This week in the BMJ
Volume 331, Number 7511, Issue of 30 Jul 2005Diagnose or rule out pulmonary embolism with confidence
Diagnose or rule out pulmonary embolism with confidence
Pulmonary angiography is the only method that can confirm or rule out pulmonary embolism. In a systematic review and meta-analysis of diagnostic studies that included 11 004 patients, Roy and colleagues (p 259) analysed 48 studies that evaluated diagnostic tests for confirming or excluding pulmonary embolism. They present pooled positive and negative likelihood ratios for strategies used to confirm or exclude pulmonary embolism, and then relate them to clinical probability. In addition they discuss clinical implications of the tests: lung scans, spiral computed tomography, leg vein ultrasonography, echocardiography, magnetic resonance angiography, and D-dimer tests.
|
![[To top]](/icons/back.gif){kind=icon}
Advice on dose adjustment in renal impairment is inconsistent
Recommendations for dose adjustment in patients with renal impairment differ greatly between the British National Formulary, Martindale: the Complete Drug Reference, American Hospital Formulary System Drug Information, and Drug Prescribing in Renal Failure. Comparing the advice that the four sources give for the 100 drugs that are most commonly prescribed in their hospital, Vidal and colleagues (p 263) found that the sources disagreed on the definition of renal impairment and differed in recommendations for specific drugs. For example, the British National Formulary and Martindale disagreed on whether the dosage needed adjusting at all for 11 drugs. Further, none of the sources clearly presented the methods and primary sources they used for their recommendations, and many criteria were described in qualitative terms.
|
Availability of emergency contraception didn't change women's practice
Over the counter availability of emergency hormonal contraception didn't change the frequency or pattern of its use by women aged 16-49 in the United Kingdom. In cross sectional surveys in 2000, 2001, and 2002, Marston and colleagues (p 271) analysed data on the contraceptive practice of women living in 7600 households. Increased availability wasn't followed by changes in women's contraceptive practice. However, more women bought emergency contraception from pharmacies rather than go to general practitioners and NHS clinics.
|
Who will investigate suspected research misconduct?
A story of suspected scientific misconduct that has lasted for over a decade is told by White on p 281. In 1992 the BMJ published a study by Dr Ram B Singh. When he submitted further studies to the BMJ peer reviewers started to voice doubts about the data. The article tells the story of the BMJ's attempts to resolve these doubts and find an authority able to investigate the suspicions. Having failed to find such an authority the BMJ decided to tell the story of its attempts to get at the truth. One outcome is the BMJ's decision to publish an Expression of Concern about the 1992 paper (p 266).
|
Richard Smith, former editor of the BMJ, discusses the need to investigate previous studies of an author whose work has been retracted (p 288). He suggests that the ideal might be an international body, together with a way of marking studies that are under suspicion in databases such as PubMed.
How to detect data fabrication
The raw data from a study submitted to the BMJ by R B Singh (see above) was sent for detailed statistical analysis. On p 267 Al-Marzouki et al describe the use of statistical methods for detecting data fabrication by comparing the baseline data from this study (the diet trial) with that from another trial. In the diet trial they show a combination of features in the baseline data (variances, means, and digit preference) that are strongly suggestive of data fabrication.
ABC of conflict and disaster
The ABC series continues with a chapter on recovery from conflict and helping restore hospital services (p 278). The authors describe phases and characteristics of medical services after the acute phase of the conflict. Two case studies, one in Azerbaijan and another in Kosovo, serve as tragic examples from which there is much to learn.
|
What's new
Keep updated
Services
Tools
Online poll
Resources
Print issues
