Advertisement

Assessing Female Suicide From a Health Equity Viewpoint, U.S. 2004–2018

      Introduction

      Geographic and urbanization differences in female suicide trends across the U.S. necessitates suicide prevention efforts on the basis of geographic variations. The purpose of this study was to assess female suicide rates by mechanism within Census divisions and by urbanicity to help inform geographically tailored approaches for suicide prevention strategies.

      Methods

      Data from 2004 to 2018 were obtained from the National Vital Statistics System (analyzed in 2021). Annual counts of female suicides were tabulated for firearm, suffocation, and drug poisoning and stratified by the U.S. Census division and urbanicity. Age-adjusted rates were calculated to describe female suicide incidence by geographic areas and urbanicity. Data were analyzed annually and by 5-year timeframes. Trends in annual female suicide rates by mechanism for 3 urbanization levels were identified using Joinpoint Regression. Annual percent change estimates were calculated for age-adjusted female suicide rates between 2004 and 2018.

      Results

      Female suicide rates by mechanism were not homogeneous within Census divisions or by urbanization levels. Suicide rates by mechanism across Census divisions within the same urbanization level varied (range=3.38–11.15 [per 100,000 person per year]). From 2014 to 2018 in large metropolitan areas in the northern divisions, rates for suffocation were higher than for firearms and drug poisoning. During the same period, in all urbanization levels in southern divisions, rates for firearms were higher than for suffocation and drug poisoning.

      Conclusions

      Female suicide mechanisms vary by urbanization level, and this variation differs by region. These results could inform female suicide prevention strategies on the basis of mechanism, urbanization, and geographic region.

      INTRODUCTION

      In the U.S., suicide ranks in the top 10 causes of death for females aged 10‒54 years and can be prevented with strategies incorporating the best available evidence that addresses a range of factors influencing suicide.
      • Stone DM
      • Holland KM
      • Bartholow BN
      • Crosby AE
      • Davis SP
      • Wilkins N.
      Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
      ,

      Web-Based Injury Statistics Query and Reporting System (WISQARS). HHS, Centers for Disease Control and Prevention. https://www.cdc.gov/injury/wisqars/index.htm. Updated May 19, 2022. Accessed April 5, 2021.

      Suicide rates have increased by 33% since 1999, potentially related to factors such as the Great Recession and the opioid overdose epidemic.
      • Kegler SR
      • Stone DM
      • Holland KM.
      Trends in suicide by level of urbanization-United States, 1999-2015.
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      Despite these increases, suicide rates saw a decline for the first time in more than a decade in 2019, with rates declining for both males and females. However, disparities by sex exist. The male suicide rate is >3 times the rate for women in the U.S.
      • Hedegaard H
      • Curtin SC
      • Warner M.
      Increase in suicide mortality in the United States, 1999–2018.
      Nevertheless, the suicide rate for females increased by approximately 50% from 1999 to 2019, compared with 25% for males during the same time period.
      • Hedegaard H
      • Curtin SC
      • Warner M.
      Increase in suicide mortality in the United States, 1999–2018.
      ,
      • Stone DM
      • Jones CM
      • Mack KA.
      Changes in suicide rates-United states, 2018-2019.
      Because males account for a large majority of suicides in the U.S., studies that look at overall suicide trends tend to reflect what is happening among males, thus overlooking subtler nuances of suicide patterns among females. This study attempts to help address this gap. Describing suicide trends can identify communities and subpopulations at increased risk and aid professionals in tailoring prevention strategies to the needs of these groups.
      Literature highlights the increasing suicide burden rural communities experience in comparison with that of their metropolitan counterparts.
      • Kegler SR
      • Stone DM
      • Holland KM.
      Trends in suicide by level of urbanization-United States, 1999-2015.
      ,
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      ,
      • Hedegaard H
      • Curtin SC
      • Warner M.
      Increase in suicide mortality in the United States, 1999–2018.
      These previous findings provide a richer picture of suicide trends by urbanization level in the U.S. In addition, previous research highlights geographic differences in suicide trends by state
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      and county.
      • Rossen LM
      • Hedegaard H
      • Khan D
      • Warner M.
      County-level trends in suicide rates in the U.S., 2005-2015.
      Suicide rates increased significantly between 1999 and 2016 in 34 states for males and in 43 states for females.
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      The overall rate declined significantly in 5 states from 2018 to 2019.
      • Stone DM
      • Jones CM
      • Mack KA.
      Changes in suicide rates-United states, 2018-2019.
      In another report looking at overall suicides, from 2005 to 2015, rates were highest in western U.S. counties, whereas rates were lowest in eastern U.S. counties.
      • Rossen LM
      • Hedegaard H
      • Khan D
      • Warner M.
      County-level trends in suicide rates in the U.S., 2005-2015.
      These geographic differences in suicide trends show the potential value of tailoring suicide prevention efforts to specific geographic settings.
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      ,
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      Understanding how people die by suicide or the mechanism used for suicide can help to inform how public health professionals and partners shape suicide prevention strategies. Literature suggests that the mechanisms used for suicide differ by sex,
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      ,
      • Pettrone K
      • Curtin SC.
      Urban-rural differences in suicide rates, by sex and three leading methods: United States, 2000–2018.
      ,
      • Sullivan E
      • Annest J
      • Luo F
      Centers for Disease Control and Prevention (CDC)
      Suicide among adults aged 35–64 years–United states, 1999–2010.
      age,
      • Sullivan EM
      • Annest JL
      • Simon TR
      • Luo F
      • Dahlberg LL
      Centers for Disease Control and Prevention (CDC)
      Suicide trends among persons aged 10–24 years–United states, 1994–2012.
      ,
      • Baker SP
      • Hu G
      • Wilcox HC
      • Baker TD.
      Increase in suicide by hanging/suffocation in the U.S., 2000-2010.
      and urbanicity.
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      Suicide rates by firearm in rural counties were almost 2 times the rates in metropolitan counties between 2001 and 2015.
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      A better understanding of differences in the mechanism of suicide within and across geographic areas can further inform suicide prevention efforts.
      This study adds to the existing literature by identifying health disparities related to the intesection of multiple social and demographic factors. It is well documented that suicide rates differ by race/ethnicity
      • Martínez-Alés G
      • Jiang T
      • Keyes KM
      • Gradus JL
      The recent rise of suicide mortality in the United States.
      ,
      • Ramchand R
      • Gordon JA
      • Pearson JL.
      Trends in suicide rates by race and ethnicity in the United States.
      and age group,
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      illustrating the important roles these factors play in risk and protection for suicide. Health disparities are differences in health status between people related to social or demographic factors such as race/ethnicity, sex or geographic region.
      This study examines suicide rates by mechanism for females within Census divisions and by urbanicity to help inform tailored approaches for suicide prevention strategies. The long-term trend of increasing suicide rates among females warrants further attention to better identify the risk factors, warning signs, and effective prevention strategies for this group. Focusing on females might inform a more gender-sensitive approach to suicide prevention and improve the well-being and health of females. By understanding the mechanisms of suicide for females, levers for change, such as policy and programmatic and behavioral change efforts, can be identified to reduce risk factors and prevent suicide in this population.
      To the authors’ knowledge, no previous study has described the overlapping and intersectional impact of geographic location and urbanicity on female suicide rates by mechanism. Painting a nuanced picture of suicide trends by geography and urbanicity can help to identify female groups who might be at increased risk for suicide. Applying a health equity lens to epidemiologic data includes identifying the disparities experienced by populations that have been socially, economically, geographically, or environmentally disadvantaged (e.g., gender); focusing on subpopulations who are more at-risk for a given outcome (e.g., females living in rural communities); and highlighting intersecting identities of a population that may increase risks for an outcome (e.g., females living in rural communities in the South).
      • Braveman P
      • Gruskin S.
      Defining equity in health.
      Focusing data analyses on under-resourced communities and groups who experience disadvantage can help to shed light on areas where inequities might be mitigated. This study can help to answer questions about who is disproportionately impacted and inform approaches for suicide prevention efforts.

      METHODS

      Study Sample

      This study is an epidemiologic descriptive analysis using data covering the period 2004–2018 (2018 was the most recent year of data available at the time of analysis) obtained from the National Center for Health Statistics (NCHS) National Vital Statistics System (NVSS). NVSS is an intergovernmental data collection and sharing system for registration of vital life events, such as births and deaths. Deaths are recorded by all 50 states. These data do not involve research on human subjects. The data use agreement was approved by NCHS. IRB approval was not required for this study.

      Measures

      Annual counts of female suicides were tabulated by mechanism for county groupings defined in the 2013 NCHS 6-level urbanization classification scheme
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      ,
      • Ingram DD
      • Franco SJ.
      2013 NCHS urban-rural classification scheme for counties.
      and by U.S. Census Bureau divisions. The 6-level classification scheme for counties was collapsed into 3 levels for the current analysis: large metropolitan (including large central metropolitan and large fringe metropolitan counties), small/medium metropolitan (including medium metropolitan and small metropolitan counties), and nonmetropolitan (including micropolitan and noncore counties). The 9 U.S. Census Bureau divisions are defined as follows: New England (CT, ME, MA, NH, RI, VT), Middle Atlantic (NJ, NY, PA), East North Central (IL, IN, MI, OH, WI), West North Central (IA, KS, MN, MO, NE, ND, SD), South Atlantic (DE, DC, FL, GA, MD, NC, SC, VA, WV), East South Central (AL, KY, MS, TN), West South Central (AR, LA, OK, TX), Mountain (AZ, CO, ID, MT, NM, NV, UT, WY), and Pacific (AK, CA, HI, OR, WA).
      Suicides by any means were identified on the basis of underlying-cause-of-death codes U03, X60–84, and Y87.0 from the ICD-10. Specific suicide mechanisms considered in this study were firearms (codes X72‒X74), suffocation (includes hanging and strangulation; code X70), and drug poisoning (codes X60‒X64) because these are the 3 leading mechanisms of suicide in the U.S.

      Web-Based Injury Statistics Query and Reporting System (WISQARS). HHS, Centers for Disease Control and Prevention. https://www.cdc.gov/injury/wisqars/index.htm. Updated May 19, 2022. Accessed April 5, 2021.

      Suicides by other means were included in the overall suicide statistics but not included in the reported results by specific mechanism owing to very low rates. Decedents aged <10 years were excluded from this analysis because suicidal intent is difficult to ascertain in young children.
      • Kegler SR
      • Stone DM
      • Holland KM.
      Trends in suicide by level of urbanization-United States, 1999-2015.

      Statistical Analysis

      Age-adjusted rates per 100,000 persons per year were calculated across and within Census divisions and (collapsed) levels of urbanization. All rates represent only females and decedents aged ≥10 years. Accompanying 95% CIs were calculated under the assumption that suicide counts follow a Poisson distribution. Data were analyzed by individual year and by 3 successive 5-year timeframes: 2004–2008, 2009–2013, and 2014–2018. Rates and CIs were calculated using the SAS, version 9.4. Overall trends in annual female suicide rates by firearm, suffocation, and drug poisoning for the 3 urbanization levels were evaluated using Joinpoint Regression (version 4.7.0.0; National Cancer Institute, Bethesda, MD).

      Joinpoint Regression Program. Version 4.7.0.0; 2019. https://surveillance.cancer.gov/joinpoint/.

      Annual percentage change (APC) estimates were obtained from the joinpoint analyses to identify and quantify significant changes in female age-adjusted suicide rates during the period 2004–2018. Statistical significance was determined by testing whether each APC estimate was different from zero at the α=0.05 level. For this analysis, a 2-sided t-test was employed, which is the default for the Joinpoint Regression program.

      RESULTS

      Table 1 shows age-adjusted female suicide rate estimates across the 3 successive 5-year periods; by Census division (all mechanisms of suicide combined); by level of urbanization (all mechanisms); and by major specific mechanism, accompanied by 95% CIs characterizing the statistical stability of the estimates. Rates are presented in chronological order from 2004 to 2008, 2009–2013, and 2014–2018 unless otherwise specified and are per 100,000 persons per year. The female suicide rate for all mechanisms combined increased over time in every Census division. For all time periods, the rate was highest in the Mountain Census division (8.32; 9.88; 11.15) and lowest in the Middle Atlantic Census division (3.48; 4.34; 5.35), where the rate was less than half of the rate in the Mountain Census division for all 3 time periods. The suicide rate for all mechanisms combined also increased over time across all urbanization levels and was highest in nonmetropolitan areas (5.91; 6.95; 8.70) across all time periods. By mechanism, firearm (1.63; 1.87; 2.17) and suffocation (1.09; 1.45; 2.01) suicide rates both increased over time, whereas the drug poisoning (1.82; 1.95; 1.94) suicide rate remained at a comparative level across the 3 time periods.
      Table 1Age-Adjusted Female Suicide Rates by Time Period, Census Division, Urbanization Level, and Mechanism, 2004–2018
      VariablesAge-adjusted suicide rate
      Age-adjusted rates per 100,000 person-years were calculated using the direct method and the year 2000 U.S. standard population. Only residents aged ≥10 years were included in the data. Suicide rates include all causes of suicide (except when stratified by specific mechanism) identified by ICD-10 underlying-cause-of-death codes U03, X60–84, and Y87.0.
      (per 100,000 person-years) (95% CI)
      Census division
      The 9 U.S. Census Bureau divisions are defined as follows: New England (CT, ME, MA, NH, RI, VT), Middle Atlantic (NJ, NY, PA), East North Central (IL, IN, MI, OH, WI), West North Central (IA, KS, MN, MO, NE, ND, SD), South Atlantic (DE, DC, FL, GA, MD, NC, SC, VA, WV), East South Central (AL, KY, MS, TN), West South Central (AR, LA, OK, TX), Mountain (AZ, CO, ID, MT, NM, NV, UT, WY), and Pacific (AK, CA, HI, OR, WA).
      2004–20082009–20132014–2018
       Mountain8.32 (8.05, 8.59)9.88 (9.60, 10.17)11.15 (10.86, 11.45)
       West North Central5.33 (5.11, 5.55)6.23 (5.99, 6.47)7.80 (7.53, 8.06)
       East South Central5.91 (5.67, 6.16)6.61 (6.36, 6.86)7.45 (7.19, 7.72)
       West South Central5.38 (5.21, 5.55)5.86 (5.69, 6.03)7.24 (7.05, 7.42)
       South Atlantic5.81 (5.68, 5.94)6.27 (6.14, 6.40)7.14 (7.00, 7.29)
       East North Central4.85 (4.72, 4.99)5.58 (5.43, 5.72)6.52 (6.36, 6.68)
       Pacific5.57 (5.42, 5.71)6.00 (5.86, 6.15)6.46 (6.31, 6.60)
       New England4.22 (4.00, 4.45)5.02 (4.78, 5.27)6.07 (5.80, 6.34)
       Middle Atlantic3.48 (3.36, 3.60)4.34 (4.21, 4.48)5.35 (5.20, 5.51)
      Urbanization level
      The 2013 National Center for Health Statistics 6-level urban‒rural classification scheme for counties included large central metropolitan (part of a metropolitan statistical area with ≥1 million population and covers a principal city), large fringe metropolitan (part of a metropolitan statistical area with ≥1 million population but does not cover a principal city), medium metropolitan (part of a metropolitan statistical area with ≥250,000 but <1 million population), small metropolitan (part of a metropolitan statistical area with <250,000 population), micropolitan (nonmetropolitan) (part of a micropolitan statistical area [has an urban cluster of ≥10,000 but <50,000 population]), and noncore (nonmetropolitan) (not part of a metropolitan or micropolitan statistical area). The 6-level urban‒rural classification scheme was collapsed into 3 categories for this analysis. Urbanization was defined as large metropolitan (large central metropolitan and large fringe metropolitan), small/medium metropolitan (medium metropolitan and small metropolitan), and nonmetropolitan (micropolitan and noncore).
       Large metropolitan4.90 (4.82, 4.97)5.44 (5.36, 5.51)6.12 (6.04, 6.20)
       Small/Medium metropolitan5.85 (5.74, 5.96)6.74 (6.63, 6.86)8.05 (7.93, 8.18)
       Nonmetropolitan5.91 (5.75, 6.06)6.95 (6.78, 7.12)8.70 (8.51, 8.89)
      Mechanism
      Firearm suicides were identified by the ICD-10 underlying-cause-of-death codes X72‒X74, suffocation suicides were identified by the ICD-10 underlying-cause-of-death code X70, and drug poisoning suicides were identified by the ICD-10 underlying-cause-of-death codes X60‒X64. These 3 specific mechanisms represent the 3 leading causes of suicide but do not cover all suicides.
       Firearm1.63 (1.60, 1.66)1.87 (1.83, 1.90)2.17 (2.14, 2.21)
       Suffocation1.09 (1.06, 1.11)1.45 (1.42, 1.48)2.01 (1.98, 2.04)
       Drug poisoning1.82 (1.79, 1.85)1.95 (1.92, 1.98)1.94 (1.90, 1.97)
      AK, Alaska; AL, Alabama; AR, Arkansas; AZ, Arizona; CA, California; CO, Colorado; CT, Connecticut; DC, District of California; DE, Delaware; FL, Florida; GA, Georgia; HI, Hawaii; IA, Iowa; ID, Idaho; IL, Illinois; IN, Indiana; KS, Kansas; KY, Kentucky; LA, Louisiana; MA, Massachusetts; MD, Maryland; ME, Maine; MI, Michigan; MN, Minnesota; MO, Missouri; MS, Mississippi; MT, Montana; NC, North Carolina; ND, North Dakota; NE, Nebraska; NH, New Hampshire; NJ, New Jersey; NM, New Mexico; NV, Nevada; NY, New York; OH, Ohio; OK, Oklahoma; OR, Oregon; PA, Pennsylvania; RI, Rhode Island; SC, South Carolina; SD, South Dakota; TN, Tennessee; TX, Texas; UT, Utah; VA, Virginia; VT, Vermont; WA, Washington; WI, Wisconsin; WV, West Virginia; WY, Wyoming.
      a Age-adjusted rates per 100,000 person-years were calculated using the direct method and the year 2000 U.S. standard population. Only residents aged ≥10 years were included in the data. Suicide rates include all causes of suicide (except when stratified by specific mechanism) identified by ICD-10 underlying-cause-of-death codes U03, X60–84, and Y87.0.
      b The 9 U.S. Census Bureau divisions are defined as follows: New England (CT, ME, MA, NH, RI, VT), Middle Atlantic (NJ, NY, PA), East North Central (IL, IN, MI, OH, WI), West North Central (IA, KS, MN, MO, NE, ND, SD), South Atlantic (DE, DC, FL, GA, MD, NC, SC, VA, WV), East South Central (AL, KY, MS, TN), West South Central (AR, LA, OK, TX), Mountain (AZ, CO, ID, MT, NM, NV, UT, WY), and Pacific (AK, CA, HI, OR, WA).
      c The 2013 National Center for Health Statistics 6-level urban‒rural classification scheme for counties included large central metropolitan (part of a metropolitan statistical area with ≥1 million population and covers a principal city), large fringe metropolitan (part of a metropolitan statistical area with ≥1 million population but does not cover a principal city), medium metropolitan (part of a metropolitan statistical area with ≥250,000 but <1 million population), small metropolitan (part of a metropolitan statistical area with <250,000 population), micropolitan (nonmetropolitan) (part of a micropolitan statistical area [has an urban cluster of ≥10,000 but <50,000 population]), and noncore (nonmetropolitan) (not part of a metropolitan or micropolitan statistical area). The 6-level urban‒rural classification scheme was collapsed into 3 categories for this analysis. Urbanization was defined as large metropolitan (large central metropolitan and large fringe metropolitan), small/medium metropolitan (medium metropolitan and small metropolitan), and nonmetropolitan (micropolitan and noncore).
      d Firearm suicides were identified by the ICD-10 underlying-cause-of-death codes X72‒X74, suffocation suicides were identified by the ICD-10 underlying-cause-of-death code X70, and drug poisoning suicides were identified by the ICD-10 underlying-cause-of-death codes X60‒X64. These 3 specific mechanisms represent the 3 leading causes of suicide but do not cover all suicides.
      Assessing suicide mechanism by urbanization level, in large metropolitan areas, annual suffocation suicide rates significantly increased from 2004 to 2018 (APC=5.2; p<0.05), annual firearm suicide rates significantly increased from 2006 to 2018 (APC=2.6; p<0.05), and annual drug poisoning suicide rates significantly increased from 2004 to 2012 (APC=2.1; p<0.05) followed by a significant decrease from 2012 to 2018 (APC= −2.4; p<0.05) (Appendix Figure 1A, available online). In small/medium metropolitan areas, annual firearm and suffocation suicide rates significantly increased over the duration of the study period (APC=3.5; p<0.05 and APC=6.8; p<0.05, respectively), whereas annual drug poisoning suicide rates significantly increased from 2004 to 2015 (APC=1.6; p<0.05), followed by a nominal (nonsignificant) decrease from 2015 to 2018 (APC= −3.6) (Appendix Figure 1B, available online). In nonmetropolitan areas, annual suicide rates for all the 3 mechanisms increased over time, with suffocation rates showing the steepest relative increase over the duration of the study period (APC=8.9; p<0.05) (Appendix Figure 1C, available online).
      Figure 1
      Figure 1Age-adjusted female suicide ratesa by Census divisionsb in large metropolitansc by mechanism,d 2004–2018.
      aAge-adjusted rates per 100,000 person-years were calculated using the direct method and the 2000 U.S. standard population. Only residents aged ≥10 years were included in the data.
      bThe 9 U.S. Census Bureau divisions are defined as follows: New England (CT, ME, MA, NH, RI, VT), Middle Atlantic (NJ, NY, PA), East North Central (IL, IN, MI, OH, WI), West North Central (IA, KS, MN, MO, NE, ND, SD), South Atlantic (DE, DC, FL, GA, MD, NC, SC, VA, WV), East South Central (AL, KY, MS, TN), West South Central (AR, LA, OK, TX), Mountain (AZ, CO, ID, MT, NM, NV, UT, WY), and Pacific (AK, CA, HI, OR, WA).
      cThe 2013 National Center for Health Statistics 6-level urban‒rural classification scheme for counties included large central metropolitan (part of a metropolitan statistical area with ≥1 million population and covers a principal city), large fringe metropolitan (part of a metropolitan statistical area with ≥1 million population but does not cover a principal city), medium metropolitan (part of a metropolitan statistical area with ≥250,000 but <1 million population), small metropolitan (part of a metropolitan statistical area with <250,000 population), micropolitan (nonmetropolitan) (part of a micropolitan statistical area [has an urban cluster of ≥10,000 but <50,000 population]), and noncore (nonmetropolitan) (not part of a metropolitan or micropolitan statistical area). The 6-level urban‒rural classification scheme was collapsed into 3 categories for this analysis. Urbanization was defined as large metropolitan (large central metropolitan and large fringe metropolitan), small/medium metropolitan (medium metropolitan and small metropolitan), and nonmetropolitan (micropolitan and noncore).
      dSuffocation suicides were identified by the ICD-10 underlying-cause-of-death code X70, firearm suicides were identified by the ICD-10 underlying-cause-of-death codes X72‒X74, and drug poisoning suicides were identified by the ICD-10 underlying-cause-of-death codes X60‒X64.
      AK, Alaska; AL, Alabama; AR, Arkansas; AZ, Arizona; CA, California; CO, Colorado; CT, Connecticut; DC, District of California; DE, Delaware; FL, Florida; GA, Georgia; HI, Hawaii; IA, Iowa; ID, Idaho; IL, Illinois; IN, Indiana; KS, Kansas; KY, Kentucky; LA, Louisiana; MA, Massachusetts; MD, Maryland; ME, Maine; MI, Michigan; MN, Minnesota; MO, Missouri; MS, Mississippi; MT, Montana; NC, North Carolina; ND, North Dakota; NE, Nebraska; NH, New Hampshire; NJ, New Jersey; NM, New Mexico; NV, Nevada; NY, New York; OH, Ohio; OK, Oklahoma; OR, Oregon; PA, Pennsylvania; RI, Rhode Island; SC, South Carolina; SD, South Dakota; TN, Tennessee; TX, Texas; UT, Utah; VA, Virginia; VT, Vermont; WA, Washington; WI, Wisconsin; WV, West Virginia; WY, Wyoming.
      Assessing suicide mechanism by Census division, differences between divisions within the same urbanization level were evident (Figures 1, 2, and 3). For example, across large metropolitan areas, the suicide rate by mechanism varied by Census division. In the New England Census division, the suffocation suicide rate (2.32) was higher than the drug poisoning (1.70) and firearm (0.42) suicide rates during 2014–2018, whereas in the East South Central Census division, the firearm suicide rate was the highest across all the 3 time periods (2.09, 2.25, 2.57) (Figure 1). In the Mountain Census division, the drug poisoning and firearm suicide rates were high compared with the suffocation suicide rate across all the 3 time periods (drug poisoning: 3.22, 3.42, 3.22; firearm: 2.51, 2.68, 3.18) (Figure 1). The same types of patterns were evident across small/medium metropolitan areas and nonmetro areas, where suicide rates by mechanism varied by Census division (Figures 2 and 3 and Appendix Table 2, available online). For instance, across nonmetropolitan areas in the East South Central, Mountain, South Atlantic, and West South Central Census divisions, the firearm suicide rate was highest during all time periods, whereas different patterns of suicide rates by mechanism occurred in the other divisions across time (Figure 3 and Appendix Table 2, available online).
      Figure 2
      Figure 2Age-adjusted female suicide ratesa by Census divisionsb in small/medium metropolitansc by mechanism,d 2004–2018.
      aAge-adjusted rates per 100,000 person-years were calculated using the direct method and the 2000 U.S. standard population. Only residents aged ≥10 years were included in the data.
      bThe 9 U.S. Census Bureau divisions are defined as follows: New England (CT, ME, MA, NH, RI, VT), Middle Atlantic (NJ, NY, PA), East North Central (IL, IN, MI, OH, WI), West North Central (IA, KS, MN, MO, NE, ND, SD), South Atlantic (DE, DC, FL, GA, MD, NC, SC, VA, WV), East South Central (AL, KY, MS, TN), West South Central (AR, LA, OK, TX), Mountain (AZ, CO, ID, MT, NM, NV, UT, WY), and Pacific (AK, CA, HI, OR, WA).
      cThe 2013 National Center for Health Statistics 6-level urban‒rural classification scheme for counties included large central metropolitan (part of a metropolitan statistical area with ≥1 million population and covers a principal city), large fringe metropolitan (part of a metropolitan statistical area with ≥1 million population but does not cover a principal city), medium metropolitan (part of a metropolitan statistical area with ≥250,000 but <1 million population), small metropolitan (part of a metropolitan statistical area with <250,000 population), micropolitan (nonmetropolitan) (part of a micropolitan statistical area has an urban cluster of ≥10,000 but <50,000 population), and noncore (nonmetropolitan) (not part of a metropolitan or micropolitan statistical area). The 6-level urban‒rural classification scheme was collapsed into 3 categories for this analysis. Urbanization was defined as large metropolitan (large central metropolitan and large fringe metropolitan), small/medium metropolitan (medium metropolitan and small metropolitan), and nonmetropolitan (micropolitan and noncore).
      dSuffocation suicides were identified by the ICD-10 underlying-cause-of-death code X70, firearm suicides were identified by the ICD-10 underlying-cause-of-death codes X72‒X74, and drug poisoning suicides were identified by the ICD-10 underlying-cause-of-death codes X60‒X64.
      AK, Alaska; AL, Alabama; AR, Arkansas; AZ, Arizona; CA, California; CO, Colorado; CT, Connecticut; DC, District of California; DE, Delaware; FL, Florida; GA, Georgia; HI, Hawaii; IA, Iowa; ID, Idaho; IL, Illinois; IN, Indiana; KS, Kansas; KY, Kentucky; LA, Louisiana; MA, Massachusetts; MD, Maryland; ME, Maine; MI, Michigan; MN, Minnesota; MO, Missouri; MS, Mississippi; MT, Montana; NC, North Carolina; ND, North Dakota; NE, Nebraska; NH, New Hampshire; NJ, New Jersey; NM, New Mexico; NV, Nevada; NY, New York; OH, Ohio; OK, Oklahoma; OR, Oregon; PA, Pennsylvania; RI, Rhode Island; SC, South Carolina; SD, South Dakota; TN, Tennessee; TX, Texas; UT, Utah; VA, Virginia; VT, Vermont; WA, Washington; WI, Wisconsin; WV, West Virginia; WY, Wyoming.
      Figure 3
      Figure 3Age-adjusted female suicide ratesa by Census divisionsb in nonmetropolitansc by mechanism,d 2004–2018.
      aAge-adjusted rates per 100,000 person-years were calculated using the direct method and the 2000 U.S. standard population. Only residents aged ≥10 years were included in the data.
      bThe 9 U.S. Census Bureau divisions are defined as follows: New England (CT, ME, MA, NH, RI, VT), Middle Atlantic (NJ, NY, PA), East North Central (IL, IN, MI, OH, WI), West North Central (IA, KS, MN, MO, NE, ND, SD), South Atlantic (DE, DC, FL, GA, MD, NC, SC, VA, WV), East South Central (AL, KY, MS, TN), West South Central (AR, LA, OK, TX), Mountain (AZ, CO, ID, MT, NM, NV, UT, WY), and Pacific (AK, CA, HI, OR, WA).
      cThe 2013 National Center for Health Statistics 6-level urban‒rural classification scheme for counties included large central metropolitan (part of a metropolitan statistical area with ≥1 million population and covers a principal city), large fringe metropolitan (part of a metropolitan statistical area with ≥1 million population but does not cover a principal city), medium metropolitan (part of a metropolitan statistical area with ≥250,000 but <1 million population), small metropolitan (part of a metropolitan statistical area with <250,000 population), micropolitan (nonmetropolitan) (part of a micropolitan statistical area [has an urban cluster of ≥10,000 but <50,000 population]), and noncore (nonmetropolitan) (not part of a metropolitan or micropolitan statistical area). The 6-level urban‒rural classification scheme was collapsed into 3 categories for this analysis. Urbanization was defined as large metropolitan (large central metropolitan and large fringe metropolitan), small/medium metropolitan (medium metropolitan and small metropolitan), and nonmetropolitan (micropolitan and noncore).
      dSuffocation suicides were identified by the ICD-10 underlying-cause-of-death code X70, firearm suicides were identified by the ICD-10 underlying-cause-of-death codes X72‒X74, and drug poisoning suicides were identified by the ICD-10 underlying-cause-of-death codes X60‒X64.
      AK, Alaska; AL, Alabama; AR, Arkansas; AZ, Arizona; CA, California; CO, Colorado; CT, Connecticut; DC, District of California; DE, Delaware; FL, Florida; GA, Georgia; HI, Hawaii; IA, Iowa; ID, Idaho; IL, Illinois; IN, Indiana; KS, Kansas; KY, Kentucky; LA, Louisiana; MA, Massachusetts; MD, Maryland; ME, Maine; MI, Michigan; MN, Minnesota; MO, Missouri; MS, Mississippi; MT, Montana; NC, North Carolina; ND, North Dakota; NE, Nebraska; NH, New Hampshire; NJ, New Jersey; NM, New Mexico; NV, Nevada; NY, New York; OH, Ohio; OK, Oklahoma; OR, Oregon; PA, Pennsylvania; RI, Rhode Island; SC, South Carolina; SD, South Dakota; TN, Tennessee; TX, Texas; UT, Utah; VA, Virginia; VT, Vermont; WA, Washington; WI, Wisconsin; WV, West Virginia; WY, Wyoming.
      Focusing on the most recent 5-year period, 2014–2018, for most northern Census divisions (i.e., New England, Middle Atlantic, East North Central), in large metropolitan areas, female suicide rates by suffocation were highest than the rates of other mechanisms (Appendix Table 1, available online), whereas in the same period for southern Census divisions (i.e., South Atlantic, East South Central, West South Central), across all urbanization levels, firearm suicide rates were highest than the rates for other mechanisms (Appendix Table 1, available online). Rates for specific mechanisms varied by urbanization level for both the Pacific and Mountain Census divisions, with the firearm suicide rate in nonmetropolitan areas of the Mountain division (5.38) the highest across all divisions and urbanization levels (Appendix Table 1, available online).

      DISCUSSION

      Findings from this study show that female suicide rates by mechanism are not homogeneous within Census divisions or urbanization levels across the U.S. The intersectional lens applied in this study helps to explain how geographic differences impact suicide rates among females. Results showed diverse trends for suicide rates by mechanism across Census divisions within the same urbanization level. Previous literature suggests that a firearm is the most common mechanism for suicide across all urbanization levels in comparison with suffocation and drug poisoning.
      • Ivey-Stephenson AZ
      • Crosby AE
      • Jack SPD
      • Haileyesus T
      • Kresnow-Sedacca MJ.
      Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
      However, previous findings also show suffocation rates increasing over the period 2001–20155 and becoming the leading method of suicide for females in urban areas by 2018.
      • Sullivan E
      • Annest J
      • Luo F
      Centers for Disease Control and Prevention (CDC)
      Suicide among adults aged 35–64 years–United states, 1999–2010.
      The findings of this study identify the geographic areas, namely the New England, Middle Atlantic, and East North Central Census divisions, showing increasing rates of suffocation seen mostly within large metropolitan areas. Authors also identified large metropolitan areas within certain Census divisions where suicide rates by suffocation were not the highest, such as East South Central and West South Central. In these areas, rates of suicide by firearm were higher than rates by drug poisoning and suffocation. Similarly, in nonmetropolitan areas in the East South Central, Mountain, South Atlantic, and West South Central Census divisions, firearm suicide rates were highest across all time periods. Furthermore, these findings can inform intersectional suicide prevention strategies for areas with higher female suicide rates by certain mechanisms.
      Underlying community-level risk factors for suicide, such as social isolation or barriers to health care, might help to explain the differences in suicide rates across geographic areas.
      • Stone DM
      • Holland KM
      • Bartholow BN
      • Crosby AE
      • Davis SP
      • Wilkins N.
      Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
      These risk factors might reflect social, cultural, and regional differences.
      World Health Organization
      Preventing suicide: a global imperative.
      Community-level strategies to address various risk factors can help to prevent suicides.
      • Stone DM
      • Holland KM
      • Bartholow BN
      • Crosby AE
      • Davis SP
      • Wilkins N.
      Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
      For instance, strengthening economic support by improving household financial security, such as through state unemployment benefits, might help to prevent suicide.
      • Cylus J
      • Glymour MM
      • Avendano M.
      Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968–2008.
      In addition, strengthening access to and delivery of suicide care by ensuring coverage of mental health conditions in health insurance policies,
      • Cunningham PJ.
      Beyond parity: primary care physicians’ perspectives on access to mental health care.
      creating protective environments by reducing access to lethal means among persons at-risk for suicide,
      • Yip PS
      • Caine E
      • Yousuf S
      • Chang SS
      • Wu KC
      • Chen YY.
      Means restriction for suicide prevention.
      ,
      • Cosgrave EM
      • Robinson J
      • Godfrey KA
      • et al.
      Outcome of suicidal ideation and behavior in a young, help-seeking population over a 2-year period.
      and promoting connectedness through community engagement
      Centers for Disease Control and Prevention
      Strategic direction for the prevention of suicidal beahvior: Promoting individual, family, and community connectedness to prevent suicidal behavior: strategic direction for the prevention of suicidal behavior.
      can aid in community-wide suicide prevention.
      • Stone DM
      • Holland KM
      • Bartholow BN
      • Crosby AE
      • Davis SP
      • Wilkins N.
      Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
      Although these strategies are beneficial for all individuals, tailoring strategies by applying female-focused, culturally, and socially appropriate approaches might help to reach those who are at higher risk for suicide, especially in areas with high female suicide rates. Findings by urbanization level show that female suicide rates by suffocation had the steepest relative increase over time across all urbanization levels than firearm and drug poisoning suicide rates. Increased access to information on how to die by suicide available through the Internet or social media might have contributed to the increase in suicide rates by suffocation during the study period.
      • Yau RK
      • Paschall MJ.
      Epidemiology of asphyxiation suicides in the United States, 2005–2014.
      Similarly, a positive association has been observed between overall suicide rates and Internet use.
      • Luxton DD
      • June JD
      • Fairall JM.
      Social media and suicide: a public health perspective.
      Nevertheless, research suggests that there are multiple reasons for the choice of method, especially accessibility, familiarity, and social acceptability, and choice of method is often influenced by cultural, religious, and social factors.
      • Martínez-Alés G
      • Jiang T
      • Keyes KM
      • Gradus JL
      The recent rise of suicide mortality in the United States.
      ,
      • Goldney RD.
      A note on the reliability and validity of suicide statistics.
      ,
      • Thomas KH
      • Beech E
      • Gunnell D.
      Changes in commonly used methods of suicide in England and Wales from 1901–1907 to 2001–2007.
      Future research might be warranted to understand the changes in mechanisms over time, including changes in access.
      Female suicide rates by drug poisoning exhibited a relatively modest but statistically significant decrease during 2012–2018 in large metropolitan areas, with the 2018 rate comparable with the 2004 rate. Determining the possible reasons for the decline requires further investigation. It should be noted that annual rates of drug overdose death among females, involving all intents, have increased.

      Web-Based Injury Statistics Query and Reporting System (WISQARS). HHS, Centers for Disease Control and Prevention. https://www.cdc.gov/injury/wisqars/index.htm. Updated May 19, 2022. Accessed April 5, 2021.

      Because classifying a death as a suicide is not a straightforward process, some suicides by drug overdose might be misclassified as deaths of undetermined intent, leading to undercounting.
      • Liu D
      • Yu M
      • Duncan J
      • Fondario A
      • Kharrazi H
      • Nestadt PS.
      Discovering the unclassified suicide cases among undetermined drug overdose deaths using machine learning techniques.
      From a gender-sensitive perspective, gender roles, such as females historically assuming caretaker roles, in conjunction with changing social and economic conditions might contribute to increased risk for suicide in females.
      • Canetto SS.
      Meanings of gender and suicidal behavior during adolescence.
      ,
      • Canetto SS.
      Women and suicidal behavior: a cultural analysis.
      Risk factors such as feeling lonely or less socially connected, substance use, and mental health problems among other factors might have contributed to changes in female suicide rates across different geographic areas and urbanization levels in the U.S.
      • McGirr A
      • Séguin M
      • Renaud J
      • Benkelfat C
      • Alda M
      • Turecki G
      Gender and risk factors for suicide: evidence for heterogeneity in predisposing mechanisms in a psychological autopsy study.
      ,

      Risk and protective factors. Centers for Disease Control and Prevention. https://www.cdc.gov/suicide/factors/index.html. May 13, 2021. Accessed June 15, 2021.

      Promoting connectedness through community engagement
      Centers for Disease Control and Prevention
      Strategic direction for the prevention of suicidal beahvior: Promoting individual, family, and community connectedness to prevent suicidal behavior: strategic direction for the prevention of suicidal behavior.
      can aid in female suicide prevention, by specifically addressing risk factors.
      This study found that female suicides predominantly involve firearms, suffocation, or drug poisoning. The study findings underscore the need for widespread adoption of a comprehensive public health approach to suicide prevention that helps people from becoming suicidal and lessens the immediate and long-term harms of people at increased risk.
      • Stone DM
      • Holland KM
      • Bartholow BN
      • Crosby AE
      • Davis SP
      • Wilkins N.
      Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
      A comprehensive approach involves multisectoral partnerships (e.g., public health, mental health, social services, schools, families), use of data for decision making, and implementation of strategies with the best available evidence to address the range of factors influencing suicide.
      • Stone DM
      • Holland KM
      • Bartholow BN
      • Crosby AE
      • Davis SP
      • Wilkins N.
      Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
      ,
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      ,

      Comprehensive suicide prevention. Centers for Disease Control and Prevention.https://www.cdc.gov/suicide/programs/csp/index.html. Updated August 19, 2021. Accessed May 15, 2022.

      This study focused solely on suicide rates and did not consider suicide attempts; however, historically, the rate of suicide attempts is higher for females than for males.
      • Ivey-Stephenson AZ
      • Demissie Z
      • Crosby AE
      • et al.
      Suicidal ideation and behaviors among high school students-youth risk behavior survey, United States, 2019.
      This finding could have different implications for prevention strategies because they relate to females.
      Centers for Disease Control and Prevention (CDC)
      Nonfatal self-inflicted injuries treated in hospital emergency departments–United States, 2000.
      Although this trend might be changing,
      • Bornheimer LA
      • Wang K
      • Zhang A
      • et al.
      National trends in non-fatal suicidal behaviors among adults in the USA from 2009 to 2017.
      additional research could be useful in assessing whether the rate of suicide attempts continues to remain higher among females. Public health professionals and other partners can use the study results to tailor prevention strategies, including counseling about access to lethal means, safe storage, and upstream strategies to prevent people from becoming suicidal.
      Understanding localized differences in suicide mechanisms might support suicide prevention efforts through means restriction strategies.
      • Sarchiapone M
      • Mandelli L
      • Iosue M
      • Andrisano C
      • Roy A.
      Controlling access to suicide means.
      Applying an intersectional lens to utilize nuanced data might help to improve tailored programs, policies, and resources for suicide prevention and to highlight the disparities and levers of change to advance health equity.

      Limitations

      The study findings are subject to at least 3 limitations. First, this analysis did not assess race/ethnicity or age because the sample would be too small for meaningful analysis given the level of stratification by other factors. Previous research has indicated that there are substantial differences in suicide rates among different race/ethnicity and age groups.
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      ,
      • Ramchand R
      • Gordon JA
      • Pearson JL.
      Trends in suicide rates by race and ethnicity in the United States.
      Second, the mechanism of death is dependent on classification by the coroner/medical examiner, and there may be variations in how this is reported.
      • Breiding MJ
      • Wiersema B.
      Variability of undetermined manner of death classification in the US.
      Third, this analysis could reflect an underreporting of suicides because suicides might be undercounted in the NVSS data.
      • Breiding MJ
      • Wiersema B.
      Variability of undetermined manner of death classification in the US.
      However, more recent evidence suggests that the number of deaths classified as being of undetermined intent has decreased in recent years.
      • Stone DM
      • Simon TR
      • Fowler KA
      • et al.
      Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
      ,
      • Baker SP
      • Hu G
      • Wilcox HC
      • Baker TD.
      Increase in suicide by hanging/suffocation in the U.S., 2000-2010.

      CONCLUSIONS

      The findings of this study highlight that female suicide mechanisms vary by urbanization level and that such variation differs across Census divisions of the U.S. These results could inform suicide prevention strategies for females on the basis of the intersecting factors of mechanism, urbanization, and geographic region.

      ACKNOWLEDGMENTS

      The National Suicide Prevention Lifeline provides 24/7 free and confidential support for people in distress, prevention and crisis resources for you or your loved one, and best practices for professionals: 1-800-273-8255.
      The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
      No financial disclosures were reported by the authors of this paper.

      CRediT AUTHOR STATEMENT

      Avital R. Wulz: Conceptualization, Formal analysis, Project administration, Software, Visualization, Writing - original draft. Gabrielle F. Miller: Conceptualization, Data curation, Methodology, Software, Validation, Writing - review and editing. Scott R. Kegler: Data curation, Software, Validation, Writing - review and editing. Ellen E. Yard: Writing - review and editing. Amy F. Wolkin: Conceptualization, Supervision, Writing - review and editing.

      Appendix. SUPPLEMENTAL MATERIAL

      REFERENCES

        • Stone DM
        • Holland KM
        • Bartholow BN
        • Crosby AE
        • Davis SP
        • Wilkins N.
        Preventing suicide: a technical package of policies, programs, and practice. National Center for Injury Prevention and Control, Division of Violence Prevention.
        Centers for Disease Control and Prevention. 2017; (Published)https://doi.org/10.15620/cdc.44275
      1. Web-Based Injury Statistics Query and Reporting System (WISQARS). HHS, Centers for Disease Control and Prevention. https://www.cdc.gov/injury/wisqars/index.htm. Updated May 19, 2022. Accessed April 5, 2021.

        • Kegler SR
        • Stone DM
        • Holland KM.
        Trends in suicide by level of urbanization-United States, 1999-2015.
        MMWR Morb Mortal Wkly Rep. 2017; 66: 270-273https://doi.org/10.15585/mmwr.mm6610a2
        • Stone DM
        • Simon TR
        • Fowler KA
        • et al.
        Vital signs: trends in state suicide rates-United states, 1999-2016 and circumstances contributing to suicide-27 states, 2015.
        MMWR Morb Mortal Wkly Rep. 2018; 67: 617-624https://doi.org/10.15585/mmwr.mm6722a1
        • Ivey-Stephenson AZ
        • Crosby AE
        • Jack SPD
        • Haileyesus T
        • Kresnow-Sedacca MJ.
        Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death-United States, 2001-2015.
        MMWR Surveill Summ. 2017; 66: 1-16https://doi.org/10.15585/mmwr.ss6618a1
        • Hedegaard H
        • Curtin SC
        • Warner M.
        Increase in suicide mortality in the United States, 1999–2018.
        NCHS Data Brief. 2020; : 1-8
        • Stone DM
        • Jones CM
        • Mack KA.
        Changes in suicide rates-United states, 2018-2019.
        MMWR Morb Mortal Wkly Rep. 2021; 70: 261-268https://doi.org/10.15585/mmwr.mm7008a1
        • Rossen LM
        • Hedegaard H
        • Khan D
        • Warner M.
        County-level trends in suicide rates in the U.S., 2005-2015.
        Am J Prev Med. 2018; 55: 72-79https://doi.org/10.1016/j.amepre.2018.03.020
        • Pettrone K
        • Curtin SC.
        Urban-rural differences in suicide rates, by sex and three leading methods: United States, 2000–2018.
        NCHS Data Brief. 2020; : 1-8
        • Sullivan E
        • Annest J
        • Luo F
        • Centers for Disease Control and Prevention (CDC)
        Suicide among adults aged 35–64 years–United states, 1999–2010.
        MMWR Morb Mortal Wkly Rep. 2013; 62 (PMCID: PMC4604925): 321-325
        • Sullivan EM
        • Annest JL
        • Simon TR
        • Luo F
        • Dahlberg LL
        • Centers for Disease Control and Prevention (CDC)
        Suicide trends among persons aged 10–24 years–United states, 1994–2012.
        MMWR Morb Mortal Wkly Rep. 2015; 64 (PMCID: PMC4584715): 201-205
        • Baker SP
        • Hu G
        • Wilcox HC
        • Baker TD.
        Increase in suicide by hanging/suffocation in the U.S., 2000-2010.
        Am J Prev Med. 2013; 44: 146-149https://doi.org/10.1016/j.amepre.2012.10.010
        • Martínez-Alés G
        • Jiang T
        • Keyes KM
        • Gradus JL
        The recent rise of suicide mortality in the United States.
        Annu Rev Public Health. 2022; 43: 99-116https://doi.org/10.1146/annurev-publhealth-051920-123206
        • Ramchand R
        • Gordon JA
        • Pearson JL.
        Trends in suicide rates by race and ethnicity in the United States.
        JAMA Netw Open. 2021; 4e2111563https://doi.org/10.1001/jamanetworkopen.2021.11563
        • Braveman P
        • Gruskin S.
        Defining equity in health.
        J Epidemiol Community Health. 2003; 57: 254-258https://doi.org/10.1136/jech.57.4.254
        • Ingram DD
        • Franco SJ.
        2013 NCHS urban-rural classification scheme for counties.
        Vital Health Stat 2, Data Evaluation and Methods Research. 2012; (PMID: 22783637): 1-65
      2. Joinpoint Regression Program. Version 4.7.0.0; 2019. https://surveillance.cancer.gov/joinpoint/.

        • World Health Organization
        Preventing suicide: a global imperative.
        World Health Organization, 2014
        • Cylus J
        • Glymour MM
        • Avendano M.
        Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968–2008.
        Am J Epidemiol. 2014; 180: 45-52https://doi.org/10.1093/aje/kwu106
        • Cunningham PJ.
        Beyond parity: primary care physicians’ perspectives on access to mental health care.
        Health Aff (Millwood). 2009; 28: w490-w501https://doi.org/10.1377/hlthaff.28.3.w490
        • Yip PS
        • Caine E
        • Yousuf S
        • Chang SS
        • Wu KC
        • Chen YY.
        Means restriction for suicide prevention.
        Lancet. 2012; 379: 2393-2399https://doi.org/10.1016/S0140-6736(12)60521-2
        • Cosgrave EM
        • Robinson J
        • Godfrey KA
        • et al.
        Outcome of suicidal ideation and behavior in a young, help-seeking population over a 2-year period.
        Crisis. 2007; 28: 4-10https://doi.org/10.1027/0227-5910.28.1.4
        • Centers for Disease Control and Prevention
        Strategic direction for the prevention of suicidal beahvior: Promoting individual, family, and community connectedness to prevent suicidal behavior: strategic direction for the prevention of suicidal behavior.
        Centers for Disease Control and Prevention, Atlanta, GA2013
        https://stacks.cdc.gov/view/cdc/5275
        Date accessed: May 23, 2022
        • Yau RK
        • Paschall MJ.
        Epidemiology of asphyxiation suicides in the United States, 2005–2014.
        Inj Epidemiol. 2018; 5: 1https://doi.org/10.1186/s40621-017-0131-x
        • Luxton DD
        • June JD
        • Fairall JM.
        Social media and suicide: a public health perspective.
        Am J Public Health. 2012; 102: S195-S200https://doi.org/10.2105/AJPH.2011.300608
        • Goldney RD.
        A note on the reliability and validity of suicide statistics.
        Psychiatr Psychol Law. 2010; 17: 52-56https://doi.org/10.1080/13218710903268014
        • Thomas KH
        • Beech E
        • Gunnell D.
        Changes in commonly used methods of suicide in England and Wales from 1901–1907 to 2001–2007.
        J Affect Disord. 2013; 144: 235-239https://doi.org/10.1016/j.jad.2012.06.041
        • Liu D
        • Yu M
        • Duncan J
        • Fondario A
        • Kharrazi H
        • Nestadt PS.
        Discovering the unclassified suicide cases among undetermined drug overdose deaths using machine learning techniques.
        Suicide Life Threat Behav. 2020; 50: 333-344https://doi.org/10.1111/sltb.12591
        • Canetto SS.
        Meanings of gender and suicidal behavior during adolescence.
        Suicide Life Threat Behav. 1997; 27: 339-351https://doi.org/10.1111/j.1943-278X.1997.tb00513.x
        • Canetto SS.
        Women and suicidal behavior: a cultural analysis.
        Am J Orthopsychiatry. 2008; 78: 259-266https://doi.org/10.1037/a0013973
        • McGirr A
        • Séguin M
        • Renaud J
        • Benkelfat C
        • Alda M
        • Turecki G
        Gender and risk factors for suicide: evidence for heterogeneity in predisposing mechanisms in a psychological autopsy study.
        J Clin Psychiatry. 2006; 67: 1612-1617https://doi.org/10.4088/jcp.v67n1018
      3. Risk and protective factors. Centers for Disease Control and Prevention. https://www.cdc.gov/suicide/factors/index.html. May 13, 2021. Accessed June 15, 2021.

      4. Comprehensive suicide prevention. Centers for Disease Control and Prevention.https://www.cdc.gov/suicide/programs/csp/index.html. Updated August 19, 2021. Accessed May 15, 2022.

        • Ivey-Stephenson AZ
        • Demissie Z
        • Crosby AE
        • et al.
        Suicidal ideation and behaviors among high school students-youth risk behavior survey, United States, 2019.
        MMWR Suppl. 2020; 69: 47-55https://doi.org/10.15585/mmwr.su6901a6
        • Centers for Disease Control and Prevention (CDC)
        Nonfatal self-inflicted injuries treated in hospital emergency departments–United States, 2000.
        MMWR Morb Mortal Wkly Rep. 2002; 51: 436-438
        • Bornheimer LA
        • Wang K
        • Zhang A
        • et al.
        National trends in non-fatal suicidal behaviors among adults in the USA from 2009 to 2017.
        Psychol Med. 2020; (In press. Online August 10)https://doi.org/10.1017/S0033291720002755
        • Sarchiapone M
        • Mandelli L
        • Iosue M
        • Andrisano C
        • Roy A.
        Controlling access to suicide means.
        Int J Environ Res Public Health. 2011; 8: 4550-4562https://doi.org/10.3390/ijerph8124550
        • Breiding MJ
        • Wiersema B.
        Variability of undetermined manner of death classification in the US.
        Inj Prev. 2006; 12: ii49-ii54https://doi.org/10.1136/ip.2006.012591