Simultaneous Co-use of Opiate: Benzodiazepine and Alcohol Among Rural and Urban Suicide Decedents

Acknowledgements: Contributors to this report included participating Violent Death Reporting System states; participating state agencies, including state health departments, vital registrars' offices, coroners' and medical examiners' offices, crime laboratories, and local and state law enforcement agencies; partner organizations, including the Safe States Alliance, National Violence Prevention Network, National Association of Medical Examiners, National Association for Public Health Statistics and Information Systems (NAPHSIS), Council of State and Territorial Epidemiologists (CSTE), and Association of State and Territorial Health Officials; federal agencies, including the Department of Justice (Bureau of Justice Statistics and the Federal Bureau of Investigation), the Department of the Treasury (Bureau of Alcohol, Tobacco, and Firearms); the International Association of Chiefs of Police; other stakeholders, researchers, and foundations, including The Joyce Foundation, the National Institute for Occupational Safety and Health, and the National Center for Health Statistics, CDC.

Disclaimer: This research uses data from NVDRS, a surveillance system designed by the Centers for Disease Control and Prevention's (CDC) National Center for Injury Prevention and Control. The findings are based, in part, on the contributions of the 42 funded states and territories that collected violent death data and the contributions of the states' partners, including personnel from law enforcement, vital records, medical examiners/coroners, and crime laboratories. The analyses, results, and conclusions presented here represent those of the authors and not necessarily reflect those of CDC. Persons interested in obtaining data files from NVDRS should contact CDC's National Center for Injury Prevention and Control, 4770 Buford Hwy, NE, MS F-64, Atlanta, GA 30341-3717, (800) CDC-INFO (232-4636). The content is solely the responsibility of the authors and does not necessarily represent the official view of the National Institute on Alcohol Abuse and Alcoholism, the National Institutes of Health, CDC, or the American Public Health Association.

Funding: This study was supported by a New Investigator Award to Dr. Bensley from the American Public Health Association. Authors were also supported by two grants from the National Institute of Alcohol Abuse and Alcoholism: T32AA007240, Graduate Research Training in Alcohol Problems: Alcohol-related Disparities, and P50AA005595, Epidemiology of Alcohol Problems: Alcohol-Related Disparities, both from the National Institute on Alcohol Abuse and Alcoholism.

Abstract (limit 250 words): 242

Manuscript (limit 4000 words): 3487

Highlights:

• Rural (vs. urban) decedents had greater risk of opiates and benzodiazepines
• No differences in alcohol and opiate co-use across rurality
• No differences in benzodiazepine, alcohol, and opiate co-use across rurality
• Simultaneous co-use among suicide decedents also varied by region

ABSTRACT

Background: Fatal suicides involving opiates are increasingly common, particularly in rural areas. While co-use in addition to opiates contributes significantly to mortality risk, it is unknown whether the prevalence of co-use relative to opiate-only deaths varies across rurality.

Methods: Data from 27 participating states with at least one opiate-related overdose suicide in the National Violent Death Reporting System (NVDRS) from 2012-2015 were used. Relative risk ratios were obtained using multinomial logistic regression, comparing opiate-only to 1) opiate & alcohol, 2) opiate and benzodiazepines, and 3) opiate, alcohol, and benzodiazepines suicides across rurality. Models were fit using robust standard errors and fixed effects for year of death adjusting for individual, county, and state-level covariates.

Results: There were 3,781 opiate-overdose suicide decedents tested for all three substances during the study period. Prevalence of co-use in decedents varied across rurality (p=0.022), with urban decedents more likely to test positive for opiates alone while rural decedents more likely test positive for opiates and benzodiazepines. Adjusting for individual factors, rural suicide decedents were 33% more likely to test positive for opiates with benzodiazepines than opiates alone relative to urban decedents (Adjusted RRR =1.33, 95% Confidence Interval (CI) = 1.08, 1.63). This association was no longer significant after adjusting for region.

Conclusions: Rural suicides are associated with increased opiate and benzodiazepine co-use but there may be regional differences, supporting the need for rural-focused interventions to support appropriate co-prescribing and better health education about risks associated with drug mixing.

KEYWORDS: suicide, rural, urban, alcohol, opiates, benzodiazepines, co-use

1. INTRODUCTION

Opiate overdose deaths tripled in the last twenty years, with over 30,000 opiate-related overdose deaths in 2015 (Rudd et al., 2016). Co-use of alcohol, benzodiazepines, and opiates simultaneously is common (Jones et al., 2014; Jones and McAninch, 2015; Witkiewitz and Vowles, 2018) and increases risk of fatal overdoses (Darke and Ross, 2002; Gressler. et al., 2017; Lembke et al., 2018; Witkiewitz and Vowles, 2018). Alcohol and benzodiazepines – both central nervous system (CNS) depressants, like opiates, with a strong abuse potential – increase opioid respiratory insufficiency (Baca and Grant, 2005), leading to increased mortality risk when used along with opiates. In addition to co-prescribed use for co-occurring conditions (e.g. anxiety and chronic pain), benzodiazepines may be used non-medically to reinforce effects of opiates (Jones et al., 2012).

Importantly, opiate use is associated with increased suicide risk. Heroin users have 14 times higher risk of fatal suicide relative to non-heroin users (Darke and Ross, 2002), and higher prescription opioid doses are associated with greater suicide risk (Ilgen et al., 2016). Among people who inject drugs, heavy alcohol use is a strong predictor of suicide risk (Fournier et al., 2017), and among those reporting opiate use, benzodiazepines co-use is a strong predictor of suicide attempts (Backmund et al., 2006). In 2014, 27.9% of all suicide decedents tested positive for acute alcohol intoxication, 32.6% for benzodiazepines, and 30.0% for opiates (Fowler et al., 2018), but the percentage of suicides involving simultaneous co-use, relative to use of opiates alone, is unknown.

In rural areas, suicide risk is higher (Searles et al., 2014) and opioid overdose rates have recently increased more steeply than in urban areas (Mack et al., 2017), with an increase of 84% between 1999-2015, relative to a 61% increase in urban areas (Mack et al., 2017). Yet rural areas often lack access to sufficient overdose prevention (i.e. naloxone) and mental health services, which may contribute to the increased mortality risk among overdoses in rural areas (Faul et al., 2015). Rural areas may also face increased risk from co-use as the majority of rural individuals reporting opiate use also report lifetime benzodiazepine use (Havens et al., 2010). There may also be regional differences, as alcohol use patterns and opiate overdose deaths are not uniformly distributed across regions: alcohol use is more common in the Midwest (Borders and Booth, 2007) and the type of opiate involved in overdoses (synthetic opioids versus heroin) varies across region (Rudd et al., 2016).

Compared to extensive literature on opiate overdoses or any-drug overdoses, little is known about the factors associated with multiple-drug co-use overdose deaths (Witkiewitz and Vowles, 2018), particularly among suicide decedents. While co-use is common and known to be associated with mortality, it is unknown which factors predict co-use (vs. opiates alone) among suicides. Yet better understanding factors associated with simultaneous co-use among fatal suicide may provide important recommendations for prevention.

The challenges of studying simultaneous co-use related overdose deaths may be one reason for the sparse literature on co-use related mortality. Specific-drug death is difficult to examine across states or jurisdictions given differences in autopsies and toxicology testing across states with different death investigation systems (Hanzlick, 2006; Kaplan et al., 2013; Larsen et al., 2008), which results in differences in how drug-related cause of death is classified on the death certificate. Most death investigations do not include toxicology reports due to differing jurisdictional requirements or available resources for death investigations (Kaplan et al., 2013; Larsen et al., 2008). Addressing these difficulties, the National Violent Death Reporting System (NVDRS) includes toxicology reports for suicide decedents, making this dataset ideal for studying specific combinations of alcohol and drug use among overdose suicide decedents. In addition, NVDRS includes a rich data set capturing many factors about violent death decedents and the circumstances surrounding their deaths collected from multiple sources, making it a valuable resources for studying correlates of simultaneous co-use related suicides.

The aim of this study was to describe the prevalence of simultaneous alcohol and benzodiazepine co-use among opiate-involved overdose suicide decedents, and compare the prevalence of simultaneous co-use related to opiate-only deaths across rurality among decedents, adjusting for potential confounders. We hypothesized that co-use of multiple CNS depressants is more strongly associated with overdose-suicide risk in rural (vs.urban) areas. Additionally, this study explores other key individual and community-level factors, including region, and whether these are additionally associated with co-use suicide deaths independent of rurality.

2. METHODS

2.1 Data source: Data from the National Violent Death Reporting System (NVDRS) were used in this study. NVDRS is the only state-based surveillance system that pools data on violent deaths from multiple sources, including death certificates, coroner/medical examiner reports, and law enforcement reports. Some participating states also collect information from secondary sources (e.g., child fatality review team data, supplemental homicide reports, and crime laboratory data). NVDRS collects and links data from these different sources, which are abstracted and entered in NVDRS. The ability to analyze linked data permits comprehensive assessment of violent deaths. Description of NVDRS has been published elsewhere (Fowler et al., 2018).

Given differences in classifying suicides and undetermined deaths across jurisdictions (Timmermans, 2005), particularly among drug-related deaths (Rockett et al., 2018), only deaths determined to be suicides were included in this study. All included decedents had opiate-related overdose listed as a cause of death, defined using both toxicology reports and cause of death determination from the medical examiner or coroner. Given differences in death reporting systems (Hanzlick, 2006; Robinson, 2017) and toxicology reporting (Kaplan et al., 2013; Larsen et al., 2008), only decedents were included that had been tested for opiate, alcohol, and benzodiazepine use and who had tested positive for opiate use.

NVDRS data collection began in 2003 with seven states; six states joined in 2004, four in 2005, and two in 2010. In 2015, NVDRS expanded to include 14 additional states for a total of 32 states in the system. In this study, data were included from the following 27 states having at least one death meeting inclusion criteria during the study period: Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, and Virginia. While NVDRS has been collecting information since 2003, the time period for this study was 2012-2015 to maximize the amount of data reported, as few deaths with all toxicology reports for all three substances were reported prior to 2012 (n=109). Data were also limited to those reports with a valid residential county.

2.2 Opiate co-use at time of death: Drug use at death was defined as: 1) opiate-only, 2) opiate and alcohol co-use, 3) opiate and benzodiazepine co-use, and 4) opiate, benzodiazepine, and alcohol co-use. Like opiate use, alcohol use and benzodiazepine use were determined by toxicology reports for decedents, defined as a positive test at time of death. Cause of death data derived from the death certificate was not used as a primary way to establish co-use at time of death because testing and reporting of substances on the death certificate can vary depending on subjective criteria such as the state death reporting system procedures (Warner et al., 2013), and individual lab processes (Slavova et al., 2015). Rurality was defined using decedent’s county of residence at time of death. 2013 Rural-Urban Continuum Codes (RUCC) were used, based on the Office of Management and Budget metropolitan and non-metropolitan categories (Parker, n.d.). Rurality was thus operationalized as urban (metropolitan) and rural (non-metropolitan).

2.3 Covariates: A number of covariates were also included in this study. Individual-level covariates were derived from NVDRS data, while county-level covariates were derived from the 2010 Census data, linked by decedent county of residence. Individual-level demographic factors such as gender, race/ethnicity, homelessness, and education were included as these are associated with substance use patterns both among living populations (Erol and Karpyak, 2015; Hasin and Grant, 2015) and among suicide decedents, patterns which vary across rurality (Caetano et al., 2013; Kaplan et al., 2013). Substance use problems (alcohol and other substance problems) were included to control for severity as these vary across rurality (Borders and Booth, 2007), and mental health variables (mental health problems and history of mental health treatment) were included as they are strongly associated with suicide risk and substance use (Logan, 2011). County-level socioeconomic status, operationalized categorically by quartiles of household poverty from 2010 Census data, was included as a covariate as this is associated with substance use patterns (Karriker-Jaffe, 2013), suicide risk (Rehkopf and Buka, 2006), and rurality (Blumenthal and Kagen, 2002). While there were not enough deaths in each participating state to include state as a fixed effect, three state-level variables were included to account for state-level differences that may confound the association between rurality and co-use. Region was included as a covariate as this is associated with reporting of substance use (Borders and Booth, 2007). State death reporting systems were included as this determines resources available for toxicology testing and reporting (Warner et al., 2013). The presence of any state-level naloxone laws was included to account for different state-level approaches to overdose prevention, as these laws indicate accessibility of naloxone to people who use opiates (Davis and Carr, 2017).

2.4 Analytic Strategy: The proportion of overdose suicide victims who tested positive for 1) opiate use only, 2) opiate and alcohol co-use, 3) opiate and benzodiazepine co-use, and 4) for all three substances were described across rurality and compared using chi-square tests of independence, overall and with pairwise comparisons using a Bonferroni correction to account for multiple comparisons (Hall and Richardson, 2016). All descriptive characteristics were compared overall with chi-square tests. Relative risk ratios comparing co-use (vs. opiate use alone) between urban and rural overdose suicide decedents were assessed using a multinomial logistic regression model, comparing opiate-only to 1) opiate and alcohol, 2) opiate and benzodiazepines, and 3) opiate, alcohol, and benzodiazepines. Models were fit using robust standard errors and fixed effects for year of death. Multiple models were ran, including 1) an unadjusted model, and 2) a model adjusted for individual-level covariates, and 3) models adjusted additionally for each county and state level covariate. Using the fully adjusted model, the significance of all variables was also assessed. All analyses were done with Stata v15 (StataCorp, 2013).

2.5: Inclusion in Analytic Sample: Of 170,758 suicide and undetermined deaths reported to NVDRS between 2003-2015, virtually all (>99.9%) of decedents had a reported state and county of death corresponding to a 2013 RUCC code (n=169,743). Of these decedents, 40.9% (n=69,506) died between 2012-2015. Of those who died between 2012 and 2015, 44.2% were tested for opiates, 57.4% were tested for alcohol use, and 35.0% were tested for benzodiazepines, and 29.9% were tested for all three (n=20,802). Of those tested for all three substances, 27.9% tested positive for opiates (n=5,796), and of those testing positive for opiates, 67.0% were confirmed suicides (n=3,885). Of the suicide decedents testing positive for opiate use, 97.3% were overdose deaths, or had opiate use listed as a cause of death (n=3,781). Therefore, 3,781 suicide decedents were included in these analyses.

3. RESULTS

As shown in Table 1, among the 3,781 suicide decedents included in these analyses, there were differences in prevalence of drug use in decedents across rurality (p=0.022). In pairwise comparisons, one significant difference across rurality was found, between use of opiates alone and opiate and benzodiazepine co-use (p = 0.007). While 32% of rural decedents used opiates only, 36% of urban decedents used opiates only. Conversely, 41% of rural decedents used a combination of opiates and benzodiazepines, relative to 34% of urban decedents. Rural decedents were also more likely to be white, have a high school education or lower, to live in a higher poverty county, and to live in a county with either a county-based mix of death investigations systems, a centralized state medical examiner, or a county coroner.

In unadjusted models and models adjusted for individual-level factors, rural suicide decedents were 33% more likely to test positive for opiates and benzodiazepines (vs. opiates alone) relative to urban suicide decedents (RRR adjusting for individual level factors =1.35, 95% Confidence Interval (CI) = 1.10, 1.67; Table 2). Importantly, this association was attenuated after adjusting for community level poverty and was no longer significant when adjusting for region in the full model (Table 2). Conversely, in the full model, rural decedents were at a marginally lower risk of opiates, alcohol, and benzodiazepine co-use (vs. opiates alone) relative to urban decedents.

In further analysis, as shown in Table 3, a number of significant factors were found to be associated with co-use. Co-use of opiates and alcohol (vs. opiates alone) was lower among younger decedents and those with a current mental health or substance use problem, lower in states with a centralized medical examiner death investigation system, and higher among decedents with a current alcohol problem. Co-use of opiates and benzodiazepines (vs. opiates alone) was higher among women, white decedents (vs. Hispanic), those with a current mental health problem or history of mental health treatment, decedents in higher income communities, and from the South (vs. Northeast), while lower in communities with a centralized medical examiner and in later years. Opiates, alcohol, and benzodiazepine co-use was lower than opiate use alone among younger decedents, Hispanic decedents, and those with a current substance use problem, while higher in women, those who were homeless, those with a current mental health problem or current alcohol problem, those in high poverty counties, those with a county medical examiner, and those in the Midwest and the South.

4. DISCUSSION

This study explores differences in simultaneous co-use among suicide decedents who died from an opiate overdose. While previous studies have established that co-use is associated with increased mortality risk, in this study we found differences in drug co-use across rurality among suicide decedents. Specifically, opiate and benzodiazepine co-use was found to be an important drug use combination contributing to mortality in rural areas.

Compared to urban suicide decedents, rural decedents were at greater risk for co-use of opiates and benzodiazepines (vs. opiates alone), in bivariate analyses and even after adjusting for individual-level factors and most community level factors. This is consistent with rural-urban differences in reported co-use among living persons and non-suicide specific decedents. For instance, simultaneous co-use of opiates and benzodiazepines was found to be higher in rural relative to urban decedents in one study in Virginia (Wunsch et al., 2009), and simultaneous non-medical use of prescription drugs has been found to be higher in rural areas relative to urban areas (Keyes et al., 2014).

Increased risk of simultaneous co-use among rural suicide decedents reflects trends in non-medical use, as well as reflecting broad rural-urban differences in access to mental health care. Previous research has indicated that rural patients are less likely to receive office-based mental health care (Mott et al., 2015) but are more likely to receive pharmacotherapy for mental health problems (Ziller et al., 2010). This may explain increased benzodiazepine co-use associated with rural opiate overdose suicide. Increased co-prescribing may also be an unintended consequence of integration of behavioral healthcare in primary care clinics in rural areas (Brown et al., 2015; Ziller et al., 2010). However, urban-rural differences remain after adjustment for a history of mental health treatment and current mental health problems, indicating that this finding is not fully explained by differences in receipt of mental health treatment.

In our study, observed rural-urban differences in opiate and benzodiazepine co-use were no longer significant after adjusting for region. Exploratory analysis revealed that decedents in the South (vs. Northeast) had a 153% higher risk of opiate and benzodiazepine co-use (vs. opiates alone). This suggests that opiate and benzodiazepine co-use among decedents may be a particularly important risk factor for suicide overdose in the South. This finding is consistent with previous studies that found higher rates of co-prescribing of opiates and benzodiazepines in rural areas in the South (McClure et al., 2017). Rural veterans in the South also had higher risk of inappropriate prescribing (Lund et al., 2013b) and were more likely to be prescribed benzodiazepines even when not clinically recommended (Lund et al., 2013a). Interestingly, there were no differences across rurality in risk of opiates and alcohol co-use nor opiates, alcohol, and benzodiazepine co-use. Future research on co-use and suicide should account for regional differences.

In addition to regional differences, we tested additional demographic and mental health factors hypothesized to be associated with co-use. Notably, both opiates and benzodiazepine co-use, and opiate, alcohol, and benzodiazepine co-use was associated with higher suicide risk (vs.opiates alone) for women (vs. men), and lower risk in non-white (vs. white) decedents. These findings reflect co-use patterns in these important subpopulations (McClure et al., 2017). Findings that co-use was related to substance use and mental health problems and history of mental health treatment was unsurprising.

Opiate and benzodiazepine simultaneous co-use was less common in later years than in 2012, which may reflect decreased co-prescribing and increased availability of high potency opiates. This may be because of changes to opiate prescribing guidelines and creation of prescription drug monitoring programs in many states.(Dowell et al., 2016; Guy et al., 2017) Findings of differences across death investigation systems were surprising, given that decedents included in this study all had a toxicology report for all three substances assessed, but may reflect differences in approach to assessing co-use related mortality. Future research is needed to compare when and how different state death investigation systems determine the need for toxicology reports in death reporting among those suspected of overdose.

While this study makes important contributions identifying factors associated with co-use among suicide decedents, there are number of significant limitations. Suicides may be underreported due to the difficulty in ascertaining intention among overdose deaths (Rockett et al., 2018). As this study was limited to NVDRS participating states and decedents with a toxicology report for all three drugs of interest, findings are not generalizable to non-NVDRS states or suicide decedents without toxicology reports (70% of deaths reported to NVDRS during the study period). While only decedents with toxicology reports for all three substances of interest were included in this study to allow comparison across rurality, there may be jurisdictional-level differences in the types of toxicology tests used that may bias results (Larsen et al., 2008). Narrative completeness of NVDRS may also vary across jurisdictions based on each state’s ability to get complete reports, differences in investigative processes within or between agencies, and different partnerships between states and local agencies. However, NVDRS provides states with coding guidelines through a coding training and manual, along with ongoing support for states. This study is also not generalizable to undetermined or non-intentional deaths. While NVDRS is a very rich dataset, it is still limited in which variables are captured. Some factors that may account for differences across rurality in simultaneous co-use patterns (such as type of opiate used, level of toxicity, and medical or non-medical prescription drug use), are not available in NVDRS, resulting in unmeasured confounding. Given the number of states reported and the small sample sizes for certain states, state fixed effects were not possible. Therefore, there may be additional unmeasured state-level confounding that is not addressed in this study.

5. CONCLUSIONS

Even given the limitations, this study makes important contributions to understanding factors associated with co-use related suicides. Rurality was associated with increased risk of opiate and benzodiazepine co-use, although this association was no longer significant after adjusting for region, suggesting that states or regions with greater rurality (such as the South) may thus be more likely to have increased mortality risk from opiate and benzodiazepine co-use. Additionally, key factors including gender, race, region, and co-morbid mental health or substance use are associated with differential simultaneous co-use related suicide risk.

This study also suggests important opportunities for future research or interventions. NVDRS has recently expanded to all states, which will provide a larger dataset in which to examine these patterns for future study. Additional research is also needed to understand the effects of policies and interventions on mortality overall and specifically suicides in rural areas. In particular, research in rural areas is needed on additional strategies to prevent mortality, including the effectiveness of increased availability of naloxone, laws limiting access to alcohol, or policies to prevent over prescribing or co-prescribing of opiates and benzodiazepines (such as prescription drug monitoring systems). Interventions that specifically target co-use among rural providers treating mental health across care settings is crucial. Given the increased risk of opiate and benzodiazepine overdoses relative to opiates alone among decedents with a history of mental health treatment, there may be an opportunity for intervention in this population. This represents decedents who had at least some contact with providers prior to death, and may provide an important opportunity for implementation of life-saving interventions. Finally, future research is needed to understand how increased availability of higher potency opiates (such as fentanyl) effects co-use behaviors and how this contributes to intentional and unintentional mortality, both in urban and rural areas.

WORKS CITED

Baca, C.T., Grant, K.J., 2005. Take-home naloxone to reduce heroin death. Addiction 100, 1823–1831. https://doi.org/10.1111/j.1360-0443.2005.01259.x

Backmund, M., Meyer, K., Meyer, K., Soyka, M., Reimer, J., Schütz, C.G., 2006. Co-Consumption of Benzodiazepines in Heroin Users, Methadone-Substituted and Codeine-Substituted Patients. J. Addict. Dis. 24, 17–29. https://doi.org/10.1300/J069v24n04_02

Blumenthal, S.J., Kagen, J., 2002. The Effects of Socioeconomic Status on Health in Rural and Urban America. JAMA 287, 109. https://doi.org/10.1001/jama.287.1.109-JMS0102-3-1

Borders, T.F., Booth, B.M., 2007. Rural, suburban, and urban variations in alcohol consumption in the United States: findings from the National Epidemiologic Survey on Alcohol and Related Conditions. J. Rural Health Off. J. Am. Rural Health Assoc. Natl. Rural Health Care Assoc. 23, 314–321. https://doi.org/10.1111/j.1748-0361.2007.00109.x

Brown, R.A., Marshall, G.N., Breslau, J., Farris, C., Osilla, K.C., Pincus, H.A., Ruder, T., Voorhies, P., Barnes-Proby, D., Pfrommer, K., Miyashiro, L., Rana, Y., Adamson, D.M., 2015. Access to Behavioral Health Care for Geographically Remote Service Members and Dependents in the U.S. Rand Health Q. 5, 21.

Caetano, R., Kaplan, M.S., Huguet, N., McFarland, B.H., Conner, K., Giesbrecht, N., Nolte, K.B., 2013. Acute Alcohol Intoxication and Suicide Among United States Ethnic/Racial Groups: Findings from the National Violent Death Reporting System. Alcohol. Clin. Exp. Res. 37, 839–846. https://doi.org/10.1111/acer.12038

Darke, S., Ross, J., 2002. Suicide among heroin users: rates, risk factors and methods. Addiction 97, 1383–1394. https://doi.org/10.1046/j.1360-0443.2002.00214.x

Davis, C., Carr, D., 2017. State legal innovations to encourage naloxone dispensing. J. Am. Pharm. Assoc. 57, S180–S184.

Dowell, D., Haegerich, T.M., Chou, R., 2016. CDC Guideline for Prescribing Opioids for Chronic Pain — United States, 2016. MMWR Recomm. Rep. 65, 1–49. https://doi.org/10.15585/mmwr.rr6501e1

Erol, A., Karpyak, V.M., 2015. Sex and gender-related differences in alcohol use and its consequences: Contemporary knowledge and future research considerations. Drug Alcohol Depend. 156, 1–13. https://doi.org/10.1016/j.drugalcdep.2015.08.023

Faul, M., Dailey, M.W., Sugerman, D.E., Sasser, S.M., Levy, B., Paulozzi, L.J., 2015. Disparity in Naloxone Administration by Emergency Medical Service Providers and the Burden of Drug Overdose in US Rural Communities. Am. J. Public Health 105, e26–e32. https://doi.org/10.2105/AJPH.2014.302520

Fournier, C., Ghabrash, M.F., Artenie, A., Roy, E., Zang, G., Bruneau, J., Jutras-Aswad, D., 2017. Association between binge drug use and suicide attempt among people who inject drugs. Subst. Abuse 1–7. https://doi.org/10.1080/08897077.2017.1389800

Fowler, K.A., Jack, S.P.D., Lyons, B.H., Betz, C.J., Petrosky, E., 2018. Surveillance for Violent Deaths — National Violent Death Reporting System, 18 States, 2014. MMWR Surveill. Summ. 67, 1–36. https://doi.org/10.15585/mmwr.ss6702a1

Gressler., L.E., Martin, B.C., Hudson, T.J., Painter, J.T., 2017. The relationship between concomitant benzodiazepine-opioid use and adverse outcomes among U.S. veterans: PAIN 1. https://doi.org/10.1097/j.pain.0000000000001111

Guy, G.P., Zhang, K., Bohm, M.K., Losby, J., Lewis, B., Young, R., Murphy, L.B., Dowell, D., 2017. Vital Signs: Changes in Opioid Prescribing in the United States, 2006–2015. MMWR Morb. Mortal. Wkly. Rep. 66, 697–704. https://doi.org/10.15585/mmwr.mm6626a4

Hall, M., Richardson, T., 2016. Basic Statistics for Comparing Categorical Data From 2 or More Groups. Hosp. Pediatr. 6, 383–385. https://doi.org/10.1542/hpeds.2015-0273

Hanzlick, R., 2006. Medical examiners, coroners, and public health: a review and update. Arch. Pathol. Lab. Med. 130, 1274–1282. https://doi.org/10.1043/1543-21

Cite this page