Lancet 2012;380:2224-2260
These days it is difficult to find a paper published in The Lancet which doesn't have the word global in the title. However the Christmas edition of this journal surpasses itself by devoting all its content to the latest findings from the global burden of disease project. It is the first time the Journal has devoted an entire issue to one study.
This project is an excellent example of big teams answering big questions – and we are talking a big team here with 486 authors from 302 institutions in 50 countries contributing to the papers. The principal findings can be summarized as fewer people are dying but more people are living with disability with chronic disease such as musculoskeletal disorders, mental health disorders and injuries the commonest causes.
One paper is the dauntingly titled “A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010” and that’s before the colon in the title. Why is this important for psychiatrists working at the coal face? Well one of the reasons for justifying increased funding in mental health services is the often quoted prediction that depression will result in the second biggest disease burden in the future. What this paper does is turn the question around and look at what risk factors are important in causing disability rather than what diseases.
What the authors conclude is that high blood pressure, tobacco smoking and alcohol use are the three leading risk factors for global disease burden. This clearly is important for substance use services and should inform policy development and funding in those areas. What about other mental health problems? The way the authors did the study was to pair up risk factors with outcomes. Most of the risk factors were traditional, some would say 19th century, items such as occupational exposure to chemicals paired with specific outcomes. The authors did not look at poverty, social class or inequality as risk factors which I think is a serious omission.
What they did look at was childhood sexual abuse and intimate partner violence paired with depression and self-harm. They found that globally intimate partner violence rated the 23rd most important risk factor and childhood sexual abuse 33rd out of the 43 factors they looked at - however for North America intimate partner violence was 22nd whilst childhood sexual abuse rose to 21st. This illustrates the importance at a population level of managing these risk factors and at an individual level asking about sexual abuse and partner violence. Not doing so is like a cardiologist not asking someone about smoking.
The other six papers in this issue by the same group look at a variety of other measures of global burden of disease. These include global death rates where self-harm (that is suicide) is the 13th most common cause of death globally contributing about 5% of deaths in adults aged 15 to 49. In another paper looking at disability life years lost, globally depressive disorders were the 11th most common cause of disease burden (up from 15th in 1990).
Tuesday 30 April 2013
Cognitive Behavioural Therapy as an Adjunct to Pharmacotherapy for Primary Care Based Patients with Treatment Resistant Depression
Lancet, Early Online Publication, December 7, 2012 doi:10.1016/S0140-6736(12)61552-9
Also available online in the Lancet is the result of the CoBaLt trial which was a trial of cognitive behaviour therapy in people who had failed to respond to antidepressants in primary care in England. Failure to respond to antidepressants is the norm with only a third of people responding fully to antidepressants. The authors recruited 469 patients into the trial with 235 randomized to usual care and 234 to cognitive behaviour therapy plus usual care. To get into the trial participants had to have been on antidepressants for longer than six weeks, have a BDI score greater than 13, be aged 18-75 and meet ICD 10 criteria for depression.
Most people in the trial had received treatment for depression for over a year. One year after starting the trial 95 participants in the intervention group (46%) met criteria for response versus 46 (22%) in the usual care group. This means that about four people need to be treated with CBT for 12 to 18 sessions to get one extra person better after one year compared to usual care. One of the most noticeable aspects of this trial is that it demonstrated the acceptability of psychological treatments with 99% of patients invited for baseline assessment agreeing to participate. This is in contrast to the flawed STAR*D trial where only a quarter of participants were willing to be randomised to CBT.
Also available online in the Lancet is the result of the CoBaLt trial which was a trial of cognitive behaviour therapy in people who had failed to respond to antidepressants in primary care in England. Failure to respond to antidepressants is the norm with only a third of people responding fully to antidepressants. The authors recruited 469 patients into the trial with 235 randomized to usual care and 234 to cognitive behaviour therapy plus usual care. To get into the trial participants had to have been on antidepressants for longer than six weeks, have a BDI score greater than 13, be aged 18-75 and meet ICD 10 criteria for depression.
Most people in the trial had received treatment for depression for over a year. One year after starting the trial 95 participants in the intervention group (46%) met criteria for response versus 46 (22%) in the usual care group. This means that about four people need to be treated with CBT for 12 to 18 sessions to get one extra person better after one year compared to usual care. One of the most noticeable aspects of this trial is that it demonstrated the acceptability of psychological treatments with 99% of patients invited for baseline assessment agreeing to participate. This is in contrast to the flawed STAR*D trial where only a quarter of participants were willing to be randomised to CBT.
The context for this trial is the introduction of IAPT in England where IAPT stands for “improving access to psychological therapies” which is a £ 500 million project to train providers in cognitive behaviour therapy for depression and anxiety. This is an attempt to address the difficulty that clinicians always encounter when recommending CBT in that there is limited access to CBT therapists.
One obvious way to alleviate this is to massively increase the number of CBT therapists which is what they are doing in England. However criticisms of IAPT are that:
- it treats people who previously would not have had treatment and who would have got better without any intervention (just the passage of time);
- IAPT workers do not manage risk so anyone who is remotely suicidal gets booted up the system paradoxically increasing waiting times in mental health service;
- the system is protocol driven with little flexibility; and
- there is no system for the introduction of new therapies (that is therapies other than CBT).
Another way to manage the lack of CBT resources is to provide computerised therapies with mental health coaches to help people progress through them. We are planning on trying the second approach in Ottawa but a few more therapists would not go amiss – one approach would be to use consumer peers who get brief training in CBT but close supervision from fully trained therapists.
Medication for Attention Deficit–Hyperactivity Disorder and Criminality
New England Journal of Medicine 2012; 367:2006-2014 November 22, 2012 doi: 10.1056/NEJMoa1203241
Seeing people with ADHD it is often difficult weighing up the pros and cons of treatment. On the one hand there are benefits with a reduction of ADHD symptoms and on the other there are the risks of side effects, over prescription and development of tolerance and addiction. One paper that may help clinicians and patients reach a conclusion is this one from Sweden published in the NEJM.
The design is a case control study nested within a cohort study where the cohort cases are people with ADHD and the controls are a general population sample. The authors found that when men with ADHD were prescribed ADHD medication (mainly methylphenidate) they committed about a third less crimes than when they were not taking medication and that for women the figure was a 41% reduction.
A question not answered by this study is what happens when the medication is stopped – does the reduction in criminality persist – does the medication cause permanent change? The authors did not find any association between being on medication in 2006 and the crime rate in 2009 which suggests that the association between reduced criminality and being on medication does not persist. So back to being crime prevention officers….
Seeing people with ADHD it is often difficult weighing up the pros and cons of treatment. On the one hand there are benefits with a reduction of ADHD symptoms and on the other there are the risks of side effects, over prescription and development of tolerance and addiction. One paper that may help clinicians and patients reach a conclusion is this one from Sweden published in the NEJM.
The design is a case control study nested within a cohort study where the cohort cases are people with ADHD and the controls are a general population sample. The authors found that when men with ADHD were prescribed ADHD medication (mainly methylphenidate) they committed about a third less crimes than when they were not taking medication and that for women the figure was a 41% reduction.
A question not answered by this study is what happens when the medication is stopped – does the reduction in criminality persist – does the medication cause permanent change? The authors did not find any association between being on medication in 2006 and the crime rate in 2009 which suggests that the association between reduced criminality and being on medication does not persist. So back to being crime prevention officers….
People who Present with Self-Harm to Hospital have a Much Higher Tate of Premature Death at Follow-Up
Lancet 380, 1568-1574 2012
One of the reasons why self-harm is important is because people who present to hospital after self-harm have a high rate of death from other causes beside suicide. So over ten years at least 10% of any cohort will die – about half of these deaths will be due to suicide and the other half to premature death for other reasons. And this is in a group of people whose average age of presentation is late 20’s early 30’s.
I can’t think of another disorder presenting in this age group with such a high mortality. So it is good to see a paper in the Lancet in the last month by Keith Hawton’s group in Oxford describing the overall mortality following presentation to hospital emergency departments with self harm.
They followed up nearly 31000 people who had presented with self-harm in various centers in the UK for a median of 6 years. They found about 6% of the cohort died before the end of follow up with deaths due to natural causes 2-7.5 times greater than expected.
The importance of this paper is that this is the largest cohort study on this topic to date. This makes me think that treating self-harm as solely a mental health problem is not that useful and that we should be paying more attention to health promotion and physical needs in this population. Clearly an issue in Ottawa is this group of people getting access to good primary care treatment.
One of the reasons why self-harm is important is because people who present to hospital after self-harm have a high rate of death from other causes beside suicide. So over ten years at least 10% of any cohort will die – about half of these deaths will be due to suicide and the other half to premature death for other reasons. And this is in a group of people whose average age of presentation is late 20’s early 30’s.
I can’t think of another disorder presenting in this age group with such a high mortality. So it is good to see a paper in the Lancet in the last month by Keith Hawton’s group in Oxford describing the overall mortality following presentation to hospital emergency departments with self harm.
They followed up nearly 31000 people who had presented with self-harm in various centers in the UK for a median of 6 years. They found about 6% of the cohort died before the end of follow up with deaths due to natural causes 2-7.5 times greater than expected.
The importance of this paper is that this is the largest cohort study on this topic to date. This makes me think that treating self-harm as solely a mental health problem is not that useful and that we should be paying more attention to health promotion and physical needs in this population. Clearly an issue in Ottawa is this group of people getting access to good primary care treatment.
The Risk of Relapse of Agitation and Psychosis Increases if you Stop Risperidone in People with Alzheimer’s
New England Journal of Medicine 2012; 367:1497-1507 October 18, 2012 DOI: 10.1056/NEJMoa1114058
The New England Journal of Medicine has a paper on the risks of relapse of agitation and psychosis in people with Alzheimer’s who have stopped risperidone. Presumably this is a reaction to the movement not to prescribe antipsychotics to people with dementia because of the increased risk of strokes and short term cognitive problems.
The New England Journal of Medicine has a paper on the risks of relapse of agitation and psychosis in people with Alzheimer’s who have stopped risperidone. Presumably this is a reaction to the movement not to prescribe antipsychotics to people with dementia because of the increased risk of strokes and short term cognitive problems.
In
a small study the authors randomized in a double blind trial 110 people with
Alzheimer’s who had responded to risperidone for “agitation” or psychosis to
either continuing risperidone for 32 weeks; continuing placebo for 32 weeks; or
a half and half group who got risperidone for 16 weeks and then placebo for 16
weeks.
Not
surprisingly the rate of relapse in the people who continued was lower than those
who continued on risperidone with about a third of people on risperidone
relapsing compared to two thirds relapsing on placebo. However none of the
groups seem to have received any psychological treatment for “agitation” so I
wasn’t sure what clinical conclusions to draw from this paper.
In
clinical practice the choice is often between offering drugs or psychological
and environmental manipulation to manage these problems. The other thing to
note, as the authors point out is that risperidone actually wasn’t that
effective in treating the symptoms it was prescribed for with between a third
and two thirds of people stopping risperidone for some reason.
Classroom Interventions to Prevent Depression do Not Work
BMJ 2012;345:e6058.
It seems medical journals these days are more willing to publish articles which do not show a difference. One example of this is a cluster randomized trial of a cognitive behavioral program to prevent depression in adolescents delivered in schools in England. In this study the authors randomly allocated teenagers by school years to either receive the cognitive behavioral program; the usual curriculum for personal, social and health education supplemented by external experts; or the usual curriculum for personal, social and health education alone.
The cognitive behavioral program consisted of nine modules and two booster sessions of 50-60 minutes each. After a year, in those at high risk of depression, there was no difference between the groups in depressive symptoms. This is disappointing but not inconsistent with other school intervention programs which although well intentioned may actually increase risky behaviors (for example suicide prevention programs in schools).
The paradox is that systematic reviews have come out in favor of school based prevention programs for depression but it seems that these analyses included studies that did not have “placebo” arms. This is clearly important in depression which has a high rate of spontaneous remission.
So do you trust the single well done randomized controlled trial or the systematic review of studies with weaker methodology – a topic which will keep epidemiologists up at the bar late into the night. It remains somewhat baffling to me why health prevention programs in schools do not work – after all schools are all about learning new things. Any ideas?
It seems medical journals these days are more willing to publish articles which do not show a difference. One example of this is a cluster randomized trial of a cognitive behavioral program to prevent depression in adolescents delivered in schools in England. In this study the authors randomly allocated teenagers by school years to either receive the cognitive behavioral program; the usual curriculum for personal, social and health education supplemented by external experts; or the usual curriculum for personal, social and health education alone.
The cognitive behavioral program consisted of nine modules and two booster sessions of 50-60 minutes each. After a year, in those at high risk of depression, there was no difference between the groups in depressive symptoms. This is disappointing but not inconsistent with other school intervention programs which although well intentioned may actually increase risky behaviors (for example suicide prevention programs in schools).
The paradox is that systematic reviews have come out in favor of school based prevention programs for depression but it seems that these analyses included studies that did not have “placebo” arms. This is clearly important in depression which has a high rate of spontaneous remission.
So do you trust the single well done randomized controlled trial or the systematic review of studies with weaker methodology – a topic which will keep epidemiologists up at the bar late into the night. It remains somewhat baffling to me why health prevention programs in schools do not work – after all schools are all about learning new things. Any ideas?
Benzodiazepines Associated with Dementia
BMJ. 2012 Sep 27;345:e6231. doi: 10.1136/bmj.e6231.
Does using benzodiazepines make it more likely that you will develop dementia? The answer according to this French study is yes. They followed 1063 people in their 70’s over 15 years – 95 began using benzodiazpeines during the follow up and 968 did not use these drugs.
At the end of follow up 30 (32%) of the benzodiazepine group had developed dementia compared to 223 (23%) non-users which translates into a hazard ratio of about 1.6. However the devil here is in the detail - as all clinicians know one of the early signs of dementia can be poor sleep or anxiety for which benzodiazepines are prescribed.
Unfortunately the authors did not control for these problems as possible confounders meaning what they have described is an association, rather than causation, which could be explained by the co-existence of sleep or anxiety problems. A good example of confounding factors in an etiological study. Despite this in my experience there are still good reasons for not prescribing benzodiazepines in the elderly which are the large increased risk of falls and the immediate cognitive confusion they often cause.
Does using benzodiazepines make it more likely that you will develop dementia? The answer according to this French study is yes. They followed 1063 people in their 70’s over 15 years – 95 began using benzodiazpeines during the follow up and 968 did not use these drugs.
At the end of follow up 30 (32%) of the benzodiazepine group had developed dementia compared to 223 (23%) non-users which translates into a hazard ratio of about 1.6. However the devil here is in the detail - as all clinicians know one of the early signs of dementia can be poor sleep or anxiety for which benzodiazepines are prescribed.
Unfortunately the authors did not control for these problems as possible confounders meaning what they have described is an association, rather than causation, which could be explained by the co-existence of sleep or anxiety problems. A good example of confounding factors in an etiological study. Despite this in my experience there are still good reasons for not prescribing benzodiazepines in the elderly which are the large increased risk of falls and the immediate cognitive confusion they often cause.
Publication Bias - Research Transparency & Accessibility
JAMA August 8th 2012 Vol 308
I’m not sure how many psychiatrists subscribe to the Jiscmail list on Evidence Based-Health but one of the topics that comes up regularly there is the subject of publication bias. This is the phenomena that studies with positive findings are more likely to get published than studies that don’t show a difference.
The reason this matters is that the treatments we prescribe may not be as effective as we think. This has been a hot topic for antidepressants in recent years with some Systematic Reviews concluding that antidepressants are no better than placebos except for the most severe forms of depression. One way to get round this is to require all treatment studies to be Registered before they start so that at least we can see if a study has not been reported. However this only applies to studies about treatment and it takes some effort to find out what hasn’t been published.
The next generation of research transparency comes with making the results of all clinical studies available online for everyone to access. An interesting twist in the accessibility argument comes from John Ioannidis writing in JAMA. He argues that data sets should be registered – including those from non randomised controlled trials – to enable authors to ask questions about studies which haven’t been done but should be. In other words the data that will help answer important clinical questions is already out there if only the datasets were accessible.
I’m not sure how many psychiatrists subscribe to the Jiscmail list on Evidence Based-Health but one of the topics that comes up regularly there is the subject of publication bias. This is the phenomena that studies with positive findings are more likely to get published than studies that don’t show a difference.
The reason this matters is that the treatments we prescribe may not be as effective as we think. This has been a hot topic for antidepressants in recent years with some Systematic Reviews concluding that antidepressants are no better than placebos except for the most severe forms of depression. One way to get round this is to require all treatment studies to be Registered before they start so that at least we can see if a study has not been reported. However this only applies to studies about treatment and it takes some effort to find out what hasn’t been published.
The next generation of research transparency comes with making the results of all clinical studies available online for everyone to access. An interesting twist in the accessibility argument comes from John Ioannidis writing in JAMA. He argues that data sets should be registered – including those from non randomised controlled trials – to enable authors to ask questions about studies which haven’t been done but should be. In other words the data that will help answer important clinical questions is already out there if only the datasets were accessible.
Risk Prediction Tools
BMJ July 24th 2012 BMJ 2012;345:e4692
We cannot predict who will kill themselves. It is that simple. Most people who commit suicide are low risk (about 90% of suicides) and most high risk people do not commit suicide. One reason often put forward for the inability to predict suicide is the low base rate of suicide – about 11 suicides per 100,000 people a year.
But what about violence and antisocial behavior which is far more common? Are we any better at prediction? A systematic review and meta-analysis in the BMJ suggest not – well not by much. The authors found 68 studies which included nearly 25,000 people of whom about a quarter offended over four years.
Tools which predicted violence and sexual offending did better than tools which predicted non-violent offending. The positive predictive value for violent offending was 0.41 and the negative predictive value was 0.91. This means that when one of the prediction tools predicted someone was going to violently offend it would be right 4 times out of 10 and when it said someone wouldn’t then it would be right about 9 times out of 10.
Useful enough in clinical practice? As the authors state:
“the view that violence, sexual, or criminal risk can be predicted in most cases is not evidence based. This message is important for the general public, media, and some administrations who may have unrealistic expectations of risk prediction for clinicians. This expectation is not as high in other medical specialties, in which the expectation that the doctor will identify the individual patient who will have an adverse event is not a primary issue whereas psychiatry, in many countries such as the UK, has developed a culture of inquiries”.
We cannot predict who will kill themselves. It is that simple. Most people who commit suicide are low risk (about 90% of suicides) and most high risk people do not commit suicide. One reason often put forward for the inability to predict suicide is the low base rate of suicide – about 11 suicides per 100,000 people a year.
But what about violence and antisocial behavior which is far more common? Are we any better at prediction? A systematic review and meta-analysis in the BMJ suggest not – well not by much. The authors found 68 studies which included nearly 25,000 people of whom about a quarter offended over four years.
Tools which predicted violence and sexual offending did better than tools which predicted non-violent offending. The positive predictive value for violent offending was 0.41 and the negative predictive value was 0.91. This means that when one of the prediction tools predicted someone was going to violently offend it would be right 4 times out of 10 and when it said someone wouldn’t then it would be right about 9 times out of 10.
Useful enough in clinical practice? As the authors state:
“the view that violence, sexual, or criminal risk can be predicted in most cases is not evidence based. This message is important for the general public, media, and some administrations who may have unrealistic expectations of risk prediction for clinicians. This expectation is not as high in other medical specialties, in which the expectation that the doctor will identify the individual patient who will have an adverse event is not a primary issue whereas psychiatry, in many countries such as the UK, has developed a culture of inquiries”.
The authors point out that the performance of risk prediction tools for violence is about the same as risk prediction instruments for cardiovascular disease events (QRISK, SCORE and Framingham). So not quite time to put in the phrase “violence prevention officer” in a psychiatrist job description but for a taste of the future perhaps a viewing of Minority Report is in order.
Downplay on Delirium?
New England Journal of Medicine July 5th 2012 Vol 367 p30-39
If
depression is the common cold of psychiatry then delirium must be the backache.
It occurs in about 10% of general hospital admissions and yet the number of
treatment trials you can count on the figures of one hand. The latest from
the very serious New England Journal of Medicine is
not a treatment trial but just adds to the need for proper studies in this
area.
The authors, all from Boston, followed 225 people aged over 60 after they
had had cardiac surgery. Anyone
who has worked in this area knows that delirium post cardiac surgery is almost
the norm; what is not so well established is what happens long term. This study
found that 40% of those who developed delirium (about half the sample) had
still not recovered their preoperative cognitive function six months after
surgery compared with about a quarter of those without delirium. As would be
expected those who developed delirium had significantly worse preoperative
cognitive function.
Delirium
is both preventable and treatable (I find haloperidol like “magic dust” in this
condition). Surely we should be doing better than this – anyone up for a large
prevention trial?
Dispelling Dementia Factors
Lancet 2012; 380:50-58
The
Lancet doesn’t publish many papers on mental health but what it does do well is
the coverage of global health issues (that is health issues not just related to
North America, Europe and Australasia). The most recent example of this is a cohort study on the
incidence and mortality of dementia in Cuba, the Dominion Republic, Venezuala,
Peru, Mexico and China.
The
authors hypothesized that because education and occupational attainment seem to
protect against dementia, then in countries with elderly populations with
limited education and who have done manual or unskilled work, that the
incidence of dementia would be higher than in developed countries. They were
able to follow up about two thirds of the original cohort of 13000 people over
three to five years.
What
they found was the incidence rate of dementia was about the same as in
countries with higher incomes. This was most striking when they used a broader
definition of dementia than the DSM IV which they found underestimated the true
incidence of dementia in middle income countries. They also found that
education and occupational attainment protected against dementia in middle
income countries just as it does in high income countries.
Welcome!
Welcome to my Psychiatry Journal Review Blog! The aim of this blog is to give a personal view of the world of research and how it affects us and our patients in mental health. It is based mainly on articles from general medical journals relevant to mental health, but I'll also be posing questions from time to time to gather insight on mental health topics.
As my blog was previously hosted on the University of Ottawa Department of Psychiatry's website, all posts appearing on April 30, 2013 are previous posts that have been migrated to my Blogger page. New posts will begin May 1, 2013!
Cheers,
Simon
Subscribe to:
Posts (Atom)