In yet another example of sensationalism posing as legitimate journalism, the Associated Press’s Carla K. Johnson penned an article over the weekend calling people with mental illness who live in nursing homes a “threat.” What kind of threat? Well, according to the article, it appears to be the usual one, drawing an unscientific and unsupported link between mental illness and violence:
Over the past several years, nursing homes have become dumping grounds for young and middle-age people with mental illness, according to Associated Press interviews and an analysis of data from all 50 states. And that has proved a prescription for violence, as Jackson’s case and others across the country illustrate.
Younger, stronger residents with schizophrenia, depression or bipolar disorder are living beside frail senior citizens, and sometimes taking their rage out on them.
Wow, that sounds like a serious problem. So let’s look at the new scientific data the AP is reporting on to back up this assertion:
Numbers obtained through the Freedom of Information Act and prepared exclusively for the AP by the Centers for Medicare and Medicaid Services show nearly 125,000 young and middle-aged adults with serious mental illness lived in U.S. nursing homes last year.
That was a 41 percent increase from 2002, when nursing homes housed nearly 89,000 mentally ill people ages 22 to 64. Most states saw increases, with Utah, Nevada, Missouri, Alabama and Texas showing the steepest climbs.
Ah, so this isn’t new scientific evidence per se, just what researchers would call “data fishing” — fishing through datasets looking for interesting (but not necessarily significant) trends.
Since the AP provided none of the actual data (you know, like you might find in an actual research study), you have no way to determine whether a 41 percent increase over 6 years (7 percent annually) is significant or not. Were the previous years worse or better? Has anything changed in mental health benefits since 2002 that explains this increase? You don’t know, because the article offers little information on these points.
So what data points does the AP offer to draw its conclusions? Oh yeah, four stories — 4 data points — out of those 125,000 people who live in nursing homes with a mental illness.
One of the cases describes a woman who had dementia — common in nursing homes — and multiple sclerosis, as well as depression. Hardly a strong example if you want to indict mental illness, as depression is a commonly-used, garden-variety diagnosis in nursing homes.
Another of the four cases describes a 62-year-old (hardly a “younger, stronger resident”) who had no identified mental illness but was described only as “incompetent to stand trial.” Well, that could be for a multitude of reasons, including dementia (again, something very commonly found in nursing home patients).
Even the lead example — a 50-year-old (again, this is a “younger, stronger” person?) — who was not identified with any particular mental illness but was identified with something that is strongly correlated with future violence risk — he had a history of aggression. Yes, this is no surprise. People with a past history of violence or aggressive behavior have a strong likelihood of future violent and aggressive behavior. It is, in fact, the only reliable and strongest predictor for future violence.
While indeed “nursing home dumping” might be a growing problem in the U.S., articles like this continue to do a disservice to the real story by indicting one population rather than the societal conditions that make such an option attractive. If society offered people with mental illness better treatment options than a nursing home, be certain they would take them in an instant.
With perhaps less finger-pointing toward the mentally ill (and especially these sensationalistic, unscientific links to violence), we as a society can find solutions to this kind of problem. Articles like this one by the Associated Press, however, do little to move us toward such solutions.
Read the full article: Mentally ill a threat in nursing homes.