User login
Ultimately, we need to find better treatments for depression and anxiety
San Francisco entrances people. Photographers capture more images of the Golden Gate Bridge than any other bridge in the world.1 And only the Nanjing Yangtze River Bridge in China surpasses the Golden Gate as a destination for dying by suicide.2 At least 1,700 people reportedly have plunged from the bridge to their deaths since its opening in 1937.3
Despite concerted efforts by bridge security, the local mental health community, and a volunteer organization – Bridgewatch Angels – suicides continue at the pace of about 1 every 2 weeks. After more than 60 years of discussion, transportation officials allocated funding and have started building a suicide prevention barrier system on the Golden Gate.
Extrapolating from the success of barriers built on other bridges that were “suicide magnets,” we should be able to assure people that suicide deaths from the Golden Gate will dramatically decrease, and perhaps cease completely.4 Certainly, some in the mental health community think this barrier will save lives. They support this claim by citing research showing that removing highly accessible and lethal means of suicide reduces overall suicide rates, and that suicidal individuals, when thwarted, do not seek alternate modes of death.
I support building the Golden Gate suicide barrier, partly because symbolically, it should deliver a powerful message that we value all human life. But will the barrier save lives? I don’t think it will. As the American Psychiatric Association prepares to gather for its annual meeting in San Francisco, I would like to share my reasoning.
What the evidence shows
The most robust evidence that restricting availability of highly lethal and accessible means of suicide reduces overall suicide deaths comes from studies looking at self-poisoning in Asian countries and Great Britain. In many parts of Asia, ingestion of pesticides constitutes a significant proportion of suicide deaths, and several studies have found that, in localities where sales of highly lethal pesticides were restricted, overall suicide deaths decreased.5,6 Conversely, suicide rates increased when more lethal varieties of pesticides became more available. In Great Britain, overall suicide rates decreased when natural gas replaced coal gas for home heating and cooking.7 For decades preceding this change, more Britons had killed themselves by inhaling coal gas than by any other method.
Strong correlations exist between regional levels of gun ownership and suicide rates by shooting,8 but several potentially confounding sociopolitical factors explain some portion of this connection. Stronger evidence of gun availability affecting suicide rates has been demonstrated by decreases in suicide rates after restrictions in gun access in Switzerland,9 Israel,10 and other areas. These studies show correlations – not causality. However, the number of studies, links between increases and decreases in suicide rates with changes in access to guns, absence of changes in suicide rates during the same time periods among ostensibly similar control populations, and lack of other compelling explanations support the argument that restricting access to highly lethal and accessible means of suicide prevents suicide deaths overall.
The installation of suicide barriers on bridges that have been the sites of multiple suicides robustly reduces or even eliminates suicide deaths from those bridges,11 but the effect on overall suicide rates remains less clear. Various studies have found subsequent increases or no changes12-14 in suicide deaths from other bridges or tall buildings in the vicinity after the installation of suicide barriers on a “suicide magnet.” Many of the studies failed to find any impact on overall suicide rates in the regions investigated. Deaths from jumping off tall structures constitute a tiny proportion of total suicide deaths, making it difficult to detect any changes in overall suicide rates. In the United States, suicides by jumping/falling constituted 1%-2% of total suicides over the last several decades.15
If we know that restricting highly lethal and accessible methods of killing reduces suicide deaths, why would I question the value of the Golden Gate suicide barrier in preventing overall suicide deaths?
Unique aspects of the bridge
The World High Dive Federation recommends keeping dives to less than 20 meters (65.5 feet), with a few exceptions.16 The rail of the Golden Gate Bridge stands 67 meters (220 feet) above the water, and assuming minimal wind resistance, a falling person traverses that distance in about 3.7 seconds and lands with an impact of 130 km/hour (81 miles per hour).17 Only about 1%-2% of those jumping from the Golden Gate survive that fall.18
A 99% likelihood of death sounds pretty lethal; however, death by jumping from the Golden Gate inherently takes place in a public space, with the opportunity for interventions by other people. A more realistic calculation of the lethality would start the instant that someone initiates a sequence of behaviors leading to the intended death. By that criteria, measuring the lethality of the Golden Gate would begin when an individual enters a vehicle or sets off on foot with the plan of going over the railing.
Unless our surveillance-oriented society makes substantially greater advances (which I oppose), we will remain unable to assess suicide lethality by starting at the moment of inception. However, we do have data showing what happens once someone with suicidal intentions walks onto the bridge.
Between 2000 and 2018, observers noted 2,546 people on the Golden Gate who appeared to be considering a suicide attempt, the San Francisco Chronicle has reported. Five hundred sixty-four confirmed suicides occurred. In an additional 71 cases, suicide is presumed but bodies were not recovered. In the 1,911 remaining instances, mental health interventions were made, with individuals taken to local hospitals and psychiatric wards, and released when no longer overtly suicidal. Interventions successfully diverted 75% (1,911/2,546) of those intending to end their own lives, which suggests that the current lethality of the Golden Gate as a means of suicide is only 25%. Even in the bridge’s first half-century, without constant camera monitoring, and a cadre of volunteers and professionals scanning for those attempting suicide, the lethality rate approached about 50%.19
We face even more difficulties measuring accessibility than in determining lethality. The Golden Gate appears to be accessible to almost anyone – drivers have to pay a toll only when traveling from the north, and then only after they have traversed the span. Pedestrians retain unfettered admittance to the east sidewalk (facing San Francisco city and bay) throughout daylight hours. But any determination of accessibility must include how quickly and easily one can make use of an opportunity.
Both entrances to the Golden Gate are embedded in the Golden Gate National Recreational Area, part of the National Park system. The south entrance to the bridge arises from The Presidio, a former military installation that housed about 4,000 people.20 Even fewer people live in the parklands at the north end of the bridge. The Presidio extends far enough so that the closest San Francisco neighborhoods outside of the park are a full 2.2 km (1.36 miles) from the bridge railing. A brisk walk would still require a minimum of about 20 minutes to get to the bridge; it is difficult to arrive at the bridge without a trek.
Researchers define impulsivity, like accessibility, inconsistently – and often imprecisely. Impulsivity, which clearly exists on a spectrum, connotes overvaluing of immediate feelings and thoughts at the expense of longer term goals and aspirations. Some suicide research appears to define impulsivity as the antithesis of planned behavior;21,22 others define it pragmatically as behaviors executed within 5 minutes of a decision,23 and still others contend that “suicidal behavior is rarely if ever impulsive.”24 Furthermore, when we assess impulsivity, we must acknowledge a fundamental difference between “impulsive” shootings and poisonings that are accomplished at home and within seconds or minutes, from “impulsive” Golden Gate Bridge suicide attempts, which require substantial travel and time commitments, and inherently involve the potential for others to intervene.
Those arguing that the bridge suicide barrier will save lives often bring up two additional sets of numbers to back up their assertions. They provide evidence that most of those people who were stopped in their attempts at suicide at the Golden Gate do not go on to commit suicide elsewhere, and that many of those who survived their attempts express regret at having tried to kill themselves. Specifically, 94% of those who were prevented from jumping from the Golden Gate had not committed suicide after a median follow-up of 26 years, according to a follow-up study published a few years ago. On the other hand, those who have made a serious suicide attempt have a substantially increased risk, relative to the general population, of dying from a later attempt,25,26 and the strongest predictor for death by suicide is having made a previous, serious suicide attempt.27
While all of these studies provide important and interesting information regarding suicide, none directly address the question of whether individuals will substitute attempts by other methods if the Golden Gate Bridge were no longer available. Many discussions blur the distinction between how individuals behave after a thwarted Golden Gate suicide attempt and how other people might act if we secured the bridge from any potential future suicide attempts. I hope that the following analogy makes this distinction clearer without trivializing: Imagine that we know that everyone who was interrupted while eating their dinner in a particular restaurant never went back and ate out anywhere, ever again. We could not conclude from this that another individual, who learned that the intended restaurant was indefinitely closed, would never dine out again. Once effective suicide barriers exist on the Golden Gate, this will likely become widely known, thereby greatly reducing the likelihood that any individuals will consider the possibility of jumping from the bridge. But it seems very unlikely that this would vanquish all suicidal impulses from the northern California population.
Lessons from patients
Two former patients of mine ended their lives by suicide from the Golden Gate. P, a solitary and lonely man in his 50s, was referred to me by his neighbor, Q, one of my long-term patients. P had a history of repeated assessments for lifelong depression, with minimal follow-up. I made a treatment plan with P that we hoped would address both his depression and his reluctance to engage with mental health professionals. He did not return for his follow-up appointment and ignored all my attempts to contact him.
P continued to have intermittent contact with Q. A decade after I had evaluated him, P was finally hospitalized for depression. Since P had no local family or friends, he asked Q to pick him up from the hospital at the time of his discharge. P asked Q to drive him to the Golden Gate Bridge, ostensibly to relish his release by partaking of the panoramic view of San Francisco from the bridge. They parked in the lot at the north end of the bridge, where Q stayed with the car at the vista point. The last that anyone saw of P was when Q noticed him walking on the bridge; nobody saw him go over and his body was not recovered.
In contrast to my brief connection with P, I worked with S over the course of 8 years to deal with her very severe attention-deficit/hyperactivity disorder and associated depression, which destroyed jobs and friendships, and estranged her from her family. She moved to Hawaii in hopes of “starting over with less baggage,” but I received a few phone calls over the next few years detailing suicide attempts, including driving her car off a bridge. Floundering in life, she returned to San Francisco and was hospitalized with suicidal ideation. The inpatient team sedated her heavily, ignored her past treatments and diagnoses, and discharged her after several days. Within a day of discharge, S’s sister called to say that S’s body had been recovered from the water below the bridge.
I don’t think that suicide was inevitable for either P or S, but I also lack any indication that either would be alive today had we installed suicide barriers on the Golden Gate years ago. Unless we eliminate access to guns, cars, trains, poisons, ropes, tall buildings and cliffs, people contemplating suicide will have numerous options at their disposal. We are likely to save lives by continuing to find ways to restrict access to means of death that can be used within seconds and have a high degree of lethality, and we should persist with such efforts. Buying a $5 trigger lock for every gun in California, and spending tens of millions on a public service campaign would cost less and may well save more lives than the Golden Gate suicide barrier. Unfortunately, we still possess very limited knowledge regarding which suicide prevention measures have an “impact on actual deaths or behavior.”28
To increase our efficacy in reducing suicide, we need to find better treatments for depression and anxiety. We also need to identify better ways of targeting those most at risk for suicide,29 improve our delivery of such treatments, and mitigate the social factors that contribute to such misery and unhappiness.
As a psychiatrist who has lost not only patients but also family members to suicide, I appreciate the hole in the soul these deaths create. I understand the drive to find ways to prevent additional deaths and save future survivors from such grief. But we must design psychiatric interventions that do the maximum good. To be imprecise in the lessons we learn from those who have killed themselves doubles down on the disservice to those lives already lost.
Dr. Kruse is a psychiatrist who practices in San Francisco. Several key details about the patients were changed to protect confidentiality.
References
1. Frommer’s Comprehensive Travel Guide, California. New York: Prentice Hall Travel, 1993.
2. “Chen Si, the ‘Angel of Nanjing,’ has saved more than 330 people from suicide,” by Matt Young, News.com.au. May 14, 2017.
3. “Finding Kyle,” by Lizzie Johnson, San Francisco Chronicle. Feb 8, 2019.
4. Beautrais A. Suicide by jumping. A review of research and prevention strategies. Crisis. 2007 Jan;28 Suppl 1:58-63. Crisis: The J of Crisis Interven Suicide Preven. 2007 Jan. (28)[Suppl1]:58-63.
5. Gunnell D et al. The global distribution of fatal pesticide self-poisoning: Systematic review. BMC Public Health. 2007 Dec. 21;7:357.
6. Vijayakumar L and Satheesh-Babu R. Does ‘no pesticide’ reduce suicides? Int J Soc Psychiatry. 2009 Jul 17;55:401-6.
7. Kreitman N. The coal gas story. United Kingdom suicide rates, 1960-71. Br J Prev Soc Med. 1976 Jun;30(2)86-93.
8. Ajdacic-Gross V et al. Changing times: A longitudinal analysis of international firearm suicide data. Am J Public Health. 2006 Oct;96(10):1752-5.
9. Reisch T et al. Change in suicide rates in Switzerland before and after firearm restriction resulting from the 2003 “Army XXI” reform. Am J Psychiatry. 2013 Sep170(9):977-84.
10. Lubin G et al. Decrease in suicide rates after a change of policy reducing access to firearms in adolescents: A naturalistic epidemiological study. Suicide Life Threat Behav. 2010 Oct;40(5):421-4.
11. Sinyor M and Levitt A. Effect of a barrier at Bloor Street Viaduct on suicide rates in Toronto: Natural experiment BMJ. 2010;341. doi: 1136/bmjc2884.
12. O’Carroll P and Silverman M. Community suicide prevention: The effectiveness of bridge barriers. Suicide Life Threat Behav. 1994 Spring;24(1):89-91; discussion 91-9.
13. Pelletier A. Preventing suicide by jumping: The effect of a bridge safety fence. Inj Prev. 2007 Feb;13(1):57-9.
14. Bennewith O et al. Effect of barriers on the Clifton suspension bridge, England, on local patterns of suicide: Implications for prevention. Br J Psychiatry. 2007 Mar;190:266-7.
15. Harvard T.H. Chan School of Public Health. 2004. “How do people most commonly complete suicide?”
16. “How cliff diving works,” by Heather Kolich, HowStuffWorks.com. Oct 5, 2009.
17. “Bridge design and construction statistics.” Goldengate.org
18. “How did teen survive fall from Golden Gate Bridge?” by Remy Molina, Live Science. Apr 19, 2011.
19. Seiden R. Where are they now? A follow-up study of suicide attempters from the Golden Gate Bridge. Suicide Life Threat Behav. 1978 Winter;8(4):203-16.
20. Presidio demographics. Point2homes.com.
21. Baca-García E et al. A prospective study of the paradoxical relationship between impulsivity and lethality of suicide attempts. J Clin Psychiatry. 2001 Jul;62(7):560-4.
22. Lim M et al. Differences between impulsive and non-impulsive suicide attempts among individuals treated in emergency rooms of South Korea. Psychiatry Investig. 2016 Jul;13(4):389-96.
23. Simon O et al. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001;32(1 Suppl):49-59.
24. Anestis M et al. Reconsidering the link between impulsivity and suicidal behavior. Pers Soc Psychol Rev. 2014 Nov;18(4):366-86.
25. Ostamo A et al. Excess mortality of suicide attempters. Psychiatry Psychiatr Epidemiol. 2001 Jan;36(1):29-35.
26. Leon A et al. Statistical issues in the identification of risk factors for suicidal behavior: The application of survival analysis. Psychiatry Res. 1990 Jan;31(1):99-108.
27. Bostwick J et al. Suicide attempt as a risk factor for completed suicide: Even more lethal than we knew. Am J Psychiatry. 2016 Nov 1;173(11):1094-100.
28. Stone D and Crosby A. Suicide prevention. Am J Lifestyle Med. 2014;8(6):404-20.
29. Belsher B et al. Prediction models for suicide attempts and deaths: A systematic review and simulation. JAMA Psychiatry. 2019 Mar 13. doi: 10.1001/jamapsychiatry.2019.0174.
Ultimately, we need to find better treatments for depression and anxiety
Ultimately, we need to find better treatments for depression and anxiety
San Francisco entrances people. Photographers capture more images of the Golden Gate Bridge than any other bridge in the world.1 And only the Nanjing Yangtze River Bridge in China surpasses the Golden Gate as a destination for dying by suicide.2 At least 1,700 people reportedly have plunged from the bridge to their deaths since its opening in 1937.3
Despite concerted efforts by bridge security, the local mental health community, and a volunteer organization – Bridgewatch Angels – suicides continue at the pace of about 1 every 2 weeks. After more than 60 years of discussion, transportation officials allocated funding and have started building a suicide prevention barrier system on the Golden Gate.
Extrapolating from the success of barriers built on other bridges that were “suicide magnets,” we should be able to assure people that suicide deaths from the Golden Gate will dramatically decrease, and perhaps cease completely.4 Certainly, some in the mental health community think this barrier will save lives. They support this claim by citing research showing that removing highly accessible and lethal means of suicide reduces overall suicide rates, and that suicidal individuals, when thwarted, do not seek alternate modes of death.
I support building the Golden Gate suicide barrier, partly because symbolically, it should deliver a powerful message that we value all human life. But will the barrier save lives? I don’t think it will. As the American Psychiatric Association prepares to gather for its annual meeting in San Francisco, I would like to share my reasoning.
What the evidence shows
The most robust evidence that restricting availability of highly lethal and accessible means of suicide reduces overall suicide deaths comes from studies looking at self-poisoning in Asian countries and Great Britain. In many parts of Asia, ingestion of pesticides constitutes a significant proportion of suicide deaths, and several studies have found that, in localities where sales of highly lethal pesticides were restricted, overall suicide deaths decreased.5,6 Conversely, suicide rates increased when more lethal varieties of pesticides became more available. In Great Britain, overall suicide rates decreased when natural gas replaced coal gas for home heating and cooking.7 For decades preceding this change, more Britons had killed themselves by inhaling coal gas than by any other method.
Strong correlations exist between regional levels of gun ownership and suicide rates by shooting,8 but several potentially confounding sociopolitical factors explain some portion of this connection. Stronger evidence of gun availability affecting suicide rates has been demonstrated by decreases in suicide rates after restrictions in gun access in Switzerland,9 Israel,10 and other areas. These studies show correlations – not causality. However, the number of studies, links between increases and decreases in suicide rates with changes in access to guns, absence of changes in suicide rates during the same time periods among ostensibly similar control populations, and lack of other compelling explanations support the argument that restricting access to highly lethal and accessible means of suicide prevents suicide deaths overall.
The installation of suicide barriers on bridges that have been the sites of multiple suicides robustly reduces or even eliminates suicide deaths from those bridges,11 but the effect on overall suicide rates remains less clear. Various studies have found subsequent increases or no changes12-14 in suicide deaths from other bridges or tall buildings in the vicinity after the installation of suicide barriers on a “suicide magnet.” Many of the studies failed to find any impact on overall suicide rates in the regions investigated. Deaths from jumping off tall structures constitute a tiny proportion of total suicide deaths, making it difficult to detect any changes in overall suicide rates. In the United States, suicides by jumping/falling constituted 1%-2% of total suicides over the last several decades.15
If we know that restricting highly lethal and accessible methods of killing reduces suicide deaths, why would I question the value of the Golden Gate suicide barrier in preventing overall suicide deaths?
Unique aspects of the bridge
The World High Dive Federation recommends keeping dives to less than 20 meters (65.5 feet), with a few exceptions.16 The rail of the Golden Gate Bridge stands 67 meters (220 feet) above the water, and assuming minimal wind resistance, a falling person traverses that distance in about 3.7 seconds and lands with an impact of 130 km/hour (81 miles per hour).17 Only about 1%-2% of those jumping from the Golden Gate survive that fall.18
A 99% likelihood of death sounds pretty lethal; however, death by jumping from the Golden Gate inherently takes place in a public space, with the opportunity for interventions by other people. A more realistic calculation of the lethality would start the instant that someone initiates a sequence of behaviors leading to the intended death. By that criteria, measuring the lethality of the Golden Gate would begin when an individual enters a vehicle or sets off on foot with the plan of going over the railing.
Unless our surveillance-oriented society makes substantially greater advances (which I oppose), we will remain unable to assess suicide lethality by starting at the moment of inception. However, we do have data showing what happens once someone with suicidal intentions walks onto the bridge.
Between 2000 and 2018, observers noted 2,546 people on the Golden Gate who appeared to be considering a suicide attempt, the San Francisco Chronicle has reported. Five hundred sixty-four confirmed suicides occurred. In an additional 71 cases, suicide is presumed but bodies were not recovered. In the 1,911 remaining instances, mental health interventions were made, with individuals taken to local hospitals and psychiatric wards, and released when no longer overtly suicidal. Interventions successfully diverted 75% (1,911/2,546) of those intending to end their own lives, which suggests that the current lethality of the Golden Gate as a means of suicide is only 25%. Even in the bridge’s first half-century, without constant camera monitoring, and a cadre of volunteers and professionals scanning for those attempting suicide, the lethality rate approached about 50%.19
We face even more difficulties measuring accessibility than in determining lethality. The Golden Gate appears to be accessible to almost anyone – drivers have to pay a toll only when traveling from the north, and then only after they have traversed the span. Pedestrians retain unfettered admittance to the east sidewalk (facing San Francisco city and bay) throughout daylight hours. But any determination of accessibility must include how quickly and easily one can make use of an opportunity.
Both entrances to the Golden Gate are embedded in the Golden Gate National Recreational Area, part of the National Park system. The south entrance to the bridge arises from The Presidio, a former military installation that housed about 4,000 people.20 Even fewer people live in the parklands at the north end of the bridge. The Presidio extends far enough so that the closest San Francisco neighborhoods outside of the park are a full 2.2 km (1.36 miles) from the bridge railing. A brisk walk would still require a minimum of about 20 minutes to get to the bridge; it is difficult to arrive at the bridge without a trek.
Researchers define impulsivity, like accessibility, inconsistently – and often imprecisely. Impulsivity, which clearly exists on a spectrum, connotes overvaluing of immediate feelings and thoughts at the expense of longer term goals and aspirations. Some suicide research appears to define impulsivity as the antithesis of planned behavior;21,22 others define it pragmatically as behaviors executed within 5 minutes of a decision,23 and still others contend that “suicidal behavior is rarely if ever impulsive.”24 Furthermore, when we assess impulsivity, we must acknowledge a fundamental difference between “impulsive” shootings and poisonings that are accomplished at home and within seconds or minutes, from “impulsive” Golden Gate Bridge suicide attempts, which require substantial travel and time commitments, and inherently involve the potential for others to intervene.
Those arguing that the bridge suicide barrier will save lives often bring up two additional sets of numbers to back up their assertions. They provide evidence that most of those people who were stopped in their attempts at suicide at the Golden Gate do not go on to commit suicide elsewhere, and that many of those who survived their attempts express regret at having tried to kill themselves. Specifically, 94% of those who were prevented from jumping from the Golden Gate had not committed suicide after a median follow-up of 26 years, according to a follow-up study published a few years ago. On the other hand, those who have made a serious suicide attempt have a substantially increased risk, relative to the general population, of dying from a later attempt,25,26 and the strongest predictor for death by suicide is having made a previous, serious suicide attempt.27
While all of these studies provide important and interesting information regarding suicide, none directly address the question of whether individuals will substitute attempts by other methods if the Golden Gate Bridge were no longer available. Many discussions blur the distinction between how individuals behave after a thwarted Golden Gate suicide attempt and how other people might act if we secured the bridge from any potential future suicide attempts. I hope that the following analogy makes this distinction clearer without trivializing: Imagine that we know that everyone who was interrupted while eating their dinner in a particular restaurant never went back and ate out anywhere, ever again. We could not conclude from this that another individual, who learned that the intended restaurant was indefinitely closed, would never dine out again. Once effective suicide barriers exist on the Golden Gate, this will likely become widely known, thereby greatly reducing the likelihood that any individuals will consider the possibility of jumping from the bridge. But it seems very unlikely that this would vanquish all suicidal impulses from the northern California population.
Lessons from patients
Two former patients of mine ended their lives by suicide from the Golden Gate. P, a solitary and lonely man in his 50s, was referred to me by his neighbor, Q, one of my long-term patients. P had a history of repeated assessments for lifelong depression, with minimal follow-up. I made a treatment plan with P that we hoped would address both his depression and his reluctance to engage with mental health professionals. He did not return for his follow-up appointment and ignored all my attempts to contact him.
P continued to have intermittent contact with Q. A decade after I had evaluated him, P was finally hospitalized for depression. Since P had no local family or friends, he asked Q to pick him up from the hospital at the time of his discharge. P asked Q to drive him to the Golden Gate Bridge, ostensibly to relish his release by partaking of the panoramic view of San Francisco from the bridge. They parked in the lot at the north end of the bridge, where Q stayed with the car at the vista point. The last that anyone saw of P was when Q noticed him walking on the bridge; nobody saw him go over and his body was not recovered.
In contrast to my brief connection with P, I worked with S over the course of 8 years to deal with her very severe attention-deficit/hyperactivity disorder and associated depression, which destroyed jobs and friendships, and estranged her from her family. She moved to Hawaii in hopes of “starting over with less baggage,” but I received a few phone calls over the next few years detailing suicide attempts, including driving her car off a bridge. Floundering in life, she returned to San Francisco and was hospitalized with suicidal ideation. The inpatient team sedated her heavily, ignored her past treatments and diagnoses, and discharged her after several days. Within a day of discharge, S’s sister called to say that S’s body had been recovered from the water below the bridge.
I don’t think that suicide was inevitable for either P or S, but I also lack any indication that either would be alive today had we installed suicide barriers on the Golden Gate years ago. Unless we eliminate access to guns, cars, trains, poisons, ropes, tall buildings and cliffs, people contemplating suicide will have numerous options at their disposal. We are likely to save lives by continuing to find ways to restrict access to means of death that can be used within seconds and have a high degree of lethality, and we should persist with such efforts. Buying a $5 trigger lock for every gun in California, and spending tens of millions on a public service campaign would cost less and may well save more lives than the Golden Gate suicide barrier. Unfortunately, we still possess very limited knowledge regarding which suicide prevention measures have an “impact on actual deaths or behavior.”28
To increase our efficacy in reducing suicide, we need to find better treatments for depression and anxiety. We also need to identify better ways of targeting those most at risk for suicide,29 improve our delivery of such treatments, and mitigate the social factors that contribute to such misery and unhappiness.
As a psychiatrist who has lost not only patients but also family members to suicide, I appreciate the hole in the soul these deaths create. I understand the drive to find ways to prevent additional deaths and save future survivors from such grief. But we must design psychiatric interventions that do the maximum good. To be imprecise in the lessons we learn from those who have killed themselves doubles down on the disservice to those lives already lost.
Dr. Kruse is a psychiatrist who practices in San Francisco. Several key details about the patients were changed to protect confidentiality.
References
1. Frommer’s Comprehensive Travel Guide, California. New York: Prentice Hall Travel, 1993.
2. “Chen Si, the ‘Angel of Nanjing,’ has saved more than 330 people from suicide,” by Matt Young, News.com.au. May 14, 2017.
3. “Finding Kyle,” by Lizzie Johnson, San Francisco Chronicle. Feb 8, 2019.
4. Beautrais A. Suicide by jumping. A review of research and prevention strategies. Crisis. 2007 Jan;28 Suppl 1:58-63. Crisis: The J of Crisis Interven Suicide Preven. 2007 Jan. (28)[Suppl1]:58-63.
5. Gunnell D et al. The global distribution of fatal pesticide self-poisoning: Systematic review. BMC Public Health. 2007 Dec. 21;7:357.
6. Vijayakumar L and Satheesh-Babu R. Does ‘no pesticide’ reduce suicides? Int J Soc Psychiatry. 2009 Jul 17;55:401-6.
7. Kreitman N. The coal gas story. United Kingdom suicide rates, 1960-71. Br J Prev Soc Med. 1976 Jun;30(2)86-93.
8. Ajdacic-Gross V et al. Changing times: A longitudinal analysis of international firearm suicide data. Am J Public Health. 2006 Oct;96(10):1752-5.
9. Reisch T et al. Change in suicide rates in Switzerland before and after firearm restriction resulting from the 2003 “Army XXI” reform. Am J Psychiatry. 2013 Sep170(9):977-84.
10. Lubin G et al. Decrease in suicide rates after a change of policy reducing access to firearms in adolescents: A naturalistic epidemiological study. Suicide Life Threat Behav. 2010 Oct;40(5):421-4.
11. Sinyor M and Levitt A. Effect of a barrier at Bloor Street Viaduct on suicide rates in Toronto: Natural experiment BMJ. 2010;341. doi: 1136/bmjc2884.
12. O’Carroll P and Silverman M. Community suicide prevention: The effectiveness of bridge barriers. Suicide Life Threat Behav. 1994 Spring;24(1):89-91; discussion 91-9.
13. Pelletier A. Preventing suicide by jumping: The effect of a bridge safety fence. Inj Prev. 2007 Feb;13(1):57-9.
14. Bennewith O et al. Effect of barriers on the Clifton suspension bridge, England, on local patterns of suicide: Implications for prevention. Br J Psychiatry. 2007 Mar;190:266-7.
15. Harvard T.H. Chan School of Public Health. 2004. “How do people most commonly complete suicide?”
16. “How cliff diving works,” by Heather Kolich, HowStuffWorks.com. Oct 5, 2009.
17. “Bridge design and construction statistics.” Goldengate.org
18. “How did teen survive fall from Golden Gate Bridge?” by Remy Molina, Live Science. Apr 19, 2011.
19. Seiden R. Where are they now? A follow-up study of suicide attempters from the Golden Gate Bridge. Suicide Life Threat Behav. 1978 Winter;8(4):203-16.
20. Presidio demographics. Point2homes.com.
21. Baca-García E et al. A prospective study of the paradoxical relationship between impulsivity and lethality of suicide attempts. J Clin Psychiatry. 2001 Jul;62(7):560-4.
22. Lim M et al. Differences between impulsive and non-impulsive suicide attempts among individuals treated in emergency rooms of South Korea. Psychiatry Investig. 2016 Jul;13(4):389-96.
23. Simon O et al. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001;32(1 Suppl):49-59.
24. Anestis M et al. Reconsidering the link between impulsivity and suicidal behavior. Pers Soc Psychol Rev. 2014 Nov;18(4):366-86.
25. Ostamo A et al. Excess mortality of suicide attempters. Psychiatry Psychiatr Epidemiol. 2001 Jan;36(1):29-35.
26. Leon A et al. Statistical issues in the identification of risk factors for suicidal behavior: The application of survival analysis. Psychiatry Res. 1990 Jan;31(1):99-108.
27. Bostwick J et al. Suicide attempt as a risk factor for completed suicide: Even more lethal than we knew. Am J Psychiatry. 2016 Nov 1;173(11):1094-100.
28. Stone D and Crosby A. Suicide prevention. Am J Lifestyle Med. 2014;8(6):404-20.
29. Belsher B et al. Prediction models for suicide attempts and deaths: A systematic review and simulation. JAMA Psychiatry. 2019 Mar 13. doi: 10.1001/jamapsychiatry.2019.0174.
San Francisco entrances people. Photographers capture more images of the Golden Gate Bridge than any other bridge in the world.1 And only the Nanjing Yangtze River Bridge in China surpasses the Golden Gate as a destination for dying by suicide.2 At least 1,700 people reportedly have plunged from the bridge to their deaths since its opening in 1937.3
Despite concerted efforts by bridge security, the local mental health community, and a volunteer organization – Bridgewatch Angels – suicides continue at the pace of about 1 every 2 weeks. After more than 60 years of discussion, transportation officials allocated funding and have started building a suicide prevention barrier system on the Golden Gate.
Extrapolating from the success of barriers built on other bridges that were “suicide magnets,” we should be able to assure people that suicide deaths from the Golden Gate will dramatically decrease, and perhaps cease completely.4 Certainly, some in the mental health community think this barrier will save lives. They support this claim by citing research showing that removing highly accessible and lethal means of suicide reduces overall suicide rates, and that suicidal individuals, when thwarted, do not seek alternate modes of death.
I support building the Golden Gate suicide barrier, partly because symbolically, it should deliver a powerful message that we value all human life. But will the barrier save lives? I don’t think it will. As the American Psychiatric Association prepares to gather for its annual meeting in San Francisco, I would like to share my reasoning.
What the evidence shows
The most robust evidence that restricting availability of highly lethal and accessible means of suicide reduces overall suicide deaths comes from studies looking at self-poisoning in Asian countries and Great Britain. In many parts of Asia, ingestion of pesticides constitutes a significant proportion of suicide deaths, and several studies have found that, in localities where sales of highly lethal pesticides were restricted, overall suicide deaths decreased.5,6 Conversely, suicide rates increased when more lethal varieties of pesticides became more available. In Great Britain, overall suicide rates decreased when natural gas replaced coal gas for home heating and cooking.7 For decades preceding this change, more Britons had killed themselves by inhaling coal gas than by any other method.
Strong correlations exist between regional levels of gun ownership and suicide rates by shooting,8 but several potentially confounding sociopolitical factors explain some portion of this connection. Stronger evidence of gun availability affecting suicide rates has been demonstrated by decreases in suicide rates after restrictions in gun access in Switzerland,9 Israel,10 and other areas. These studies show correlations – not causality. However, the number of studies, links between increases and decreases in suicide rates with changes in access to guns, absence of changes in suicide rates during the same time periods among ostensibly similar control populations, and lack of other compelling explanations support the argument that restricting access to highly lethal and accessible means of suicide prevents suicide deaths overall.
The installation of suicide barriers on bridges that have been the sites of multiple suicides robustly reduces or even eliminates suicide deaths from those bridges,11 but the effect on overall suicide rates remains less clear. Various studies have found subsequent increases or no changes12-14 in suicide deaths from other bridges or tall buildings in the vicinity after the installation of suicide barriers on a “suicide magnet.” Many of the studies failed to find any impact on overall suicide rates in the regions investigated. Deaths from jumping off tall structures constitute a tiny proportion of total suicide deaths, making it difficult to detect any changes in overall suicide rates. In the United States, suicides by jumping/falling constituted 1%-2% of total suicides over the last several decades.15
If we know that restricting highly lethal and accessible methods of killing reduces suicide deaths, why would I question the value of the Golden Gate suicide barrier in preventing overall suicide deaths?
Unique aspects of the bridge
The World High Dive Federation recommends keeping dives to less than 20 meters (65.5 feet), with a few exceptions.16 The rail of the Golden Gate Bridge stands 67 meters (220 feet) above the water, and assuming minimal wind resistance, a falling person traverses that distance in about 3.7 seconds and lands with an impact of 130 km/hour (81 miles per hour).17 Only about 1%-2% of those jumping from the Golden Gate survive that fall.18
A 99% likelihood of death sounds pretty lethal; however, death by jumping from the Golden Gate inherently takes place in a public space, with the opportunity for interventions by other people. A more realistic calculation of the lethality would start the instant that someone initiates a sequence of behaviors leading to the intended death. By that criteria, measuring the lethality of the Golden Gate would begin when an individual enters a vehicle or sets off on foot with the plan of going over the railing.
Unless our surveillance-oriented society makes substantially greater advances (which I oppose), we will remain unable to assess suicide lethality by starting at the moment of inception. However, we do have data showing what happens once someone with suicidal intentions walks onto the bridge.
Between 2000 and 2018, observers noted 2,546 people on the Golden Gate who appeared to be considering a suicide attempt, the San Francisco Chronicle has reported. Five hundred sixty-four confirmed suicides occurred. In an additional 71 cases, suicide is presumed but bodies were not recovered. In the 1,911 remaining instances, mental health interventions were made, with individuals taken to local hospitals and psychiatric wards, and released when no longer overtly suicidal. Interventions successfully diverted 75% (1,911/2,546) of those intending to end their own lives, which suggests that the current lethality of the Golden Gate as a means of suicide is only 25%. Even in the bridge’s first half-century, without constant camera monitoring, and a cadre of volunteers and professionals scanning for those attempting suicide, the lethality rate approached about 50%.19
We face even more difficulties measuring accessibility than in determining lethality. The Golden Gate appears to be accessible to almost anyone – drivers have to pay a toll only when traveling from the north, and then only after they have traversed the span. Pedestrians retain unfettered admittance to the east sidewalk (facing San Francisco city and bay) throughout daylight hours. But any determination of accessibility must include how quickly and easily one can make use of an opportunity.
Both entrances to the Golden Gate are embedded in the Golden Gate National Recreational Area, part of the National Park system. The south entrance to the bridge arises from The Presidio, a former military installation that housed about 4,000 people.20 Even fewer people live in the parklands at the north end of the bridge. The Presidio extends far enough so that the closest San Francisco neighborhoods outside of the park are a full 2.2 km (1.36 miles) from the bridge railing. A brisk walk would still require a minimum of about 20 minutes to get to the bridge; it is difficult to arrive at the bridge without a trek.
Researchers define impulsivity, like accessibility, inconsistently – and often imprecisely. Impulsivity, which clearly exists on a spectrum, connotes overvaluing of immediate feelings and thoughts at the expense of longer term goals and aspirations. Some suicide research appears to define impulsivity as the antithesis of planned behavior;21,22 others define it pragmatically as behaviors executed within 5 minutes of a decision,23 and still others contend that “suicidal behavior is rarely if ever impulsive.”24 Furthermore, when we assess impulsivity, we must acknowledge a fundamental difference between “impulsive” shootings and poisonings that are accomplished at home and within seconds or minutes, from “impulsive” Golden Gate Bridge suicide attempts, which require substantial travel and time commitments, and inherently involve the potential for others to intervene.
Those arguing that the bridge suicide barrier will save lives often bring up two additional sets of numbers to back up their assertions. They provide evidence that most of those people who were stopped in their attempts at suicide at the Golden Gate do not go on to commit suicide elsewhere, and that many of those who survived their attempts express regret at having tried to kill themselves. Specifically, 94% of those who were prevented from jumping from the Golden Gate had not committed suicide after a median follow-up of 26 years, according to a follow-up study published a few years ago. On the other hand, those who have made a serious suicide attempt have a substantially increased risk, relative to the general population, of dying from a later attempt,25,26 and the strongest predictor for death by suicide is having made a previous, serious suicide attempt.27
While all of these studies provide important and interesting information regarding suicide, none directly address the question of whether individuals will substitute attempts by other methods if the Golden Gate Bridge were no longer available. Many discussions blur the distinction between how individuals behave after a thwarted Golden Gate suicide attempt and how other people might act if we secured the bridge from any potential future suicide attempts. I hope that the following analogy makes this distinction clearer without trivializing: Imagine that we know that everyone who was interrupted while eating their dinner in a particular restaurant never went back and ate out anywhere, ever again. We could not conclude from this that another individual, who learned that the intended restaurant was indefinitely closed, would never dine out again. Once effective suicide barriers exist on the Golden Gate, this will likely become widely known, thereby greatly reducing the likelihood that any individuals will consider the possibility of jumping from the bridge. But it seems very unlikely that this would vanquish all suicidal impulses from the northern California population.
Lessons from patients
Two former patients of mine ended their lives by suicide from the Golden Gate. P, a solitary and lonely man in his 50s, was referred to me by his neighbor, Q, one of my long-term patients. P had a history of repeated assessments for lifelong depression, with minimal follow-up. I made a treatment plan with P that we hoped would address both his depression and his reluctance to engage with mental health professionals. He did not return for his follow-up appointment and ignored all my attempts to contact him.
P continued to have intermittent contact with Q. A decade after I had evaluated him, P was finally hospitalized for depression. Since P had no local family or friends, he asked Q to pick him up from the hospital at the time of his discharge. P asked Q to drive him to the Golden Gate Bridge, ostensibly to relish his release by partaking of the panoramic view of San Francisco from the bridge. They parked in the lot at the north end of the bridge, where Q stayed with the car at the vista point. The last that anyone saw of P was when Q noticed him walking on the bridge; nobody saw him go over and his body was not recovered.
In contrast to my brief connection with P, I worked with S over the course of 8 years to deal with her very severe attention-deficit/hyperactivity disorder and associated depression, which destroyed jobs and friendships, and estranged her from her family. She moved to Hawaii in hopes of “starting over with less baggage,” but I received a few phone calls over the next few years detailing suicide attempts, including driving her car off a bridge. Floundering in life, she returned to San Francisco and was hospitalized with suicidal ideation. The inpatient team sedated her heavily, ignored her past treatments and diagnoses, and discharged her after several days. Within a day of discharge, S’s sister called to say that S’s body had been recovered from the water below the bridge.
I don’t think that suicide was inevitable for either P or S, but I also lack any indication that either would be alive today had we installed suicide barriers on the Golden Gate years ago. Unless we eliminate access to guns, cars, trains, poisons, ropes, tall buildings and cliffs, people contemplating suicide will have numerous options at their disposal. We are likely to save lives by continuing to find ways to restrict access to means of death that can be used within seconds and have a high degree of lethality, and we should persist with such efforts. Buying a $5 trigger lock for every gun in California, and spending tens of millions on a public service campaign would cost less and may well save more lives than the Golden Gate suicide barrier. Unfortunately, we still possess very limited knowledge regarding which suicide prevention measures have an “impact on actual deaths or behavior.”28
To increase our efficacy in reducing suicide, we need to find better treatments for depression and anxiety. We also need to identify better ways of targeting those most at risk for suicide,29 improve our delivery of such treatments, and mitigate the social factors that contribute to such misery and unhappiness.
As a psychiatrist who has lost not only patients but also family members to suicide, I appreciate the hole in the soul these deaths create. I understand the drive to find ways to prevent additional deaths and save future survivors from such grief. But we must design psychiatric interventions that do the maximum good. To be imprecise in the lessons we learn from those who have killed themselves doubles down on the disservice to those lives already lost.
Dr. Kruse is a psychiatrist who practices in San Francisco. Several key details about the patients were changed to protect confidentiality.
References
1. Frommer’s Comprehensive Travel Guide, California. New York: Prentice Hall Travel, 1993.
2. “Chen Si, the ‘Angel of Nanjing,’ has saved more than 330 people from suicide,” by Matt Young, News.com.au. May 14, 2017.
3. “Finding Kyle,” by Lizzie Johnson, San Francisco Chronicle. Feb 8, 2019.
4. Beautrais A. Suicide by jumping. A review of research and prevention strategies. Crisis. 2007 Jan;28 Suppl 1:58-63. Crisis: The J of Crisis Interven Suicide Preven. 2007 Jan. (28)[Suppl1]:58-63.
5. Gunnell D et al. The global distribution of fatal pesticide self-poisoning: Systematic review. BMC Public Health. 2007 Dec. 21;7:357.
6. Vijayakumar L and Satheesh-Babu R. Does ‘no pesticide’ reduce suicides? Int J Soc Psychiatry. 2009 Jul 17;55:401-6.
7. Kreitman N. The coal gas story. United Kingdom suicide rates, 1960-71. Br J Prev Soc Med. 1976 Jun;30(2)86-93.
8. Ajdacic-Gross V et al. Changing times: A longitudinal analysis of international firearm suicide data. Am J Public Health. 2006 Oct;96(10):1752-5.
9. Reisch T et al. Change in suicide rates in Switzerland before and after firearm restriction resulting from the 2003 “Army XXI” reform. Am J Psychiatry. 2013 Sep170(9):977-84.
10. Lubin G et al. Decrease in suicide rates after a change of policy reducing access to firearms in adolescents: A naturalistic epidemiological study. Suicide Life Threat Behav. 2010 Oct;40(5):421-4.
11. Sinyor M and Levitt A. Effect of a barrier at Bloor Street Viaduct on suicide rates in Toronto: Natural experiment BMJ. 2010;341. doi: 1136/bmjc2884.
12. O’Carroll P and Silverman M. Community suicide prevention: The effectiveness of bridge barriers. Suicide Life Threat Behav. 1994 Spring;24(1):89-91; discussion 91-9.
13. Pelletier A. Preventing suicide by jumping: The effect of a bridge safety fence. Inj Prev. 2007 Feb;13(1):57-9.
14. Bennewith O et al. Effect of barriers on the Clifton suspension bridge, England, on local patterns of suicide: Implications for prevention. Br J Psychiatry. 2007 Mar;190:266-7.
15. Harvard T.H. Chan School of Public Health. 2004. “How do people most commonly complete suicide?”
16. “How cliff diving works,” by Heather Kolich, HowStuffWorks.com. Oct 5, 2009.
17. “Bridge design and construction statistics.” Goldengate.org
18. “How did teen survive fall from Golden Gate Bridge?” by Remy Molina, Live Science. Apr 19, 2011.
19. Seiden R. Where are they now? A follow-up study of suicide attempters from the Golden Gate Bridge. Suicide Life Threat Behav. 1978 Winter;8(4):203-16.
20. Presidio demographics. Point2homes.com.
21. Baca-García E et al. A prospective study of the paradoxical relationship between impulsivity and lethality of suicide attempts. J Clin Psychiatry. 2001 Jul;62(7):560-4.
22. Lim M et al. Differences between impulsive and non-impulsive suicide attempts among individuals treated in emergency rooms of South Korea. Psychiatry Investig. 2016 Jul;13(4):389-96.
23. Simon O et al. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001;32(1 Suppl):49-59.
24. Anestis M et al. Reconsidering the link between impulsivity and suicidal behavior. Pers Soc Psychol Rev. 2014 Nov;18(4):366-86.
25. Ostamo A et al. Excess mortality of suicide attempters. Psychiatry Psychiatr Epidemiol. 2001 Jan;36(1):29-35.
26. Leon A et al. Statistical issues in the identification of risk factors for suicidal behavior: The application of survival analysis. Psychiatry Res. 1990 Jan;31(1):99-108.
27. Bostwick J et al. Suicide attempt as a risk factor for completed suicide: Even more lethal than we knew. Am J Psychiatry. 2016 Nov 1;173(11):1094-100.
28. Stone D and Crosby A. Suicide prevention. Am J Lifestyle Med. 2014;8(6):404-20.
29. Belsher B et al. Prediction models for suicide attempts and deaths: A systematic review and simulation. JAMA Psychiatry. 2019 Mar 13. doi: 10.1001/jamapsychiatry.2019.0174.