This week in the BMJ
Volume 331, Number 7515, Issue of 3 Sep 2005More melanomas are caught early in the US
More melanomas are caught early in the US
In the United States, the incidence of melanoma more than doubled between 1986 and 2001 in people aged 65 and older, but this was probably caused by increased diagnostic scrutiny and not an increase in the true occurrence of the disease. A population based ecological study in nine geographical areas of the US by Welch and colleagues (p 481) found a 2.5-fold increase in the average biopsy rate and a 2.4-fold increase in the average incidence of melanoma. The extra cases were mostly in early stages of the disease, and the mortality from melanoma remained stable.
|
![[To top]](/icons/back.gif){kind=icon}
Flocculant-disinfection gives best results in Kenya
Flocculant-disinfection seems to be the best solution to infected drinking water in areas of Kenya where drinking water is obtained from ponds, rivers, and springs that are regularly contaminated by human and animal faeces. In a three arm cluster randomised controlled trial that included 6650 people and 605 family compounds in Kenya, Crump and colleagues (p 478) found that, compared with the usual practice of water collection, using flocculant-disinfectant reduced the prevalence of diarrhoea in children under 2 years by 25% and sodium hypochlorite reduced it by 17%. Flocculant-disinfectant also reduced water turbidity significantly when compared with sodium hypochlorite and control.
|
Smoking ban proposed in the UK may worsen health inequalities
In the United Kingdom, the proposed smoking ban, which can exempt pubs that don't serve catered food, may worsen health inequalities. A modelling study by Woodall and colleagues based on the borough of Telford and Wrekin (p 488) predicts that smoking could still be permitted in two thirds of the pubs in deprived areas, and about a quarter in more affluent areas. If members' clubs are included in the model, four fifths of establishments in deprived areas and two fifths in more affluent areas could be exempt from the smoking ban.
|
Nurse led programmes don't improve COPD
Nurse led interventions for management of chronic obstructive pulmonary disease (COPD) don't have a detectable effect on mortality, disability, or health related quality of life and psychological wellbeing. In a systematic review of nine randomised controlled trials, Taylor and colleagues (p 485) found no evidence that nurse led interventions improve management of this chronic disease. The evidence on hospital readmissions was equivocal, and the evidence on some important outcomes such as patients' adherence to treatment was extremely weak or absent.
|
Hypertension can be monitored by patients
Patients with hypertension can monitor their blood pressure at least as well as doctors and nurses. McManus and colleagues (p 493) randomised 441 people in eight general practices in south Birmingham to the self monitoring intervention or usual practice. After one year, no differences in systolic or diastolic blood pressure, anxiety, health behaviours, or number of prescribed drugs were found between the groups, and patients who self monitored lost more weight, rated self monitoring better than being monitored by a doctor or a nurse, and consulted less often. More than 90% of all patients attended follow up at one year.
Systematic reviews should assess quality of RCT interventions
Systematic reviews of randomised controlled trials that assess complex interventions should also assess the quality of interventions tested in the trials, argue Herbert and Bø on page 507. They present a systematic review that assessed effect of training pelvic floor muscle exercises on urinary incontinence in pregnancy to show how the pooled estimate of all trials changes after exclusion of a trial where exercise wasn't supervised. Analyses of quality of intervention should be specified in the review protocol, and they should focus on the effects that this quality could have on the effects of the intervention, say the authors.
What's new
Keep updated
Services
Tools
Online poll
Resources
Print issues
