Suicides Among American Indian or Alaska Native Persons — National Violent Death Reporting System, United States, 2015–2020
Weekly / September 16, 2022 / 71(37);1161–1168
Deborah Stone, ScD1; Eva Trinh, PhD1; Hong Zhou, MPH1; Laura Welder, DrPH1; Pamela End of Horn, DSW2; Katherine Fowler, PhD3; Asha Ivey-Stephenson, PhD1 (View author affiliations)View suggested citation
What is already known about this topic?
Suicide is preventable. It disproportionately affects American Indian or Alaska Native (AI/AN) persons. Previous studies have examined suicide characteristics and circumstances among non-Hispanic AI/AN only in a limited number of states.
What is added by this report?
Comparison of 2015–2020 suicides among all AI/AN and non-AI/AN decedents in 49 states, Puerto Rico, and the District of Columbia found that AI/AN suicide decedents had higher adjusted odds of a range of relationship and alcohol or other substance use problems, and reduced odds of known mental health conditions and treatment than did non-AI/AN suicide decedents.
What are the implications for public health practice?
Culturally relevant comprehensive public health approaches to suicide prevention are needed to address systemic and long-standing inequities among AI/AN persons.
Views equals page views plus PDF downloads
Compared with the general U.S. population, American Indian or Alaska Native (AI/AN) persons, particularly those who are not Hispanic or Latino (Hispanic) AI/AN, are disproportionately affected by suicide; rates among this group consistently surpass those among all other racial and ethnic groups (1). Suicide rates among non-Hispanic AI/AN persons increased nearly 20% from 2015 (20.0 per 100,000) to 2020 (23.9), compared with a <1% increase among the overall U.S. population (13.3 and 13.5, respectively) (1). Understanding characteristics of suicide among AI/AN persons is critical to developing and implementing effective prevention strategies. A 2018 report described suicides in 18 states among non-Hispanic AI/AN persons only (2). The current study used 2015–2020 National Violent Death Reporting System (NVDRS) data among 49 states, Puerto Rico, and the District of Columbia to examine differences in suicide characteristics and contributing circumstances among Hispanic and non-Hispanic AI/AN populations, including multiracial AI/AN. Results indicated higher odds across a range of circumstances, including 10 of 14 relationship problems (adjusted odds ratio [aOR] range = 1.2–3.8; 95% CI range = 1.0–5.3) and six of seven substance use problems (aOR range = 1.2–2.3; 95% CI range = 1.1–2.5), compared with non-AI/AN persons. Conversely, AI/AN decedents had reduced odds of having any current known mental health condition, any history of mental health or substance use treatment, and other common risk factors (aOR range = 0.6–0.8; 95% CI = 0.2–0.9). Suicide is preventable. Communities can implement a comprehensive public health approach to suicide prevention that addresses long-standing inequities affecting AI/AN populations (3).
NVDRS is a state-based surveillance system that collects information from death certificates, coroner or medical examiner reports, and law enforcement reports on the characteristics and circumstances of violent deaths, including suicides (4). Data in this report are from the District of Columbia, Puerto Rico, and 49 U.S. states participating in NVDRS during 2015–2020*; some jurisdictions did not participate for the entire period because they were not yet funded or because they did not achieve data completion thresholds (Supplementary Table, https://stacks.cdc.gov/view/cdc/121071) (4). Analyses were limited to decedents aged ≥10 years, because determining suicide intent in young children can be difficult (5). AI/AN persons are defined in NVDRS as persons with origins among any of the original peoples of North America and who maintain cultural identification through tribal affiliation or community recognition (Alaska Natives are included among this group) (6). For this study, characteristics and circumstances of suicide were compared among decedents with any AI/AN identification, similar to a recent analysis of homicides among AI/AN persons (7). Rural-urban commuting area codes were used to determine nonmetropolitan and metropolitan geographic areas. All comparisons between AI/AN and non-AI/AN persons were examined using Pearson’s chi-square tests (with p<0.05 considered statistically significant) and logistic regression analyses, controlling for age and sex to estimate aORs with 95% CIs. Analyses were conducted using SAS (version 9.4; SAS Institute). This analysis was reviewed by CDC and was conducted consistent with applicable federal law and CDC policy.†
During 2015–2020, a total of 3,397 suicides among AI/AN persons and 179,850 suicides among non-AI/AN persons were recorded in NVDRS (Table 1). Approximately three quarters (74.6%) of AI/AN suicide decedents were aged ≤44 years, compared with less than one half (46.5%) of non-AI/AN decedents. The highest percentage of AI/AN suicides (46.9%) occurred among persons aged 25–44 years, whereas among non-AI/AN persons, the largest percentage (35%) occurred among persons aged 45–64 years. Nearly 45% of AI/AN suicide decedents (compared with 18.7% of non-AI/AN suicide decedents) lived in nonmetropolitan areas. AI/AN suicide decedents had higher odds of dying by hanging, strangulation, or suffocation (aOR = 1.8) and lower odds of dying from a firearm injury (aOR = 0.7) compared with non-AI/AN decedents. AI/AN suicide decedents also had higher odds of dying in a natural area (e.g., field; aOR = 1.4) or supervised facility (e.g., prison; aOR = 2.0) compared with non-AI/AN suicide decedents.
The circumstances of suicide were known for 86% of AI/AN and 89% of non-AI/AN decedents (Table 2). AI/AN decedents were more likely than were non-AI/AN decedents to disclose suicidal intent before death (aOR = 1.2) and to have had previous suicidal thoughts or plans (aOR = 1.1), but they were less likely to leave a note (aOR = 0.7). Nearly 55% of AI/AN suicide decedents experienced any relationship problems or losses before their death, compared with 42.2% of non-AI/AN decedents (aOR = 1.4). AI/AN decedents had increased odds of an additional nine of 14 relationship problems, including higher odds of intimate partner problems (aOR = 1.4), family relationship problems (aOR = 1.2), other relationship problems (aOR=1.4), interpersonal violence victimization (aOR = 2.7) and perpetration (aOR = 1.6) within the preceding month, suicide of a friend or family member (aOR = 1.6), and arguments or conflicts preceding death (aOR = 1.6). Conversely, AI/AN suicide decedents had decreased odds of physical health, job, and financial problems than did non-AI/AN decedents (aOR range = 0.6–0.8).
Approximately one third of AI/AN (31.8%) and non-AI/AN suicide decedents (29.7%) had experienced a crisis within the preceding 2 weeks or anticipated a crisis in the upcoming 2 weeks; AI/AN decedents had higher odds of having experienced crises involving intimate partners and recent suicide of friends or family members as well as crises involving criminal legal problems than did non-AI/AN decedents (aOR range = 1.2–3.8). In addition, AI/AN decedents had higher odds of six of seven alcohol or substance use problems including any current substance use problem (aOR = 2.0), a current alcohol (aOR = 2.3) or other substance use problem (aOR = 1.6), reported alcohol use hours before death (aOR = 1.9), and crises involving alcohol (aOR = 1.6). Among persons released from an institution within the month preceding death (196), 9.2% of AI/AN decedents had been in residential substance use treatment, compared with 5.5% of non-AI/AN decedents. The prevalences of known mental health diagnoses (41.5%; [aOR = 0.7]) and history of mental health or substance use treatment (29.5%; [aOR = 0.7]) were lower among AI/AN decedents than among non-AI/AN decedents (49.2% and 35.1%, respectively).
Toxicology testing was performed for 66.6% of AI/AN suicide decedents and 61.1% of non-AI/AN decedents (Table 3). Overall, AI/AN decedents had higher odds than did non-AI/AN decedents of receiving a positive test result for at least one substance (aOR = 1.2), blood alcohol concentration of ≥0.08 g/dL (aOR = 2.3), amphetamines (aOR = 1.5), and marijuana (aOR = 1.5). Conversely, AI/AN decedents had lower odds than did non-AI/AN decedents of receiving a positive test result for opioids (aOR = 0.5), benzodiazepines (aOR = 0.4), cocaine (aOR = 0.5), antidepressants (aOR = 0.6), antipsychotics (aOR = 0.7), and barbiturates (aOR = 0.3).
Analyses of characteristics of and circumstances preceding suicide among AI/AN and non-AI/AN persons in participating NVDRS jurisdictions during 2015–2020 identified many differences, including higher odds of relationship and substance use problems and lower odds of physical, job, and financial problems; known mental health conditions; and any history of mental health or substance use treatment among AI/AN decedents compared with non-AI/AN decedents. Although direct comparison of circumstances between studies is not possible, these findings suggest a similar pattern observed in a previous analysis of suicide in 18 states among non-Hispanic AI/AN persons compared with non-Hispanic White populations, during 2003–2014 (2). Those findings also indicated higher odds of relationship and alcohol problems and reduced odds of known mental health problems, current or past mental health or substance use treatment, and physical, job, or financial problems. Toxicology results from the earlier study also followed the same pattern as those observed in the current study, including higher odds of positive alcohol, amphetamine, and marijuana toxicology results among AI/AN decedents, and reduced odds of positive opioid and antidepressant test results, compared with non-AI/AN decedents.
The current study found higher odds of suicide among AI/AN persons across a range of relationship problems related to intimate partners, family, other relationships, interpersonal violence victimization and perpetration, and death of friends or family members by suicide. Similarly, more alcohol and other substance use circumstances, including those of an acute and more chronic nature, were observed in this study, as were criminal problems, although the nature of these problems was unknown. According to previous NVDRS reports, approximately one half of persons who die by suicide do not have a known mental health condition (4). This study found that only 41.5% of AI/AN suicide decedents had a known mental health condition. This might be the result of less available or accessible mental health services, especially in rural areas, and therefore fewer diagnoses. Post-hoc analyses controlling for metropolitan status did not change these results, suggesting possible contribution of other factors.
Suicide prevention efforts among AI/AN populations must consider the context and consequences of current inequities as well as historical trauma, including intergenerational transmission, that continue to affect AI/AN persons, families, and communities today (8). Suicide is a complex problem with multiple contributing circumstances that affect different communities differently. A comprehensive public health approach to suicide prevention (3), with attention to strategies that aim to reduce health inequities among AI/AN persons, is needed. These strategies might include strengthening access to and delivery of culturally relevant care, including telehealth for mental health concerns and well-being, increasing training and hiring of AI/AN providers, promoting community engagement and cultural traditions, increasing coping and problem-solving skills (e.g., American Indian Life Skills Training),§ increasing training to recognize and respond to suicide risk, making postvention programs (activities that reduce risk and promote healing after a suicide death) more available to AI/AN survivors of suicide loss (3), and promoting the 988 Suicide and Crisis Lifeline (persons who are thinking about suicide or who know someone who is thinking about suicide, should call 988).¶
The findings in this report are subject to at least four limitations. First, participation in NVDRS states increased during the analysis period; therefore, not all jurisdictions contributed data equally during all years. Second, deaths among AI/AN persons are prone to racial and ethnic misclassification, leading to potential underestimation of AI/AN suicides (9). However, the analysis included any decedent with noted AI/AN ancestry, including multiracial AI/AN, irrespective of Hispanic ethnicity, allowing for a more inclusive understanding of AI/AN suicide characteristics and circumstances. Third, NVDRS does not yet include tribal affiliation, and results might vary by tribe. Finally, circumstance data in NVDRS rely upon reporting by next-of-kin and other informants who knew the decedent, and their knowledge and willingness to share information about the decedent and circumstances preceding suicide. This might overestimate or underestimate this information.
Prevention of suicide is possible (3). Identification of new evidence-based programs, evaluation of existing AI/AN programs, and tailoring of other effective programs to prevent suicide among AI/AN persons is needed. Programs can benefit from holistic indigenous evaluation, which takes into consideration AI/AN cultural values and practices, such as storytelling (10). Addressing AI/AN-specific risk and promoting the many protective factors among AI/AN persons can save lives.
Carter Betz, Division of Violence Prevention, National Center for Injury Prevention and Control, CDC; Mark Stevens, Division of Injury Prevention, National Center for Injury Prevention and Control, CDC.
Corresponding author: Deborah Stone, email@example.com, 770-488-3942.
1Division of Injury Prevention, National Center for Injury Prevention and Control, CDC; 2Indian Health Service, Rockville, Maryland; 3Division of Violence Prevention, National Center for Injury Prevention and Control, CDC.
All authors have completed and submitted the International Committee of Medical Journal Editors form for disclosure of potential conflicts of interest. No potential conflicts of interest were disclosed.
* Florida is not included because pilot data were collected only during the study period.
† 45 C.F.R. part 46, 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.
- CDC. Injury prevention & control. WISQARS: Web-based Injury Statistics Query and Reporting System. Atlanta, GA: US Department of Health and Human Services, CDC; 2021. Accessed July 25, 2022. https://www.cdc.gov/injury/wisqars/index.html
- Leavitt RA, Ertl A, Sheats K, Petrosky E, Ivey-Stephenson A, Fowler KA. Suicides among American Indian/Alaska Natives—National Violent Death Reporting System, 18 states, 2003–2014. MMWR Morb Mortal Wkly Rep 2018;67:237–42. https://doi.org/10.15585/mmwr.mm6708a1 PMID:29494572
- Stone DM, Holland KM, Bartholow B, Crosby AE, Davis S, Wilkins N. Preventing suicide: a technical package of policies, programs, and practices. Atlanta, GA: US Department of Health and Human Services, CDC; 2017. https://www.cdc.gov/suicide/pdf/suicideTechnicalPackage.pdf
- Wilson RF, Liu G, Lyons BH, et al. Surveillance for violent deaths—National Violent Death Reporting System, 42 states, the District of Columbia, and Puerto Rico, 2019. MMWR Surveill Summ 2022;71(No. SS-6):1–40. https://doi.org/10.15585/mmwr.ss7106a1 PMID:35588398
- Crepeau-Hobson F. The psychological autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 2010;14:24–34. https://doi.org/10.1080/13811110903479011 PMID:20112141
- CDC. National Violent Death Reporting System web coding manual version 5.4.1. Atlanta, GA: US Department of Health and Human Services, CDC; 2021. https://www.cdc.gov/violenceprevention/pdf/nvdrs/nvdrsCodingManual.pdf
- Petrosky E, Mercer Kollar LM, Kearns MC, et al. Homicides of American Indians/Alaska Natives—National Violent Death Reporting System, United States, 2003–2018. MMWR Surveill Summ 2021;70(No. SS-8):1–19. https://doi.org/10.15585/mmwr.ss7008a1 PMID:34793415
- Evans-Campbell T. Historical trauma in American Indian/Native Alaska communities: a multilevel framework for exploring impacts on individuals, families, and communities. J Interpers Violence 2008;23:316–38. https://doi.org/10.1177/0886260507312290 PMID:18245571
- Arias E, Heron M, Hakes J; National Center for Health Statistics; US Census Bureau. The validity of race and Hispanic-origin reporting on death certificates in the United States: an update. Vital Health Stat 2 2016;172:1–21. PMID:28436642
- LaFrance J, Nichols R. Reframing evaluation: defining an indigenous evaluation framework. Renfrew, Canada: Canadian Evaluation Society, Canadian Journal of Program Evaluation; 2010. https://nnlm.gov/sites/default/files/2021-08/indigenous%20eval.pdf
Suggested citation for this article: Stone D, Trinh E, Zhou H, et al. Suicides Among American Indian or Alaska Native Persons — National Violent Death Reporting System, United States, 2015–2020. MMWR Morb Mortal Wkly Rep 2022;71:1161–1168. DOI: http://dx.doi.org/10.15585/mmwr.mm7137a1.
MMWR and Morbidity and Mortality Weekly Report are service marks of the U.S. Department of Health and Human Services.
Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services.
References to non-CDC sites on the Internet are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of pages found at these sites. URL addresses listed in MMWR were current as of the date of publication.
All HTML versions of MMWR articles are generated from final proofs through an automated process. This conversion might result in character translation or format errors in the HTML version. Users are referred to the electronic PDF version (https://www.cdc.gov/mmwr) and/or the original MMWR paper copy for printable versions of official text, figures, and tables.
Questions or messages regarding errors in formatting should be addressed to firstname.lastname@example.org.