ACCIDENTAL DRUG OVERDOSE FATALITIES
ALLEGHENY COUNTY, PA
2007-2011: COMPARISON AND RECOMMENDATIONS
by
Ashley Lynn Parsons
BA, Alma College, 2013
Submitted to the Graduate Faculty of
the Department of Epidemiology
Graduate School of Public Health in partial fulfillment
of the requirements for the degree of
Master of Public Health
University of Pittsburgh
2014
UNIVERSITY OF PITTSBURGH
GRADUATE SCHOOL OF PUBLIC HEALTH
This essay is submitted
by
Ashley L. Parsons
on
December 2, 2014
and approved by
Essay Advisor:
Nancy W. Glynn, PhD
Assistant Professor of Epidemiology
Department of Epidemiology
Graduate School of Public Health
University of Pittsburgh
Essay Reader:
Anthony Fabio, PhD, MPH
Assistant Professor of Epidemiology
Department of Epidemiology
Graduate School of Public Health
University of Pittsburgh
______________________________________
______________________________________
Essay Reader:
A. Everette James, III, JD, MBA
______________________________________
Professor of Health Policy and Management
Director, Pitt Health Policy Institute
Department of Health Policy and Management
Graduate School of Public Health
University of Pittsburgh
ii
Copyright © by Ashley L. Parsons
2014
iii
Nancy W. Glynn, PhD
ACCIDENTAL DRUG OVERDOSE FATALITIES
ALLEGHENY COUNTY, PA
2007-2011: COMPARISON AND RECOMMENDATIONS
Ashley L. Parsons, MPH
University of Pittsburgh, 2014
ABSTRACT
A report of the drug overdose fatalities in Allegheny County between years 2007-2011 with
comparison to peer counties is important to the field of Public Health, because of the
implications these findings have on health policy, future monitoring, and targeted interventions
to decrease drug use within the county. All deaths of Allegheny County residents coded as X40X499 under the International Statistical Classification of Disease and Related Health Problems,
10th edition (ICD-10) between 2007-2011 were analyzed to observe and assess trends. A total of
1,055 Allegheny County residents died from accidental poisonings in this time period. Of these
deaths 335 were of residents residing within the city limits for Pittsburgh, while 720 were of
residents living in the remainder of the county (area not considered to be the City of Pittsburgh).
Males had substantially higher age-adjusted fatality rates than females, with average fatality rates
for 2007-2011 of 24.3 and 10.1 per 100,000 people, respectively. The largest proportion of
deaths within Allegheny County was due to accidental poisonings by and exposure to narcotics
and hallucinogens (ICD-10: X42). When comparing Allegheny County to the United States and
its peers of Philadelphia, Cook, and Wayne County, Allegheny County had lower age-adjusted
iv
rates than Philadelphia County, but higher rates than the US, Cook County, and Wayne County.
When considering the average fatality rates, Allegheny County had similar poisoning rates
between whites and blacks (17.7 and 17.8 per 100,000 people, respectively). This trend was not
seen in any other location observed. After analyzing and comparing Allegheny County
accidental poisoning fatalities for 2007-2011 it appears that the county is experiencing a drug use
problem that warrants investigation and intervention. Future steps to decrease drug use and
subsequent death of Allegheny County residents include policy change, monitoring of changes in
drug death patterns (movement from urban areas to suburban and rural), consideration of new
intervention techniques, and education for residents and physicians. By monitoring future drug
overdose fatalities within the county, there is potential to decrease the public health burden of
drug use through implementation of targeted interventions and education programs.
v
TABLE OF CONTENTS
PREFACE ..................................................................................................................................... X
DISCLAIMER............................................................................................................................. XI
1.0
INTRODUCTION................................................................................................................ 1
1.1 DRUG ABUSE IN THE UNITED STATES ............................................................. 1
1.2 ACCIDENTAL POISIONING & EXPOSURE TO NOXIOUS SUBSTANCES
FATALITIES ........................................................................................................................ 3
1.3 PUBLIC HEALTH SIGNIFICANCE ....................................................................... 4
1.4 OBJECTIVE ................................................................................................................ 5
2.0
METHODS ........................................................................................................................... 6
2.1 STUDY POPULATION .............................................................................................. 6
2.2 ALLEGHENY COUNTY DEATH RECORDS ....................................................... 7
2.2.1
Oracle Database ............................................................................................... 8
2.2.2
Age-Adjustment ............................................................................................... 8
2.3 CDC WONDER ........................................................................................................... 8
2.4 DATA ANALYSIS....................................................................................................... 9
3.0
RESULTS ........................................................................................................................... 10
4.0
DISCUSSION ..................................................................................................................... 23
APPENDIX: TECHNICAL NOTES ......................................................................................... 31
vi
BIBLIOGRAPHY ....................................................................................................................... 34
vii
LIST OF TABLES
Table 1. Age-Adjusted Rates (per 100,000 people) of Overdose Fatalities, 2007-2011 by
Location………………………………………………………………………………..15
Table 2. ICD-10 Mortality Classification Definitions…………………………………………...21
Table 3. Drug Overdose Fatalities Allegheny County Residents, 2007-2011 by Specific Cause of
Death…………………………………………………………………………………..22
viii
LIST OF FIGURES
Figure 1. External Causes of Death Classification Diagram…………………………………..…7
Figure 2. Drug Overdose Fatlaities of Allegheny County Residents, 2007-2011 by Sex…….…11
Figure 3. Drug Overdose Fatalities of Allegheny County Residents, 2007-2011 by Location of
Residency………………………………………………………………………………12
Figure 4. Drug Overdose Fatalities, 2007-2011: A Comparison Among Peer Counties………...13
Figure 5. Drug Overdose Fatalities in Whites, 2007-2011 by Location…………………………16
Figure 6. Drug Overdose Fatalities in Blacks, 2007-2011by Location……………………….…17
Figure 7. Drug Overdose Fatalities in Males, 2007-2011 by Location………………………….18
Figure 8. Drug Overdose Fatalities in Females, 2007-2011 by Location……………………….19
ix
PREFACE
I would like to thank my friends and family for their continuous support and understanding
during my time in Pittsburgh and throughout my graduate education. My deepest gratitude goes
to my advisor Dr. Nancy Glynn for her guidance, patience, and reassurance for the entirety of my
graduate career. I would also like to extend my appreciation to my essay committee of Dr. Tony
Fabio and Mr. Everette James and to my internship preceptor Mr. John Kokenda for their
constructive feedback and encouragement throughout this process. All of your support has paved
the way for my future as a Public Health professional and I am truly grateful for all of your time
and effort towards my success.
x
DISCLAIMER
The Allegheny County Health Department, Office of Epidemiology and Biostatistics provide
these data through a cooperative agreement with the Pennsylvania Department of Health, which
requires the following disclaimer:
"These data were supplied by the State Health Data Center, Pennsylvania
Department of Health, Harrisburg, Pennsylvania.
The Pennsylvania
Department of Health specifically disclaims responsibility for any analyses,
interpretations or conclusions."
The Bureau of Health Statistics and Research strictly maintain the confidentiality of death
certificates. All published reports of vital statistics contain only aggregated data. Individually
identifiable information is only released, as provided for in Act 66, Vital Statistics Law of 1953,
for health research purposes and to government agencies in pursuance of their official duties
upon completion, review and approval of an application for access to protected data. Certified
copies are also released to family members or legal representatives for determination of personal
or property rights.
xi
1.0
1.1
INTRODUCTION
DRUG ABUSE IN THE UNITED STATES
Historically, the United States has had a large problem with illicit drug use and resulting
fatalities; however, since 2005 heroin and cocaine use has plateaued, while prescription drug
abuse has grown.1 Between the years of 1999 and 2010 the death rate from prescription opioid
use in the United States has increased four-fold. This increase of 4,030 to 16,651 deaths exceeds
the combined death toll for cocaine and heroin overdoses in the same time period.2, 3 Likewise, in
2010, prescription opioids were involved in over 16,000 overdose deaths; heroin was implicated
in only 3,036.4 A large issue with this shift is that the population that has been most affected by
fatalities are those using Medicaid5. The National Institute on Drug Abuse points out that
individuals who are 65 years and older are more likely to improperly use their medication and
are at highest risk for fatalities due to drug interactions and compounded health problems.6
In 2010, the National Survey on Drug Use and Health (NSDUH) found that an estimated
2.4 million Americans used prescription drugs for non-medical purposes for the first time in the
past year. Of these prescription drug abusers, the group with highest rate of abuse was young
adults aged 18 to 25 years. Young adults were followed by adolescents aged 12 to 17, who
accounted for one-third of the drug abusers in 2010. With the exception of the age group 12 to 17
1
years, males had higher rates of drug abuse than females. Commonly abused prescription drugs
included opioids, drugs acting as depressants of the central nervous system, and stimulants.6
The National Institute on Drug Abuse reports that in the year 2012, 23.9 million
Americans aged 12 years and older (9.2% of the population) had used an illicit drug or abused
psychotherapeutic medicines in the past month. This rate increased from 2002, when the rate of
abuse was 8.3%. The most recent increases in illicit drug use reflect an increase in the use of
marijuana. NSDUH shows that most people use drugs for the first time as teenagers—with 2.8
million new users of illicit drugs in 2012. Although new drug users are commonly young adults
and adolescents (52% of new users in 2012 were 18 years of age or younger), drug use is
increasing rapidly among people in their 50’s. This trend is partially due to the aging of the
“baby boomer” generation, which has had much higher rates of illicit drug use than previous
cohorts.6
The U.S Department of Health and human Services’ Substance Abuse and Mental Health
Services Administration (SAMHSA) collects substance abuse trends via survey every year. In
2013, the rate of substance dependence or abuse was higher in males aged 12 or older than the
rate for females (10.8 percent in males, 5.8 percent in females). However, when examining the
age group of 12 to 17 years, dependence and abuse among males was similar to the rate in
females for 2013 (5.3 percent and 5.2 percent, respectively). Also in 2013, the rate of substance
dependence or abuse was higher in whites than blacks (8.4 percent vs. 7.4 percent) among
individuals aged 12 years or older. The 2013 SAMHSA survey also found that rates of illicit
drug or alcohol dependence and abuse was highest in the West (8.9 percent), followed by similar
rates in the Northeast and Midwest (8.3 and 8.2 percent), and lowest in the South (7.8 percent).
Illicit drug or alcohol dependence or abuse was highest in large and small metropolitan areas (8.6
2
and 8.4 percent), and lowest in nonmetropolitan areas (6.6 percent). 7 These trends indicate that
drug abuse and dependency poses the largest burden to white males living in urban areas.
As drug abuse rates increase, it would be expected that treatment rates would increase
proportionally. However, this is not the case. In 2012, only 2.5 million individuals received
treatment, which is approximately 10% of the individuals who needed treatment. Likewise, of
those treated, less than 14% were for substances normally associated with high rates of fatalities
due to overdose.6
1.2
ACCIDENTAL POISIONING & EXPOSURE TO NOXIOUS SUBSTANCES
FATALITIES
In the United States, nine out of ten poisonings are accidental drug overdoses. This type of
external cause of death has surpassed motor vehicle crash fatalities and is now the leading cause
of injury death.8, 9 In regards to demographics, in 2012 whites had the highest death rate for
poisoning fatalities compared to blacks. It was also found that in 2011, the majority of
emergency department visits (56%) were for males and 82% were for people 21 and older.
Likewise, men were 59% more likely than women to die from a drug overdose. The highest
death rate for poisoning fatalities in 2012 was to individuals aged 45 to 59, while the lowest
death rate for this time period was in children who were less than 15. These trends are also
represented in abuse tendencies in these age groups, with adolescents having lower rates of drug
use than middle-aged individuals.4 Historically, rural and impoverished counties tend to have
higher prescription drug overdose fatalities than higher income areas.10, 11
3
1.3
PUBLIC HEALTH SIGNIFICANCE
Drug abuse and poisoning fatalities pose a huge threat to public health and to society as a whole.
In 2012, the FBI reported that the highest number of arrests came from drug abuse violations
(1,552,432 arrests).12 There is also evidence that shows that drug users are more likely to commit
crimes and disrupt society than nonusers. This increase in violence stems from the nature of the
drugs that are abused, but is also due to drug trafficking itself. However, how much crime results
from drugs is hard to quantify.13
Drug abuse is not only problematic because of the implications it has on crime, but also
because of its effect on health care costs. In 2007, it was reported that prescription-opioid abuse
cost insurers an estimated $72.5 billion. This cost is similar to that of chronic disease treatments
for illnesses such as asthma and HIV.4 Also, in 2009, it was reported that 1.2 million emergency
department visits were associated with prescription drug miss use or abuse, which was a 98.4%
increase over emergency department visits in 2004. Comparatively, one million emergency
department visits were due to illicit drug use.14 With poisoning fatalities increasing across the
board and different groups being disproportionately affected, it is crucial that public health
leaders and policy makers examine ways in which the rate of drug abuse can be decreased. The
first step in changing these outcomes is to examine the areas in which poisoning fatalities are
highest and determine which groups are most affected.
4
1.4
OBJECTIVE
The objective of this report is to provide an analysis of the trends of accidental drug fatalities in
Allegheny County, Pennsylvania between the years 2007 and 2011. Likewise this report will
compare the rates of poisoning deaths to Allegheny County residents with those of the United
States and peer counties for the same time period. This comparison will allow for the
interpretation of the death rates and examine demographic trends. Results of these analyses will
help to inform and influence health practice and health policy regarding drug access and fatalities
in Allegheny County and in the state of Pennsylvania.
5
2.0
2.1
METHODS
STUDY POPULATION
The data for this research is from the Allegheny County Health Department, Epidemiology and
Biostatistics department, office of Vital Statistics in Pittsburgh, PA. The study population
consists of decedents who were residents of Allegheny County and died between the years 2007
and 2011. The subset of the population utilized for this research consists of all individuals whose
cause of death was considered to be an accidental poisoning or death due to inhalation of a
noxious substance as classified by the International Statistical Classification of Diseases and
Related Health Problems, 10th edition (ICD-10: X40 to X499), Figure 1.
6
Figure 1. External Causes of Death Classification Diagram
2.2
ALLEGHENY COUNTY DEATH RECORDS
The variables for this research come from the decedents’ death certificate. These certificates are
completed by medical professionals or during an autopsy at a coroner’s office. Once completed,
they are sent to the Pennsylvania Department of Health in Harrisburg, PA, for disbursement to
the corresponding county health departments.
The Allegheny County Health Department
receives a physical copy of all death certificates and a digital file to be used for analytical
purposes.
7
2.2.1 Oracle Database
The digital death certificate information is stored using ORACLE (Version 11G, Release 2).
Within the ORACLE database, a program to abstract necessary records has been created. This
program identified race, sex, and location (Allegheny, City of Pittsburgh, and Remainder of
County) information for all records classified as X40-X499 between 2007 and 2011.
2.2.2 Age-Adjustment
Crude number of fatalities were inputted into an excel program to calculate the age-adjusted
rates. Age-specific fatality rates were calculated using the 2010 bridged population and age
adjusted using the U.S. 2000 Standard Million Population. Adjusted rates were calculated in
order to make comparisons across different populations (See Appendix).
2.3
CDC WONDER
Allegheny County fatality rates were compared with fatality rates for the United States and other
counties of similar size and social structure, as determined by the Centers for Disease Control
and Prevention. The comparable counties selected include Philadelphia County in Pennsylvania,
Wayne County in Michigan, and Cook County in Illinois. Age-adjusted rates for the United
States and peer counties were derived using the CDC’s Wonder Database utilizing the US 2000
Standard Million Population. The time period of 2007-2011 was used for all counties and ICD-
8
10 cause of death classification of “accidental poisoning and inhalation of noxious substances”
(ICD-10: X40-X499) was used to derive all rates.
2.4
DATA ANALYSIS
Fatality rates due to accidental poisonings for Allegheny County were analyzed based on
location of the decedents’ residence. Three locations were used including all of Allegheny
County (the comparator), the City of Pittsburgh, and the Remainder of the County (all of
Allegheny County that is not classified as “City of Pittsburgh”)—see Appendix. Median age at
death and age-adjusted rates were compared based on location, sex, and race (black vs. white).
Age-adjusted rates based on aggregated 2007-2011 data for the United States, Philadelphia
County, Wayne County, and Cook County were compared with Allegheny County to examine
differences and disparities.
9
3.0
RESULTS
Between the years 2007 and 2011, 1,055 Allegheny County Residents died from accidental
poisonings or inhalation of noxious substances. Of the poisoning deaths, 335 (31.8%) were from
residents living within the city limits of Pittsburgh, while 720 (68.3%) occurred in Allegheny
County residents living outside of the city (Remainder of the County). The lowest age-adjusted
rate in the five-year period was 12.5 per 100,000 people (year 2007) and the highest age-adjusted
rate was in 2011, with 20.2 deaths per 100,000 people (an increase of 61.6%).
This upward trend was found in both sexes throughout the time period, as shown in
Figure 2. Males had higher rates than females in all years; however, there was a larger increase
in rates of fatalities between years 2010 and 2011 in women than in men (10.6 to 14.7 per
100,000 and 24.5 to 26.0 per 100,000, respectively).
10
Age-Adjusted Rate (per 100,000 people)
30.0
25.0
20.0
All
15.0
Males
Females
10.0
5.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 2. Drug Overdose Fatalities of Allegheny County Residents, 2007-2011 by Sex
Figure 3 shows differences between the City of Pittsburgh and the Remainder of the
County (all of Allegheny County that is not classified as “City of Pittsburgh”), in regards to
overdose fatalities. Overall, Allegheny County experienced an increase in death of its residents
from accidental poisonings between 2007-2011 with rates increasing from 12.5 to 20.2 deaths
per 100,000 people. It was also observed that there was an increase between 2007 and 2008 (12.5
to 17.9 per 100,000 people) followed by a decline and plateau between 2008-2010 (17.9, 17.0,
and 17.4 per 100,000 people, respectively), and an increase from 17.4 to 20.2 deaths per 100,000
people between years 2010 and 2011. The Remainder of the County followed a similar trend as
Allegheny County age-adjusted rates; however, the fatalities within the City of Pittsburgh were
constant about 22 per 100,000 people during years 2007-2009 and declined in 2010 (19.4 per
11
100,000 people). After 2010, there was a spike in drug fatalities for City of Pittsburgh residents
with age-adjusted rates increasing from 19.4 to 26.0 per 100,000 people.
Age-Adjusted Rate (per 100,000 people)
30.0
25.0
20.0
Allgheny County
15.0
somewhere on your graph
City of Pittsburgh
Remainder
10.0
5.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 3. Drug Overdose Fatalities in Allegheny County Residents, 2007-2011 by
Location of Residency
When comparing average age-adjusted rates of accidental drug fatalities between the
United States, Allegheny County, Philadelphia County, Cook County, and Wayne County;
Allegheny County surpassed all areas except Philadelphia County for the time period observed,
Figure 4. Allegheny County had higher fatality rates between 2007 and 2011 than the United
States or Cook County, with rates between 12.5 and 20.2 per 100,000 people. Between 2008 and
2011, Allegheny County had higher age-adjusted rates than Wayne County.
12
Age-Adjusted Rate (per 100,000 people)
30.0
25.0
20.0
Allegheny County, PA
USA
15.0
Philadelphia County, PA
Cook County, IL
10.0
Wayne County, MI
5.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 4. Drug Overdose Fatalities, 2007-2011: A Comparison Among Peer Counties
A comparison of all age-adjusted rates between Allegheny County, the City of Pittsburgh,
and the Remainder of the County is represented in Table 1. The city of Pittsburgh had higher
rates than the Remainder of the County (average 22.2 and 15.8 per 100,000 people, respectively.
In Allegheny County, the fatality rates increased between years 2007 and 2011, this increase was
also seen in the Remainder of the County with rates increasing from 9.9 in 2007 to 19.0 per
100,000 people in 2011. The City of Pittsburgh remand constant around its average fatality rate
of 22.2 per 100,00 people for years 2007 through 2010 and increased in year 2011 (26.0
poisoning deaths per 100,000 people). For whites, the City of Pittsburgh had an average fatality
rate of 24.2 per 100,000 people, while the Remainder of the County was 16.5. Males were higher
than females in all locations, but for the City of Pittsburgh, Males were much higher (33.6 per
13
100,000 people) than the Remainder of the County (22.0 per 100,000). This trend was also seen
in the female population with the City of Pittsburgh having a fatality rate of 11.2 per 100,000
people and a fatality rate of 9.9 per 100,000 people for the Remainder of the County.
14
Table 1. Age Adjusted Rates, Overdose Fatalities 2007 to 2011 by Location
All Drug
Fatalities
White
Black
Males
Females
Location
Allegheny County, PA
City of Pittsburgh
Remainder of Allegheny County
USA
Philadelphia County, PA
Cook County, IL
2007
12.5
21.3
9.9
9.9
22.5
9.5
2008
17.9
22.1
17.3
10.2
24.7
9.8
2009
17.0
22.2
15.6
10.3
22.8
9.2
2010
17.4
19.4
17.3
10.6
21.4
8.9
2011
20.2
26.0
19.0
10.5
27.5
9.3
Average
Rate
17.0
22.2
15.8
10.3
23.8
9.3
Wayne County, MI
13.5
12.8
15.4
13.4
15.9
14.2
Allegheny County, PA
City of Pittsburgh
Remainder of Allegheny County
USA
12.9
23.5
10.3
10.7
18.6
24.5
17.8
11.3
17.6
22.4
16.5
11.4
17.9
19.4
18.2
11.9
21.3
31.1
19.5
13.0
17.7
24.2
16.5
11.7
Philadelphia County, PA
Cook County, IL
30.0
7.8
32.2
8.4
29.2
7.9
29.8
9.1
37.3
9.0
31.7
8.4
Wayne County, MI
13.9
14.1
17.3
16.6
20.4
16.5
Allegheny County, PA
13.9
18.7
17.7
20.4
18.4
17.8
City of Pittsburgh
Remainder of Allegheny County
USA
Philadelphia County, PA
Cook County, IL
19.2
7.8
8.5
18.1
16.4
19.9
17.5
7.3
20.9
15.8
22.8
12.1
7.3
19.5
15.1
23.3
17.7
7.3
16.2
11.0
14.8
21.4
8.0
21.9
12.7
20.0
15.3
7.7
19.3
14.2
Wayne County, MI
14.3
12.4
14.3
11.2
11.2
12.7
Allegheny County, PA
City of Pittsburgh
Remainder of Allegheny County
USA
20.2
35.7
15.5
13.1
25.6
33.8
24.1
13.6
25.3
36.3
22.0
13.5
24.5
26.8
24.4
13.8
26.0
35.4
24.2
15.1
24.3
33.6
22.0
13.8
Philadelphia County, PA
33.0
36.8
34.1
31.3
42.9
35.6
Cook County, IL
13.8
14.4
14.0
13.7
14.4
14.1
Wayne County, MI
18.7
16.5
21.1
15.6
19.9
18.4
Allegheny County, PA
City of Pittsburgh
Remainder of Allegheny County
USA
Philadelphia County, PA
Cook County, IL
Wayne County, MI
5.4
7.8
4.7
6.6
13.3
5.3
8.7
10.6
10.8
10.7
6.8
14.0
5.3
9.4
9.2
8.8
9.6
7.1
12.7
4.6
10.1
10.6
11.9
10.5
7.5
12.6
4.3
11.4
14.7
16.8
14.1
8.2
13.7
4.4
12.1
10.1
11.2
9.9
7.2
13.3
4.8
10.3
15
Table 1 also shows a comparison of all age-adjusted rates between Allegheny County
with the United States, Philadelphia County, Cook County, and Wayne County. Among whites,
Allegheny County had rates higher than the United States and Cook County. Allegheny County
and Wayne County had similar trends of overdose fatalities between 2007 and 2011 among the
White population, Figure 5. Allegheny County had lower rates than Philadelphia County, which
exceeded all areas observed for rates of overdose fatalities in whites and had the highest rate for
all locations in 2010 (37.3 deaths per 100,000 people).
Age-Adjusted Rate (per 100,000 people)
40.0
35.0
30.0
25.0
Allegheny County, PA
USA
20.0
Philadelphia County, PA
15.0
Cook County, IL
Wayne County, MI
10.0
5.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 5. Drug Overdose Fatalities in Whites, 2007-2011 by Location
16
Allegheny County blacks had much higher fatality rates between 2007 and 2011 for
poisonings than the United States (Allegheny County ranged from 13.9 to 20.4 per 100,000
people and United States ranged from 7.3 to 8.5 per 100,000 people), Figure 6. This trend is
similar to Philadelphia County between years 2007 and 2011. The average age-adjusted rate for
accidental drug overdose deaths in Allegheny County was 17.8 and for Philadelphia County it
was 19.3 per 100,000 people. The City of Pittsburgh had the highest average age-adjusted drug
overdose rate for blacks during this time period, with 20 per 100,000 people.
Age-Adjusted Rate (per 100,000 people)
25.0
20.0
15.0
Allegheny County, PA
USA
Philadelphia County, PA
10.0
Cook County, IL
Wayne County, MI
5.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 6. Drug Overdose Fatalities in Blacks, 2007-2011 by Location
17
When comparing males and females, Figure 7 and Figure 8, males had substantially
higher rates of drug overdose fatalities than females in Allegheny County and all other locations
observed for all years. The highest rate for females was seen in the City of Pittsburgh in 2011
with 16.8 per 100,000 people. Allegheny County followed the same trends as the United States,
Philadelphia County, and Wayne County with an increase of drug overdose age-adjusted death
rates for Females over time.
50.0
Age-Adjusted Rate (per 100,000 people)
45.0
40.0
35.0
30.0
Allegheny County, PA
USA
25.0
Philadelphia County, PA
20.0
Cook County, IL
15.0
Wayne County, MI
10.0
5.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 7. Drug Overdose Fatalities in Males, 2007-2011 by Location
18
Age-Adjusted Rate (per 100,000 people)
16.0
14.0
12.0
10.0
Allegheny County, PA
USA
8.0
Philadelphia County, PA
6.0
Cook County, IL
Wayne County, MI
4.0
2.0
0.0
2007
2008
2009
Year of Death
2010
2011
Figure 8. Drug Overdose Fatalities in Females, 2007-2011 by Location
Table 3 gives descriptions for all ICD-10 classifications for accidental poisonings and
inhalation of noxious substances. As shown in Table 4, the largest number of deaths due to
accidental overdoses for Allegheny County residents occurred in X42 (accidental poisoning by
and exposure to narcotics and hallucinogens) and X44 (Accidental poisoning by and exposure to
other and unspecified drugs, medicaments, and biological substances). Classification X42
accounted for 46.2% of deaths within the county and X44 accounted for 43.3%. In Allegheny
County the median age at death for all poisonings was 39.2 years. For the City of Pittsburgh and
the Remainder of the County the median ages of death were 45.3 and 38.8 years, respectively. In
Allegheny County, ICD-10 classification X42 (Accidental poisoning by and exposure to
narcotics and hallucinogens) had the youngest median age at death with 37.7 years, while X47
19
(Accidental poisoning by and exposure to other gases and vapors) had the oldest with 65.3 years
of age at death and accounted for 3% of poisoning deaths within the county.
20
Table 2. ICD-10 Mortality Classification Definition15
X40
4-amniophenol derivatives, nonsteroidal anti-inflammatory drugs (SAID), pyrazolone derivatives
X41
Accidental poisoning by and exposure to antiepileptic, sedative-hypnotic, antiparkinsonism and psychotropic
drugs: Antidepressants, barbiturates, hydantoin derivatives, iminostilbenes, methaqualone compounds,
neuroleptics, psychostiumlants, succinimides and oxazolidinediones, tranquillizers
X42
Accidental poisioning by and exposure to narcotics and hallucingoens: Cannabis (derivatives), cocaine, codeine,
heroin, lysergide (LSD), mescaline, methadone, morphine, opium (alkaloids)
X43
Accidental poisoning by and exposure to other drugs acting on the autonomic nervous system
X44
Accidental poisoning by and exposure to other and unspecified drugs, medicaments, and biological substances
X45
Accidental poisoning by and exposure to alcohol
X46
Accidental poisoning by and exposure to organic solvents and halogenated hydrocarbons and their vapours
X47
Accidental poisoning by and exposure to other gases and vapors
X48
Accidental poisoning by and exposure to pesticides
X49
Accidental poisoning by and exposure to other and unspecified chemicals and noxious substances
21
Table 3. Drug Overdose Fatalities, 2007-2011 by Cause of Death
Cause
Number of Deaths
Allegheny City of
County
Pittsburgh Remainder
Median Age at Death (years)
Allegheny City of
County
Pittsburgh Remainder
Proportion of Drug Deaths* (%)
Allegheny City of
County
Pittsburgh Remainder
X40
X41
X42
X43
X44
X45
X46
X47
X48
X49
6
26
487
0
455
47
1
31
0
2
3
5
160
0
141
18
1
7
0
0
3
21
327
0
314
29
0
24
0
2
49.0
45.9
37.7
45.0
46.9
49.5**
65.3
44.5
46.7
46.2
45.5
39.6
45.0
49.5**
75.7
-
52.5
45.8
36.8
45.1
48.3
65.0
47.0**
0.57
2.46
46.16
0.00
43.13
4.45
0.09
2.94
0.00
0.19
0.90
1.49
47.76
0.00
42.09
5.37
0.30
2.09
0.00
0.00
0.42
2.92
45.42
0.00
43.61
4.03
0.00
3.33
0.00
0.28
Total
1055
335
720
39.2
45.3
38.8
100.00
100.00
100.00
*Proportion of drug deaths was calculated by dividing the crude number of fatalities for a particular cause by the total number of deaths for that location
**Median age based on average age group in n<2
22
4.0
DISCUSSION
Overall, from 2007 to 2011, Allegheny County had fatalities due to accidental poisonings at
higher rates than peer counties and at higher rate than the United States. Fatality rates for
accidental poisonings were highest in males and whites. The calculated average ages at death
points to the largest problem for overdose fatalities being in middle-aged individuals. The death
certificates reviewed for this report indicate that there is a drug problem within Allegheny
County that is increasing in the suburban and rural areas of the county. There are several
different aspects that should be examined and evaluated in order to combat this issue. This
includes evaluation of prescribing tendencies of Allegheny County physicians, current
medication laws and policies, examining reasons for changes in types of drug fatalities, and
consideration of future implications of prescription drug monitoring and intervention programs.
Nationwide, there has been a large increase in prescriptions for opioids between the years
1999 to 2010. One reason for this uptick in prescribing has been due to a greater emphasis on
treating pain in the medical community.16,
17
There has also been an increase in industry-
sponsored advertising, which has included instances of educational brochures suggesting that
physicians should ignore signs of opioid dependence in low-risk patients.18 In addition to a
change in attitude and industry intervention, long-term opioid use has been shown to lead to
tolerance. A solution to tolerance in a patient with chronic pain is to increase opioid dose;
however, there is little known about a maximum safe dose of opioids. It has been shown that
23
there is a linear relationship between dose and likelihood of death, with increasing dosages
resulting in an increased chance of overdose fatality.19 An increase in poisoning fatalities in the
United States has increased at the same rate and magnitude as the increase in prescribing of
opioids, which suggests that the primary driving force of fatalities comes from prescription
drugs.
Within Allegheny County, males have the highest rate of fatalities. This was represented
in analysis of the county death records and was shown when the City of Pittsburgh and the
Reminder of the County were separated. Although whites had slightly higher rates than blacks
for each individual year, average rates of fatalities for blacks and whites were similar for
Allegheny County. This trend was not observed in the other areas analyzed. However, it should
be noted that when looking at only the City of Pittsburgh, whites had fatality rates from
poisoning fatalities higher than their black counterparts. When looking at age at death, the
median age for poisonings in the county was 39.2 years. This indicates that the population most
at risk for drug overdose fatalities is middle-age individuals. Lastly, the data analyzed shows
that, on average, the City of Pittsburgh has higher fatality rates than Allegheny County (as a
whole) and higher rates than the Remainder of the County. The trends of fatalities for Allegheny
County are consistent with the SAMHSA findings of substance abuse and dependence for the
United States.7 This suggests that drug abuse and dependence lead to future poisoning fatalities.
It was found that for 2007 to 2011, Allegheny County had higher fatality rates than peer
counties in Detroit, Michigan and Chicago, Illinois. One reason for this difference could be the
result of the structure of Allegheny County in comparison to the more urbanized Wayne and
Cook counties. Research has shown that counties with rural areas tend to have higher
prescription drug overdose fatalities in comparison to entirely urban and suburban counties.10, 11
24
Another possible explanation for these dissimilar rates is that these counties come from states
with differing laws. Wayne County follows Michigan law, which was established in 1988 to
monitor drug schedules II through V (see Appendix), Michigan has the strictest laws of these
states examined in this report and may explain why Wayne County has the second lowest drug
fatality rates of the peer counties examined.20 Since 1961, Cook County (Illinois) has monitored
scheduled II drugs and has the lowest peer county rate.21 Pennsylvania has the newest laws; with
schedule II drugs being monitored as of 2002 and new policy—passed in the fall of 2014—that
will require monitoring of schedules III through V.20,22 Philadelphia and Allegheny County, both
in Pennsylvania, had the highest rates in all subgroups, which could be a result of having newer
drug monitoring laws than other states. States not only hold the power to regulate the use and
prescribing tendencies of opioid pain relievers, but also are responsible for financing and
regulating health care for Medicaid populations—who are at a greater risk for overdose.5
Therefore, it is important that states take responsibility for decreasing risk of overdose within
their jurisdiction.
Recent evidence has shown that the types of drug use associated with fatalities have
changed over time. A study by Paulozzi and colleagues showed that prescription drugs have
replaced heroine and cocaine as the leading causes of drug fatalities. This article also comments
that drug overdoses are no longer an “urban problem”.11 Although the data for years 2007 to
2011 does not show that Allegheny County has experienced a complete shift of drug fatalities
from the City of Pittsburgh (urban) to the Remainder of the County (suburban and rural), there
has been a recent increase in drug fatalities in the suburban and rural portions of the county that
was not seen in previous years. It should also be noted that Allegheny County had higher fatality
rates due to accidental poisonings than counties that contain large metropolitan areas such as
25
Chicago and Detroit. This indicates that a shift in where and by whom drugs are being used is
happening.
A nationwide shift of drug use from large cities to smaller cities and non-metropolitan
areas, such as Allegheny County poses several problems. The first being that increased drug use
in a population has been shown to lead to an increase in other types of injury. This is because
drug abuse is a risk factor for both suicide and motor vehicle crashes, which are already more
common in rural areas.24, 25, 26 Another issue with drug use expanding out of urban populations is
that physicians in suburban and rural areas are not always accustomed to treating pain and may
be targeted by drug manufacturers, as discussed earlier.24, 27, 28
This report utilized standardized data programs and data collection, which conferred
several strengths and introduced some limitations. Death certificate data has conflicting views of
it’s quality as a data source due to errors in completing the certificates and missing information.29
Although errors within the certificates the Allegheny County Vital Statistics data set inevitably
exists, it should be noted that the strengths of this data collection method outweigh the possible
flaws. One such strength is that all deceased Allegheny County residents have a death certificate
that is submitted to the Allegheny County Health Department, making this sample nearly perfect.
Another strength is that the deaths examined were due to accidental causes, which means that a
coroner reviewed the deaths and completed the death certificate. Because the deaths are
accidental, the majority of the sample received toxicology to determine the particular cause of
death; thus leading to potentially more accurate data than typical death certificates. Another
possible source of limitations comes from age-adjustment. Since age-adjustment relies on rates,
it can be affected by population size; however, age-adjustment is a necessary step for comparison
26
and allowed for accurate evaluation of the differences in drug fatalities between Allegheny
County and peer counties with different age structures.
This report has indicated that Allegheny County has a drug-poisoning problem that
warrants further evaluation and intervention. An implication of these results is that Allegheny
County needs to utilize drug-monitoring programs in order to provide targeted interventions for
populations at highest risk for drug fatalities. Based on the records reviewed, middle-aged men
living in the City of Pittsburgh are at highest risk for death from a drug overdose. With this being
said, it is important that the female and suburban/rural populations be monitored due to the
emerging trend of abuse for these populations. A drug-monitoring program, such as the one
passed by Pennsylvania state representatives in September 2014 under Senate Bill 1180, will
allow for public health professionals to determine where misuse and abuse is happening and by
what population subgroups. Other types of data that should be monitored include Naloxone
administration, hospital admissions relating to drug use, and arrest data. By monitoring different
types of drug use data, public health officials will be able to understand what populations are
using drugs and which populations are dying from overdose. Because populations who are
abusing drugs are changing, it is important that prevention efforts shift accordingly in order to
focus on populations that were not previously considered at high risk.11 By knowing to whom
and where the problem exists, targeted interventions in Allegheny County can be created.
A type of intervention that should be considered by treatment centers in Allegheny
County is the use of medication-assisted technologies (MATs). Despite their safe and cost
effective ability to reduce risk of overdose, MATs are underutilized. MATs consist of treatments
such as methadone, buprenorphine, and naltrexone to rehabilitate drug addicts to a life free of
addictive substances. MATs have not only been successful in improving patient’s social
27
functioning, but have also been shown to reduce risks of infectious disease transmission and
engagement in criminal activities.30, 31 Currently, this type of treatment is not heavily utilized;
less than half of private-sector treatment programs adopt this practice and of those that do, only
34.4% of patients receive them.28 Likely reasons as to why MATs are not relied upon are that
there are provider skepticism, inadequate doses which result in relapse, and limitations due to
policy and regulation. Such regulations are placed by insurance providers and limit the amount
of dosages, provide minimal counseling, and have strict authorization requirements that impede
their use.31, 32 If Allegheny County wishes to see lower rates of drug fatalities, it must consider
interventions based on the intervention’s success and cost-effectiveness and work to overcome
barriers to such treatments.
Although a prescription drug-monitoring bill is an important step for Pennsylvania to
decrease its rate of drug fatalities, it is only the first step in a series of necessary policy and
attitude changes for the state. Another important change to legislation comes from the
amendments to Senate Bill 1164, which has been coined the “Good Samaritan Law”. This bill
allows for Good Samaritan immunity from prosecution for certain drug crimes to individuals
who are seeking help for a companion who overdoses on drugs. This bill will also allow for the
expansion of naloxone—an opioid antagonist—to individuals who have a risk of overdosing and
their friends and family members.23 This bill not only helps to overcome the problem of drug
overdose fatalities occurring due to fear of seeking medical attention because of potential legal
problems, but also allows for the widespread use of Naloxone in instances of drug overdose. In
the years 1996 to 2010, Overdose Education trained 50,000 potential bystanders with Naloxone
Distribution (OEND) programs in the United States. This program resulted in the reversal of
over 10,000 opioid overdoses.33 Likewise, simulations have shown that Naloxone distribution to
28
known heroin abusers increases quality of life years and is a cost-effective way to decrease death
from drug abuse.34 Even though the bill has been passed and Naloxone has shown to be an
effective life-saving tool, there are still many barriers to its practical use. Pushback of Naloxone
distribution programs include questions on whether lay responders will be able to use the opioid
antagonist effectively, misconceptions that Naloxone will increase opioid use and promote a
“false safety” among drug users, and that it will reduce interaction with hospital emergency
departments.35
It is important that education programs within Allegheny County follow the passing of
this bill in order to communicate that there is currently no significant evidence that untrained
rescuers are any different than trained rescuers in regards to management of overdose events. In
addition to this, there is no evidence that opioid use increases after Naloxone administration.
Under the specifications of the bill, it is required that those who use Naloxone are transported
immediately to a medical facility for further observation.23,35 Lastly, the study of OEND
participants found that educating family and friends about overdose through Naloxone education
was viewed as having more value and was more acceptable than discussing how to reduce drug
use behaviors—which could potentially lead to a future decrease in drug use.36
Allegheny County is currently in a state of high demand for education and intervention
programs for drug abuse. Based on data from 2007 to 2011 on specific cause of death, the
biggest contributor to drug fatalities are from narcotic and hallucinogen use and abuse. Current
legislative amendments offer the potential for promising change within the state, but must be
followed with further monitoring and education programs. Likewise, future continuous
monitoring of drug overdose fatalities by public health institutions such as the Allegheny County
Health Department and the University of Pittsburgh will allow for targeted interventions for high
29
risk and emerging risk populations in order to reduce drug use and poisoning fatalities within
Allegheny County. This report provided a necessary first step in addressing the problem of drug
abuse in Allegheny County. Comparison to similar counties across the United States showed
that, despite its similarity to other urban areas, Allegheny County has a problem with drug use
that exceeds the nation and its peers. Combatting these trends in drug abuse in Allegheny County
needs to be a major public health priority.
30
APPENDIX: TECHNICAL NOTES
Definition of Key Terms
DRUG OVERDOSE: Ingestion or application of a drug or other substance in quantities that was
greater than recommended or prescribed. This event may result in a toxic state or death.
FATALITY: Unexpected death
CAUSE OF DEATH: The causal agents or underlying causes that result in death. Underlying
causes are any diseases, injuries, or accidents, which eventually lead to death. The International
Classification of Disease, Tenth Revision (ICD-10), codes causes of death or underlying causes.
See Table 2.
MEDIAN AGE AT DEATH: An indicator of premature mortality. Is the exact midpoint of ages
of deaths that occurred in a designated time period.
RACE: Bridged population estimates for Allegheny County are used to calculate the rates from
2007 and 2008. These estimates are used from bridging the 31 race categories used in the 2000
census as directed in the 1997 Office of Management and Budget standards for the collection of
data on race and ethnicity.
31
Drug Schedules37
According to the United States Drug Enforcement Administration:

Schedule I: Defined as drugs with no currently accepted medical use and a high
potential for abuse. This classification is used for the most dangerous drugs of all
the drug schedules with potentially severe psychological or physical dependence.
o Examples: heroin, LSD, marijuana, ecstasy, methaqualone, and peyote

Schedule II: Substances or chemicals that have a high potential for abuse, but less
abuse potential than Schedule I drugs. Use of Schedule II drugs can potentially
lead to severe psychological or physical dependence.
o Examples: cocaine, methamphetamine, methadone, Demerol, OxyContin,
fentanyl, Adderall, and Ritalin

Schedule III: Drugs with moderate to low potential for physical and psychological
dependence. Abuse potential is less than Schedule I and Schedule II drugs, but
more addictive than Schedule IV drugs.
o Examples: Vicodin, Tylenol with codeine, anabolic steroids, testosterone

Schedule IV: Substances or chemicals that have a low potential for abuse and low
risk of dependence
o Examples: Xanax, Soma, Darvon, Darvocet, Valium, Ativan, Talwin,
Ambien

Schedule V: Defined as substances with the lowest potential for abuse of the
Scheduled drugs and consist of preparations containing limited quantities of
certain narcotics.
32
o Examples: Robitussin AC, Lomotil, Motofen, Lyrica, Parepectolin
Formulas for Rates
AGE ADJUSTED MORTALITY RATE = Age-specific mortality rate multiplied by the weight
for that age group in the standard population

Age adjusting allows for control over any
differences that may exist in the underlying
age structures of different populations. Ageadjusted death rates cannot be compared to
crude rates and cannot be interpreted as an
absolute measure of mortality. All ageadjusted rates used in this essay are adjusted
using the direct method with the U.S 2000
Standard Million Population.
Allegheny County Census Tracts
For this report, an Allegheny County Census tract less than or equal to 40000 were considered
“City of Pittsburgh” and a tract greater than 4000 but less than 80000 were considered
“Remainder of the County”.
33
BIBLIOGRAPHY
1. National Drug Intelligence Center. (2001). United States-Canada Border Drug Threat
Assessment. Report No. 2002-R0423-001. Johnstown, PA.
2. Jones, C. M., Mack, K. A., & Paulozzi, L. J. (2013). Pharmaceutical overdose deaths, United
States, 2010. JAMA, 309(7), 657–9. doi:10.1001/jama.2013.272
3. Centers for Disease Control and Prevention (CDC). Primary care and public health initiative:
balancing pain management and prescription opioid abuse: educational module.
http://www.cdc.gov/primarycare/materials/opoidabuse/. Published October 24, 2012.
4. Addressing prescription drug abuse in the United States: current activities and future
opportunities. Atlanta: Centers for Disease Control and Prevention. (2014). Home and
Recreational Safety. Retrieved October 30, 2014, from
http://www.cdc.gov/homeandrecreationalsafety/overdose/facts.html
5. Centers for Disease Control and Prevention. Overdose deaths involving prescription opioids
among Medicaid enrollees - Washington, 2004-2007. (2009). MMWR. Morbidity and
Mortality Weekly Report, 58(42), 1171–5. Retrieved from
http://www.ncbi.nlm.nih.gov/pubmed/19875978
6. National Institute on Drug Abuse (NIDA). (2014). The Science of Drug Abuse & Addiction.
Retrieved September 20, 2014, from http://www.drugabuse.gov/
7. Substance Abuse and Mental Health Services Administration, Results from the 2013 National
Survey on Drug Use and Health: Summary of National Findings, NSDUH Series H-48,
HHS Publication No. (SMA) 14-4863. Rockville, MD: Substance Abuse and Mental Health
Services Administration, 2014.
8. Vital signs: overdoses of prescription opioid pain relievers and other drugs among women-United States, 1999-2010. (2013). MMWR. Morbidity and Mortality Weekly Report, 62(26),
537–42. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23820967
9. Centers for Disease Control and Prevention (CDC): National vital statistics reports: deaths:
final data for 2009. National Vital Statistics Report 2011, 60(3):81–82. 34
10. Hall, A. J., Logan, J. E., Toblin, R. L., Kaplan, J. A., Kraner, J. C., Bixler, D., Crosby, A.,
Paulozzi, L. J. (2008). Patterns of abuse among unintentional pharmaceutical overdose
fatalities. JAMA, 300(22), 2613–20. doi:10.1001/jama.2008.802
11. Paulozzi, L. J., & Xi, Y. (2008). Recent changes in drug poisoning mortality in the United
States by urban-rural status and by drug type. Pharmacoepidemiology and Drug Safety,
17(10), 997–1005. doi:10.1002/pds.1626
12. Federal Bureau of Investigation. (2012). Persons Arrested. Retrieved October 30, 2014, from
http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2012/crime-in-the-u.s.-2012/personsarrested/persons-arrested
13. U.S. Department of Justice. (1994). Fact Sheet: Drug-Related Crime. NCJ-149286. Retrieved
from http://www.bjs.gov/content/pub/pdf/DRRC.PDF
14. Substance Abuse and Mental Health Services Administration. Highlights of the 2009 Drug
Abuse Warning Network (DAWN) findings on drug- related emergency department visits. The
DAWN Report. Rockville, MD: US Department of Health and Human Services, Substance
Abuse and Mental Health Services Administration; 2010. Available at http://oas.
samhsa.gov/2k10/dawn034/edhighlights.htm 15. ICD-10 : International statistical classification of diseases and related health problems. Geneva:
World Health Organization, 2011. Print.
16. Van Zee, A. (2009). The promotion and marketing of OxyContin: commercial triumph,
public health tragedy. American Journal of Public Health, 99(2), 221–7.
doi:10.2105/AJPH.2007.131714
17. The use of opioids for the treatment of chronic pain. A consensus statement from the
American Academy of Pain Medicine and the American Pain Society. (1997). The
Clinical Journal of Pain, 13(1), 6–8. Retrieved from
http://www.ncbi.nlm.nih.gov/pubmed/9084947
18. Stanos SP, M. G. (2012). Appropriate Use of Opioids in Chronic Pain: Caring for Patients
and Reducing Risks. Englewood, CO.
19. Dowell, D., Kunins, H. V, & Farley, T. a. (2013). Opioid analgesics--risky drugs, not risky
patients. Jama, 309(21), 2219–20. doi:10.1001/jama.2013.5794
20. National Alliance for Model State Drug Laws. (2006b). Michigan’s Prescription Drug
Monitoring Program (PDMP). Retrieved from http://www.namsdl.org/library/2BBB4C5965BE-F4BB-A41D0E849476B7FF/
21. National Alliance for Model State Drug Laws. (2006a). Illinois’ Prescription Drug
Monitoring Program (PDMP). Retrieved from http://www.namsdl.org/library/2B9E7FA81372-636C-DD2E9959BB3C9BFE/
35
22. National Alliance for Model State Drug Laws. (2006c). Pennsylvania’s Prescription Drug
Monitoring Program (PDMP). Retrieved from http://www.namsdl.org/library/2C1AAA081372-636C-DD16A12271F62285/
23. Providing for prescription drug monitoring; creating the ABC-MAP Board; establishing the
Achieving Better Care by Monitoring All Prescriptions Program; and providing for
unlawful acts and penalties, PA S.1180, Act No. 191 (2014).
24. Eberhardt MS, Ingram DD, Makuc DM, et al. (2001). Urban and Rural Health Charbook,
Health, United States, 2001. Hyattsville, MD.
25. Rivara, F. P., Mueller, B. A., Somes, G., Mendoza, C. T., Rushforth, N. B., & Kellermann,
A. L. (1997). Alcohol and illicit drug abuse and the risk of violent death in the home.
JAMA, 278(7), 569–75. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9268278
26. Walsh, J. M., de Gier, J. J., Christopherson, A. S., & Verstraete, A. G. (2004). Drugs and
driving. Traffic Injury Prevention, 5(3), 241–53. doi:10.1080/15389580490465292
27. Breuer, B., Pappagallo, M., Tai, J. Y., & Portenoy, R. K. (2007). U.S. board-certified pain
physician practices: uniformity and census data of their locations. The Journal of Pain :
Official Journal of the American Pain Society, 8(3), 244–50.
doi:10.1016/j.jpain.2006.08.009
28. United States General Accounting Office. (2003). OxyContin Abuse and Diversion and
Efforts to Address the Problem. Report No. GAO-04-110. Washington, DC.
29. Glasser, J. H. (1981). The quality and utility of death certificate data. American Journal of
Public Health, 71(3), 231–3. Retrieved from
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1619790&tool=pmcentrez&ren
dertype=abstract
30. Schwartz, R. P., Gryczynski, J., O’Grady, K. E., Sharfstein, J. M., Warren, G., Olsen, Y.,
Mitchell S.G., Jaffe, J. H. (2013). Opioid agonist treatments and heroin overdose deaths in
Baltimore, Maryland, 1995-2009. American Journal of Public Health, 103(5), 917–22.
doi:10.2105/AJPH.2012.301049
31. Knudsen, H. K., Abraham, A. J., & Roman, P. M. (2011). Adoption and implementation of
medications in addiction treatment programs. Journal of Addiction Medicine, 5(1), 21–7.
doi:10.1097/ADM.0b013e3181d41ddb
32. Clark, R. E., & Baxter, J. D. (2013). Responses of state Medicaid programs to buprenorphine
diversion: doing more harm than good? JAMA Internal Medicine, 173(17), 1571–2.
doi:10.1001/jamainternmed.2013.9059
36
33. Community-based opioid overdose prevention programs providing naloxone - United States,
2010. (2012). MMWR. Morbidity and Mortality Weekly Report, 61(6), 101–5. Retrieved
from http://www.ncbi.nlm.nih.gov/pubmed/22337174
34. Coffin, P. O., & Sullivan, S. D. (2013). Cost-effectiveness of distributing naloxone to heroin
users for lay overdose reversal in Russian cities. Journal of Medical Economics, 16(8),
1051–60. doi:10.3111/13696998.2013.811080
35. Doe-Simkins, M., Quinn, E., Xuan, Z., Sorensen-Alawad, A., Hackman, H., Ozonoff, A., &
Walley, A. Y. (2014). Overdose rescues by trained and untrained participants and change in
opioid use among substance-using participants in overdose education and naloxone
distribution programs: a retrospective cohort study. BMC Public Health, 14(1), 297.
doi:10.1186/1471-2458-14-297
36. Sherman, S. G., Gann, D. S., Tobin, K. E., Latkin, C. A., Welsh, C., & Bielenson, P. (2009).
“The life they save may be mine”: diffusion of overdose prevention information from a city
sponsored programme. The International Journal on Drug Policy, 20(2), 137–42.
doi:10.1016/j.drugpo.2008.02.004
37. United States Drug Enforcement Administration. (n.d.). Drug Scheduling. Retrieved October
1, 2014, from http://www.dea.gov/druginfo/ds.shtml
37