Sunday, 27 May 2007

There's a lot of it about

More evidence in the last week of the prevalence of mental health problems in general hospitals. This time it’s a report from the Agency for Healthcare Research and Quality (AHRQ) in the USA who describe hospital care for adults age 18 and older with mental health or substance abuse problems who were treated for those conditions in community (as opposed to speciality) hospitals in 2004. These are the most numerous hospitals in the USA (about 5000 institutions) – they are non Federal general hospitals but exclude speciality psychiatric or substance abuse treatment facilities.

A number of interesting facts and figures.

  • Most psychiatric care occurs in the general hospitals, that is 44 million days of care to patients with mental health and substance abuse disorders compared to 27 million days of care in psychiatric facilities.

  • Hospital stays for adults with only secondary MHSA diagnoses* were 20 percent longer than adults with no MHSA diagnosis (5.4 versus 4.5 days).

  • Adults with only secondary MHSA diagnoses were the most likely to be admitted through the ED (63.6 percent). They were 1.4 times more likely to be admitted through the ED than those with no MHSA diagnosis (45.0 percent) and 1.3 times more likely than those with only a principal MHSA diagnosis (50.7 percent).

  • The costs of stays with only secondary MHSA diagnoses were slightly less than stays with no MHSA diagnosis ($8,500 versus $8,900).

  • Care for adults with only secondary MHSA diagnoses accounted for roughly 18 percent of hospital stays and total hospital costs and 20 percent of all days in the hospital.

  • The top four secondary diagnoses were mood disorders, substance abuse disorders, schizophrenia and other psychoses (surprisingly) and delirium. No mention of somatoform disorders.

*Secondary MHSA disorders are people diagnosed with a principle non mental health diagnosis (for example myocardial infarction) but also have a secondary mental health diagnosis.

So to summarise about 1:6 admissions to community hospitals in the USA also have a mental health disorder, which costs slightly less to treat than those without a mental health disorder but they have a 20% longer admission. Presumably they don’t have increased costs because the increased length of stay is due to placement issues? The reassuring thing is that this confirms previous work that about 1:5 people admitted to hospital also have a mental health or substance abuse problem – there’s a lot of it about.

Sunday, 13 May 2007

The fundamental attribution error or why patients get blamed for their problems

One of the things I notice working in acute mental health services is the tendency some teams have for blaming patients for their problems. This usually takes three forms. Firstly patients are said to have a “personality disorder” – a bit like having blue eyes it can’t be changed. Secondly that their substance abuse stops them being treated and lastly that they don’t turn up to appointments. Therefore the patients are really to blame for not getting better. However this view of patients is based on a well recognised fallacy in reasoning called the fundamental attribution error. This is based on experimental studies that have repeatedly demonstrated that as humans we attribute much more weight to “personality” factors that to environmental factors in explaining people’s behaviour. In fact the main determinant of people’s behaviour seems to be external environmental factors rather than “personality”.

The classic experiment that demonstrated this was described in 1967 by Jones and Harris. Subjects listened to pro- and anti-Fidel Castro speeches (it just had to be the USA….). Subjects were asked to rate the pro-Castro attitudes of the speakers. When the subjects believed that the speakers freely chose the positions they took (for or against Castro), they naturally rated the people who spoke in favour of Castro as having a more positive attitude toward Castro. However when the subjects were told that the speaker's positions were determined by a coin toss, they still rated speakers who spoke in favour of Castro as having a more positive attitude towards Castro than those who spoke against him. The subjects were unable to see the speakers as mere debaters performing a task chosen for them by circumstance; they could not refrain from attributing some disposition of sincerity to the speakers.

A further demonstration of the impact of environment on people’s behaviour (as opposed to personality) is shown by Philip Zimbardo’s classic experiments in August 1971 looking at the behaviour of “normal college students” put in to the roles of guards and prisoners. Some guards turned into sadists, despite knowing that the “prisoners” had done nothing wrong and the experiment had to be stopped after six days. "Many of the normal, healthy mock prisoners suffered such intense emotional stress reactions that they had to be released in a matter of days; most of the other prisoners acted like zombies totally obeying the demeaning orders of the guards; the distress of the prisoners was caused by their sense of powerlessness induced by the guards who began acting in cruel, dehumanizing and even sadistic ways. The study was terminated prematurely because it was getting out of control in the extent of degrading actions being perpetrated by the guards against the prisoners - all of whom had been normal, healthy, ordinary young college students less than a week before." (Zimbardo has just written a book “The Lucifer Effect: Understanding how good people turn evil” which makes links between this phenomena and such abuses as the torture that occurred in Abu Ghraib).

So in clinical practice I've become wary of blaming the patient for their problems – their behaviour is far more likely to be due to their environment than their “personality”. But of course understanding their environment requires time and interest to enquire about. Something that may be lacking in acute clinical practice.


Jones, E. E. & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology 3, 1–24

Zimbardo P. The Lucifer Effect: Understanding how good people turn evil. Random House. 2007 ISBN-10: 1400064112