Increases in Heroin Overdose Deaths — 28 States, 2010 to 2012

Rose A. Rudd, MSPH; Len J. Paulozzi, MD; Michael J. Bauer, MS; Richard W. Burleson, MBA, MPH; Rick E. Carlson, MPH, MA; Dan Dao, MPH; James W. Davis, MA; Jennifer Dudek, MPH; Beth Ann Eichler, MS; Jessie C. Fernandes, MPH; Anna Fondario, MPH; Barbara Gabella, MSPH; Beth Hume, MPH; Theron Huntamer; Mbabazi Kariisa, PhD; Thomas W. Largo, MPH; JoAnne Miles, MPH; Ashley Newmyer, MPH; Daniela Nitcheva, PhD; Beatriz E. Perez, MPH; Scott K. Proescholdbell, MPH; Jennifer C. Sabel, PhD; Jessica Skiba, MPH; Svetla Slavova, PhD; Kathy Stone, MBA; John M. Tharp, MPH; Tracy Wendling, DrPH; Dagan Wright, PhD; Anne M. Zehner, MPH

Disclosures

Morbidity and Mortality Weekly Report. 2014;63(39):849-854.

Abstract and Introduction

Introduction

Nationally, death rates from prescription opioid pain reliever (OPR) overdoses quadrupled during 1999–2010, whereas rates from heroin overdoses increased by <50%.* Individual states and cities have reported substantial increases in deaths from heroin overdose since 2010. CDC analyzed recent mortality data from 28 states to determine the scope of the heroin overdose death increase and to determine whether increases were associated with changes in OPR overdose death rates since 2010. This report summarizes the results of that analysis, which found that, from 2010 to 2012, the death rate from heroin overdose for the 28 states increased from 1.0 to 2.1 per 100,000, whereas the death rate from OPR overdose declined from 6.0 per 100,000 in 2010 to 5.6 per 100,000 in 2012. Heroin overdose death rates increased significantly for both sexes, all age groups, all census regions, and all racial/ethnic groups other than American Indians/Alaska Natives. OPR overdose mortality declined significantly among males, persons aged <45 years, persons in the South, and non-Hispanic whites. Five states had increases in the OPR death rate, seven states had decreases, and 16 states had no change. Of the 18 states with statistically reliable heroin overdose death rates (i.e., rates based on at least 20 deaths), 15 states reported increases. Decreases in OPR death rates were not associated with increases in heroin death rates. The findings indicate a need for intensified prevention efforts aimed at reducing overdose deaths from all types of opioids while recognizing the demographic differences between the heroin and OPR-using populations. Efforts to prevent expansion of the number of OPR users who might use heroin when it is available should continue.

In February, 2014, CDC invited state health departments to submit data from their mortality files for the period 2008–2012 if they judged those files to be substantially complete and if the causes of death had been coded by the International Classification of Diseases, 10th Revision. Participating states had the option of submitting resident deaths or deaths that occurred in the state. States submitted annual counts of deaths with an underlying cause of drug overdose of any intent (codes X40–X44, X60–X64, X85, Y10–Y14). They also submitted counts of subsets of the overdose deaths, those involving heroin (T40.1) and those involving OPR (T40.2–T40.4). States also provided the demographic distributions of these types of overdoses.

CDC calculated annual heroin and OPR death rates per 100,000 using bridged-race population estimates^† for each state and for the combined 28 participating states.^§ Because examination of state rates revealed pronounced increases in heroin death rates for most states in the study after 2010, CDC calculated changes in rates by demographic characteristics for the period of increasing rates only from 2010 to 2012. The correlation of change in state heroin overdose death rates with change in state OPR overdose death rates was examined both overall and for specific demographic subgroups. Statistical significance of changes in rates was tested using z-tests when rates were based on 100 or more deaths and examination of confidence intervals from gamma distributions when rates were based on fewer than 100 deaths. A weighted Pearson's correlation coefficient was used to examine the correlation between state level heroin and OPR death rate changes, with weights proportional to the state's 2012 population. Test results with p≤0.05 were considered statistically significant.

The death rate from heroin overdose doubled in the 28 states from 2010 to 2012, increasing from 1.0 to 2.1 per 100,000 population, reflecting an increase in the number of deaths from 1,779 to 3,635 ( Table ). Comparing the same years, the death rate from OPR overdose declined 6.6%, from 6.0 to 5.6 per 100,000, a decline from 10,427 to 9,869 deaths. The overall drug overdose death rate increased 4.3%, from 13.0 to 13.6. Heroin death rates increased after 2010 in every subgroup examined. Heroin death rates doubled for males and females, whereas OPR death rates declined 12.4% in males and were unchanged in females. Heroin death rates increased for all age groups, whereas OPR death rates declined for age groups <45 years. OPR death rates increased for persons aged 55–64 years. Heroin death rates doubled in non-Hispanic whites and Hispanic whites, and nearly doubled in blacks. OPR death rates decreased 8% in non-Hispanic whites and remained level in all other races/ethnicities. The Northeast and South had much larger heroin overdose death increases (211.2% and 180.9%, respectively), than the Midwest and West (62.1% and 90.7%, respectively). OPR death rates declined only in the South.

Comparing 2010 to 2012, trends in heroin and OPR overdose death rates varied widely by state. Of the 28 states, five states had increases in OPR death rates, seven states had decreases, and 16 states had no change in the OPR death rate. Of the 18 states with heroin overdose death rates based on at least 20 deaths, none had a decline (Figure 1). Increases in heroin overdose death rates were significantly associated with increases in OPR death rates (r = 0.47, p = 0.05). Similar patterns in the death rates for males and non-Hispanic whites, the two populations with the largest numbers of heroin deaths, also were observed, but the associations were not significant.

(Enlarge Image)

Figure 1.

Absolute change in heroin overdose death rates* compared with change in prescription opioid pain reliever (OPR) overdose death rates — 18 states, 2010 to 2012 ^†
* Rate change per 100,000 persons (r = 0.47, p = 0.05). Rates based on fewer than 20 deaths in a year are considered unstable and not shown. Marker is proportional in size to the 2012 population of the state it represents.
^† Arizona, Colorado, Florida, Illinois, Indiana, Kentucky, Massachusetts, Michigan, Missouri, Nevada, New Mexico, New York, North Carolina, Ohio, Oregon, Utah, Virginia, and Washington.

In 2012, the age group with the highest heroin overdose death rate was aged 25–34 years, and the age group with the highest OPR overdose death rate was aged 45–54 years. The racial/ethnic population with the highest death rate for both heroin and OPR was non-Hispanic whites (Figure 2). The death rate for heroin among males in 2012 was almost four times that among females, whereas the death rate for OPR among males was 1.4 times that among females.

(Enlarge Image)

Figure 2.

Death rates* from overdoses of heroin or prescription opioid pain relievers (OPRs), by age group and race/ethnicity — 28 states, 2012
* Crude (unadjusted) rate per 100,000 population. Based on bridged-race population estimates for 28 states, available at http://wonder.cdc.gov/bridged-race-v2012.html.

*Additional information available at http://wonder.cdc.gov/mcd.html.
^†Additional information available at http://wonder.cdc.gov/bridged-race-v2012.html.
^§Alabama, Arizona, Colorado, Florida, Illinois, Indiana, Iowa, Kansas, Kentucky, Massachusetts, Michigan, Minnesota, Missouri, Montana, Nebraska, Nevada, New Hampshire, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Virginia, Washington.

1 2

Next Section

Abstract and Introduction
Discussion
References
Sidebar

Table. Annual number of deaths and death rates* from overdoses of heroin or prescription opioid pain relievers (OPRs), by selected characteristics — 28 states, 2008–2012
Characteristic	Year					Change from 2010 to 2012^†
Characteristic	2008	2009	2010	2011	2012	Absolute rate change	% change
No. of drug overdose deaths ^§ overall	21,922	22,787	22,472	23,792	23,732	—	—
Heroin	1,786	2,058	1,779	2,679	3,635	—	—
OPR	9,480	10,303	10,427	10,393	9,869	—	—
Drug overdose death rates overall	12.9	13.3	13.0	13.7	13.6	0.6	4.3
Heroin^¶	1.0	1.2	1.0	1.5	2.1	1.0	101.7
OPR**	5.6	6.0	6.0	6.0	5.6	-0.4	-6.6
Sex
Male
Heroin	1.7	2.0	1.7	2.5	3.3	1.7	99.0
OPR	7.0	7.3	7.4	7.1	6.5	-0.9	-12.4
Female
Heroin	0.4	0.4	0.4	0.6	0.9	0.5	110.9
OPR	4.2	4.8	4.7	4.9	4.8	0.1	2.2
Age group (yrs)
15–24
Heroin	1.2	1.3	1.2	1.9	2.3	1.1	86.3
OPR	4.1	4.2	4.3	3.8	3.1	-1.2	-28.1
25–34
Heroin	2.2	2.7	2.4	3.7	5.1	2.7	109.1
OPR	8.6	9.2	9.8	9.5	8.4	-1.4	-14.5
35–44
Heroin	1.8	2.0	1.8	2.6	3.5	1.7	92.6
OPR	9.7	10.6	10.5	10.5	9.9	-0.6	-5.9
45–54
Heroin	1.8	2.1	1.5	2.2	3.2	1.8	119.6
OPR	12.0	12.5	12.2	12.3	11.9	-0.3	-2.5
55–64
Heroin	0.7	0.8	0.7	1.0	1.3	0.7	102.1
OPR	5.2	6.4	6.7	6.7	7.3	0.6	8.7
Race/Ethnicity ^††
White, non-Hispanic
Heroin	1.1	1.3	1.2	1.8	2.4	1.2	101.9
OPR	6.9	7.4	7.6	7.5	7.0	-0.6	-8.0
White, Hispanic
Heroin	1.0	1.0	0.7	1.0	1.4	0.7	102.6
OPR	2.8	2.5	2.6	2.6	2.5	0.0	-0.6
Black
Heroin	0.8	0.9	0.7	1.0	1.4	0.7	89.3
OPR	1.8	2.2	2.1	2.0	2.2	0.1	2.7
American Indian/Alaska Native
Heroin	0.9	1.0	0.9	1.2	1.4	0.6	63.9
OPR	6.2	7.1	6.0	6.2	6.2	0.3	4.5
U.S. Census region ^§§
Northeast
Heroin	1.0	1.2	0.9	1.8	2.7	1.9	211.2
OPR	4.1	4.3	4.3	4.8	4.6	0.3	7.5
Midwest
Heroin	1.3	1.5	1.6	2.0	2.6	1.0	62.1
OPR	3.7	4.2	4.3	4.2	4.1	-0.2	-4.7
West
Heroin	1.5	1.6	1.2	2.1	2.3	1.1	90.7
OPR	8.2	8.5	7.9	8.2	7.9	0.1	0.7
South
Heroin	0.6	0.7	0.4	0.6	1.0	0.7	180.9
OPR	6.9	7.6	7.9	7.2	6.6	-1.3	-16.3

*Crude rate per 100,000 population. Based on bridged-race population estimates for 28 states, available at http://wonder.cdc.gov/bridged-race-v2012.html. Because deaths might involve both heroin and OPRs, some deaths are included in both categories.
^†Change is in bold if statistically significant (p<0.05). Rate and percentage change might not match calculations based on table data because of rounding.
^§Deaths with underlying causes of unintentional drug poisoning (X40–X44), suicide drug poisoning (X60–X64), homicide drug poisoning (X85), or drug poisoning of undetermined intent (Y10–Y14), as coded in the International Classification of Diseases, 10th Revision.
^¶Drug overdose deaths, as defined, that had heroin (T40.1) as a contributing cause.
**Drug overdose deaths, as defined, that had other opioids (T40.2), methadone (T40.3), or other synthetic narcotics (T40.4) as contributing causes.
^††Persons of black and American Indian/Alaska Native race include Hispanic and non-Hispanic ethnicity. Persons of other races/ethnicities or with missing race information on the death certificate are not included.
^§§ Northeast: Massachusetts, New Hampshire, New York, Rhode Island. Midwest: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, Ohio. West: Arizona, Colorado, Montana, Nevada, New Mexico, Oregon, Utah, Washington. South: Alabama, Florida, Kentucky, North Carolina, Oklahoma, South Carolina, Virginia.