As designs for seven permanent barriers on campus bridges move toward completion, Prof. Garrett Glasgow, U.C. Santa Barbara, political science, said there is no reason to believe that the barriers will save lives, as Cornell has repeatedly suggested.
In a forthcoming study in the Journal of Science and Medicine, Glasgow develops an alternative test for assessing whether barriers simply lead to method substitution — the transfer of suicides elsewhere when one location is blocked off.
Surveying 3116 counties across the United States, Glasgow’s paper found “no positive relationship between [suicide hot spots] and the overall number of suicides” of a given area.
While one will “never be able to prove that barriers will never save anyone ever,” Glasgow said if barriers were effective at reducing overall suicide rates, “the presence of an unsecured suicide-jumping site [would] lead to a higher local suicide rate in comparison to areas without such a site.”
This means even if the barriers reduce the incidence of suicide at the specific site at which they are installed, there is no evidence that they decrease overall suicides rates in a given area.
Gannett Director of Mental Health Initiatives Tim Marchell ’82 and Associate Director of Communications at Gannett Sharon Dittman disagreed with Glasgow on several fronts, particularly for linking the barriers’ effectiveness explicitly with an area’s suicide rate.
“When a barrier reduces the average number of suicides per year at a bridge from one to zero … the impact on the annual suicide rate is not statistically significant. But having 20 fewer deaths over 20 years from that bridge would be a meaningful outcome,” Marchell said in a statement.
Glasgow responded that the manner in which these suicides occurred should not be considered particularly important.
“I don’t think leading 20 people over 20 years to commit suicide by a means other than jumping is a meaningful outcome,” Glasgow said. “If this response is actually meant to argue that a barrier might save 20 people over 20 years, then yes, that would be a meaningful outcome, but there’s no scientific evidence to give us confidence that this would happen.”
Marchell, however, said that several potential effects of jumping suicides off bridges — including psychological trauma to bystanders and risk of suicide contagion — did make eliminating jumping deaths significant.
Marchell and Dittman argued that even if an area’s overall suicide rate does not change, individual lives may be saved.
“Means restriction is an important part of suicide prevention, whether or not it leads to a statistically significant decrease in local suicide rates,” Marchell said. Marchell added that suicide rates for Tompkins County would not be necessarily be reflective of changes at Cornell.
Like Marchell, Dittman said the barriers could still save lives, even if they do not lower local suicide rates.
“He’s talking about rates in a region,” Marchell said of Glasgow. “We’re talking about potential for saving lives in this community.”
While Marchell and Glasgow agreed that no study has proven the efficacy of bridge barriers one way or another, they differed on the significance of the literature on the subject.
“[Glasglow] specifically states that his methodology ‘may eventually allow researchers to determine if means restriction at suicide-jumping sites reduces total suicides,’” Marchell said, reading Glasgow’s forthcoming study. “In other words, in his view the jury is still out.”
Glasgow called Marchell’s statement “partly true.”
While admitting “we may never know the answer,” Glasgow pointed out that “every other study on this topic has produced results similar to mine.”
“No study on suicide barriers has been able to demonstrate a lifesaving effect, and most of them openly acknowledge this,” Glasgow said.
Marchell responded by pointing to Harvard University’s Public Health Means Matter website, which states “research has indicated that erecting a bridge barrier does not result in more jumps from nearby sites.”
“Is there a commensurate increase in deaths by other methods? Evidence is weaker on this point,” the Harvard study states. “Some studies have found no increase, one found some evidence for males and none for females, others either did not examine this or did not have enough statistical power to examine this (since jumps do not make up a large proportion of suicides in most areas).”
Glasgow is not alone in questioning Cornell’s interpretation of the extant data on bridge barriers and suicide prevention.
In January, Dan Jost ’05 published an article in Landscape Architecture that doubted the University’s justification for building the barriers.
In the article, Jost evaluates the impact of various bridge barrier studies cited by Cornell. For instance, the University has cited an article on a bridge barrier placed on the Duke Ellington Bridge in Washington D.C. in 1986.
Jost quotes Tommy Bruce, Cornell’s vice president of communications, as saying that although a barrier placed on the Duke Ellington Bridge reduced suicides at that site, the nearby Taft Bridge did not see an increase in suicides. According to Jost, Bruce viewed this as evidence that the Duke Ellington Bridge barrier saved lives.
“But it seems few people have read the actual report on this subject,” Jost writes. He also quotes the article as saying the “‘the data are not sufficient’” to conclude that the bridge barriers were an effective means of preventing suicide.
“I will leave others to explain why the researchers Cornell has employed didn’t seem to dig deeply enough to see their conclusions in that article were unjustifiable,” Jost told The Sun.
Last July, University of Toronto Prof. Anthony Levitt, psychiatry, published a study on a bridge barrier placed at the Bloor Street Viaduct in Toronto.
Levitt found that while previous studies — including many cited by Cornell — “concluded that little, if any, substitution of [suicide] location occurred … these studies lacked statistical power because of the relatively small yearly decreases in numbers of suicides at each bridge as well as low rates of suicide in general.”
“No study of a suicide prevention barrier has shown a statistically significant drop in overall rates of suicide in the vicinity,” Levitt’s study states.
“There’s plenty of people at Cornell who could read these studies and know what they say, but a lot of times these things are done by administrators or deans,” Glasgow said. “People worry it’s a liability issue or a bad public relations issue and say, ‘lets do whatever it takes to get this built and then it’s not our issue anymore.’”
Is Cornell Different?
Marchell said that, in terms of barrier construction, Cornell is “unique in several ways.”
He said that, “unprecedented in any community in the country,” jumping suicides account for nearly half of all suicides by Cornell students.
“In young adults, suicide is often an ambivalent act. Barriers buy time by slowing down the thoughts of desperate, impulsive individuals,” Marchell said, arguing Glasgow’s study does not address these particularities of Cornell.
Still, Jost viewed Cornell’s distinctive environment as an argument against the barriers and for preserving the original view of the gorges.
“If suicides will continue anyway, we ought to focus on preserving the beauty and life-affirming aspects of the gorges, their ability to relax us and refocus our minds,” Jost said. “Viewing Cornell’s gorges from its bridges was, for many students, their only daily connection with these wonders of nature.”
Marchell said that the barriers have already saved lives since being placed on campus last spring.
“We have at least two documented cases in which bystanders were able to successfully intervene in a suicide attempt because the barriers slowed down individuals who were attempting to jump,” Marchell said.
Thursday night The Sun received an e-mail from Chris Hoeber ’69, who said he was in a similar position more than 40 years ago.
“One day in December ’67 or January ’68 I sat on the ledge on the … bridge and had a very long conversation with myself: I think that it lasted at least a half hour and it involved many passers-by. To jump, or not to jump?” Hoeber said. “I was drunk. I had at least a 50-50 chance of falling off, even if I didn’t want to jump.”
Hoeber, who said he went on to have a “truly wonderful life,” added he was “lucky to have survived that night. When I read about recent student suicides I can put myself in the place of those students.”
Speaking from experience, Hoeber questioned the ability of “people who are addressing legal liability, but who don’t know what the hell they are talking about with respect to preventing suicide,” to make a decision about the bridge barriers.
“If I wanted to kill myself there are obviously many ways to do it, and flinging myself from a bridge or from the side of a bridge both would have worked,” Hoeber said. “There is no way in hell that I would have flung myself from the side of a bridge and tumbled over and over down the walls of a gorge … As an engineer, I know that that could be just as effective as jumping from the bridge in the middle of the gorge, but the decision making process would have been a lot different.”