Online First articles may not be available until 09:00 (UK time) Friday.

Press releases Monday 17 to Friday 21 November 2008

Please remember to credit the BMJ as source when publicising an article and to tell your readers that they can read its full text on the journal's web site (http://www.bmj.com).

(1) Patients with depressive disorders or schizophrenia more likely to re-attempt suicide
(2) Older people should have the flu jab this winter, warn experts
(3) The government's obesity strategy, more of the same rhetoric
(4) Screening for colorectal cancer detects unrecognised disease
(5) Hospitals could be fined millions of pounds even if they reduce infection risk

(1) Patients with depressive disorders or schizophrenia more likely to re-attempt suicide
(Research paper: Risk of suicide after suicide attempt according to coexisting psychiatric disorder: Swedish cohort study with long term follow-up)
http://www.bmj.com/cgi/content/full/337/nov18_3/a2205
(Research paper: Hospital admissions for self harm after discharge from psychiatric inpatient care: cohort study)
http://www.bmj.com/cgi/content/full/337/nov18_3/a2278
(Editorial: Suicide risk after a suicide attempt)
http://www.bmj.com/cgi/content/short/337/nov18_3/a2512

Men and women who have tried to kill themselves and are suffering from unipolar disorder (major depression), bipolar disorder (manic depression) or schizophrenia are at a very high risk of committing suicide within a year of their first attempt, concludes a study published today on bmj.com.

This is the first time research has identified a link between specific psychiatric disorders and increased suicide risk in such a large study of people who have attempted suicide. The authors call for prevention programmes to target these high risk groups.

It is well known that there is a 30-40 times increased risk of death from suicide in individuals who have previously attempted suicide compared with the general population. But little is known about the impact of coexisting psychiatric disorders on the risk of suicide in this group.

Dag Tidemalm and colleagues from the Karolinska Institutet in Stockholm studied nearly 40 000 individuals (53% women) who were admitted to hospital in Sweden following a suicide attempt during 1973-82. They analysed how many suicides were completed during the 30 year follow-up and if the risk varied with type of psychiatric disorder.

The authors found that schizophrenia and unipolar/bipolar disorder were the strongest predictors of completed suicide throughout the follow-up period. In patients suffering from unipolar/bipolar disorder, 64% of all suicides in men and 42% of suicides in women occurred within the first year of follow-up; the matching figures for schizophrenia were 56% in men and 54% in women.

Death from suicide occurred mostly within the five years after the initial suicide attempt.

People suffering with most other psychiatric disorders had a lower but still significantly increased risk of suicide. Interestingly, individuals suffering from adjustment disorder, post-traumatic stress disorder and alcohol abuse (men only) were not at significantly increased risk of re-attempting suicide compared to suicide attempters without a psychiatric diagnosis at baseline.

The authors call for patients who have unipolar/bipolar disorder or schizophrenia and previous suicidal behaviour to be given more intensive after-care, especially in the first few years after trying to kill themselves.

In an accompanying editorial, Dr Udo Reulbach from the National Suicide Research Foundation in Ireland and Professor Stefan Bleich from the Medical School of Hanover in Germany, explain that suicide is one of the 10 leading causes of death worldwide with predictions of 1.5 million people dying from suicide each year by 2020. Therefore, they say, suicide prevention must be made a health service and public health priority on medical, ethical and cost effectiveness grounds.

In another research paper, Professor David Gunnell from the University of Bristol and colleagues report that non-fatal self-harm may occur in over 10% of adults discharged from psychiatric inpatient care in England and Wales, and that the risk is greatest in the first month. Patients who had previous self-harming behaviour were at the greatest risk. Others at increased risk included women, the young, and those with depression, personality disorders and substance misuse.

These findings suggest the need to develop interventions to reduce the risk of fatal and non-fatal self-harm in the weeks immediately after hospital discharge - these might include improved discharge arrangements and clear crisis plans and lines of communication with specialist staff.

Contacts:
Suicide paper: Professor Bo Runeson, Department of Clinical Neuroscience, Karolinska Institutet, Division of Psychiatry, Stockholm, Sweden
Email: bo.runeson@ki.se

(2) Older people should have the flu jab this winter, warn experts
(Editorial: Influenza vaccine in the over 65s)
http://www.bmj.com/cgi/content/short/337/nov18_3/a2545

Despite recent doubts about its effectiveness, the influenza vaccine does give valuable protection against illness, hospital admission and death caused by influenza, and people over 65 should have the flu jab this winter, say experts on bmj.com today.

Several prominent media articles have suggested that the flu vaccination programme for the over 65s is not worthwhile. Rachel Jordan from the University of Birmingham and Jeremy Hawker from the Health Protection Agency review the evidence surrounding the ongoing controversy.

The annual influenza vaccination campaign is one of the most successful public health programmes in the UK, with 78% of people over 70 receiving the jab, the highest in Europe.

However, according to the authors, it is notoriously difficult to find out the full effectiveness of the influenza vaccine. For instance, the vaccine may not fully match the strain of a particular year, influenza seasons vary in timing and strength, and few studies measure influenza specific outcomes.

Previous trials have shown that the vaccine is effective in preventing influenza in those over 60. But recently, some researchers have questioned the validity and plausibility of findings for other less specific benefits provided by the vaccine, such as a reported reduction in hospital admissions for influenza and pneumonia by 27% and particularly all cause deaths by 48%. This would imply that most deaths in winter were due to influenza.

Large, routinely collected datasets analysed in some of these studies had limited patients' details which did not include functional status, disease severity, or smoking status. Other studies which tried to allow for this, have shown benefits to be smaller but significant, whereas one reported no clear effect on all cause deaths. Influenza activity was, however, relatively low in several of these years.

Despite these methodological problems and disagreements, most experts agree that although more evidence is needed on the precise effectiveness of the vaccine, it has some benefit and the vaccination programmes should continue, say the authors.

"Influenza causes substantial mortality and morbidity in older people, and even if the more general all cause mortality benefit may be lower than previously thought, the current vaccine programme offers older people valuable protection against illness, hospital admissions, and mortality specific to influenza, should they be exposed this winter," say the authors.

They conclude by calling for resources to develop more effective vaccines, better forms of delivery of the vaccine for the elderly, and ensuring that more healthcare workers who come into contact with elderly people are vaccinated.

Contact:
Rachel Jordan, Unit of Public Health, Epidemiology and Biostatistics, University of Birmingham, Birmingham, UK
Email: r.e.jordan@bham.ac.uk 

Nigel Hawkes chronicles a decade of the UK government's attempts to tackle obesity, including its latest bid to turn the tide on obesity "which is so smothered in jargon" that it is hard to understand. He says: "The danger of wrapping an issue such as obesity up in the language of sociology and systems analysis is that it all begins to seem impossibly complicated. It is as if one needs to solve all the problems of society in order to tackle one relatively small sub-problem. So the Government's strategy includes food supply, education, the design of towns, primary care, walking kids to school, tackling false perceptions, counselling, old Uncle Tom Cobbleigh and all."

Contact:
Nigel Hawkes, freelance journalist
Email: nigel.hawkes1@btinternet.com 

(4) Screening for colorectal cancer detects unrecognised disease
(Research paper: Test, episode, and program sensitivities of screening for colorectal cancer as a public health policy in Finland)
http://www.bmj.com/cgi/content/full/337/nov20_2/a2261
(Editorial: Screening for colorectal cancer)
http://www.bmj.com/cgi/content/short/337/nov20_2/a2207

Screening for colorectal cancer detects four out of ten cancers and should be carefully designed to be more effective, according to a study published today on bmj.com.

About one in 20 people in the UK develop bowel cancer during their lifetime. It is the third most common cancer in the UK and the second leading cause of cancer deaths in Europe and the US.

Previous screening trials have show that faecal occult blood testing can reduce the risk of dying by about 16%. More than 50 countries have introduced screening programmes, but their effectiveness in a public health setting is not clear.

Dr Nea Malila and colleagues from the Finnish Cancer Registry examined whether Finland's national colorectal cancer screening programme could detect unrecognised disease. They studied 106 000 people aged 60-64 to test how sensitive screening was in identifying unrecognised disease at three levels - the faecal occult blood test (test to detect small traces of blood in faeces that may indicate disease at an early stage), screening episode, and the national screening programme.

A national screening programme for colorectal cancer began in Finland in 2004 as a public health policy in 22 volunteer municipalities and grew to 161 municipalities by 2006.

Nationally it was decided to split the 106 000 people into two groups - a screening group which received faecal occult blood tests kits by mail and a control group which received the routine health services available in the country.

Anyone whose test indicated blood was contacted so a full colonoscopy could take place.

The researchers found that the sensitivity (accuracy) of the test was 55% when considering cancers that developed after positive tests. The sensitivity from screening episodes was 51% and sensitivity was 38% for the national screening programme.

Roughly, four out of ten colorectal cancers were detected thanks to the organised colorectal cancer screening programme and the researchers concluded that the sensitivity of the Finnish programme was "adequate if relatively low."

The study also presents a model of how to implement a new programme using the principles of experimental design to provide good evidence on effectiveness.

The researchers say: "The sensitivity of the Finnish screening programme for colorectal cancer at the first round was adequate even if relatively low. Programme sensitivity in Finland was sufficient to justify continuation of the programme."

In an accompanying editorial, Joan Austoker and Paul Hewitson from the University of Oxford, say that in addition to programme sensitivity there are other important factors that should be taken into consideration when evaluating a cancer screening programme.

Contacts:
Nea Malila, Director of the Mass Screening Registry, Finnish Cancer Registry, Helsinki, Finland
Email: nea.malila@cancer.fi 

Editorial: Joan Austoker, Cancer Epidemiology, University of Oxford, Oxford, UK
Email: joan.austoker@ceu.ox.ac.uk 

(5) Hospitals could be fined millions of pounds even if they reduce infection risk
(Analysis: Fairness of financial penalties to improve control of Clostridium difficile control)
http://www.bmj.com/cgi/content/short/337/nov20_2/a2097

NHS Hospital Trusts that are successful in reducing Clostridium difficile risks in line with government targets still have a 50% chance of paying a financial penalty every year, and around a 95% chance of being fined over three years, warn researchers on bmj.com today.

Sarah Walker and colleagues from research institutes in Oxford, Cambridge, and London, say a recent initiative to encourage trusts to reduce C difficile infections have resulted in a perverse incentive where, in the worst case situation, trusts can be heavily penalised if they go over the set target by just a single extra case, and, in general, penalties are very hard to avoid completely.

Penalties may involve millions of pounds - for example, a 2% fine of a large trust with a revenue of £500 million could be around £10 million.

The authors argue that the current system, introduced as part of the 2008-9 National Health Service contract for acute services, may de-incentivise trusts from performing well and encourage hospitals to automatically reserve funds to pay penalties, whether or not they achieve the target. Such reserve funds cannot be used in delivering vital services to patients.

The authors explain that the problem with the current penalty system is the way targets and financial penalties are applied to trusts. The NHS contract does not distinguish between the underlying risk of each new patient getting C difficile, which they want to reduce by about 10% a year, and the actual count of cases observed in a particular trust - this is only an approximate measure of the underlying risk, particularly in smaller trusts. Even if this underlying risk is reduced by the right amount, for a relatively rare event like C difficile the play of chance means that the actual number of cases seen may be higher or lower than this.

In addition, the penalties are uneven. In the worst case scenario, a trust with 199 cases of C difficile cases in the previous year and 199 cases in the actual year will escape penalty. However, if the trust reaches 200 cases it is automatically fined 2% of its revenue because the 10% target reduction (179 cases) has not been achieved by a margin of 10% or more.

Thus a single case could cost a trust millions of pounds. Realistically, trusts will have to exceed the target risk reduction by 5-15% to avoid being fined. According to the authors, the efforts a trust puts into reducing infection rates and improving C difficile control, such as enhanced cleaning and hand washing are not taken into account in this scheme.

A fairer system, and one supported by the Healthcare Commission, would be to estimate a baseline number of cases for each trust from several years data, calculate target risk reductions for the next three years, and then only penalize Trusts where there was strong statistical evidence that they have not met these targets, say the authors. The authors conclude that while there is a strong case for using incentives to reduce C difficile cases in NHS trusts, the system to achieve this aim should be fairer and better designed so as not to penalise trusts who are working hard to reduce infection rates and meet targets, but experience normal year-to-year variation in their number of cases.

Contacts:
Alison Barnes, Press Office, Oxford Radcliffe Hospitals Trust, Oxford, UK
Tel: +44 (0)1865 228932 

Or
Jonathan Wood, Oxford University Press Office, Oxford, UK
Tel: +44 (0)1865 280530

Or
Laure Thomas, Medical Research Council Press Office, London, UK
Tel: +44 (0)207 670 5139

FOR ACCREDITED JOURNALISTS

Embargoed press releases and articles are available from:

Public Affairs Division, BMA House, Tavistock Square London WC1H 9JR

(contact: pressoffice@bma.org.uk)

and from:

the EurekAlert website, run by the American Association for the Advancement of Science (http://www.eurekalert.org)