User login
Recently, a lot has been in the news about the increasing rates of suicide in all communities, including among African American youth, and two high-profile celebrities. Now that we have a CEO in the White House who has made the phrase “fake news” part of the national lexicon (and as a former CEO myself), I feel compelled to take a critical, clinical look at the way the suicide story has been reported.
CEOs tend to be unique people, and many of them are fond of hyperbole – as it promotes “followship” in employees and fosters business deals. I interpret fake news as the kind of information, or maybe spin is a better word, promulgated by CEOs.
While following a research letter published recently in JAMA Pediatrics – “Age-Related Racial Disparity in Suicide Rates Among U.S. Youths From 2001 Through 2015” (2018 May 21. doi: 10.001/jamapediatrics.2018.0399) – it occurred to me that this struck me as fake news. But as I thought about it, I realized that the conclusions in the research letter would be better characterized as perhaps misleading news. My basis for reaching those conclusions is rooted in the lessons I learned as a 2-year member of the Institute of Medicine’s Board on Neuroscience and Behavioral Health, Committee on Pathophysiology & Prevention of Adolescent & Adult Suicide. In fact, the report we produced was the first one referenced in the research letter. Unfortunately, however, the research letter’s authors seemed to miss the IOM report’s major take-away messages.
For example, the research letter authors compared the suicide rates among black children and white children in this way: “However, suicide rates increased from 1993 to 1997 and 2008 to 2012 among black children aged 5 to 11 years (from 1.36 to 2.54 per million) and decreased among white children of the same age (from 1.14 to 0.77 per million).” That sentence supports the conclusions of the IOM’s “Reducing Suicide” report, as it confirms that those are very low base rates. However, because the base rates are so low in most populations, it is difficult to determine scientifically whether a significant rise or decrease in rates occurred.
To quote page 377 of IOM report: “The base rate of completed suicide is sufficiently low to preclude all but the largest of studies. When such studies are performed, resultant comparisons are between extremely small and large groups of individuals (suicide completers versus non–suicide completers, or suicide attempters versus non–suicide attempters). Use of suicidal ideation as an outcome can increase incidence and alleviate the problem to some extent; however, it is unclear whether suicidal ideation is a strong predictor of suicide completion. Using both attempts and completions can confound the analysis since attempters may account for some of the suicides completed within the study period. Because the duration of the prevention studies is frequently too brief to collect sufficient data on the low frequency endpoints of suicide or suicide attempt, proximal measures such as changes in knowledge or attitude are used. Yet the predictive value of these variables is unconfirmed.”
Further, according to page 410 of the report: “As the statistical analysis above points out, at a suicide rate of 10 per 100,000 population, approximately 100,000 participants are needed to achieve statistical significance in an experimental context. In studying suicide among low-risk groups, the numbers needed are even greater.”
Let me break this down a bit. , because the numerator is so small and the dominator is so large. Let me put it this way – if the black female suicide rates are 2/100,000, and those rates quadrupled (sounds impressive, doesn’t it?) then there would be 8/100,000 black female suicides; the difference between 2 and 8 per 100,000 is not really a significant difference because the base-rates are so small. But to say the rates quadrupled sounds scary and impressive. “Figures don’t lie, but liars can figure.”
So, the premise of the research letter is whack.
I am not impressed that the rates of black children aged 5-7 increased from 1.36/1,000,000 to 2.54/1,000,000. I am not even sure those two numbers are significantly different, much less have clinical relevance. I have tried to make this point before, but it always gets lost by the hyperbolic press – which continues to yell about suicides in the United States rising by 30% or doubling, even quadrupling. The low base rates make drawing firm conclusions from this data like spitting into the ocean. I understand that one suicide is one suicide too many. But this is not science.
The characterizations about soaring U.S. suicide rates are not exactly fake news. Instead, I would call these interpretations misleading science and news.
Dr. Bell is staff psychiatrist at Jackson Park Hospital’s surgical-medical/psychiatric inpatient unit; clinical professor emeritus, department of psychiatry, University of Illinois at Chicago; former director of the Institute for Juvenile Research (the birthplace of child psychiatry), and former president/CEO of the Community Mental Health Council, all in Chicago. He also serves as chair of psychiatry at Windsor University, St. Kitts.
Recently, a lot has been in the news about the increasing rates of suicide in all communities, including among African American youth, and two high-profile celebrities. Now that we have a CEO in the White House who has made the phrase “fake news” part of the national lexicon (and as a former CEO myself), I feel compelled to take a critical, clinical look at the way the suicide story has been reported.
CEOs tend to be unique people, and many of them are fond of hyperbole – as it promotes “followship” in employees and fosters business deals. I interpret fake news as the kind of information, or maybe spin is a better word, promulgated by CEOs.
While following a research letter published recently in JAMA Pediatrics – “Age-Related Racial Disparity in Suicide Rates Among U.S. Youths From 2001 Through 2015” (2018 May 21. doi: 10.001/jamapediatrics.2018.0399) – it occurred to me that this struck me as fake news. But as I thought about it, I realized that the conclusions in the research letter would be better characterized as perhaps misleading news. My basis for reaching those conclusions is rooted in the lessons I learned as a 2-year member of the Institute of Medicine’s Board on Neuroscience and Behavioral Health, Committee on Pathophysiology & Prevention of Adolescent & Adult Suicide. In fact, the report we produced was the first one referenced in the research letter. Unfortunately, however, the research letter’s authors seemed to miss the IOM report’s major take-away messages.
For example, the research letter authors compared the suicide rates among black children and white children in this way: “However, suicide rates increased from 1993 to 1997 and 2008 to 2012 among black children aged 5 to 11 years (from 1.36 to 2.54 per million) and decreased among white children of the same age (from 1.14 to 0.77 per million).” That sentence supports the conclusions of the IOM’s “Reducing Suicide” report, as it confirms that those are very low base rates. However, because the base rates are so low in most populations, it is difficult to determine scientifically whether a significant rise or decrease in rates occurred.
To quote page 377 of IOM report: “The base rate of completed suicide is sufficiently low to preclude all but the largest of studies. When such studies are performed, resultant comparisons are between extremely small and large groups of individuals (suicide completers versus non–suicide completers, or suicide attempters versus non–suicide attempters). Use of suicidal ideation as an outcome can increase incidence and alleviate the problem to some extent; however, it is unclear whether suicidal ideation is a strong predictor of suicide completion. Using both attempts and completions can confound the analysis since attempters may account for some of the suicides completed within the study period. Because the duration of the prevention studies is frequently too brief to collect sufficient data on the low frequency endpoints of suicide or suicide attempt, proximal measures such as changes in knowledge or attitude are used. Yet the predictive value of these variables is unconfirmed.”
Further, according to page 410 of the report: “As the statistical analysis above points out, at a suicide rate of 10 per 100,000 population, approximately 100,000 participants are needed to achieve statistical significance in an experimental context. In studying suicide among low-risk groups, the numbers needed are even greater.”
Let me break this down a bit. , because the numerator is so small and the dominator is so large. Let me put it this way – if the black female suicide rates are 2/100,000, and those rates quadrupled (sounds impressive, doesn’t it?) then there would be 8/100,000 black female suicides; the difference between 2 and 8 per 100,000 is not really a significant difference because the base-rates are so small. But to say the rates quadrupled sounds scary and impressive. “Figures don’t lie, but liars can figure.”
So, the premise of the research letter is whack.
I am not impressed that the rates of black children aged 5-7 increased from 1.36/1,000,000 to 2.54/1,000,000. I am not even sure those two numbers are significantly different, much less have clinical relevance. I have tried to make this point before, but it always gets lost by the hyperbolic press – which continues to yell about suicides in the United States rising by 30% or doubling, even quadrupling. The low base rates make drawing firm conclusions from this data like spitting into the ocean. I understand that one suicide is one suicide too many. But this is not science.
The characterizations about soaring U.S. suicide rates are not exactly fake news. Instead, I would call these interpretations misleading science and news.
Dr. Bell is staff psychiatrist at Jackson Park Hospital’s surgical-medical/psychiatric inpatient unit; clinical professor emeritus, department of psychiatry, University of Illinois at Chicago; former director of the Institute for Juvenile Research (the birthplace of child psychiatry), and former president/CEO of the Community Mental Health Council, all in Chicago. He also serves as chair of psychiatry at Windsor University, St. Kitts.
Recently, a lot has been in the news about the increasing rates of suicide in all communities, including among African American youth, and two high-profile celebrities. Now that we have a CEO in the White House who has made the phrase “fake news” part of the national lexicon (and as a former CEO myself), I feel compelled to take a critical, clinical look at the way the suicide story has been reported.
CEOs tend to be unique people, and many of them are fond of hyperbole – as it promotes “followship” in employees and fosters business deals. I interpret fake news as the kind of information, or maybe spin is a better word, promulgated by CEOs.
While following a research letter published recently in JAMA Pediatrics – “Age-Related Racial Disparity in Suicide Rates Among U.S. Youths From 2001 Through 2015” (2018 May 21. doi: 10.001/jamapediatrics.2018.0399) – it occurred to me that this struck me as fake news. But as I thought about it, I realized that the conclusions in the research letter would be better characterized as perhaps misleading news. My basis for reaching those conclusions is rooted in the lessons I learned as a 2-year member of the Institute of Medicine’s Board on Neuroscience and Behavioral Health, Committee on Pathophysiology & Prevention of Adolescent & Adult Suicide. In fact, the report we produced was the first one referenced in the research letter. Unfortunately, however, the research letter’s authors seemed to miss the IOM report’s major take-away messages.
For example, the research letter authors compared the suicide rates among black children and white children in this way: “However, suicide rates increased from 1993 to 1997 and 2008 to 2012 among black children aged 5 to 11 years (from 1.36 to 2.54 per million) and decreased among white children of the same age (from 1.14 to 0.77 per million).” That sentence supports the conclusions of the IOM’s “Reducing Suicide” report, as it confirms that those are very low base rates. However, because the base rates are so low in most populations, it is difficult to determine scientifically whether a significant rise or decrease in rates occurred.
To quote page 377 of IOM report: “The base rate of completed suicide is sufficiently low to preclude all but the largest of studies. When such studies are performed, resultant comparisons are between extremely small and large groups of individuals (suicide completers versus non–suicide completers, or suicide attempters versus non–suicide attempters). Use of suicidal ideation as an outcome can increase incidence and alleviate the problem to some extent; however, it is unclear whether suicidal ideation is a strong predictor of suicide completion. Using both attempts and completions can confound the analysis since attempters may account for some of the suicides completed within the study period. Because the duration of the prevention studies is frequently too brief to collect sufficient data on the low frequency endpoints of suicide or suicide attempt, proximal measures such as changes in knowledge or attitude are used. Yet the predictive value of these variables is unconfirmed.”
Further, according to page 410 of the report: “As the statistical analysis above points out, at a suicide rate of 10 per 100,000 population, approximately 100,000 participants are needed to achieve statistical significance in an experimental context. In studying suicide among low-risk groups, the numbers needed are even greater.”
Let me break this down a bit. , because the numerator is so small and the dominator is so large. Let me put it this way – if the black female suicide rates are 2/100,000, and those rates quadrupled (sounds impressive, doesn’t it?) then there would be 8/100,000 black female suicides; the difference between 2 and 8 per 100,000 is not really a significant difference because the base-rates are so small. But to say the rates quadrupled sounds scary and impressive. “Figures don’t lie, but liars can figure.”
So, the premise of the research letter is whack.
I am not impressed that the rates of black children aged 5-7 increased from 1.36/1,000,000 to 2.54/1,000,000. I am not even sure those two numbers are significantly different, much less have clinical relevance. I have tried to make this point before, but it always gets lost by the hyperbolic press – which continues to yell about suicides in the United States rising by 30% or doubling, even quadrupling. The low base rates make drawing firm conclusions from this data like spitting into the ocean. I understand that one suicide is one suicide too many. But this is not science.
The characterizations about soaring U.S. suicide rates are not exactly fake news. Instead, I would call these interpretations misleading science and news.
Dr. Bell is staff psychiatrist at Jackson Park Hospital’s surgical-medical/psychiatric inpatient unit; clinical professor emeritus, department of psychiatry, University of Illinois at Chicago; former director of the Institute for Juvenile Research (the birthplace of child psychiatry), and former president/CEO of the Community Mental Health Council, all in Chicago. He also serves as chair of psychiatry at Windsor University, St. Kitts.