Why it is time for a public health approach to preventing suicide

Transcription

1 Why it is time for a public health approach to preventing suicide Matthew Miller, MD, MPH, ScD Associate Director, Harvard Injury Control Research Center

2 Background GGB: Leading suicide location worldwide More than 1200 suicides since 1937 Approximately one suicide every 2-3 weeks 4 foot railing, no other barrier Not always the uncontested leader among sites worldwide the Eiffel Tower, St. Peter s Basilica, the Sydney Harbor and Arroyo Seco Bridges, Mt. Mihara

3 Barriers to Suicide and Suicide Prevention: Public Opinion and the Golden Gate Bridge All respondents (n=2638) Male (n=1310) Female (n=1328) Current Smoker (n=548) Not a Current Smoker (n=2087) Personally own a firearm (n=735) Live in a household with guns but do not personally own any firearms (n=296) Live in a household without firearms (n=1560) Percent who think ALL Jumpers would find another way to kill themselves if a foolproof anti-suicide barrier were put up on the Golden Gate Bridge Multivariate Odds Ratio (95% CI) Miller M, Azrael D, Hemenway D. Belief in the inevitability of suicide: results from a national survey. Suicide and Life Threatening Behavior (1):1-11.

4 Simplistic model of some causes of fatal and non-fatal suicidal behaviour GENETIC FACTORS PSYCHIATRIC DISORDERS PERSONALITY FACTORS Aggression Impulsivity Hopelessness Cultural meaning of suicide/ role models/media influences AVAILABILITY OF METHOD Method likely to be lethal SUICIDAL BEHAVIOUR SUICIDE Life Events/Crises Method unlikely to be lethal NON-FATAL ATTEMPT Modified from Keith Hawton, PhD

5 Causes of Violent Death, Worldwide ~ one million suicides annually worldwide 16 per 100,000, or one death every 40 seconds. Suicide is now among the three leading causes of death among those aged years Armed Conflict 20% Homicide 30% Suicide 50% Source: WHO, Violence and Health, Report of the Secretariat, November 2001 How many SPH offer courses on suicide prevention?

6 Leading Causes of Death, US, 2002

7 2003 Deaths Suicides 31,484 Firearm Suicides 16,907 (54%) Homicides 17,732 Firearm Homicide 11,920 (67%)

8 HICRC 2004 Firearms Survey Households with Kids Firearm prevalence 39% 38% Households without Kids In Households with Guns: % Unlocked % Loaded % Loaded & Unlocked 48% 22% 5% 56% 31% 12% Hepburn L, Miller M, Azrael D, Hemenway D. Firearm ownership in the US: findings a national survey. Preliminary findings

9 Non-fatal Self-Harm vs. Suicide, United States Non-fatal Self-Harm (n=411,128) Suicide (n=31,484) Cut/pierce 22% Poisoning 18% Cutting/piercing 2% Drowning 1% Fall 2% Drown/Near-drown* 0% Fall 1% Firearm* 1% Poisoning 75% Inhalation/ suffocation 1% Suffocation 21% Firearm 55% WISQARS Non-Fatal Injury Reports (2003 Data) seen in ER regardless of disposition WISQARS Fatal Injury Reports (2003 Data)

10 Relationship of attempted suicide to completed suicide A B Attempted Suicides Completed Suicides AB <10% of A AB= 25%-30% of B Source: Maris R.W. (1981) Pathways to suicide: A survey of self-destructive behaviors. Reprinted in Maris, Berman, Silverman, eds. Comprehensive Textbook on Suicidology, 2000.

11 Suicide Attempts and Case Fatality Rate by Method Seven Northeast States, Percent of suicide acts by methods All Methods Firearm Poisoning/ Cutting/ Piercing Suffocation/ Hanging All Other Methods 100% 5% 85% 5% 5% Case Fatality Rate All Methods Firearm Poisoning / Cutting / Suffocation / Hanging All Other Methods CFR Source: Miller, Azrael, Hemenway, Annals of Emer Med, 2004.

12 Guns used in suicide Handgun 46% Rifle or Shotgun 54% Handgun 60% Rifle or Shotgun 40% Youth 17 and under Adults 18 and over The vast majority (~90%) of firearms used in suicides come from the victims home

13 A ACUTE RISK (only) SUICIDE RISK No risk x Risk period y B ACUTE ON CHRONIC RISK Chronic risk x Period of extra risk y From: Keith Hawton, PhD

14 Means Restriction as a Suicide Prevention Strategy Suicide acts are often impulsive Crises are often temporary 90% of survivors of near-lethal suicide attempts do not commit suicide thereafter

15 Impulsivity SURVIVORS OF SELF-INFLICTED SHOOTING (De Moore et al, 1994) 33 survivors of self-inflicted firearm injuries Most common reason for shooting was interpersonal conflict with partner or family members (21 patients) Act almost always described as impulsive All used firearms available in household and availability was often reason given for choice of this method

16 Impulsivity Factors associated with the medical severity of suicide attempts in youths and young adults (Swahn MH, Potter LB, 2001) 25% of 153 survivors of near lethal suicide attempts acted within 5 minutes of the impulse to do so 71% acted within one hour Sri Lanka Pesticide Suicides (Eddleston, 2006 JCT) 268 self-poisonings, 85% noted choice of poison dictated by what was readily at hand Little premeditation

17 On the bridge, Baldwin counted to ten and stayed frozen. He counted to ten again, then vaulted over. I still see my hands coming off the railing, he said. As he crossed the chord in flight, Baldwin recalls, I instantly realized that everything in my life that I d thought was unfixable was totally fixable except for having just jumped. Tad Friend. Jumpers. The New Yorker (2003)

18 CRISIS same-day Did suicide occur within 24 hours of a very recent crisis or precipitating event? e.g. arrest, job loss, argument, relationship break-up, legal trouble, police pursuit, bad report card Purpose is to identify suicides that may have an impulsive element

19 Crisis, same day Victim was a 16 year-old boy who had been suspended from school for misbehavior. His mother grounded him and they got in a loud argument. He ran from the room, slammed the door, went to the basement and shot himself with his father s rifle. No known mental health history.

20 CRISIS % same day by age Data % Age group N=1,671 CT, ME, UT, WI, Allegheny County, San Francisco County

21 Lifetime Risk of Suicide Among Five Populations of Psychiatric Patients From: Bostwick: Am J Psychiatry, Volume 157(12).December

22 Prognosis after attempted suicide

23 Prognosis after attempted suicide

24 Prognosis after attempted suicide

25 Does availability of method affect suicide rates?

26 THE COAL GAS STORY (Kreitman, 1976) Percentage of CO in domestic gas, United Kingdom Percent CO Year

27 THE COAL GAS STORY (Kreitman, 1976) Sex-specific suicide rates by mode of death: England & Wales Males Females Rate per 100, Total Non CO CO Rate per 100, Total Non CO CO Year Year

28 Suicide Rates United States, Suicide Rate % Adults in Households with Guns Source: Natl. Center for Health Statistics Age-adjusted to 1940 U.S. population

29 Figure 2. Household gun ownership levels and rates of firearm and non-firearm suicide mortality: United States, Year Rolling Average of Household Gun Ownership Suicide Rate (per 100,000 population) Year 3 Yr. Rolling Average Household Gun Ownership Non-Firearm Suicide Rate Firearm Suicide Rate Miller M, Azrael D, Hepburn L, Hemenway D, Lippmann S. Injury Prevention In Press.

30 Table 1. Longitudinal Association between Household Gun Ownership and Suicide Mortality: Percent Decrease in Mortality Rate for each 10% Decrease in Household Firearm Ownership Level ( ), adjusted for age, unemployment, poverty, per capita alcohol consumption and region of the country % Decrease in Firearm Suicide Rate (95% Confidence Interval) % Decrease in Non-Firearm Suicide Rate (95% Confidence Interval) % Decrease in Overall Suicide Rate (95% Confidence Interval) Total Population 4.2% (2.3% 6.1%) *** 0.3% (-1.4% 2.3%) 2.5% (1.4% 3.6%) *** Males 3.4% (2.8% 4.1%) *** 1.1% (-0.7% 2.9%) 2.3% (1.7% 2.9%)*** Females 3.4% (1.9% 4.9%) *** -0.5%(-1.9% 1.9%) 1.0% (0.4% 1.6)** Children (0-19 years of age) 8.3% (6.1% 10.5%) *** -0.4.% (-2.9% 2.1%) 4.1% (2.3% 5.9%) *** Miller M, Azrael D, Hepburn L, Hemenway D, Lippmann S. Injury Prevention

31 Why might declines occur? Improved life circumstances Improved mental health Improved mental health treatment (SSRIS?) Cohort characteristics Changes in social norms Coding changes Changes in population (e.g. large influx of immigrants with lower suicide rates; change in age distribution) In most of these scenarios, we would expect to see attempts and completions by all methods going down.

32 12-month prevalence of suicide related behaviors, year olds (National Co-morbidity Survey) NCS NCS P Value Ideation 2.8% 3.3% 0.44 Gesture 0.3% 0.2% 0.24 Attempt 0.4% 0.6% 0.44 Kessler et al. JAMA. 2005;293:

33 Rates of Household Firearm Ownership and Rates of Suicide, Firearm Suicide and Non-Firearm Suicide across 7 North East States, Percent of Households with Firearms Suicides per Firearm Suicides per Non-Firearm Suicides per Vermont 42% Maine 41% New Hampshire 30% Connecticut 17% Rhode Island 13% Massachusetts 12% New Jersey 12% All 7 States 17% Suicide attempt rate with non-firearm methods did not significantly correlate with firearm prevalence Miller M, Azrael D, Hemenway D. Firearms and Suicide in the Northeast. Journal of Trauma. 2004; 57:

34 Firearms and Suicide across the 50 US States, Dependent Variable Firearm and non-firearm suicide mortality rates Independent Variables Household Firearm Prevalence (BRFSS 2002) Alcohol and illicit substance abuse and dependence- Serious mental illness Unemployment Poverty and urbanization Miller M, Lippmann S, Azrael D, Hemenway D. Journal of Trauma In Press.

35 Suicides by Age Group: The 15 U.S. States with the Highest vs. the 6 U.S. States with the Lowest Average Household Gun Ownership* ( ) 5-19 Year Olds year olds High-Gun States (Total Pop., : 115 Million) 47% of Households have firearms Low-Gun States (Total Pop., : 118 Million) 15% of Households have firearms Mortality Rate Ratio (High Gun : Low Gun) Firearm Suicide Non-Firearm Suicide Total Suicide Firearm Suicide Non-Firearm Suicide Total Suicide Miller, Azrael, Hemenway. Journal of Trauma In Press

36 Correlation Coefficients: rates of suicide, firearm suicide, non-firearm suicide, lifetime major depression, lifetime suicidal thoughts and firearm ownership, US Census regions Suicide Rate Firearm suicide rate Non-firearm suicide rate Major depression Suicidal thoughts Household Handgun prevalence Psychiatric disorder prevalence estimates by census region, (National Co-morbidity Study, n=48,000). Lifetime major depression (n=8098) and ever seriously considered suicide (n=5877) were reported by 17% and 13% of respondents nationally. Mean household firearm and handgun ownership rates come from the General Social Surveys. Controlled for per capita alcohol consumption, unemployment, college education from NCS Hemenway D, Miller M. Association of rates of household handgun ownership, lifetime major depression, and serious suicidal thoughts with rates of suicide across US census regions. Injury Prevention 2002; 8:

37 Seriously Considered Suicide in Past 5 Years HICRC National RDD Survey, 2004 Personally Own Firearms (n=762) Households with Firearms (n=1075) No Firearms (n=1639) 2.8% 2.9% 4.9% Miller et al

38 Case Control: Brent et al Five overlapping studies of adolescents in Western PA (no one study with more than 70 suicides) Results common to all studies: Firearms in the home strongly associated with risk of suicide (OR=2 to 5) Brent et al

39 Case Control: Kellermann et al Suicide in the HOME All ages Large study (803 suicides) Results: Firearms in the home strongly associated with risk of suicide The association is strongest for adolescents/young adults and for individuals without a known psychiatric history Risk greater if gun stored loaded or unlocked Risk is elevated for all members of the household Kelermann, NEJM, 1992

40 Case Control: Cummings et al RQ: Is the purchase of handgun from licensed dealer associated with suicide, risk vary with time since purchase, risk extend to family members? Results: Suicides more likely than controls to live in household in which a family member had purchased a handgun (RR=1.9). RR=2.0 if an individual himself purchased the gun and 1.5 if the handgun was purchased by another family member Risk greatest within first year, persists even after 5 years Cummings et al, AJPH, 1997

41 Storage Practices and Suicide Risk, 5-19 year olds Firearm Storage Practice Cases (N=106) No. (%) Controls (N=480) Adjusted Odds Ratio (95% CI) Gun unloaded 66% 91% 0.30 ( ) Gun locked 32% 58% 0.27 ( ) Ammunition locked 24% 48% 0.39 ( ) Grossman et al, JAMA, February 9, 2005.

42 Variable Risk of Suicide in the Home in Relation to Various Patterns of Gun Ownership Type of guns in the home Adjusted Odds Ratio 95% Confidence Interval One or more handguns Long guns only No guns in the home Loaded guns Any gun kept loaded All guns kept unloaded No guns in the home Locked guns Any guns kept unlocked All guns kept locked up No guns in the home Source: Kellerman et al NEJM

43 Do Interventions work? Parents of 106 Adolescents with Major Depression Asked if firearms in the home 27% Yes 73% No Told about the increase risk of suicide conveyed by guns in the home Not counseled Follow up 2 yrs Follow up 2 yrs 27% removed guns from home 73% retained guns in home 17% acquired guns Brent et al. 2000: JAACAP

44 Are we trying hard enough? Survey of ED Nurses in Illinois (n=527) Experience Recent experience with suicidal adolescents Ever received means restriction training Provided means restriction education to parents Worked in units where means restriction is standard practice Percentage 80% (n = 407) 24% (n = 122) 28% (n = 136) 18% (n = 89) Grossman et al, 2003: JAPNA

45 Are we delivering the right message? Poison center chart review F/U call to caretakers, 6-24 months after suicide attempt by adolescent

46 Potentially lethal means available in the home at the time of the attempt 75% prescription medication 86% OTC medications 2% street drugs 25% firearms 95% had some medication available at time of SA

47 At follow-up ~10% reported that injury prevention education was given about restricting access medication 86% locked up medications (vs. 32% of those w/o IP education) No one (0%) reported being told that firearms in the home increase the risk of suicide 76% of the adults were not worried about the possibility of a repeat suicide attempt by their adolescent

48 Active Opposition In early 2006, two separate but virtually identical bills were introduced in the Virginia and West Virginia legislatures Each bill would have prohibited a physician from asking a patient if he or she owned firearms for the purpose of counseling that patient about ways to reduce risks associated with firearms Penalties for violation of the bills included revocation of a physician's license to practice Counseling about firearms: proposed legislation is a threat to physicians and their patients. Vernick JS, Teret SP, Smith GA, Webster DW. Pediatrics 2006; 118(5):

49 Passive Opposition

50 Barriers to Suicide and Suicide Prevention: Public Opinion and the Golden Gate Bridge Percent who think ALL Jumpers would find another way to kill themselves if a foolproof anti-suicide barrier were put up on the Golden Gate Bridge Multivariate Odds Ratio (95% CI) All respondents (n=2638) 34% NA Male (n=1310) 37%*** 1.05 (0.86, 1.28) Female (n=1328) 31% Reference Current Smoker (n=548) 43%*** 1.55 (1.25, 1.93)*** Not a Current Smoker (n=2087) 32% Reference Personally own a firearm (n=735) 48%*** 2.10 (1.70,2.59)*** Live in a household with guns but do not personally own any firearms (n=296) 36%*** 1.36 (1.02,1.81)* Live in a household without firearms (n=1560) 26% Reference An additional 40% thought that most jumpers would have found another way to complete suicide Miller M, Azrael D, Hemenway D. Belief in the inevitability of suicide: results from a national survey. Suicide and Life Threatening Behavior (1):1-11.

51 Conclusions Suicide is preventable Means Matter Interventions can change behavior Message infrequently delivered/ received Active and Passive Opposition The Deadly Breach Opportunity for success

52 Questions?

53 Discordance Between Parent and Child Responses to Questions about Household Firearms Know Location Handled All children 39% 22% Household firearms All locked 46% 17% Some unlocked 30% 29% Parent Report Discussed gun safety w child 40% 21% Never discussed gun safety w child 39% 29% Baxley and Miller, 2006: APAM

54 21 Grounds for Revocation of a Medical License in Virginia 1. False statements, fraud, deceit 2. Substance abuse 3. Intentionally or neglig. harming a patient 4. Mental or physical incapacity 5. Conviction for moral turpitude Other egregious conduct Vernick, Teret, Smith, Webster. Under review by Pediatrics.

55 VA HB 1531 Sought to Add # Oral or written inquiry to a patient concerning the possession, ownership, or storage of firearms, where such inquiry has no relationship to the practice of the healing arts or the medical condition of the patient, and is for the purpose of gathering statistics or to justify patient counseling, unless such inquiry is the subject of a request, or related to a medical complaint, made by the patient.

56 HB 1531: What Happened Next Jan 20: Bill introduced Feb 13: Passed VA State Assembly, Feb 14-22: Lobbying by NRA and AAP Feb 23: Defeated in VA Senate Comm., 9-6 Feb 24: Virtually identical bill introduced in West Virginia Mar 19: WV session closes with no action

57 Implications of the Bills Medical Malpractice The First Amendment Dr. Timothy Wheeler: This was not a bad showing for a first try at a boundary violation bill.

58 Case Control: Cornwell et al All Suicides, 65 years and older Large study (86 cases and 86 community controls matched on age, sex, county of residence) Results: Presence of a firearm in the home was associated with increased risk for suicide, even after controlling for psychiatric illness. Elevated risk was accounted for by access to handguns rather than long guns RR more pronounced in men than women Storing the weapon loaded and unlocked were independent predictors of suicide Cornwell, Am J Geriatric Psychiatry, 2002

59 Case Control: Wiebe All Suicides drawn from the 1993 National Mortality Followback Survey Large study (1,959 cases and 13,535 controls from National health interview Survey matched on age, sex, race) Results: Presence of a firearm in the home was associated with increased risk for suicide: 66% of suicides lived in households with guns vs. 37% of controls OR=3.4 (3.1, 3.9) Elevated risk was accounted for by access to handguns rather than long guns Wiebe, AEM,41(6): , 2003

60 Guns used in suicide Among 35 gun suicide victims aged 17 and under, the owner of the gun was documented in 26 cases. In 92% of cases (24/26), the gun came from the family. Owner # % Parent Other family Self, received from parent % Self, source unknown Total *Among adults as well, firearms used in suicide are predominantly household guns

61 Rates of Household Firearm Ownership and Rates of Suicide, Firearm Suicide and Non-Firearm Suicide across 7 North East States, Percent of Households with Firearms Suicides per Firearm Suicides per Non-Firearm Suicides per Vermont 42% Maine 41% New Hampshire 30% Connecticut 17% Rhode Island 13% Massachusetts 12% New Jersey 12% All 7 States 17% Miller M, Azrael D, Hemenway D. Firearms and Suicide in the Northeast. Journal of Trauma. 2004; 57:

62 Rates of Self-harm (per 100,000), 7 Northeast States, , by State and Method (per 100,000 population) across 7 North East States, listed in order of firearm prevalence, highest to lowest Suicide Attempt Rate All Methods Suicide Attempt Rate Firearm Suicide Attempt Rate Poisoning/ Cutting/ Piercing Suicide Attempt Rate Suffocation/ Hanging Suicide Attempt Rate All Other Methods Vermont Maine New Hampshire Connecticut Rhode Island Massachusetts New Jersey Percent of suicide attempts by method 100% 5% 85% 5% 5% Miller M, Azrael D, Hemenway D. Firearms and Suicide in the Northeast. Journal of Trauma. 2004; 57:

63 Baxley & Miller (2006)

64 Suicide Rates by Method Israel Eng & Wales The Netherlands US Norw ay Scotland Canada New Zealand Australia Denmark Suicide Deaths/100,000 Firearm Poisoning Suffocation All Other Source: Lois A. Fingerhut, MA, Centers for disease Control and Prevention, Unpublished

65 Youth alcohol use Tested Positive for Alcohol (83% tested) 35% 5% <18 yrs Adults

66

67

68

69 The Association between Changes in Household Firearm Ownership and Rates of Suicide in the United States, Dependent Variable Firearm and non-firearm suicide mortality rates from WISQARS, 4 Census regions, Independent Variables Percentage of individuals living in households with firearms (gun prevalence) General Social Survey (GSS) Age Per capita alcohol consumption - National Institute on Alcohol Abuse and Alcoholism Unemployment Bureau of Labor Statistics Poverty and urbanization Census Current Population Survey Analysis GEE modeling to account for serial auto-correlation Regional fixed effects controlled for cross sectional, time invariant differences among the four census regions. The association between changes in household firearm ownership and rates of suicide in the United States, M Miller, D Azrael, L Hepburn, D Hemenway, S J Lippmann. Injury Prevention 2006;12:

70 Rates of Self-harm (per 100,000), 7 Northeast States, , by State and Method (per 100,000 population) across 7 North East States, listed in order of firearm prevalence, highest to lowest Suicide Attempt Rate All Methods Suicide Attempt Rate Firearm Suicide Attempt Rate Poisoning/ Cutting/ Piercing Suicide Attempt Rate Suffocation/ Hanging Suicide Attempt Rate All Other Methods Vermont Maine New Hampshire Connecticut Rhode Island Massachusetts New Jersey Percent of suicide attempts by method 100% 5% 85% 5% 5% Miller M, Azrael D, Hemenway D. Firearms and Suicide in the Northeast. Journal of Trauma. 2004; 57: