Abstract
This study used data from the National Vital Statistics System Mortality to evaluate trends in US drug overdose deaths involving opioids certified as unintentional, suicide, or undetermined intent.
Increasing rates of opioid overdose and suicide deaths have fueled interest in defining the extent to which these 2 urgent public health problems are syndemic. Although it has been assumed that approximately one-quarter of opioid overdose deaths are intentional,1 prior research has not evaluated the national distribution of opioid-related deaths by manner. Reevaluation of officially classified poisoning deaths of undetermined intent suggests that 21% may be misclassified suicides.2 We evaluated trends in US drug overdose deaths involving opioids certified as unintentional, suicide, or undetermined intent.
Methods
Drug overdose deaths involving opioids of persons ages 15 years and older were identified in the National Vital Statistics System mortality multiple cause-of-death data using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes X40-44 (unintentional), X60-64 (suicide), or Y10-Y14 (undetermined). Homicides were excluded. The sensitivity of death certificates for suicide is high. Undetermined overdose deaths tend to more closely resemble suicides than unintentional deaths. Among deaths with drug overdose as the underlying cause, those involving opioids were identified based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes T40.0 (opium), T40.1 (heroin), T40.2 (other opioids), T40.3 (methadone), T40.4 (other synthetic narcotics), and T40.6 (other and unspecified narcotics). Trends from 2000 to 2017 by manner of death were examined using negative binomial regression with population counts as an offset (Stata version 15 SE, StataCorp). Wald tests were used to compare groups and evaluate trends, with 2-sided P < .05 considered significant. Ethical approval was not required because deidentified publicly available data were used.
Results
In 2017, there were 47 506 total opioid deaths, excluding homicides: 43 036 unintentional deaths (90.6% [95% CI, 90.3%-90.9%]), 1884 suicides (4.0% [95% CI, 3.8%-4.1%]), and 2586 deaths of undetermined intent (5.4% [95% CI, 5.2%-5.7%]). Compared with decedents of undetermined intent, suicide decedents tended to be older (mean age, 51.7 vs 42.7 years), female (56.3% vs 34.9%), and white (89.1% vs 75.5%) (all P < .001).
Between 2000 and 2017, opioid-related deaths per 100 000 persons increased from 2.20 to 13.21 (trend: 9.19 [95% CI, 7.91-10.46]) for unintentional deaths and from 0.27 to 0.58 (trend: 0.30 [95% CI, 0.21-0.38]) for suicides (Table). There was no significant trend in deaths of undetermined intent (from 0.51 to 0.79 per 100 000 persons; trend: 0.02 [95% CI, −0.08 to 0.13]) (Figure). During this period, the percentage of opioid-related deaths that was unintentional increased from 73.8% to 90.6%, suicides decreased from 9.0% to 4.0%, and deaths of undetermined intent decreased from 17.2% to 5.4% (all P < .001). Significant increases in opioid-related suicide rates were observed for males, females, all racial/ethnic groups, and all age groups, except between the ages of 35 and 44 years (Table).
Table. Rates and Trends of Unintentional, Suicide, and Undetermined Intent Drug Overdose Deaths Involving Opioids in the United States per 100 000 Persons, 2000 and 2017a.
| Unintentional Deaths per 100 000 Persons | Suicide Deaths per 100 000 Persons | Undetermined Deaths per 100 000 Persons | |||||||
|---|---|---|---|---|---|---|---|---|---|
| 2000 | 2017 | Linear Trend (95% CI) | 2000 | 2017 | Linear Trend (95% CI) | 2000 | 2017 | Linear Trend (95% CI) | |
| Total No. of deaths | 6190 | 43 036 | 757 | 1884 | 1446 | 2586 | |||
| Total | 2.20 | 13.21 | 9.19 (7.91 to 10.46) | 0.27 | 0.58 | 0.30 (0.21 to 0.38) | 0.51 | 0.79 | 0.02 (−0.08 to 0.13) |
| P value | <.001 | <.001 | .64 | ||||||
| Sex | |||||||||
| Male | 3.44 | 18.56 | 11.92 (10.05 to 13.79) | 0.25 | 0.51 | 0.21 (0.12 to 0.30) | 0.75 | 1.05 | −0.04 (−0.18 to 0.10) |
| P value | <.001 | <.001 | .55 | ||||||
| Female | 1.01 | 8.02 | 6.62 (5.65 to 7.60) | 0.28 | 0.64 | 0.38 (0.27 to 0.48) | 0.28 | 0.55 | 0.10 (0.01 to 0.18) |
| P value | <.001 | <.001 | .03 | ||||||
| Age, yb | |||||||||
| 15-24 | 1.46 | 8.78 | 6.65 (5.52 to 7.77) | 0.11 | 0.22 | 0.05 (0.01 to 0.10) | 0.28 | 0.44 | −0.05 (−0.14 to 0.04) |
| P value | <.001 | .01 | .29 | ||||||
| 25-34 | 3.13 | 27.18 | 20.89 (18.00 to 23.79) | 0.24 | 0.43 | 0.10 (0.01 to 0.20) | 0.74 | 1.40 | 0.28 (0.10 to 0.47) |
| P value | <.001 | .03 | .002 | ||||||
| 35-44 | 5.34 | 25.27 | 14.40 (11.96 to 16.84) | 0.54 | 0.61 | 0.00 (−0.15 to 0.15) | 1.37 | 1.35 | −0.5 (−0.73 to 0.27) |
| P value | <.001 | .99 | <.001 | ||||||
| 45-54 | 4.25 | 21.57 | 14.58 (12.03 to 17.13) | 0.60 | 1.11 | 0.44 (0.22 to 0.67) | 0.93 | 1.37 | −0.06 (−0.30 to 0.18) |
| P value | <.001 | <.001 | .62 | ||||||
| 55-64 | 1.09 | 14.72 | 14.10 (12.23 to 15.97) | 0.36 | 1.15 | 1.03 (0.81 to 1.24) | 0.18 | 1.15 | 0.75 (0.60 to 0.91) |
| P value | <.001 | <.001 | <.001 | ||||||
| ≥65 | 0.21 | 2.40 | 1.98 (1.70 to 2.27) | 0.16 | 0.74 | 0.58 (0.47 to 0.70) | UR | 0.24 | 0.16 (0.12 to 0.20) |
| P value | <.001 | <.001 | <.001 | ||||||
| Race/ethnicityc | |||||||||
| White, non-Hispanic | 2.24 | 16.60 | 12.90 (10.30 to 15.50) | 0.35 | 0.83 | 0.48 (0.29 to 0.66) | 0.55 | 0.97 | 0.11 (−0.08 to 0.30) |
| P value | <.001 | <.001 | .25 | ||||||
| Black, non-Hispanic | 2.40 | 11.56 | 5.19 (4.07 to 6.32) | 0.09 | 0.15 | 0.07 (0.03 to 0.11) | 0.77 | 1.20 | 0.10 (−0.07 to 0.26) |
| P value | <.001 | .001 | .25 | ||||||
| Other, non-Hispanic | 0.53 | 3.00 | 2.32 (1.78 to 2.85) | UR | 0.17 | 0.09 (0.02 to 0.15) | UR | 0.12 | 0.00 (−0.05 to 0.06) |
| P value | <.001 | .009 | .91 | ||||||
| Hispanic | 2.19 | 6.35 | 2.78 (1.95 to 3.61) | 0.07 | 0.16 | 0.07 (0.02 to 0.11) | 0.20 | 0.15 | −0.11 (−0.16 to 0.06) |
| P value | <.001 | .004 | <.001 | ||||||
| Opioid typed | |||||||||
| Heroin | 0.60 | 4.53 | 3.54 (2.59 to 4.50) | 0.01 | 0.04 | 0.02 (0.01 to 0.03) | 0.04 | 0.18 | 0.16 (0.11 to 0.21) |
| P value | <.001 | <.001 | <.001 | ||||||
| Methadone | 0.28 | 0.87 | 0.44 (−0.07 to 0.96) | 0.03 | 0.03 | −0.02 (−0.05 to 0.01) | 0.05 | 0.08 | −0.05 (−0.11 to 0.01) |
| P value | .09 | .12 | .08 | ||||||
| Other synthetic | 0.17 | 8.05 | 4.04 (3.04 to 5.04) | 0.07 | 0.15 | 0.05 (0.01 to 0.10) | 0.04 | 0.52 | 0.24 (0.17 to 0.31) |
| P value | <.001 | .03 | <.001 | ||||||
| Other | 1.41 | 4.30 | 3.38 (1.90 to 4.86) | 0.18 | 0.44 | 0.29 (0.12 to 0.45) | 0.42 | 0.33 | −0.17 (−0.33 to 0.02) |
| P value | <.001 | .001 | .03 | ||||||
Abbreviation: UR, unreliable rate (based on <20 deaths).
Excludes assaultive overdoses. Linear trends (95% CIs) were estimated using predictive margins from negative binomial regressions with population counts as an offset; year was rescaled to vary from 0 to 1 so that values in the Table represent the absolute change in the rates from 2000 to 2017. Data are from the National Center for Health Statistics National Vital Statistics System.
Results not shown for persons younger than 15 years.
Race and ethnicity are reported on death certificates by an informant, typically the next of kin.
Drug overdose deaths involving opioids were identified based on International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes T40.0 (opium), T40.1 (heroin), T40.2 (other opioids), T40.3 (methadone), T40.4 (other synthetic narcotics), and T40.6 (other and unspecified narcotics).
Figure. Trends in Drug Overdose Deaths Involving Opioids in the United States per 100 000 Persons by Intent, 2000-2017.
Data are from the National Center for Health Statistics National Vital Statistics System and exclude assaultive overdose deaths. The linear trend is 9.19 for unintentional deaths (P < .001), 0.30 for suicide (P < .001), and 0.02 for undetermined intent deaths (P = .64).
Discussion
Between 2000 and 2017, the percentage of opioid-related deaths that were suicides declined from 9.0% to 4.0%. Despite this proportionate decline, the rate of opioid-related suicides as well as unintentional deaths significantly increased. Recent increases in overdose deaths involving heroin and nonmethadone synthetic opioids, including fentanyl,3 likely contributed to these trends.
Limitations include that proportionately fewer drug overdose deaths did not have information on the specific drug during the study period (decreasing from approximately 25% to 12%), related to increasing toxicological screening,4 which may have contributed to the increase in opioid-related deaths over time. Also, the validity of coroner or medical examiner determination of manner of death is uncertain5 and may have changed over time. In some cases, discerning the intent of overdose deaths may be difficult and an unknown proportion classified as undetermined intent may have been suicides. With greater public appreciation of the opioid epidemic since 2010, certifiers may be more likely to consider opioid overdose deaths as unintentional, thereby contributing to the proportionate declines in suicides and undetermined deaths.
While the public health crises of opioid overdose deaths and suicide do not appear closely linked, it is nevertheless possible that the national increase in suicides may be partly related to increasing opioid overdose deaths. A better understanding of the contribution of suicidal intent to opioid overdoses could inform suicide prevention efforts, especially considering the high risk of suicide following nonfatal opioid overdose.6
Section Editor: Jody W. Zylke, MD, Deputy Editor.
References
- 1.Oquendo MA, Volkow ND. Suicide: a silent contributor to opioid-overdose deaths. N Engl J Med. 2018;378(17):1567-1569. doi: 10.1056/NEJMp1801417 [DOI] [PubMed] [Google Scholar]
- 2.Donaldson AE, Larsen GY, Fullerton-Gleason L, Olson LM. Classifying undetermined poisoning deaths. Inj Prev. 2006;12(5):338-343. doi: 10.1136/ip.2005.011171 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Spencer MR, Warner M, Bastian BA, Trinidad JP, Hedegaard H. Drug Overdose Deaths Involving Fentanyl, 2011-2016: National Vital Statistics Reports, Vol 68, No. 3. Hyattsville, MD: National Center for Health Statistics; 2019. [PubMed] [Google Scholar]
- 4.Hedegaard H, Miniño AM, Warner M. Drug Overdose Deaths in the United States, 1999-2017: NCHS Data Brief, No. 329. Hyattsville, MD: National Center for Health Statistics; 2018. [Google Scholar]
- 5.Bohnert ASB, McCarthy JF, Ignacio RV, Ilgen MA, Eisenberg A, Blow FC. Misclassification of suicide deaths: examining the psychiatric history of overdose decedents. Inj Prev. 2013;19(5):326-330. doi: 10.1136/injuryprev-2012-040631 [DOI] [PubMed] [Google Scholar]
- 6.Olfson M, Crystal S, Wall M, Wang S, Liu SM, Blanco C. Causes of death following nonfatal opioid overdose. JAMA Psychiatry. 2018;75(8):820-827. doi: 10.1001/jamapsychiatry.2018.1471 [DOI] [PMC free article] [PubMed] [Google Scholar]
