International Journal of Drug Policy - 2014

Volume 25 Issue 3 May 2014

Issue link:

Contents of this Issue


Page 72 of 153

A. Siegler et al. / International Journal of Drug Policy 25 (2014) 569–574 573 were no significant demographic differences among the four cate- gories of non-home overdoses, and these were collapsed into one "not home" category for statistical analysis (Table 1). Approximately three-quarters of the sample (74% or 1954) were males, having an age-adjusted mortality rate nearly three times that of females. Slightly less than half of the sample (n = 1224) were non-Hispanic whites, less than one-third (n = 783) were Hispanic, and approximately one-fifth (n = 573) were non-Hispanic black. Whites had the highest age-adjusted mortality rate. Decedents' ages ranged from 15 to 82, with a median age of 45 years. Indi- viduals with the highest overdose mortality rate were those aged 45–54 years. Two-thirds of the sample (67% or 1782) had a high school diploma or more education. While the largest proportion of decedents lived in the borough of Brooklyn (30% or 784), the Bronx and Staten Island had the highest age-adjusted mortality rates. Over one-third of the decedents (39% or 1022) came from neigh- borhoods with very high poverty levels. As neighborhood poverty increased, rates of overdose mortality increased as well, with very high poverty neighborhoods having more than double the rates of low poverty neighborhoods. When stratifying by setting of overdose, three-quarters of the analytic sample (2000) overdosed at home. Of these decedents, approximately 89% (1770) overdosed in their own home and the remainder overdosed in others' homes. There were 632 decedents who overdosed outside of the home. Of these decedents, 42% (n = 268) overdosed in an institutional residence, followed by 29% in a public indoor setting, 26% in an outdoor setting, and 3% in other settings (n = 181, n = 162, and n = 21, respectively). Among each demographic group, the majority of overdoses occurred in the home. In bivariate analysis, female gender was asso- ciated with overdosing in the home, compared to male gender (OR 1.79). Non-Hispanic black and Hispanic races were associated with overdosing outside of the home (OR 0.59 and 0.74, respectively). Decedents aged 25–84 were more likely to overdose outside of the home compared to individuals under 25 (ORs from 0.40 to 0.52). College-educated individuals had twice the risk of overdosing at home compared to individuals without a high school diploma (OR 2.10). Compared to residents of Brooklyn, residents of Manhat- tan were more likely to overdose outside of the home (OR 0.68), while residents of Staten Island were more likely to overdose inside the home (OR 2.05). Compared to individuals who live in neigh- borhoods with very high poverty rates, individuals in wealthier neighborhoods are more likely to overdose at home (ORs from 1.46 to 1.86). No significant interactions were observed when age and education were tested with other demographic variables. Heroin was the most common opioid in overdoses among all settings and involved in 63% of all overdoses. Heroin also had the highest age-adjusted mortality rate among all single drug types and drug combinations. Cocaine and alcohol were the second and third most common substances in all settings except institutions, involved in 50% and 43% of all overdoses overall, respectively. In institutional residences, methadone was the second most com- mon drug after heroin, involved in 48% of institution overdose deaths, and cocaine was the third most common, involved in 47%. Benzodiazepines were most commonly found in combina- tion with opioid analgesics (benzodiazepines were in 53% of opioid analgesic-related deaths), and also were found in combination with methadone (44%) and heroin (35%) (data not presented). The most common combination of drugs in overdoses occurring at home was opioid analgesics and benzodiazepines, whereas in all other non-home settings, heroin and methadone was the most common drug combination. In bivariate analysis, heroin was significantly associated with overdosing outside of the home (OR 0.61). Opioid analgesics, benzodiazepines, and combined opioid analgesics with benzodiazepines were significantly associated with overdosing inside the home (ORs 1.91, 1.35, 2.58, respectively). In a multivariate analysis, after adjusting for significant demo- graphics and drug types, gender, age, education, borough of residence, as well as the drug types heroin and combined use of opi- oid analgesics and benzodiazepines remained significant predictors of overdose setting. Females, college graduates, and residents of Staten Island had significantly greater odds of overdosing at home compared to males (AOR 1.68), individuals who did not complete high school (AOR 2.13), and residents of Brooklyn (AOR 1.71). Resi- dents of Manhattan had lower odds of overdose at home compared to residents of Brooklyn (AOR 0.69), as did individuals ages 35–64 compared to those ages 15–24 (AORs 0.40–0.48). Decedents who overdosed on heroin were less likely to overdose at home, com- pared to decedents who had not used heroin (AOR 0.76). Concurrent use of opioid analgesics and benzodiazepines nearly doubled the odds of overdosing at home compared to decedents with neither or only one of these medications (AOR 1.92). Of note, neither opioid analgesics nor benzodiazepines alone were significantly associated with overdose setting, after controlling for demographic character- istics. Discussion We found that three-quarters of unintentional drug poisoning deaths occurred inside the home, which is consistent with the highest estimates in the NYC literature (Cerdá et al., 2013) and higher than the proportions found in other locations such as San Francisco (Davidson et al., 2003) and New South Wales, Australia (Darke, Ross, Zador & Sunjic, 2000). Our study further identified that female gender, college education, Staten Island residence, and combined opioid analgesic/benzodiazepine were significantly associated with overdosing at home. In contrast, ages 35–64 (com- pared to ages 15–24), residence in Manhattan, and heroin use were associated with overdosing outside of the home. Despite these find- ings, fatal overdoses were more likely to occur in the home than elsewhere for every demographic group and drug type. Because most opioid overdoses occur inside the home, opioid overdose response programs can most efficiently address the epi- demic by both reducing the risk of overdose in the home and targeting those who may be in the home at the time of an over- dose for overdose response training. To reduce risks in the home, programs that minimize opportunities for misuse and diversion of opioids such as safe storage and safe disposal of opioid analgesics should be encouraged. Physician awareness can be raised regarding the risks of overdose and physicians should be encouraged to uti- lize PMP to review patient histories and reduce risky prescribing practices of opioid analgesics and benzodiazepines. Patient aware- ness can be raised regarding the risks of polydrug use, particularly opioids with benzodiazepines. To prepare potential witnesses of opioid overdose, legal take- home naloxone, prescribed by a medical professional for use by a trained layperson, is a key element of effective response. Layperson administration of naloxone was legalized in New York State in 2006 and utilization of naloxone by community members to reverse opi- oid overdose occurred throughout the study period in NYC. One way of increasing the availability of naloxone is through co-prescription of naloxone with opioid analgesics, for safe-keeping in patients' medicine cabinets. Calling emergency services when a person over- doses is another critical element of effective response. Rates of calling may be improved by protecting callers from arrest for drug possession. A Good Samaritan law offers this protection and in New York was passed after the study period ended, in 2011. Good Samar- itan laws would make little difference, if, for example, the majority of overdoses occurred in institutions, where staff were trained to call emergency services and would not fear arrest, underscoring the importance of understanding overdose setting and the ways in which setting has implications for policy.

Articles in this issue

view archives of International Journal of Drug Policy - 2014 - Volume 25 Issue 3 May 2014