Corrected US opioid-involved drug poisoning deaths and
mortality rates, 1999–2015
Christopher J. Ruhm
Frank Batten School of Leadership and Public Policy, University of Virginia, Charlottesville, VA, USA
Background and Aims
Most prior estimates of opioid-involved drug poisoning mortality counts or rates are understated
because the speciﬁc drugs leading to death are frequently not identiﬁed on death certiﬁcates. This analysis provides
corrected national estimates of opioid and heroin/synthetic opioid-involved counts and mortality rates, as well as changes
over time in them from 1999 to 2015.
Data on drug poisoning deaths to US residents from 1999 to 2015, ob-
tained from the Centers for Disease Control and Prevention (CDC) Multiple Cause of Death (MCOD) ﬁles, were used with the
drugs involved in fatal overdoses imputed when not identiﬁed on the death certiﬁcates.
The ofﬁcial CDC ﬁgure
that 33 091 drug deaths involved opioids in 2015 is an undercount, with the actual number being approximately
39 999. Corrected counts and rates of any opioid and heroin/synthetic opioid-involved drug deaths are 20–35% higher
in every year than reported ﬁgures. The corrections almost always raise the changes estimated to have occurred since
1999, with the largest differences observed in 2011 for any opioids (5677 deaths and 1.7 per 100 000) and in 2015
for heroin/synthetic opioids (3228 deaths and 1.0 per 100 000). However, percentage growth since 1999 is sometimes
slower when based on corrected rather than reported fatality data, and with sensitivity to the choice of base years.
Death certiﬁcate reports understate the prevalence of and changes over time in opioid and
heroin/synthetic opioid-involved drug mortality in the United States. Adjustments imputing the drugs involved for cases
where none are identiﬁed on the death certiﬁcates are likely to provide more accurate estimates.
Keywords Drug deaths, fatal overdoses, heroin, mortality rates, opioids, synthetic opioids.
Correspondence to: Christopher J. Ruhm, Frank Batten School of Leadership and Public Policy, University of Virginia, 235 McCormick Road, PO Box 400893,
Charlottesville, VA 22904-4893, USA. E-mail: email@example.com
Submitted 6 November 2017; initial review completed 27 November 2017; ﬁnal version accepted 20 December 2017
The number of Americans dying from drug poisonings
(excluding alcohol) increased from 16 849 in 1999 to
52 404 in 2015 . Drug overdoses have been the lead-
ing cause of injury deaths in the United States since
2009 . The rapid growth in drug mortality originally
involved prescription opioids such as OxyContin, al-
though often in combination with other drugs [3,4],
but with heroin and synthetic opioids (particularly fenta-
nyl) now playing a more important role . However, our
understanding of the nature of the fatal drug epidemic is
limited for several reasons, one of which is that the drugs
involved in these deaths are frequently unspeciﬁed on
death certiﬁcates. This analysis provides information on
how estimates of the numbers of fatalities and mortality
rates from drug deaths involving opioids of all types and
heroin or synthetic opioids change when accounting for
the under-reporting resulting from this lack of speciﬁcity.
Although this source of ascertainment bias has been
recognized previously [6,7] and some efforts have been
made to correct for it [8,9], prior analyses do not provide
year-speciﬁc estimates of the differences between reported
versus corrected numbers of deaths or mortality rates
and do not cover the most recent years of currently avail-
able data during which deaths involving fentanyl, and its
analogues have risen dramatically.
National quantity of opioid and heroin/synthetic opioid-
involved drug fatalities plus mortality rates per 100 000
population were analysed for each year between 1999
and 2015, examining differences between statistics based
only on information provided on the death certiﬁcates
compared to those correcting for incomplete reporting of
drug involvement. Counts of drug deaths to US residents
were obtained from the Centers for Disease Control and
Prevention Multiple Cause of Death (MCOD) ﬁles .
© 2018 Society for the Study of Addiction Addiction, 113,1339–1344