UCSB Study Debunks CalTrans
Ummmm, several day old news but the just-released UCSB study discredits the CalTrans proposal to install so-called suicide prevention barriers on the Cold Spring Bridge -- or is it now the Chumash Highway Bridge, doh!
It seems more important to me to carry forward carefully-considered, facts-based, and effective suicide prevention measures and also consider any traffic calming techniques that could save lives from driving towards the casino as well. A newly-formed citizens group is calling upon CalTrans to withdraw its proposal to spend up to $1,000,000 to install such barriers. What do you think?
"there is no evidence that a suicide prevention barrier on the Cold Spring Bridge would save lives,"
It seems more important to me to carry forward carefully-considered, facts-based, and effective suicide prevention measures and also consider any traffic calming techniques that could save lives from driving towards the casino as well. A newly-formed citizens group is calling upon CalTrans to withdraw its proposal to spend up to $1,000,000 to install such barriers. What do you think?
Labels: Cold Spring Bridge, San Marcos Pass
posted by Sara De la Guerra | 9/26/2007 11:51:00 PM
Our Community Guidelines ask that you use the Blogger/Open ID login or the
more private Name/URL option when making comments. Comments using "Anonymous" will not be published. For more info, see choosing an identity.
35 Comments:
Glasgow is way more sloppy than Seiden was... Seiden interviewed >500 folks who were suicidal on the GG Bridge, and concluded most didn't go on to commit suicide in a different way.
Glasgow, who didn't interview even 5 suicidal folks, did a sloppy statistical analysis. He didn't even check every bridge he studied to see if it had a suicide barrier!
The Glasgow study doesn't help at all. It is sophistry and semantics, and can't hold a candle to Seiden.
Here we go again with those studies and statistics.
If we narrowly focus on preventing a mere 1 of the 43 lives that were ended by jumping off the bridge then I would presume the 1 million dollar price tag would still be determined not to be worth it. And, if you calculate prevented recovery costs and the trauma and mental health to the recovery team then it wouldn't be worth the price tag either.
The real issue here is our preoccupation with asthetics (the view) and our preoccupation while seated in the automobile or what one would expect in the way of "Vistas."
I'd suggest eliminating the production and associated costs of 20 parking spaces in a jumbo sized environmentally friendly and esthetically pleasing parking garage on the UCSB campus (20 x $50,000 = 1 million dollars) and transfer that to this suicide prevention project. It's only a couple tenths of a mile and the design of the barriers will probably be nearly transparent. The view isn't noticeable while chatting on the cell phone anyway so put up the barrier and lets save a life!
I read the study and it is the study that is flawed, not the barrier concept. Obviously the study was conducted by someone with an agenda, or comissioned by someone with an agenda.
I thought it was interesting that the study determined the only way that a bridge suicide barrier can be deemed successful is through a drop in the suiicide rate. This is flawed because it presumes that other factors such as the unemployment rate, weather, release of patients from mental institutions would remain constant.
The other big flaw in the study is the belief that you can't trust the data from previous surveys done that stated that suicide barriers result in a major reduction in the number of people who will reattempt suicide. The study says that years of scientific survey research is meaningless because it is possible some of the people surveyed might not have gone through with their earlier suicide attempts if the barriers were not there. That they might have just been "cries for help."
Holding the earlier surveys to a standard of "if they REALLY meant it than they'd be dead" would mean it is impossible to conduct most scientific research on suicides because of an obviously ridiculous standard.
The coveted "peer review" in academia apparently is through anonymous blog comments here.
This is a good analysis by one professor, but it needs a traditional review to improve the analysis and make its conclusions more believable.
Imagine all the safety improvements that actually could be enacted with the same amount of money. Of course, asphalt cracks and guard rails are not as sexy as a giant fence along a historic classic bridge.
I agree that the study was marginally interesting, but flawed, and certainly not the be-all-end-all in terms of assessing the worthiness of the project. I hope the CalTrans project goes through.
As I've said before, if anyone holds "the view" so precious, just pull into the turnout less than a quarter mile past the bridge, stop, and enjoy the view all you want. It's pretty spectacular.
Hi all, Garrett Glasgow here.
Well, here we go. I was wondering when the civilized discussion would stop and the distortions of my research and ad hominem attacks would begin. The answer, apparently, is the day after you appear on TV.
Rather than point out the obvious lack of understanding of both past and current research in the posts above or address the ad hominem attacks, I'll simply invite all interested people to read my report for themselves.
To summarize the report here:
(1) Preventing suicide at a particular location is not the same thing as preventing suicide.
(2) The only way we will know if we are preventing suicide is by observing a drop in the suicide rate.
(3) None of the studies presented by Caltrans (nor any other study I could locate on the topic) demonstrate a statistically significant drop in the suicide rate after barriers are installed.
(4) Therefore, we don't know if a suicide prevention barrier will save lives.
Details, including a discussion of some of the studies and issues raised in posts above, are in the study:
http://www.polsci.ucsb.edu/faculty/glasgow/barrier_report.pdf
What a joke......this "study" by a fellow UCSB professor of Marc McGinness [the main voice opposing the barrier] does nothing to counter the multitude of studies demonstrating that barriers do prevent suicides. Would mean much more coming from a mental health expert. It's not about "debunking Cal Trans" its about [unsuccessfully attempting to] debunk extensive mental health studies.
Try again.
UCSB Study? How did UCSB conduct their research? How long did they spend in collecting it and where did they collect it from. I really questions the validity of this "so called" study. Furthermore, as a tax payer, I say that any deterent to get someone to not be able to commit suicide is worth spending money on.
This study truely does not make any sense in terms of determining the effectiveness of barriers. I am appalled that the news press would report on a self published study(not published or reviewed by a scientific or peer-reviwed journal) as fact. There is a reason that research has standards to be accepted as meaningful. This is to protect the public from misinformation and politics disguised as fact and "science." I hope people visit his site, read the study and realize the flaws in his reasoning and analysis.
There is no evidence that reducing cigarette smoking saves lives either.
That's because substitution happens for cigarettes too; when people stop smoking, a small fraction take up drugs like cocaine and speed, and then that small fraction suffers a very high mortality rate. The overall mortality rate stays the same.
So, should we stop our program of denying cigarettes to under-18 people?
Hello, Garrett Glasgow again.
My intention was to ignore all the personal attacks, but I feel I must set the record straight on one thing. I was not commissioned to write my report, nor have I received any funding or compensation of any kind in relation to my report. My only reason for becoming involved in this debate was a personal, scientific interest in the topic and the desire to provide the community with better information before making an important decision.
Thanks,
Garrett
Legislating my aesthetics, even if I want to commit suicide while looking at a beautiful historic bridge, is government at it's worst. How about installing some bungee cords for people who might change their mind on the way down?
Leave things alone unless there is some really big reason to change them. You just increase social brownian movement in the body politic and we're way too busy fixing things already.
To rebut Glasgow's 4 points:
(1) If you prevent suicide at a given spot you might be preventing suicides
(2) The number of suicides off bridges is so small compared to suicides by other methods that it is impossible to use overall suicide rates to detect the impact of eliminating suicides off bridges
His (3) is correct
His (4) is wrong, because his (2) is wrong.
More specifically: the overall rate of suicide for all types of falls (including off buildings and cliffs) is 0.24 per 100,000. that is only about 2.4% of the total suicide rate.
Glasgow sees a change in suicide rate of about 5 per 100,000 in a correlation plot with number of bridges, and Glasgow suggests that that this correlation has something to due with suicide by jumping.
Well, 5 per 100,000 is about 20 times higher than the *total rate of suicide by all falls*. It does not make sense that his correlation has anything to do with jumpers... there just too few jumpers.
No statistical study of the sort he proposes could have sufficient sensitivity to show the effect he seeks. His idea is ineffective.
Garrett,
I don't want to attack you but I would like a response from you regarding your #2 point.
You say that the way to determine if a suicide barrier is successful is to observe the suicide rate in the general area.
A couple questions:
1. Can you define the "area"?
2. How can you assume that a change in rate following the erection of the area is a direct result of the barrier?
3. If you can't trust the study that deals directly with suicidal individuals because you think they might not have gone through with it, doesn't that negate almost all research on suicide prevention in the world?
Isn't it a fact that many factors contribute to suicide? The status of the local economy, weather conditions and so forth? And do you disagree that a bridge like Cold Springs with a reputation for successful suicide attempts constitutes an attractive nuissance?
Also since you attended the suicide forum at the Faulkner Gallery and no doubt saw the video where law enforcement officers nearly lost their lives trying to keep a man from pulling them with him over the side of the bridge. Do you think that the lives of those and other officers are worth saving?
A response to these questions and arguments would be most appreciated sir. And thank you for engaging in this discussion. You could have just put the study out there and stayed silent, but you have chosen to engage the community further and deserve credit for this.
While we are at it, there is no proof the billions spent on cholesterol drugs matter either.
One of medicines larger frauds along with testing for cholesterol numbers and treating those numbers, but never asking for the final score.
Amazing stupidity we have on lots of things that divert our time and money.
Hi, Garrett Glasgow here. I'm happy to talk about the details of the study with anyone interested in a real discussion.
To address 8:27pm's concerns:
(1) Yes, if you prevent suicides at a specific location you might be saving lives. Or you might not (they might just go somewhere else or use a different method). We need some other kind of evidence to know for sure.
(2) It may be impossible to detect changes in the suicide rate due to bridge barriers for the reasons you suggest. Past research has pointed this out, and I specifically address this point in my study. This is why I came at the problem from a different angle, looking at the relationship between the suicide rate and tall bridges without suicide barriers. We have a lot more data here, and thus a better chance of detecting if such unsecured bridges pose a temptation or danger to suicidal individuals (a necessary condition for suicide barriers to be effective).
I didn't find any relationship between bridges and suicides, so we're back to square one -- we don't know if the barrier will work or not.
One other note: the scatterplot in my study showed a *negative* relationship between the number of bridges and the overall number of suicides, not a positive relationship. In my study I state that this doesn't seem to be a plausible causal relationship (bridges preventing suicides), and in fact this relationship disappears once you account for the unemployment rate.
Garret Glasgow again (I should just get an account, huh?). Now looking at 10:53pm's post. Your post isn't an attack at all, but set of reasonable questions I'm happy to try to answer.
Ideally, we would define the area where we'd expect to see a change in the suicide rate as the geographic region where most jumpers or potential jumpers come from. Right now I'm using state level data since that's all the CDC will release publicly -- I think the county level would be better, but the CDC doesn't make it available online for privacy reasons. I've written to them to request permission to use it, and if I get it I'll release a study looking at the relationship between bridges and suicides at the county level.
Even just looking at the state level I did get one plausible result -- the rate of suicide *by jumping* is higher in states with more bridges, so the state level data seems to be adequate for at least this initial study.
On your point 2, we can't assume any change in the suicide rate we see is solely due to the barrier -- suicide rates will naturally fluctuate slightly up and down from year to year for a number of reasons. However, if the barrier is saving lives the suicide rate should fluctuate around a lower rate after the barrier is installed, and with enough data, we could detect this difference with a test for statistical significance. However, as 8:27pm points out, we might not have enough data to do this.
I don't think anything I've done or said negates previous suicide research. Many of those studies are quite valuable, and have produced some great findings on how to save lives. However, on the narrow question of the likely effectiveness of suicide barriers these studies are inconclusive.
I completely agree that many factors contribute to suicide, which is why we see those fluctuations in suicide rates I described above. I also agree that the Cold Spring Bridge is a likely place for people to commit suicide -- but of course we want to know if building a barrier there just drives suicide victims to a new location that might also be attractive psychologically (The Courthouse tower? The bluffs looking out to sea?).
And yes, I did see the video where the police officer almost lost her life. However, police officers risk their lives every day, often in more mundane ways we don't think about -- getting out of their car on the side of the highway to talk to a motorist, for example. Could the money we would spend on the barrier be better spent on some other project that would do more for officer safety (widening the highway shoulders in dangerous areas, for instance)?
As I've said before, I believe the community needs to have a discussion about the best way to spend that money and save lives.
OK, past my bedtime. Thanks for those last couple of posts -- robust and respectful debate on this very important topic is exactly what we need.
I want to agree with Anon 10:54.
Just thinking about myself (how selfish of me!)--it's much more important to put a fence around the freezer compartment of the refrigerator where I keep my ice cream, then it is to put a fence around the bridge to oblivion. I hardly ever think about the bridge except when I see Bill Dewey's picture, and on the other hand, I can't keep my mind off the ice cream. This is a cry for help.
Seiden showed that suicidal folks found on the Golden Gate Bridge simply did not find other means of suicide. What is hard to understand about that?
Seiden is very powerful evidence that a barrier would therefore deter the Bridge-seekers from *any* subsequent suicide. Glasgow and others pick away at this, say it ain't airtight, but it is damn powerful evidence.
On another blog someone pointed out that the County spends $41 million a year on Mental Health Services, and that this bridge barrier, taken over is its lifetime, works out to $0.03 million a year, like a thousand times less.
WTF? Why aren't we arguing about how the $41 million/year at County Mental Health Services could be spent best?
Since it is such small beer compared to County Mental Health services, and also because there is powerful evidence from Seiden that it works, let's just put the darned barrier on the Cold Springs Bridge and Move On.
But it seems to me some folks want to debate the issue until they weary do-ers and achievers who will turn to something less effective and with less scrutiny on it. It's not about getting clarity but it's about muddying the waters, like the oil companies do when it comes go Global Warming.
Just build the barrier and Move On.
Considering Cal Trans apparently spent $5 million on a hundred yards of on-ramp at Storke Rd., what makes anyone think they put up more than a volleyball net under the bridge for a million?
More likely, they'll start the project and then stop for a few months. Then startup up again then stop and complain they've run out of money then start again and somewhere a few years down the road, we'll have some ugly artifact on the bridge that is ineffective and cost $7 million and some politician will want to take positive credit for it...Maybe even pass a bill to rename it the Chumash Trampoline
Just a thought but would the money save more lives if it was given to mental health services? When there are obviously shrinking budgets statewide why not use it to address issues people have before they drive to the bridge.
I think you should check your facts on this story. UCSB has not backed this report and they did not commission it either. This "study" isn't a true study but a report by a professor that is completely one sided and not statisically factual. PLEASE look into real studies and follow up on this story. You can go to the American Association of Suicidology, the American Psychological Association, the CDC or even the Santa Barbara Coronor's website (connected to the Santa Barbara Sherriff's website).
So, if we are to believe this 'study', then I suppose that all the locations around the world that have built barriers to prevent suicide including: Bloor Street Viaduct (Toronto), Eiffel Tower (Paris), Empire State Building and Brooklyn Bridge (NY), Clifton Bridge (UK) - to name a few - were wrong to do so.
And while we're at it, I suppose that all the mental health experts from the Nat'l Dept of Health & Human Services who created the National Strategy for Suicide Prevention were incorrect in saying that 'restriction of means' is one of eleven things that can be done to prevent suicide - including building physical barriers on high structures.
Learn the facts. Listen to mental health experts from around the world. Don't be swayed by flawed, inflammatory data manipulated by people with an obvious agenda.
Thanks for reading.
Hi, and welcome to our anonymous internet smear campaign! What we lack in understanding, we make up for in invective! Together we can drown out dissent!
Don't ask questions, don't even think. Just build the barrier and Move On.
Whoever drew a parallel between this debate and the global warming debate got it exactly right.
In both cases we have special interests trying to cast doubt on legitimate scientific research to protect their financial interests.
So, which side has a million dollar project at stake they want to protect? Hmmm...
Garrett... I don't think you established that your technique has or could have sensitivity to the effect you seek; the jumping rate is just too small.
It's like taking out a pocket magnifying glass to prove/disprove the existence of amoebas. Not seeing amoebas with a pocket magnifying glass proves that the amoebas are not 5 microns or bigger. It doesn't prove or disprove the existence of real amoebas.
Your study is like that. It does not have sensitivity to address the effect you seek. Not only does you study not verify a barrier would help, *by design* your study *could never* have provided verification.
Statistical sensitivity can be computed based on a data set. You would have done well to report that and also to report the expected effect due to bridge jumping, and then compared the two to let the reader know at the outset: are you looking for amoebas with a pocket magnifying glass or a microscope?
8:58pm and is County Mental Health trying to protect is $41 million a year budget, which adds up to billions over the lifetime of the barrier?
Thanks Garrett. You brought something really new to the table, and don't let yourself be brought down by the trolls.
I think 7:01am anon my have a point, but one must start somewhere.
Hi, Garrett Glasgow here. Thanks for the kind words from those of you interested in a real discussion on this topic.
7:01am, you do raise a valid concern, and one that I did spend a lot of time thinking about. If a study isn't sensitive enough to detect reasonable effect sizes, it will be misleading to report you didn't find anything.
It's not accurate to say my study couldn't detect changes in the suicide rate by design -- with enough data we can detect these changes.
This is one reason why I examined the relationship between bridges and suicide across states and over time -- this give us a lot more observations than previous studies (around 1300), and thus a lot more statistical power than previous approaches.
Note further that the relationship between bridges and suicides by jumping was positive, as expected, but the relationship between bridges and the overall suicide rate was *negative*. Both findings were statistically significant without control variables, but insignificant once I considered the unemployment rate. It's hard to imagine how this pattern of findings could simply be due to inadequate statistical power.
The UCSB quarter starts getting busy on Monday, so if I'm unable to find time to come back here and continue the discussion, it's not for lack of interest or respect for what anyone has to say.
"There is no evidence that reducing cigarette smoking saves lives either."
That's not true.
"That's because substitution happens for cigarettes too; when people stop smoking, a small fraction take up drugs like cocaine and speed, and then that small fraction suffers a very high mortality rate. The overall mortality rate stays the same."
No, it doesn't of course, because the number of mortalities from cocaine and speed are much much lower than the number of mortalities from cigarettes.
Wrong comparison, jqb. What matter is not the number of mortalities from cigarettes.
It's whether cigarette quitters (a much smaller group than total mortalities) live any longer, on average.
On average, a few die fast because they can't deal with the tobacco withdrawal and turn to drugs and alcohol. And that rapid mortality negates the benefits for the more numerous group who live longer.
Let em jump. It will cost a whole lot less to employ some illegal aliens working for county parks & rec to scrape 20 nutcases off the canyon floor in the next 10 years than it will to build the $1 million (then $2, then $4, then $6 million by the time it's finally installed) dollar barrier. Your mental health services costs will go down, as there will be less nutcases to deal with since they've checked out.
Suicide is the most selfish act a person can undertake, and I have very little sympathy for those who even try to perpetrate the act. My brother-in-law checked out nearly three years ago and not once did I feel bad for him. I felt bad for my wife and my father-in-law, since they had to bear the brunt of the loss. Screw the suicidal, let the selfish bastards nosedive.
Yup MC, then all those relatives can feel just as bad as you wife and father-in-law.
While we're at it, let's deny antibiotics to everyone who gets an infection. Anyone stupid enough to not wash their hands or to get near a sick person doesn't deserve those antibiotics.
Come to think of it, why the heck should anyone be allowed to get any medical care. It's their own darned fault they got sick. Let `em rot and die.
While we're at it, let's deny computers to lamebrains. Then we don't have to deal with moronic posts about denying medicine to sick people.
MC, great suggestion. Your own posts would then disappear as well!
Denying aid to the suicidal, which is your lamebrainwave, is identical to denying medicine to sick people.
Post a Comment
<< Home