Why does the Treatment Advocacy Center (TAC) misrepresent psychological research?
For instance, in its post on its website titled, “STUDY: Manic Symptoms Linked to Specific Criminal Acts,” the unattributed and undated article suggests that a new study was released that demonstrated a causal link between manic symptoms, and well, specific criminal acts.
But when I read the study, and compared it with what was in the article on the TAC website, I saw a complete misunderstanding (or misrepresentation, whether intentional or not) of the new study.
It now makes me question the validity of any information published by the Treatment Advocacy Center on their website, because it appears their bias — to drive home the mistaken idea that mental illness = increased risk of violence — affects their ability to even deliver research news objectively.
It’s pretty clear whoever authored this post is not a researcher and probably shouldn’t be trying to interpret and disseminate research results. The post begins with the mistaken suggestion that the new study (Christopher et al, 2012) found:
Individuals with bipolar disorder are “more than twice as likely as the general population to commit violent crimes and nearly five times as likely to be arrested, jailed or convicted of an offense other than drunk driving,” authors of a new study on the association between manic symptoms and criminal acts report.
While indeed the current authors write that, that’s actually just background information in the study — it does not refer to any new data.
But rather than just take what a researcher says at face value, we do something here other websites don’t provide — a critical analysis. Let’s look at those two statements first, since they set the stage (both for TAC’s and the current researchers’ article).
The current authors (Christopher et al, 2012) write in the introduction to their study:
Persons with bipolar disorder, in particular, are more than twice as likely as the general population to commit violent crimes…
The reference for this statement comes from a single study (Fazel et al., 2010), that examined “violent crime” (which also included crimes such as simply threatening another person), and bipolar disorder in Sweden. (Whether Sweden is like the rest of the world’s population of people with bipolar disorder or who commit violent crimes is an exercise I leave to the reader). Here’s what they actually found:
During follow-up, 314 individuals with bipolar disorder (8.4%) committed violent crime compared with 1312 general population controls (3.5%). The risk was mostly confined to patients with substance abuse comorbidity. [Emphasis added]
That means that the vast majority of the increased doubling of the risk is not from bipolar disorder alone, but rather from someone who has a drug abuse or alcohol problem, who also happens to have bipolar disorder. That’s a big difference, and one conveniently overlooked by the current study’s authors (and duly reported by TAC).
And the second part of the statement:
… and nearly five times as likely to be arrested, jailed, or convicted of an offense other than drunk driving.
This comes from Calabrese and colleagues’ (2003) study of 1,167 subjects from an epidemiologic study of bipolar prevalence using the Mood Disorder Questionnaire (MDQ) to assess for bipolar symptoms. This is an important distinction to note — these were not individuals actually diagnosed with bipolar disorder, but rather were simply assessed with a self-report screening measure they filled out on their own. Whether a person was arrested, jailed or convicted for offenses (of any nature or severity, except for DUIs) was also based upon self-report, not actual jail or court records.
The researchers reported that “MDQ-positive women reported more disruption in social and family life, while MDQ-positive men reported being jailed, arrested, and convicted for crimes.”
But here’s the catch — the researchers never specifically asked about substance or alcohol abuse. Since we know that such abuse is the primary determinant of violent crime and criminal behavior when combined with certain kinds of mental illness, that oversight is significant. It is a confound that means we cannot draw any meaningful conclusions from their findings regarding criminality and bipolar disorder. (Furthermore, it’s unclear why the researchers arbitrarily removed driving while under the influence of alcohol from their results, given its serious nature. They provided no rationale for doing so.)
So both statements that the researchers simply repeated in the current study (without any qualifications) are less than accurate, when you delve into their research support.
But let’s get on with the actual study, shall we?
Prevalence of Involvement in the Criminal Justice System During Severe Mania and Associated Symptomatology
The current study used the NESARC, “the largest U.S. epidemiologic survey to assess psychiatric disorders according to the DSM-IV criteria.” The study used a structured diagnostic interview to generate DSM-IV diagnoses for major axis I and axis II (personality) disorders, a reliable method used by researchers to diagnose disorders in large groups of people.
Here’s what they found:
Among NESARC wave 1 respondents (N=43,093), a total of 42,079 (97.7%) had valid responses to the questions in the mania section and, of these, 1,044 (2.5%) met specified criteria for having experienced at least one episode of mania.
Of these, 135 persons (13.0%) had legal involvement during the episode that they identified as the most severe in their lifetime.
What is “legal involvement”? Is that the same thing as committing a crime or going to jail? Is it the same thing as committing “specific criminal acts,” or a violent crime?
Well, no. It’s one of those fuzzy terms that researchers use when they want to make something seem like a bigger problem that it is. Here’s how they defined it:
Legal involvement was defined as being arrested, held at the police station, or put in jail, during the manic episode that the respondent identified as the most severe in his or her lifetime.
So in America, where you are innocent until proven guilty, researchers who are pursuing their own agenda define things a little differently. These aren’t people who actually were found guilty of committing a crime — they were simply people who may have had a run-in with the police.
The data the researchers do not provide are data that would put that 13 percent into some sort of context. How many people who did not have a manic episode also had “legal involvement?”
Sadly, the researchers do not report that number. An inquiry asking the researchers about this missing data was not returned.
While it’s interesting to see that 13 percent of people who report manic symptoms had some sort of legal involvement — serious or not — it’s a number that exists in a vacuum. It also demonstrates once again that the vast majority of individuals with bipolar disorder and mania had no legal problems.
Other demographic statistics, while not significant, also point in the direction of existing data. For instance, if you’re Black, you’re 35 percent more likely in this study to have reported legal involvement in the study. (Being African-American puts you at greater risk in general for incarceration in America.) If you have less than a high school education, you’re at 45 percent greater risk of having legal involvement.
But when all was said and done, and demographics were taken into account, this is what the researchers ultimately found that had the strongest statistical power (e.g., the most robust results):
When adjusted for demographic and clinical variables not in potential temporal conflict with the most severe lifetime manic episode, being male and having a first manic episode at age 23 or younger were associated with a higher risk of legal involvement.
Surprise! Being a young man — who have an incarceration rate 9 to 11 times that of women — is the strongest predictor of legal involvement. Also have social indiscretions and having both social and occupational impairment are strong risk factors for legal involvement, according the researchers’ data. Again, neither of which are surprising.
And being manic? Well, given the definition of mania (which is different than actually being diagnosed with bipolar disorder, an important distinction blurred by the researchers), it’s little surprise people with more energy, inflated self-esteem, flight of ideas and distractibility might find themselves at occasional odds with societal norms and laws. It’s like noting that you’re more likely to be pulled over for a DUI after you’ve drank too much and then try and drive.
I can’t really answer why TAC misrepresents the psychological research, and doesn’t bother to delve into it more deeply to examine the conclusions drawn by researchers. While I suspect it may be related to their own advocacy agenda, it could also just be due to sloppy reporting on their part.
What the study clearly shows is that manic symptoms are not linked to any specific criminal acts.
Read the blog post yourself: STUDY: Manic Symptoms Linked to Specific Criminal Acts
Calabrese, Joseph R.; Hirschfeld, Robert M. A.; Reed, Michael; Davies, Marilyn A.; Frye, Mark A.; Keck, Paul E., Jr.; Lewis, Lydia; McElroy, Susan L.; McNulty, James P.; Wagner, Karen D. (2003). Impact of bipolar disorder on a U.S community sample. Journal of Clinical Psychiatry, 64, 425-432.
Christopher, P.P, McCabe, P.J., Fisher, W.H. (2012). Prevalence of Involvement in the Criminal Justice System During Severe Mania and Associated Symptomatology. Psychiatric Services, doi: 10.1176/appi.ps.201100174
Fazel, Seena; Lichtenstein, Paul; Grann, Martin; Goodwin, Guy M.; Långström, Niklas; (2010). Bipolar disorder and violent crime: New evidence from population-based longitudinal studies and systematic review. Archives of General Psychiatry, 67, 931-938.
Sheldon, CT, Aubry, TD, Arboleda-Florez, J., Wasylenki, D., & Goering, PN. (2006). Social disadvantage, mental illness and predictors of legal involvement. International Journal of Law and Psychiatry, 29, 249-256.