Saturday, June 04, 2011

Psychiatrists and Media Statistics

Thank you to Angela Wilson and her Twitter feed for pointing me in the direction of this story.

The Palm Beach Post News is covering an investigation of prescribing habits of psychiatrists working in the juvenile justice system. The implication is that doctors who receive Medicaid funding are overprescribing atypical antipsychotics for detained juveniles. While the data looks compelling at first glance, I agree with Angela that it is incomplete.

The article implies there is a correlation between the number of scripts written and the amount of money paid to the doctors. This may be true, but there are no correlation statistics in the article to support this conclusion. They list the doctor's name, how many scripts he writes and how much money he received from the pharmaceutical industry, but there are no correlation measurements whatsoever.

The other problem with the article is that the number of scripts written is not placed into any context: How many hours per week does each doc spend in a juvenile facility? How many patients is he seeing? Is there a difference in the diagnoses per patient group (some docs may be seeing sicker kids)?

I'm not saying that that the pharmaceutical industry doesn't influence prescribing habits, or that it's good to overmedicate kids. My complaint is with how this story is presented. It's incomplete, so obviously incomplete (and I'm NOT a statistics maven!) that even I can pick it up. Sloppy, inflammatory journalism won't help these kids.


Rob Lindeman said...

What? You mean you have to get consent to administer chemical restraints?

"In at least 40 cases since 2008, DJJ files have lacked proof that parents or guardians gave consent before children were put on medications, were apprised of possible side effects or told that doctors were adding drugs or adjusting dosages.

At Impact House, a residential program in Jacksonville, one child was put on drugs, taken off them, or had dosages and combinations adjusted six times in a six-month period. He began on Risperdal and ultimately was put on Seroquel. His files contained no evidence that anybody ever had notified his parents to gain consent."

Meg said...

Of course it is incomplete. The article says it is part 1 of a two part series. Part 2 is on Sunday.

ClinkShrink said...

Maybe Part 2 will be better.

a psychiatrist who learned from veterans said...

I'm not saying that that the pharmaceutical industry doesn't influence prescribing habits,

I feel like there ought to be something like another blog post:

In related news, the state of Florida has passed a law saying that to be an M.D., you must have, on a standard IQ test a score of 65 or greater. On the other hand, those with a score of 80, will be assigned to another profession where the handling of complex information is preferred.

If the government wants to provide balance, which is fine, it can pay for speakers to talk to doctors about side effects and other issues which may not be talked about by the drug companies.

Meg said...

Actually, I checked the dates. The series came out in May, so the "Sunday" referred to was a couple weeks ago.

I read the entire series. It is true that the article reports no correlation coefficients, let alone regression coefficients, but even if it did, those statistics would not establish causality. (The mantra of all statistics professors is "Correlation does not prove causality." My personal favorite example comes from Tufte's classic textbook. Based on annual data from the 1920s and 1930s, the correlation between the number of radios in Great Britain and the number of "registered mental defectives" in Great Britain was 0.99, almost perfect correlation. But no responsible statistician would conclude causality from that correlation.)

In any event, few reporters have the statistical chops to do the kind of sophisticated statistical analysis that would be required to analyze this type of dataset. And few newspaper readers have the statistical sophistication to interpret them.

That said, I think the correlation coefficient and regression coefficient statistics are beside the point.

In most government agencies with which I am familiar, standard conflict of interest rules forbid officials with responsibility for making major government purchasing decisions from accepting cash payments from vendors.

Whether or not it influences their purchasing decisions, accepting large cash payments from government vendors strikes me as a completely unacceptable conflict of interest for anyone paid by the state to make purchasing decisions on the state's behalf.

If accepting such large payments from vendors doing business with the government does not violate Florida laws, then, in my opinion, those laws need to change and the newspaper has done a service by highlighting that need.

Sunny CA said...

It does not take advanced statistics to interpret this:

"In 2007, for example, DJJ bought more than twice as much Seroquel as ibuprofen. Overall, in 24 months, the department bought 326,081 tablets of Seroquel, Abilify, Risperdal and other antipsychotic drugs for use in state-operated jails and homes for children. "
"Paula....saw a lot of Seroquel. Every morning ...she said, one of her bunk mates, 17-year-old Desire, would take several hundred milligrams. Every evening, she would swallow several hundred more,..."By the time we all woke up at 6 o'clock the next the morning, I mean, she's still laying in bed and won't get up. We usually would have to push her out of the door by the time we went to breakfast"Paula said Desire didn't hallucinate or hear voices. "She just seemed like a normal kid who would be in the 'hood. You know, trying to show up everybody. It just seemed normal to me," Paula said. However, she added, her friend had a penchant for acting out.
"She was definitely one of the more rampant ones, so I guess they were trying to, like, kind of calm her down and keep her subdued throughout the time she was there."
"Antoinette Appel, a Tamarac neuropsychologist, said she was suspicious of the amounts of drugs used by DJJ. "They're not allowed to put kids in restraints, so they put kids in restraints this way," Appel said."

Duane Sherry said...

This has been going on for years, in Medicaid, foster care, and juvenile justice systems.

From the Alliance for Human Research Protection -

You wrote that sloppy journalism "won't help these kids."

What will help?

Aggresive prosecution of some of these doctors, for the obvious, Medicaid Fraud -

Psychiatry will collapse.
In fact, it is in the process of collapsing.

Justice will complete the process.

A message to psychiatrists who like to illegally drug kids...

Don't want to go to prison?
Don't break the law!

Duane Sherry, M.S.

Duane Sherry said...

Bad link on the Medicaid Fraud Initiative.

bHopefully, this one will take -

Duane Sherry, M.S.

Dinah said...

I imagine these kids are difficult to deal with and some may get put on these medications when they aren't indicated, and kept on them when they aren't helpful. I don't know that, but it wouldn't surprise me and I could imagine many scenarios.

Because a chart doesn't indicate that parents weren't notified doesn't mean it didn't happen. This is't to defemd DJJ or "Impact House" but to say agree with Clink that media statistics don't make something so.

There are too many Meg commenters and I'm never sure who is talking. I think I actually know two of them.

LWHLFV: When do we take the IQ tests? Oh, yeah, I don't live in Florida.

SunnyCA: Maybe the DJJ uses tylenol more than advil? It doesn't help journalistic credibility to use a juvenile delinquent's quotes as a rationale for saying that too much of anything is being done...and I don't think these children are the regular kids from the 'hood.

Duane: I've checked with the other two bloggers here and no one is taking your comments down. We fully understand that you don't like psychiatrists or psychiatry and I'm not sure what there is to be gained by repeatedly putting these comments up. Ditto to Rob. Respectful challenge is one thing in an intellectual exchange, but you are both rude.

Anonymous said...


You said,

""It doesn't help journalistic credibility to use a juvenile delinquent's quotes as a rationale for saying that too much of anything is being done...and I don't think these children are the regular kids from the 'hood.""

You might want to rethink what I feel are extremely unfair assumptions:

""Paula was 14 and living in Palm Beach County when police hauled her in as a runaway. At the station, she kicked over a table and got arrested, she said. Though it was her first offense, a juvenile judge committed her to a program. ""

Yeah, it isn't good to kick over a table at a police station but I would hardly classify someone like this as a juvenile delinquent. Kind of tough when you are 14 to be picked up by the police for being a runaway.

Her observations on what she observed are just as legitimate as a psychiatrist or any other so-called expert.


Duane Sherry, M.S. said...


Two comments disappeared off the blog.

As far as being "rude."

Has it occured to you that some of us (a growing number) don't like the injury your profession has caused, and continues to cause?

Children do NOT need to be given psychiatric drugs, Dr. Miller.

The term "over-use" is understandable when it comes to informed adults. There's room for some healthy debate in their use. But children are "off-limits. Period.

I find it interesting that you see me as "rude"... Even more interesting that you call yourself "fun" and some of your colleages "nice."

Nice and fun.

How refreshing.

Duane Sherry, M.S.

ClinkShrink said...

Meg: That's funny, we both missed the date on that article. I was going to look for the second part tomorrow.

Separate from government vendors--and I agree with you about that--accepting handouts from the pharmaceutical industry alone is a reason to cleanup the system. I think the story would have been stronger by simply sticking to that point. Adding a list of meaningless uninterpretable numbers detracted from the force of the story. While I understand about journalists lacking statistical knowledge, they still should have made an effort to place the prescription issue in a broader context. In the second part of the series there is only one brief paragraph that acknowledges other possible causes of changes in prescribing patterns:

"Medical experts cautioned against drawing conclusions from Mhatre's prescribing, saying a number of variables might have been at work.

For instance, he might have seen more patients in the spring of 2010 than he did in previous months."

That was it.

Again, I agree it was an important story that serves a useful purpose. Stretching a diagnosis to please a drug company is as bad as stretching a diagnosis to make sure the hospital gets paid for an admission.

Rob Lindeman said...

"Because a chart doesn't indicate that parents weren't notified doesn't mean it didn't happen. This is't to defemd DJJ or "Impact House" but to say agree with Clink that media statistics don't make something so. "

Rather than attempting to defend the indefensible, it's preferable to deny that the indefensible is happening at all.

Meg O said...

I'll add an initial to distinguish myself from any other Megs. (The above Megs are me.)

There is a great deal of disturbing stuff going on juvenile homes across the country these days, some of it having nothing to do with the use of drugs. See, for example, this article:

NY Governor Andrew Cuomo and his predecessor Governor Paterson have both highlighted the issue that our state is spending large amounts of money to send children to juvenile detention facilities far from their homes. For both humane and budgetary reasons, their arguments that children whose infractions are non-violent would be better served by local (non-residential) programs in their communities rather than shipping them off to remote facilities in far corners of the state.

However, those residential detention facilities are major employers in those remote communities, and their elected representatives have staunchly fought efforts to downsize or close them.

There are a lot of people with an economic stake in the treatment of troubled children, and unfortunately that can distort decision making.

For the $190,000 it costs to institutionalize one child far from home, we could purchase an awful lot of local professional help for the child and his/her family and school.

Meg O said...

Oops, my previous link got cutoff. Here is an article describing concerns about the use of dangerous non-drug methods of restraint in juvenile justice facilities.

Anonymous said...

Meg O,

A social worker whom Bob Whitaker interviewed in his book, Anatomy of an Epidemic, grew in in a residential home before the widespread practice of prescribing meds came into play. He said he was beaten.

Interestingly, he said he preferred growing up in that type of environment vs. being chemically restrained.

On a more positive note, Whitaker talks about Seneca Center, in San Leandro, CA, which actually takes kids off of medications even though they are classified as the the most troubled.

Here is a novel thought - The psychiatrist in charge actually tracks their medication history and notices how their behavior worsened after being put on a med.

He'll have people suggest a drug and he actually looks at their medication history and says, no, I don't think that is going to be helpful based on what has previously happened.

After the kids are taken off of med, they do become more aggressive for a time. But then the behavior problems abate and the kids come alive and form relationships with staff.

The center has anecdotal information that the students go on to do very well but have not done any formal studies.


Duane Sherry, M.S. said...

The link to PsychRights is too long to take.

For anyone interested in reading the Medicaid Fraud Initiative (Against the Psychiatric Drugging of Children/Youth, go to -

If you spend time on the site, you'll find it. The bottom line is that is a lot of Medicaid Fraud taking place in psychiatry.

Attorney, Jim Gottstein outlines what constitutes Medicaid Fraud, and how to end it.

Our children deserve better than labels and drugs. And it's time our court system addressed this problem, by prosecuting "off label" prescribers, for conditions that these drugs were never approved for.

It's against the law, and it needs to stop.

Duane Sherry

Sarebear said...
This comment has been removed by the author.
Anonymous said...

I think this exert sums up the crucial issues perfectly. It is why focusing on whether the Palm Beach Post News failed to put the statistics in context is totally missing the big picture.

""Unlike other states, including California, Massachusetts and Minnesota, Florida lacks rules requiring drug makers to disclose payments to doctors licensed by the state.

Such measures would create needed transparency, medical ethicists said.

"I think it's because we worry that paying and marketing and advertising and detailing can shape behavior," said Arthur Caplan, director of the University of Pennsylvania's Center for Bioethics. "I don't think that's a hypothetical, because we know that they do."

Even better, said Campbell, the Harvard professor, would be if Florida government medical contractors refused all offers of money from drug companies.

"Physicians who care for patients who are vulnerable populations need to be particularly above reproach when it comes to this behavior," Campbell said. "I personally believe it's completely inappropriate for physicians to moonlight as drug reps. You either sell drugs or you practice medicine. But you can't do both." ""