The Board that oversees San Francisco’s Golden Gate Bridge has authorized the installation of a $76 million suicide prevention net. The board members hope to end the bridge’s status as the Western world’s leading suicide hotspot: Over 1,600 people have leapt to their deaths from the bridge since it was constructed in 1937.
The goal is unquestionably noble. But does imposing barriers to suicide stop desperate people from taking their lives or simply shift those tragedies elsewhere? A half century of experience and evidence supports an optimistic view.
History’s most successful suicide barrier was, like penicillin, a serendipitous British discovery. When the nation’s ovens were upgraded from carbon monoxide-producing coal to gas in the 1960s, the suicide rate fell by about one-third. Use of other means of suicide rose slightly, but nowhere near enough to cancel out the beneficial effect of a readily accessible means of suicide by suffocation being removed from virtually every British home.
The oven conversion experience indicates that some suicide attempts are impulsive rather than planned and can therefore be prevented by erecting simple barriers. Further support for this concept came from a study of 515 individuals who were stopped from jumping from the Golden Gate Bridge by police or passers by. In a long term follow-up study, only 6% of these individuals were found to have died by suicide.
As this lesson was implemented at a range of suicide hotspots around the world, it also became clear that some suicidal people have strong preferences about a particular place that they wish to end their lives. The installation of a suicide prevention fence at the Duke Ellington Bridge in D.C. for example, did not result in thwarted jumpers moving to the adjacent Taft Bridge to take their lives.
However, after research studies linked the installation of barriers to reductions in suicides at hotspots such as the Muenster Terrace in Berne, Switzerland and the Memorial Bridge in Augusta, Maine, a major study in Toronto raised doubt about their value. The study was conducted at the Bloor Street Viaduct, the second most lethal suicide hotspot in North America. After the installation of a barrier, suicides at the Viaduct dropped from about 9 per year to none, but jumping suicides at other bridges and buildings in Toronto went up by a comparable amount.
Because suicide by jumping is a mercifully rare event, most studies of barriers have small samples, making findings unstable and the difference between the Toronto study and other research unsurprising. Statistically, a more reliable result would come from combining the findings across all prior studies. When Dr. Jane Pirkis of the University of Melbourne led such a “meta-analysis” in 2013, she and her colleagues found that on average barriers reduce suicides by 86% at the barrier site, and that jumping suicides at other nearby sites rise by 44%. The net benefit is a 28% decrease in suicides by jumping per year.
Dr. Pirkis’ findings bode well for the success of San Francisco’s suicide barrier, which is expected to be installed in about three years. Even if the net has only the average level of effectiveness, it would have saved a life a month in 2013 alone, as well as sparing the families of the deceased years of mental and emotional anguish.
Keith Humphreys is the Mental Health Policy Director and a Professor of Psychiatry at Stanford University. Follow him on Twitter @KeithNHumphreys