Clinical and Health Affairs
Nonfatal Suicide Attempts and Other Self-Inflicted Harm
Beltrami County Youths, 2002-2006
By Jon Roesler, M.S., Matthew Petcoff, Arsalan Azam, Sara Westberg, Mark Kinde, M.P.H., and Alex Crosby, M.D.
ABSTRACT
Following a highly publicized murder-suicide by a high school student in Beltrami County, Minnesota, the Minnesota Department of Health initiated an investigation into suicide attempts and self-inflicted harm (SA/SIH) among youths in the county between 2002 and 2006. This article summarizes the results of that effort, which found an annualized rate of 356 hospital-treated SA/SIH per 100,000 population among Beltrami County residents ages 10 to 24 years—a rate more than 50% higher than that of other youths in Minnesota or the United States. In addition, the rate of SA/SIH for American Indian youths in Beltrami County was 2.5 times higher than that for white youths. An examination of medical history found 52% of the youths in this study had previously attempted suicide. This article also discusses several modifiable risk factors that were identified and potential interventions.
During 2004, community leaders in Beltrami County, Minnesota, noted increasing reports of suicide attempts and suicides among the county’s youths. Beltrami County is located in a rural, economically disadvantaged part in northern Minnesota; it includes portions of two Ojibwa reservations. Nineteen percent of the county’s residents are American Indian, 77% are white. On March 21, 2005, a local high school student killed himself after killing nine others. Later that year, community leaders convened a task force to address suicidal behavior; in 2007, the task force contacted the Minnesota Department of Health with a request for assistance.
A preliminary investigation showed that Beltrami County had the highest overall suicide rate of any Minnesota county from 1996 to 2005 and higher-than-expected rates for nonfatal, hospital-treated suicide attempts and other self-inflicted harm (SA/SIH).1
In February 2007, the Minnesota Department of Health initiated an investigation to characterize SA/SIH in Beltrami County using a retrospective case series. This article summarizes that effort and the results.
Methods
Individual cases were identified from Uniform Billing Form (UB-92) administrative datasets of unduplicated hospital-treated injury visits provided annually by the Minnesota Hospital Association to the Department of Health for epidemiologic surveillance. In addition to being treated in a hospital, cases had to meet the following criteria: 1) be between 10 and 24 years old at admission; 2) be discharged between 2002 and 2006; 3) be a Beltrami County resident; 4) have an ICD-9-CM external cause code of E950.0-E959 or a health status factor code of V62.84; and 5) have an ICD-9-CM diagnostic code indicating an acute injury, including 800-904.99, 910-994.99, 995.5-995.9, or 995.80-995.85. The cases selected for medical record abstraction included all of those that involved in-patient hospitalization and a simple random sample of those that were treated in the emergency department (ED); sampled cases were weighted accordingly.
Clinical record confirmation was required for cases to be considered true positives and be included in the study; the clinical record had to indicate that 1) the patient was a Beltrami County resident, 2) the patient sustained an acute physical injury or poisoning, 3) the abstracted information included the first hospital treatment for the injury, and 4) the injury was confirmed or suspected to be intentionally self-inflicted. Age-adjusted SA/SIH rates were calculated using sex and race (white, American Indian) and further divided into three age groups: 10 to 14, 15 to 19, and 20 to 24 years (no cases of SA/SIH were reported in children younger than 10). Where applicable, toxicology test results were included. Because of time limitations, the person’s history of previous risk factors was based on self-report in the medical record.
Findings
Between 2002 and 2006, 234 cases of SA/SIH were reported in youths 10 to 24 years of age in Beltrami County; all of the 84 hospitalized cases and a sample of 53 of the 150 ED cases were selected for abstraction. Of the 84 hospitalized cases, the hospital could not locate charts of six, and four were false positives using the clinical record case criteria. Of the 53 sampled ED cases, two charts were unable to be located, eight were false positives, and 14 were actually hospitalized cases. The positive predictive value for SA/SIH was 91% (95% CI: 86%, 96%). Because of misclassification with regard to ED treatment, analyses were done of the combined hospitalized and ED-only treated cases; age-adjusted rates were calculated from the true positives.
The overall SA/SIH rate for youths was 356/100,000 population. Female American Indian youths had the highest rate at 774.4 per 100,000 population (535.9, 982.9). The rate for young male American Indians was 371.1/100,000 (225.8, 516.8); for white female youths, 375.1 (205.3, 460.0); and for white male youths, 84.6 (45.5, 162.8). The SA/SIH risk ratios for female American Indian youths compared with young American Indian males, young white females, and young white males were 2.1 (1.3, 3.8), 2.1 (1.5, 3.0), and 9.2 (5.4, 15.7), respectively. Youths ages 9 to 15 years had a rate of 553.9 (449.9, 657.9), twice that of those ages 10 to 14 years (278.5 [196.2, 360.8] or 20 to 24 years (226.4 [165.4, 287.4]). Many cases had a documented history of prior SA/SIH as well as other risk factors such as alcohol or substance misuse and mental illness. At the time of admission, 42% (32%, 53%) of cases tested positive for alcohol and/or illicit drugs (principally cannabinoids/THC); no cases tested positive for methamphetamine or phencyclidine (Table).
Discussion
This investigation revealed a substantial burden of SA/SIH among American Indian youths, particularly females and those ages 15 to 19 years. The annualized rate of SA/SIH among Beltrami County residents ages 10 years to 24 years was more than 50% higher than that of other youths in Minnesota or the United States.2,3 In addition, the rate for American Indian youths in Beltrami County was 2.5 times higher than that for their white peers.
A prior history of SA/SIH was identified in more than half of the cases. Two-thirds of American Indian youths had used a mood-altering substance at the time of admission, compared with 18% of white youths. Along with the mechanism of SA/SIH, the presence of drug use and a history of SA/SIH, reveal complex combinations of risk factors.
Exposure to suicide or suicidal behaviors within one’s family or one’s peer group can result in an increase in suicide and suicidal behaviors.4 Similarly, a major correlate of youth suicide is an earlier attempt. It has been previously documented that fatal and nonfatal suicidal behaviors are more common among American Indian adolescents compared with adolescents overall, and suicide attempts among American Indian youths are linked to risk factors such as having friends or relatives who had attempted or completed suicide.5 A previous study of SA/SIH found that 35% of patients had self-reported previous SA/SIH.6 The 52% of cases with a prior history of SA/SIH was higher than expected in both white and American Indian youths. Although this investigation focused on Beltrami County, similar rates of suicide have been noted among youths in adjoining counties.1
Administrative hospital discharge data were shown to be useful in initially identifying the problem. However, medical record abstraction was needed to obtain additional information about risk factors and outcomes. Indeed, these findings are subject to at least four limitations. First, in Beltrami County, the peak in SA/SIH occurred in 2004 and the peak in suicides occurred in 2005. The current investigation was not initiated until 2007. Such time lags pose a significant challenge to those who conduct public health interventions. This illustrates the importance of timely recognition of youths who might be at risk for suicide and self-inflicted harm so that local authorities can intervene in a timely manner. Second, although the administrative hospital discharge data were shown to have a good positive predictive value for identification of SA/SIH overall, the misclassification of outpatient and inpatient status was problematic. Third, the description of race/ethnicity and suicide-related factors in the medical record was sketchy or missing, thus limiting its usefulness. Fourth, information that could separate an event in which an individual used poisoning as the primary mechanism of self-inflicted injury from one in which the use of substances was combined with another method was often lacking. This limited our ability to determine the exact role of substances in some events.
The high rate of previous suicidal behavior among youths seen in the hospital following an attempt points to the need for a comprehensive outpatient treatment plan before discharge. The lack of continuity of care for individuals who have exhibited self-harming behaviors has been shown to be a major factor in subsequent negative health outcomes.7 To help maintain continuity of care following discharge and address other risk factors such as substance misuse, mental illness, and history of SA/SIH, the Minnesota Department of Health has worked with leaders from Beltrami County to fund and initiate follow-up programs for youths who have engaged in SA/SIH. This included a program modeled after an existing one for survivors of hospitalized traumatic brain injury in Minnesota.8 The latter intervention involves phone calls to patients and their families in order to help them solve problems, identifying resources needed for recovery, and connecting youths and their families with existing services after hospital discharge; the program can be adapted for specific cultural communities. This initiative represents a new application of an existing intervention strategy but shares many characteristics of those specifically designed to help individuals exhibiting suicidal behavior.9
The Department of Health has also provided hospital-based staff training and community-based education on suicide prevention, as well as a suicide prevention grant to a coalition in Beltrami County to support local awareness and strengthen community-based education. These latter components are consistent with other successful programs that have been used in American Indian and Alaska Native communities;10 they are comprehensive and include public awareness, promoting protective factors, and community outreach.
Mortality data confirm a continuing decline in suicides through 2008. We have documented decreasing trends of hospital-treated self-inflicted harm through 2006; however, preliminary analysis of 2007 data indicates that hospital-treated SA/SIH increased. Although suicide mortality was the driving impetus behind this study, SA/SIH has returned to 2004 levels. This serves as the call to action for community-based practitioners to screen for risk factors for suicide in this patient population and follow up with individuals who have tried harming themselves.
Conclusion
Suicide attempters are frequently known to the medical system and often present with a constellation of risk factors. It is important for these patients to be appropriately referred. As described in this article, systems for follow-up may need to be enhanced or established. Although this need may be felt most critically in Beltrami County, the importance of connecting survivors to services following their hospital visit is needed throughout Minnesota. MM
Jon Roesler is epidemiologist supervisor, Matthew Petcoff is a graduate student intern, Arsalan Azam was a student intern, Sara Westberg was medical records abstract coordinator, and Mark Kinde is the injury unit director with the Minnesota Department of Health, Health Promotion and Chronic Disease Division, Center for Health Promotion, Injury and Violence Prevention Unit. Alex Crosby is a medical epidemiologist with the Division of Violence Prevention in the Centers for Disease Control and Prevention’s National Center for Injury Prevention and Control.
REFERENCES
1. Roesler J. Suicide and nonfatal, self-inflicted harm, Beltrami County, 1990-2005: preliminary epidemiologic investigation. Saint Paul, MN: Minnesota Department of Health, Injury and Violence Prevention Unit; Sept. 2007. Available at http://www.health.state.mn.us/injury/pub/BeltramiSandSIH.pdf. Accessed July 10, 2009.
2. MIDAS: Minnesota Injury Data Access System. Saint Paul, MN: Minnesota Department of Health. 1998-2006. Available at: http://www.health.state.mn.us/injury/midas/ub92/index.cfm. Accessed July 10, 2009.
3. WISQARS: Web-based Injury Statistics Query and Reporting System. Atlanta, GA: Center for Disease Control and Prevention. 1999-2005. Available at: http://www.cdc.gov/injury/wisqars/index.html. Accessed July 10, 2009.
4. Gould MS, Greenberg T, Velting DM, Shaffer D. Youth suicide risk and preventive interventions: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry. 2003;42(4);386-405.
5. Borowsky IW, Resnick MD, Ireland M, Blum RW. Suicide attempts among American Indian and Alaska Native youth: risk and protective factors. Arch Pediatr Adolesc Med. 1999;153(6):573-80.
6. Miranda R, Scott M, Hicks R, Wilcox HC, Harris Manfakh JL, Shaffer D. Suicide attempt characteristics, diagnoses, and future attempts: comparing multiple attempters to single attempters and ideators. J Am Acad Child Adolesc Psychiatry. 2008;47(1):32-40.
7. U.S. Department of Health and Human Services. Mental health: culture, race, and ethnicity—a supplement to mental health. 2001. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services; 2001. Available at: http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=hstat5.chapter.971. Accessed July 10, 2009.
8. Seymour L, Roesler J, Kinde M. Connecting kids! Effective recruitment for resource facilitation in the pediatric population. J Head Trauma Rehabil. 2008; 23(4):264-70.
9. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. 2005;294(5):563-70.
10. May PA, Serna P, Hurt L, DeBruyn LM. Outcome evaluation of a public health approach to suicide prevention in an American Indian tribal nation. Am J Public Health. 2005; 95(7):1238–44.