finally, the study, in part, depended on a state-sponsored pilot PMP to determine the source of methadone. Therefore, some of the data may be incomplete if all distributing agencies were not involved or if decedents provided false information when obtaining prescriptions. Decedents who were prescribed opioids outside the jurisdiction of the PMP, from OTPs in bordering states, or from the Department of Veteran's Affairs may not have been captured.


This study attempts to determine the source of methadone in overdose deaths and reports rich details about 61 methadone-related deaths. The findings underscore the potential lethality of methadone when used nonmedically, especially when mixed with other psychoactive substances. The study illustrates the difficulty in monitoring the distribution, legal or illegal, of scheduled medications. Finally, the findings disprove our hypothesis that most methadone-related deaths are from physician prescription for analgesia, thus increasing concern about the nonmedical use of methadone. Given the predominance of illicit use of methadone causing overdose deaths, public health interventions may be best targeted to try to reduce diversion or prevent overdose in those who use illicit methadone.

The current study highlights potential risks in those who are prescribed methadone for analgesia: older patients and those who are prescribed, or may illicitly obtain, antidepressants. Physicians should exercise special care when prescribing methadone to these populations. Moreover, given the known risk factor of social isolation and overdose, another aim of overdose prevention in the older population may be the promotion of various outreach programs (eg, communal living or a buddy system). Such measures may reduce the risk of potentially fatal overdose events, particularly when patients for whom they are prescribed misuse their medications.

Finally, physicians may not be able to have an impact upon the problem of illicit methadone abuse, but it is within their grasp to carefully prescribe, monitor, and educate their patients who are prescribed methadone. Patients should be advised to protect prescribed medications by locking them in a safe place and never sharing them. States can aid physicians in this work by providing physicians with regular reports from PMPs that may show signs of diversion or misuse. Further research is needed to measure the effectiveness of PMPs to predict risk for overdose deaths. In addition, continued research combining OCME, PMP, physician, and OTP information may better elucidate risk factors associated with methadone overdose deaths, thus ultimately developing targeted interventions for harm reduction and prevention.


The authors thank the Commonwealth of Virginia Office of the Chief Medical Examiner, Western District, The Virginia Prescription Monitoring Program, Pantops Clinic OTP, The New River Valley Treatment Center OTP, and the Clinch Valley Treatment Center OTP for their willingness to participate in this study.

Supported by The National Institutes of Health, National Institute on Drug Abuse (K23 DA019809 [PTK] and RO3 DA019047 [MJW]).



Methadone-overdose decedent characteristics, western Virginia 2004
OverallSource of Methadone

Characteristicn = 61Illicit (n = 41)OTP (n = 3)Prescription (n = 17)P-valuea
Mean age, years (SE)36 (1.3)33 (1.6)25 (0.9)44 (1.6)<0.001
 Female22 (36%)14 (34%)1 (33%)7 (41%)0.875
 Caucasian58 (95%)38 (93%)3 (100%)17 (100%)0.463
 Rural27 (44%)19 (46%)2 (67%)6 (35%)0.539
History of substance abuse33 (54%)23 (56%)3 (100%)7 (41%)0.153
Cause of Death
 Polysubstance overdose*37 (61%)24 (58%)1 (33%)12 (71%)0.424
Concomitant Drugs on Toxicology
 Any47 (77%)31 (76%)2 (67%)14 (82%)0.778
 Antidepressantsb14 (23%)5 (12%)0 (0%)9 (53%)0.002
 Benzodiazepinesc22 (36%)18 (44%)1 (33%)3 (18%)0.165
 Other prescribed opiatesd18 (30%)15 (37%)0 (0%)3 (18%)0.183
 Ethanol11 (18%)7 (17%)1 (33%)3 (18%)0.778
 Cocainee8 (13%)4 (10%)0 (0%)4 (24%)0.290
aWald x2 test obtained from simple linear regression.
*Drug classes detected include antidepressants, benzodiazepines, opioids, alcohol, and cocaine.
bAntidepressants include mirtazapine, sertraline, clomipramine, citalopram, amitriptyline, venlafaxine
cBenzodiazepine parent drug and metabolites include Diazepam, Nordiazepam, Clonazepam, Alprazolam, and Desoxychlordiazepam.
dOther opiates include hydrocodone, oxycodone, morphine, and hydromorphone.
eCocaine or cocaine metabolites.


  • Baker DD, Jenkins AJ. A comparison of methadone, oxycodone, and hydrocodone related deaths in Northeast Ohio. J Anal Toxicol. 2008;32:165–170. [PubMed]
  • Ballesteros M, Budnitz D, Sanford C, Gilchrist J, Agyekum G, Butts J. Increase in deaths due to methadone in North Carolina. JAMA. 2003;290:40. [PubMed]
  • Barrett D, Luk A, Parrish RG, Jones TS. An investigation of medical examiner cases in which methadone was detected, Harris County, Texas, 1987-1992. J Forensic Sci. 1996;41(3):442–448. [PubMed]
  • Cairns A, Roberts I, Benbow E. Characteristics of fatal overdose in Manchester, 1985-1994. Br Med J. 1996;313:264–165. [PMC free article] [PubMed]
  • Caplehorn JR, Dalton SY, Haldar F, Petrenas AM, Nisbet JG. Methadone maintenance and addicts' risk of fatal heroin overdose. Subst Use Misuse. 1996;31:177–196. [PubMed]
  • Darke S, Hall W. Heroin overdose: research and evidence-based intervention. J Urban Health. 2003;80(2):189–200. [PMC free article] [PubMed]
  • Dickerson ED. Methadone: the question or the answer for US opioid therapy and pharmaco-economics? Support Care Cancer. 2001;9:646–647. [PubMed]
  • Eap CB, Buclin T, Baumann P. Interindividual variability of the clinical pharmacokinetics of methadone: implications for the treatment of opioid dependence. Clin Pharmacokinet. 2002;41:1153–1193. [PubMed]
  • Federal Register Standards for Defining Metropolitan and Micropolitan Statistical Areas That the Office of Management and Budget. Dec 272000. [November 8, 2009]. Federal Register web site. Available at:
  • Fingerhut LA. Increases in poisoning and methadone-related deaths: United States, 1999-2005. 2008. [October 29, 2009]. CDC web site. Available at:
  • Gagajewski A, Apple FS. Methadone-related deaths in Hennepin County, Minnesota: 1992-2002. J Forensic Sci. 2003;48(3):668–671. [PubMed]
  • Gronbladh L, Ohlund L, Gunne L. Mortality in heroin addiction: impact of methadone treatment. Acta Psychiatr Scand. 1990;82:223–227. [PubMed]
  • Hall AJ, Logan JE, Toblin RL, et al. Patterns of abuse among unintentional pharmaceutical overdose fatalities. JAMA. 2008;300(22):613. [PubMed]
  • Inturrisi C, Colburn W, Kaiko R, Houde R, Foley K. Pharmacokinetics and pharmacodynamics of methadone in patients with chronic pain. Clin Pharmacol Ther. 1987;41(4):392–400. [PubMed]
  • Intrussi CE, Verebely K. Disposition of methadone in man after a single oral dose. Clin Pharmacol Ther. 1972;13:923–930. [PubMed]
  • Joseph H, Stancliff S, Langrod J. Methadone maintenance treatment (MMT): a review of historical and clinical issues. Mt Sinai J Med. 2000;67(5,6):347–364. [PubMed]
  • Katz N, Houle B, Fernandez K, et al. Update on prescription monitoring in clinical practice: a survey study of prescription monitoring program administrators. Pain Med. 2008;9(5):587–594. [PubMed]
  • Katz N, Panas L, Kim M, et al. Usefulness of prescription monitoring programs for surveillance---analysis of Schedule II opioid prescription data in Massachusetts, 1996-2006. Pharmacoepidemiol Drug Saf. 2010;19(2):115–123. [PubMed]
  • Leavitt S. Addiction Treatment Forum. 3rd. Mundelein, IL: Clinco Communications; 2005. Methadone-drug interactions.
  • Leavitt S, Krantz M. Addiction Treatment Forum. Mundelein, IL: Clinco Communications; 2003. Cardiac considerations during MMT.
  • Lehder D, Arria A, Artigani E, Wish E. Alcohol and drug-related overdose deaths in Maryland: 1997-2001. 2002. [November 8, 2009]. Center for Substance Abuse Research web site. Available at:
  • McLellan AT, Turner B. Prescription opioids, overdose deaths, and physician responsibility. JAMA. 2008;300(22):2672–2673. [PubMed]
  • Mikolaenko I, Robinson A, Davis G. A review of methadone deaths in Jefferson County, Alabama. Am J Forensic Med Pathol. 2002;23(3):299–304. [PubMed]
  • Moolhan E, Umbricht A, Epstein D. Therapeutic drug monitoring in methadone maintenance. J Addict Dis. 2001;20(2):55–73. [PubMed]
  • National Institutes of Health . Eur Addict Res. Vol. 5. 1999. NIH consensus panel recommends expanding access to and improving methadone treatment programs for heroin addiction; pp. 50–51. [PubMed]
  • Oregon Department of Human Services Methadone deaths (and distribution) on the rise. Jul 152003. [November 8, 2009]. Oregon web site. Available at:
  • Oregon Department of Human Services Opioid-related poisoning deaths in Oregon. Sep 292009. [October 29, 2009]. Oregon web site. Available at:
  • Paulozzi L, Logan J, Hall A, McKinstry E, Kaplan J, Crosby A. A comparison of drug overdose deaths involving methadone and other opioid analgesics in West Virginia. Addiction. 2009;104:1541–1548. [PubMed]
  • Roxane Laboratories . Dolophine® Hydrochloride tablets (Methadone Hydrochloride Tablets USP) Columbus, OH: Roxane Laboratories Inc; 2000. package insert.
  • Roxane Laboratories . Dolophine® Hydrochloride tablets (Methadone Hydrochloride Tablets USP) Columbus, OH: Roxane Laboratories; Nov 17, 2006. package insert.
  • Shah N, Lathrop S, Landen M. Unintentional methadone-related overdose death in New Mexico (USA) and implications for surveillance, 1998-2002. Addiction. 2005;100:176–188. [PubMed]
  • Shields LBE, Hunsaker JC, Corey TS, Ward MK, Stewart D. Methadone toxicity fatalities: a review of medical examiner cases in a large metropolitan area. J Forensic Sci. 2007;52(6):1389–1395. [PubMed]
  • Sinha S, Callan E, Akpunonu P, Baehren D, Marco C. How does use of a statewide prescription monitoring program affect emergency department prescribing behaviors? Ann Emerg Med. 2009;54(3):S14–S15. abstract.
  • Sorg M, Greenwald M. Maine drug-related mortality patterns: 1997-2002. 2002. [November 8, 2009]. Maine web site. Available at:
  • Stitzer M, Griffiths R, McLellan T, Grabowski J, Hawthorne J. Diazepam use among methadone maintenance patients: patterns and dosages. Drug Alcohol Depend. 1981;8:189–199. [PubMed]
  • Substance Abuse and Mental Health Services Administration, Center for Substance Abuse Treatment Methadone Associated Mortality Report of a National Assessment. 2004. [May 2007]. SAMHSA web site. Available at:
  • Substance Abuse and Mental Health Services Administration National Survey on Drug Use and Health. 2008. [November 8, 2009]. SAMHSA web site.
  • Substance Abuse and Mental Health Services Administration Division of Pharmacologic Therapies. Follow directions how to use methadone safely. Apr2009. [November 8, 2009]. SAMHSA web site. Available at:
  • US Department of Justice, Drug Enforcement Administration Automation of reports and consolidation order's system (ARCOS): Retail Drug Summary Reports, 1998-2002. 2004. [November 8, 2009]. USDOJ web site. Available at:
  • US Government Accountability Office Report to Congressional Requesters Methadone-Associated Overdose Deaths. Mar2009. [October 30, 2009]. SAMHSA web site. Available at:
  • Wheeler Clinical use of methadone. Am J Hosp Palliat Care. 2000;17(3):196–203. [PubMed]
  • Wolf BC, Lavezzi WA, Sullivan L, Flannagan L. Methadone-related deaths in Palm Beach County. J Forensic Sci. 2004;49(2):375–378. [PubMed]
  • Wolff K. Characterization of methadone overdose: clinical considerations and the scientific evidence. Ther Drug Monit. 2002;24:457–470. [PubMed]
  • Wunsch MA, Nakamoto K, Behonick G, Massello W. Opioid deaths in rural Virginia: a description of a high prevalence of accidental fatalities involving prescribed medications. Am J Addict. 2009;18(1):5–14. [PMC free article] [PubMed]