Wrong on almost every level

It's not common to find a newspaper that's wrong on almost every level, but there's one here:

Gloom lifts as Sad affects fewer people. Aberdeen Press and Journal (13/1/11).

 

It reports on the number of people being diagnosed with Seasonal Affective Disorder (SAD) in Aberdeen, and comments that: "The dark nights are taking their toll on fewer people in the north-east, it emerged yesterday." This assumption comes from the finding that fewer people are being diagnosed with SAD than a few years ago.

 

Of course, the number of diagnoses doesn't mean that people no longer have the disorder since a diagnosis doesn't necessarily occur in everyone with a particular condition. Also, one often finds that it's only a small number of specialists who might make regular use of a SAD diagnosis. Many psychiatrists might make a diagnosis of major depression rather than SAD. Same suffering but different diagnosis. In DSM-IV, SAD is not a unique diagnosis - it is a 'specifier' of major depression. SAD doesn't have a unique diagnostic category in ICD-10 either - there are provisional criteria but it would normally be diagnosed with the same code as the majority of depression.

 

So, reductions in rates of diagnosis may reflect nothing other than variations in the use of certain diagnostic codes, or a reduced preference for specifying descriptors such as "seasonal affective disorder".

 

"The health board said the drop was a reflection of increased awareness among GPs." Awareness of what, exactly? If GPs are now more aware of SAD, then they should be more alert to the symptoms and making the diagnosis more frequently. This isn't the case, according to the information coming from the Freedom of Information request. So, can we assume that GPs are more aware of the condition, but making fewer diagnoses?

 

Apparently not. NHS Grampian would like to believe that "...patients may be 'self-diagnosing', after becoming aware of the symptoms and taking action to treat themselves". Even if patients aren't treating themselves, NHS Grampian "has a supply of light boxes which can be loaned to patients for the winter. The lamps are designed to give natural light indoors, helping to reduce hormonal changes." Nice idea, but there's not much evidence of light effects on specific hormonal levels (Full-Spectrum Fluorescent Lighting Effects on People: A Critical Review.) We don't know either whether more people are using the light boxes.

 

The article goes on to say that: "Other ways to alleviate Sad symptoms include herbal remedies such as St John’s wort, light exercise and eating whole grain foods, root vegetables and fresh fruit. All of these help to give sufferers energy." Unfortunately, the combination of 'Seasonal Affective Disorder' and 'vegetables' or 'fruit' or 'grain' in PubMed doesn't identify any articles. There are a few studies that involve exercise AND light therapy, but "...only weak preliminary evidence exists for use of SJW in SAD". This statement came from a paper in the Journal of Alternative and Complementary Medicine in 2009. When an alternative therapy journal is saying there's not much evidence for St John's Wort in SAD you can be pretty sure there's no evidence. So, there's minimal to no evidence that any of these things are of much benefit in treating SAD.

 

Finally, we've got a comment from Mary Scanlon, the health spokeswoman for the Scottish Conservatives: "It is also encouraging that methods are being used which reduce the prescription of antidepressants." She is clearly clueless when it comes to understanding the figures or the treatment of depression. The number of antidepressants saved by not giving them to a dozen fewer people with SAD (many of whom wouldn't have had them in the first place) is not going to make much dent in the Scottish Government's asinine attempt to reduce antidepressant prescribing. The antidepressant HEAT target is now a psychological therapies access target, so perhaps it's only Mary Scanlon who thinks it's important.

Stigma among people with mental health problems

A great deal has been written about the stigma associated with mental illness. Much of it has focused on stigma present in doctors, psychiatrists, and other professionals (for example: NORDT, C., ROSSLER, W. & LAUBER, C. (2006) Attitudes of Mental Health Professionals Toward People With Schizophrenia and Major Depression. Schizophrenia Bulletin, 32, 709-714).

But we often forget that doctors and patients/ service users are part of the general public, and attitudes held by doctors are often similar to those held by the 'general public'. One might expect that those with experience of mental health problems might have had their attitudes modified by their experience, and that they would be less stigmatising.

This is unlikely, however, and someone with a history of chronic depression, who has lots of contact with other service users, was telling me that there remains an implicit 'hierarchy' of illnesses. The hierarchy operates by partially accounting for changes in behaviour or level of group activity. For example, if someone hasn't taken part in social activity, the extent to which the rest of the group 'disapproves', or excuses their absence depends on the person's condition.

The hierarchy is:

1. bipolar disorder
2. schizophrenia
3. depression
4. alcohol problems
5. drug problems

That is, those with bipolar disorder have a greater leeway for behaviour - the reason given was that it is somewhat "trendy". Those with drug problems come second place to alcohol problems. Guess addictions are always disadvantaged in this respect.

So, despite professional attitudes often being critiqued, such apparent stigma is perhaps equally present among service users...

More misunderstandings about antidepressant prescribing in Scotland

Just before Christmas most papers in Scotland were reporting on recent data published by the Information Service Division (ISD) of NHS Scotland. ISD crunch most the health service information in Scotland. In many cases, they make useful and helpful information available on their website. Sometimes, there's an attempt at understanding and interpreting the data. But not always.

Here's a typical report from the Scottish Television website. The headline of 'One in ten adults' on antidepressants is followed by the comment that "One in ten adults in Scotland are thought to be on antidepressants, according to official figures published." They don't say who thinks this, however. The ISD data release doesn't say this because they don't know how many patients actually took the antidepressants dispensed: "It should be noted that the proportion of the population taking any of the drugs listed is purely an estimate; the actual patient numbers are unknown."

How can they not know how many people took the antidepressants? Well, lets look at how the numbers are calculated. The first problem is that quantities of antidepressant drugs are often reported as Defined Daily Doses (DDDs). A DDD is defined by the World Health Organisation as:

"...the assumed average maintenance dose per day for a drug used for its main indication in adults."

The DDD for different drugs varies. For example, it is 75mg/ day for Amitriptyline and 20mg/ day for Fluoxetine. The starting dose for Fluoxetine is 20mg/ day so you're getting one DDD from day one. With Amitriptyline there may be a more gradual dose increase, meaning that you're not getting a full DDD until later on.

That's not the only problem, though. For most people, 75mg/ day of Amitripytline (or other tricyclic drug) is sub-therapeutic. That means that you need a bigger dose to get maximum benefit. For many people, a therapeutic dose is around 150mg/ day (two DDDs); and for some people, it might be even higher. Some even consider a proper trial of a tricyclic antidepressant (TCA) to be at least 200mg/ day (2 1/2 DDDs). A study done in Scotland in 1996 found that in primary care, tricyclic antidepressants were given in subtherapeutic doses (MacDonald, 1996). This confirmed a similar study that reported that only 13% of patients received a TCA for depression at an effective dose (Donoghue, 1996).

What does this mean for understanding Defined Daily Doses? Well, it tells us that if we see the total number of DDDs of antidepressants per year increasing, this increase could be due to a number of things (or a combination):

1. The same number of patients getting higher doses of drugs. In many cases, this might mean patients getting effective doses of antidepressants - this has to be a good thing.
2. The same number of patients getting antidepressants for longer. Since antidepressant prescription in primary care tends to be for short periods, ensuring that patients get treatment that is guideline-based and prevents recurrence has to be a good thing.
3. More people getting the same quality of prescribing. This isn't necessarily a good thing.
4. Different types of drugs being prescribed. A dose of 50mg of Amitriptyline is only 2/3 of a DDD yet 20mg of Fluoxetine is 1 DDD. Simply changing the type of drug being given can increase the number of DDDs by 1/3 or more. Over the last 10 years, we've certainly seen a preference in primary care for Selective Serotonin Reuptake Inhibitors (SSRIs) over TCAs. Again, this is in accordance with good clinical practice and guideliness - SSRIs are equally effective; better tolerated; and safer in overdose.

The point is that counting DDDs tells us nothing about numbers of patients or the quality of prescribing. The Scottish Government have assumed that more DDDs are bad because it means more people getting antidepressants. Let's remember that the Scottish Government made a 2007 Election Manifesto pledge to "...to reduce the use of anti depressants by 10 per cent by 2009." This was translated into a HEAT target shortly after. Unfortunately, they didn't really understand whether the current antidepressant usage was too high, too low, or just right. They simply made a rather ill-informed pledge. This point was made by the Public Audit Committee last year when they made this clear to the Government:

"However the target did not take account of initial prescribing levels in each NHS board. It
was also unclear what the correct rate of prescribing should be and whether reducing the rate of prescribing was always appropriate."

and:

"...we do not know whether the difference in DDD levels is due to more people receiving antidepressants and/or prescribing at a higher dose and/or prescribing for a longer duration”. The Scottish Government also acknowledged in written evidence that “information regarding the number of people on antidepressants is not currently collected so we do not know whether the number of people taking antidepressants has increased'"

That's a pretty explicit indictment on the Government not really knowing what it was talking about when it set the HEAT target. However, this is still the case. ISD do not know any better than anyone else whether increases in the total number of antidepressants reflects better treatment or not. We don't even know whether the antidepressants are being prescribed for depression, anxiety, pain, nocturnal enuresis, or a range of valid indications. The Scottish Government has generally tried to move away from its daft HEAT target and has tried to say that the target was about better treatment all along.

Fortunately, people have looked at the possibility that antidepressants are being prescribed for longer periods. Tony Kendrick and colleagues examined whether observed rises in antidepressants were due to the drugs being prescribed for longer periods, rather than in more people. It seemed that it was the case that prescriptions were for longer periods. Over time, the number of new cases of depression over time actually falling, and prescribing for longer periods (as per guidelines) increasing (Moore, 2009). See image below.

There remain many myths about antidepressant prescribing. Some of them have entered the population consciousness because of daft government targets and the media haven't really helped to increase understanding of some of the issues. After all, 'depression treatment getting a bit better over time' is not as good a story as 'Scotland struggles to kick its antidepressant habit'.

References
MACDONALD, T. M., MCMAHON, A. D., REID, I. C., FENTON, G. W. & MCDEVITT, D. G. (1996) Antidepressant drug use in primary care: a record linkage study in Tayside, Scotland. BMJ, 313, 860-861.

DONOGHUE, J., TYLEE, A. & WILDGUST, H. (1996) Cross sectional database analysis of antidepressant prescribing in general practice in the United Kingdom, 1993-5. BMJ, 313, 861-862. [PDF]

MOORE, M., YUEN, H. M., DUNN, N., MULLEE, M. A., MASKELL, J. & KENDRICK, T. (2009) Explaining the rise in antidepressant prescribing: a descriptive study using the general practice research database. BMJ, 339, b3999. [Full Paper]

Irish Government attempt to 'regulate' suicide prevention groups

A recent article in the Irish Times suggested that the Irish Minister of State for Mental Health, John Moloney was looking at regulating suicide prevention groups in Ireland.

According to the article, Ireland spends 5.5 million Euros on suicide prevention each year, and there are around 500 different groups who have some activity relating to suicide prevention. These will include national groups such as the Samaritans, but also includes lots of smaller regional and community groups; most of which have a degree of uncertainty over remit, governance, and performance management. Last year there were just over 500 suicides in Ireland. That means that there is a suicide prevention group for each suicide in Ireland.

Why would there be so many groups? Well, suicide is undoubtedly an emotive subject and there is often a response among communities that 'something must be done'. We find it hard to react to apparent suicide clusters in a detached, rational way and most people struggle to distentangle tragedy from statistical chance. The apparent spike in suicides at the FoxConn manufacturing plant in China led people to associate poor working conditions with ripe conditions for suicides. However, the rate of suicide at the plant was somewhat less than the rate in Scotland, for example. The number of suicides seen was no higher than would be expected from a large population (FoxConn had around 300,000 employees - the same number as a sizeable town).

So, should there be fewer groups trying to tackle suicide? This is an area where there is disappointingly little evidence to support typical approaches. Of course, one can't count things that didn't happen, so it can always be argued (often by the charities themselves) that they were preventing suicides. However, when you start comparing areas with lots of such groups with those that don't you generally don't see any evidence that the presence or activity of such groups is having a detectable effect on suicide rates. They may be offering support for those in distress, but they don't seem to have much of an effect on actual numbers of suicides.

For example, if we look at how many people phoning The Samaritans were actually suicidal, it turns out to be a small proportion of all calls. A presentation from The Samaritans at a Choose Life conference in 2010  reported 148,000 contacts, of which 84% were by telephone. That makes 124,320 telephone contacts. Only 20% of those were in a 'suicidal crisis', which means that 28,864 were suicidal when they phoned The Samaritans. The risk of suicide in those who are suicidal is, of course, higher than the general population but it's still relatively low, and being in 'suicidal crisis' covers everything from people standing on the edge of a bridge to someone contemplating what the point of life is and thinking that they would sometimes rather be dead.

Knowing how many suicidal people will die by suicide is difficult. However, we know that between 0.5% and 1.8% of people who have self-harmed will die by suicide in the year following the incident of self-harm (Owens, 2002). If we assumed that the risks were the same, then between 124 and 520 people will die by suicide in the year following their call to The Samaritans.

But the risks aren't the same. Suicidal thinking is reasonably common. The lifetime prevalence of suicidal ideation, plans, and attempts is 9.2% (Nock, 2008). The percentage of people who will die by suicide is (assuming a rate of 20 per 100,000) is 0.02%. Therefore, suicidal ideation is at least 460 times more common than suicide. If we therefore take 28,864 people with suicidal thinking, only 62 of those will die by suicide in the following year. That's a rate of just over one per week.

We don't know how protective a call to The Samaritans is, but the effect is unlikely to last a year. Indeed, we know that almost half of all suicide attempts have less than 10 minutes between thinking and acting (Diesenhammer, 2009), so the person would have to call The Samaritans pretty quickly in order for them to prevent at least 50% of all suicides/ attempts.

If the ability of The Samaritans and other such telephone crisis lines to prevent suicides seems pretty weak, what is their point? The reasons for calling, according to the presentation cited above, are:

• 55% sad or low
• 51% lonely and isolated
• 40% anxious all the time
• 39% have mental health problems
• 35% family problems
• 35% bereavement

The Samaritans undoubtedly provides a source of support to those with a range of mental health issues, but they're unlikely to be preventing suicide. First, because at least 50% of suicides are very impulsive in that there isn't much time between thinking about it and acting upon these thoughts. Second, because they're not going to be having contact with large numbers of people at very high risk of suicide. Most callers are lonely, sad, or anxious; not suicidal.

Perhaps the Irish have a point in trying to bring some form of regulation to the field? If they're spending 5.5 million euros on a range of groups who are unlikely to be preventing suicide, then what is the state getting for their money? Some groups will be providing support for those bereaved by suicide, and others will be supporting people with a range of mental health problems, but this is a different activity to actually preventing suicide. As yet, there is precious little evidence for these activities in preventing suicide. Indeed, the only things which have compelling evidence to reduce suicide rates are "Physician education in depression recognition and treatment and restricting access to lethal methods reduce suicide rates." (Mann, 2005) Perhaps our (suicide prevention) money could be better spent elsewhere...


References
OWENS, D., HORROCKS, J. & HOUSE, A. (2002) Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of Psychiatry, 181, 193-199)

NOCK, M. K., BORGES, G., BROMET, E. J., ALONSO, J., ANGERMEYER, M., BEAUTRAIS, A., BRUFFAERTS, R., CHIU, W. T., DE GIROLAMO, G., GLUZMAN, S., DE GRAAF, R., GUREJE, O., HARO, J. M., HUANG, Y., KARAM, E., KESSLER, R. C., LEPINE, J. P., LEVINSON, D., MEDINA-MORA, M. E., ONO, Y., POSADA-VILLA, J. & WILLIAMS, D. (2008) Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. British Journal of Psychiatry, 192, 98-105.

DEISENHAMMER, E. A., ING, C. M., STRAUSS, R., KEMMLER, G., HINTERHUBER, H. & WEISS, E. M. (2009) The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Journal of Clinical Psychiatry, 70, 19-24.

MANN, J. J., APTER, A., BERTOLOTE, J., BEAUTRAIS, A., CURRIER, D., HAAS, A., HEGERL, U., LONNQVIST, J., MALONE, K., MARUSIC, A., MEHLUM, L., PATTON, G., PHILLIPS, M., RUTZ, W., RIHMER, Z., SCHMIDTKE, A., SHAFFER, D., SILVERMAN, M., TAKAHASHI, Y., VARNIK, A., WASSERMAN, D., YIP, P. & HENDIN, H. (2005) Suicide prevention strategies: a systematic review. JAMA, 294, 2064-74.

Costs of obesity drugs

The Courier, a newspaper based in Dundee, Scotland recently reported that NHS Tayside were spending 19 times more on obesity drugs than it did in 1999. My goodness, 19 times more. They cite number of drugs, including: ORLISTAT, SIBUTRAMINE (REDUCTIL) and RIMONABANT (ACCOMPLIA). Of course, Sibutramine has been removed from the market in most European countries because of safety issues, so one would hope that NHS Tayside aren't prescribing much of this anymore.

However, most of these drugs weren't widely available in 1999. Orlistat and Sibutramine were only recommended by NICE in 2001,and Rimonabant has only been available since 2008.

So, it would seem likely that NHS Tayside were not prescribing many drugs in 1999 that had neither been recommended by bodies such as NICE, or hadn't actually been released yet. I've done a bit of research and found that the amount of money spent by the consumer on hybrid cars is massively greater than it was in 1950. Should we be alarmed?

Mental Welfare Commission turn down Freedom of Information request

There's an interesting exchange between a Mr McLean and the Mental Welfare Commission for Scotland (MWC) at What Do They Know.com?

Essentially, Mr McLean wishes to know "...the number of reported (alleged or  proven) cases of abuse (physical, sexual, verbal, and mental) of detained/sectioned patients in Scotland reported by anybody." Okay, the request is strangely formal, and there is some ambiguity about how aspects of abuse are defined, but some might consider this a reasonable request. One of the MWC's statutory duties is to protect the interests of those with mental disorder in Scotland (although the Commission tend to focus on detained patients). Their website states that they believe that everyone with a mental illness should: "have the right to live free from abuse, neglect or discrimination". Nothing to disagree with that statement.

Mr McLean has made a request for numbers (not patient details, obviously) of cases of abuse of detained patients in Scotland. Since detained patients come very much within the remit of the Commission, one might imagine that they would/ should know how many cases there have been involving alleged abuse. Of course, if the Commission haven't been notified then they might not have such information, but they should be able to provide the number of cases reported to the Commission in recent years.

However, the Commission has invoked section 12(1) of the Freedom of Information (Scotland) Act 2002 which means that they consider it too costly to provide such information. There is a general limit of £600, so if a public body thinks it will cost more than £600 to find the information, they don't have to provide it.

Of course, there may be some people who think that a public body whose duties include monitoring of abuse of people with mental illness should be able to count the number of cases of abuse of people with mental illness on its records without it costing more than £600. The Commission will probably have some kind of database, and as most people know, a database that can't output such simple counts of relevant records isn't up to much. From what I've heard, the Commission has adopted a relatively 'paperless' office and forms and documents tend to be scanned into the system.

Despite some of the ambiguous language of the request, I don't think it should be too hard to provide Mr McLean with the numbers of cases of abuse (however defined) of detained patients in Scotland that it has on its records, and the argument that it would cost too much seems more than a little spurious.