Pete Earley Now Has A Problem

And his problem is me. Or more accurately, my willingness to be skeptical about a statistic he cited in his Washington Post op-ed of May 12. In it, he asserted that "people with mental illnesses kill law enforcement officers at a rate 5.5 times greater than the rest of the population." With that stat, he turned every schizophrenic and bipolar I into a pariah and cop-killer, especially since his op-ed has since run in almost 200 newspapers, according to Google News.

He cited the nonprofit Treatment Advocacy Center as his source for that data. So I called TAC and asked them to get back to me with their sourcing for that data. They didn't call me back.

I poked around their site and discovered the 5.5 number on one of their facts and figures pages. This will get boring and technical for a few moments. Sorry, population statistics are sucky that way.

The number was footnoted so I clicked on it and learned that the 5.5 number was for 1998, which is to say that Earley used 1998 data to describe 2006. Using TAC's database--they are so obsessed with acts of violence committed by the mentally ill that they have a searchable database--in 2005, there were 4 murders of cops by people with mental illnesses. So that 5.5 number is now something under a rate of 2 times the general population.

It gets weirder from there. Here's how the TAC website describes where that 5.5 number was arrived at:

"The Surgeon General estimated that 2.6 percent of the U.S. population has a severe and persistent mental illness. [U.S. Department of Health and Human Services. Mental health: A report of the Surgeon General (1999).] Therefore, there were 7 million people with severe and persistent mental illnesses in the United States in 1998. [Bureau of Justice Statistics (2001). U.S. Census Bureau. Monthly estimates of the United States population. Available at http://www.census.gov/population/ estimates/nation/intfile1-1.txt. (Visited May 4, 2001).] People with mental illnesses committed at least eight of the 61 police homicides that occurred in 1998 (13 percent) at a rate of 11 per 10 million. [Treatment Advocacy Center. Preventable tragedies database. Available at http://www.psychlaws.org/ep.asp. (Visited June 26, 2003).] Fifty-three police homicides were committed by the rest of the population of 263 million, at a rate of two per 10 million. [Brown, Jodi M., and Patrick A. Langan. Policing and homicide, 1976-98: Justifiable homicide by police, police officers murdered by felons.] Thus, people with mental illnesses committed police homicides at a rate 5.5 times greater than the general population."

So TAC calculated the rate using a small subset, between 5 percent and 10 percent, of the total number of Americans with mental illness. That 2.6 percent prevalence of severe and persistent mental illness appears to be pasted together from the Surgeon General's 1999 report--combining the number of Americans with schizophrenia, bipolar disorder I and non-affective psychosis to arrive at 2.6 percent. Since that report appeared in 1999, it's fairly safe to assume that those numbers were pieced together in 1997 and 1998 (the government works slowly), and that they are themselves estimates from earlier in the 1990s.

Here and now in 2006, we all just got painted with a mid-1990s brush. Or, we were pelted with apples and oranges.

I could walk you all through why TAC's rate calculations are bogus--in 2006 we know that the percentage of severe and persistent mental illness is somewhat higher than the 2.6 percent of the 1990s, which used 1.3 percent as the prevalence of bipolar I, so that would decrease the 5.5 on its own. That 1.3 figure was the consensus number that came about in the early-1990s, and works out to about 3 million Americans. I think we all know, as do researchers, that we are now looking at something closer to 10 million, although the pharma companies talk as if it’s closer to 15 million.

There is also no age-adjustment in the rate for the general population of 263 million (it doesn't make sense to include in that general population number the many millions of youngsters and elderly who wouldn't be able to murder a cop unless they used a rattle or a cane), so that would drive the rate within the general population up and, conversely, drive the rate among those with severe and persistent mental illness down. In population stats, you’ve got to account for how your data might be skewed by its prevalence in sub-groups of the broader population.

But Earley didn't say "severe and persistent mental illness" in his copy. He said "people with mental illnesses." That's a much larger group than the 7 million upon which the rate of 5.5 times was based. Apples to describe oranges. For a reporter cum advocate who is actively trying to help us, that’s a remarkable breach of faith.

Estimates of the prevalence of mental illnesses in America vary widely from 10 percent to 25 percent, or about 30 million to 75 million Americans with schizophrenia, depression, the various kind of bipolar disorder, anxiety disorders and AD/HD. If you take the number of murders of cops killed by people with mental illness and calculate a rate using those numbers--and Earley specifically referred to that population--then you'd end up with a rate of the mentally ill killing cops that would be less than the general population. Not that I think there is anything "general" about the population that kills cops.

By now, easily more than a million people have read what Earley wrote. In his op-ed, he confirmed many peoples' built-in assumptions about the mentally-ill: we are fucked-up and only forced compliance with medication will get the kind of results society expects. What's more, we are now demonized as cop killers. That's fucking offensive to me, especially since I have friends who are cops. Its doubly offensive because the data he used doesn't even support his argument.

So what's Earley's point? His broad argument in other articles he's written of late as well as his new book Crazy is that America's system of grappling with mental illness is broken and a disgrace. He's right. He's right that too often it is too difficult to get care for the mentally ill and that as a result our jails are filled with the mentally ill and so are the streets of this country. The trouble with Earley's thesis is that it comes from the point of view of the father of a son with bipolar disorder who, when his son melted down, found it impossible to force the mental health system in Virginia to force his son to take meds and comply with treatment, because his son was resisting every step of the way. I can understand his position. Over the years, I've had a few friends develop mental illness. Some of them have refused treatment, refused to even acknowledge their condition, and I have had to cut off contact with them. Not fun.

I am weary of the family crowd who are very influential in setting policy goals, laws and research priorities in this country. Too often, they see the story of mental illness in this country through the stray cop killer, the son or daughter who killed themselves and the like. What they never acknowledge is that for every schizophrenic run amok, there is a John Nash. For every bipolar running naked down the streets, there is a you and a me. For every cop gunned down by someone flying off their meds, there are hundreds of police officers in this country living with depression and bipolar disorder who must suck it up and stay in the shadows.

Still, I agree with Earley that we need a global fix for this problem, post-haste. The trouble, though, isn't that we cannot force enough people into treatment. The big overarching problem is that we have a paradigm of treatment in this country that isn't especially effective. Psych meds work less than 50 percent of the time and are very rough on peoples' bodies and cognitive functioning most of the time. Some meds have been linked to suicides and diabetes. What's more, it's well known in the mental health world that patients can play by the rules, take their meds religiously, avoid booze and crack and still have the very kinds of problems that we want them to take meds to prevent. Especially chilling is that increased openness about treating mental illness has led to tens of millions of Americans taking psychotropic medications each day of their lives to the tune of $40 billion a year in sales. But, measuring our success by changes in the rate of suicide over time, the ultimate index of mental illness, it's easy to see that we haven't made much progress. The rate of suicide in this county, most pronounced among adult males, remains virtually unchanged since the late-1950s.

It's unfair to argue for people like me and like Earley's son to be forced into a system of treatment that produces spotty results and expect that to be the answer. It's OK to expect us to take meds and do all the other complicated existential handsprings in order to make life play out right—and work with us on making that work. But to call for forced medication for the many based upon bad outcomes with the few would return us to a system of vrtual confinement not too dissimilar from the bad old days of the snake pit state mental hospitals where hundreds of thousands of Americans were confined for life.

But Earley wants this system fixed and so do I. You'll never fix it by demonizing us, especially if you use bogus statistics and flawed assumptions.