BMJ  2005;331:987-988 (29 October), doi:10.1136/bmj.331.7523.987

Short cuts

What's new in the other general journals

Terror attacks in US linked to fewer admissions for self poisoning in Canada

The impact of the terrorist attacks in America on 11 September 2001 spread around the world, but there are few data on the psychological aftermath outside the affected areas. Researchers now report that in the three days after the attack, hospital admissions for self poisoning fell by nearly two thirds in Ontario, Canada, several hundreds of kilometres from New York. Careful tracking of all admissions for self poisoning in September between 1988 and 2003 show a dramatic drop between 11 and 13 September 2001, which is distinct from more normal admission patterns during the three days before or three days after that period. Instead of the expected 36 admissions during the three days after the attacks, there were only 13, including far fewer men (8% in 2001 v 46% during the same period in other years combined).

View larger version (53K): [in this window] [in a new window]   Credit: JAMA

 

The authors think that the people of Ontario, and possibly elsewhere, were briefly distracted from their personal distress by the scale of the tragedy unfolding in a neighbouring country. The distraction may have been enough to prevent some of the most vulnerable from attempting self harm or suicide.

JAMA 2005;294: 1900-1[Full Text]

Physical abuse of boys is common among poor urban families in Philadelphia

After randomly dialing over 13 000 households in Philadelphia county, researchers eventually found 298 men eligible for their study of physical abuse in childhood. Of those men, 197 (66%) completed the survey, which included validated questionnaires about childhood abuse and parental bonding. Half (51%, 100/197) the men surveyed had experienced at least one form of physical abuse during childhood, such as being hit with something, kicked, punched, or burnt. Three quarters of abuse in this study was committed by parents, more often the mother than the father. Unsurprisingly, abused boys were more likely than others to develop symptoms of depression and post-traumatic stress disorder as adults.

The survey targeted an area of Philadelphia known for domestic violence against girls and women. It was urban and poor. Even so, the authors were concerned at the high rates of abuse they found. Doctors and other social and health professionals should not forget that boys are abused too, they say. More should be done to find out exactly what abuse teaches boys about conflict resolution in their own families, and what it does to their mental health as adults.

Ann Intern Med 2005;143: 581-6[Abstract/Full Text]

Bariatric surgery: there's room for improvement

Bariatric surgery can be an effective treatment for obesity, although only about 0.6% of US citizens who might qualify for surgery get it. Three studies, all based on routinely collected data, help to identify the kind of people having surgery and what is happening to them. The first study found that, since 1998, people having bariatric surgery have got older, richer, and better insured, and are more likely to be women. They are also being discharged earlier, with fewer complications, and they have a low risk of death in hospital (0.1% to 0.2%). The other two studies were less upbeat. One from California reported that a fifth of patients are readmitted to hospital the year after a Roux-en-Y bypass, and a substantial proportion are admitted in the second (18%) and third (15%) years after surgery (usually for surgery related problems, such as ventral hernia or a revision). The final study showed that beneficiaries of the Medicare programme have a worryingly high risk of death after bariatric surgery, reaching 4.6% by the end of the first year. Risk of death was linked to increasing age, being male, and surgeons who do fewer bariatric procedures than average.

View larger version (41K): [in this window] [in a new window]   Credit: JAMA

 

Bariatric surgery remains an effective and enduring treatment for many patients, says a linked editorial (1960-3). These studies show that some populations are more vulnerable to the risks of surgery than others, but "the studies... must be seen as opportunities for improvement... not as a support for exclusionary practices by payers for patients in dire need."

JAMA 2005;294: 1909-17, 1918-24, 1903-8[Abstract/Full Text]

Ghost writers and guest authors blur accountability and bias research

Shady authorship arrangements are bad for medical research, bad for medical journals, bad for researchers' credibility, and bad for anyone being treated in accordance with the results of published clinical trials (hopefully most patients these days). "Ghost" writers and "guest" authors are the two biggest headaches, write two editors, in a plea to authors and sponsors to be clear about who has done what in a paper and who has paid. Because ghost writers, usually professionals employed by drug manufacturers, cannot exist without willing guest authors to take the credit, the editors warn all academics to avoid signing off articles they haven't written. Or if they must, to make sure the writer is acknowledged in the manuscript along with their financial and other vested interests. An easy test that potential authors can apply if they are unsure is to ask the question, "Have I been close enough to this research to take the heat when and if the results are questioned?" If not, don't sign.

Ann Intern Med 2005;143: 611-2[Full Text]

blockers underperform as first line treatment for hypertension

Many authorities producing guidelines on the treatment of uncomplicated hypertension will have to rethink their endorsement of blockers after a meta-analysis found that these drugs don't protect people from stroke as well as other alternatives, such as thiazide diuretics, angiotensin converting enzyme inhibitors, and calcium antagonists. In a linked comment, a leading hypertension researcher thinks this may be the end for blockers as first line treatment for hypertension, except for patients with coronary heart disease.

The meta-analysis, which included over 120 000 patients in 18 trials, reports that ß blockers reduce blood pressure but don't seem to save lives or prevent heart attacks when compared with placebo. blockers looked better than placebo at preventing strokes (relative risk for stroke 0.81, 95% CI 0.71 to 0.93), but in head to head comparisons with other drugs they looked considerably worse (relative risk 1.16, 95% CI 1.04 to 1.30).

Apart from their obvious clinical implications, these findings may also change the way new drugs are evaluated. New drugs that are compared with blockers will probably seem more effective than they would if compared with other established treatments for hypertension.

Lancet 2005 doi: 10.1016/S0140-6736 (05)67573-3 published on line October 18

Comment: doi: 10.1016/S0140-6736(05) 67575-7

Newer antipsychotic drugs may increase the risk of death in elderly people with dementia

Atypical antipsychotic drugs have already been linked to an increased risk of stroke in elderly people with dementia. A recent meta-analysis indicates that these drugs can also increase the risk of death (odds ratio 1.54, 95% CI 1.06 to 2.23). After a great deal of effort, researchers found nine unpublished trials and six published trials comparing atypical antipsychotic drugs with placebo in these vulnerable patients. All the trials were short (the longest lasted only 26 weeks), and all but one were funded by the drugs industry. Overall 3.5% of participants taking drugs and 2.3% taking placebo died during treatment, a significant difference that translates to about one extra death for every 100 people treated with an antipsychotic. The analysis included over 5000 patients in trials of aripiprazole, olanzapine, quetiapine, and risperidone. The extra deaths became obvious only when all drugs were analysed together.

View larger version (46K): [in this window] [in a new window]   Credit: JAMA

 

It's unclear why most of the data on these drugs remain unpublished, and the authors urge investigators to make their data public. Longer term trials are also needed to further characterise the harm atypical antipsychotics may be doing.

JAMA 2005;294: 1934-43[Abstract/Full Text]

Scientists move one step closer to the genes responsible for language

The genetic basis of language is extremely complex and largely unknown. FOXP2, a gene on the long arm of chromosome 7, is already known to be involved; now scientists from Canada think they have found a second genetic locus for language—a duplication of 26 or 27 genes that are also on chromosome 7.

Scientists have been looking for this duplication for a long time. Deletion of the same genes causes a relatively common developmental disorder known as Williams-Beuren syndrome, and many deletion disorders have a reciprocal duplication disorder. The duplication finally surfaced in an 8 year old boy who presented early in childhood with a catalogue of developmental problems including severe delay in expressive language. In direct contrast to this boy's language deficits, children with a deletion affecting the same 26 or 27 genes (the Williams-Beuren syndrome) have good language skills relative to their overall intellectual ability. The obvious conclusion is that this area of chromosome 7 is critical to the development of expressive language. The next step is to find out which of the many duplicated genes are responsible. The authors think the important bit of the entire duplication probably lies somewhere in a small region of nine genes.

N Engl J Med 2005;353: 1694-1701[Abstract/Full Text]

Acute lung injury kills as many US citizens as breast cancer or HIV

To find out the true public health impact of acute lung injury and acute respiratory distress syndrome, researchers painstakingly screened 4251 patients who were ventilated in hospitals in King County, Washington, between April 1999 and July 2000. They found 1113 cases. That's an estimated incidence of 86 per 100 000 person years, substantially higher than the authors were expecting. Their figures translate to almost 200 000 cases of acute lung injury in the United States every year, accounting for a total of 3.6 million days in hospital.

Thirty eight per cent of the entire cohort died before they left hospital. But the risk of death went up dramatically with age reaching 60% for patients aged 85 or over. Overall, the authors estimate that 74 500 people die from acute lung injury in the United States each year, an amount similar to the number of adults that die each year from breast cancer or diseases associated with HIV. The cohort was dominated by older people who were hospitalised with severe infections, often pneumonia.

This epidemiological picture, already worse than expected, is likely to deteriorate further as the US population ages, write two observers (1736-8). A nationwide strategy to tackle it is overdue.

N Engl J Med 2005;353: 1685-93[Abstract/Full Text]

Inhaled insulin looks promising for people with type 2 diabetes

Overweight people with type 2 diabetes should be encouraged to lose weight. If they can't, and oral treatments aren't enough, inhaled insulin is a relatively new therapeutic option. In one brief randomised trial, adding inhaled insulin to regular treatment with two oral drugs reduced participants' mean serum concentration of haemoglobin A_1c from 9.2% to 7.3% over 12 weeks. Another group of 104 people switched to inhaled insulin alone. Their diabetic control also improved (9.3% to 7.9%). Both treatment groups did significantly better than the 99 controls who simply carried on with their two oral drugs. Participants had a mean body mass index of 30 kg/m².

View larger version (45K): [in this window] [in a new window]   Credit: ANNALS OF INTERNAL MEDICINE

 

Predictably, people who used the new inhaler had more hypoglycaemic episodes than those who did not (one or two a month compared with 0.1 in the control group). All but one of the episodes were mild. Weight gain is an equally unwelcome side effect of insulin, particularly in a population who are already overweight. In this trial inhaled insulin was associated with an average weight gain of nearly 3 kg.

Ann Intern Med 2005;143: 549-58[Abstract/Full Text]

---

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?