BMJ No 7068 Volume 313

This week in BMJ Saturday 23 November 1996

Inadequate studies leave questions about cervical samplers unanswered

We still do not know which sampling device is best for cervical screening. Buntinx and Brouwers (p 1285) report a meta-analysis of randomised and quasi-randomised studies to determine which methods are related to the highest yield of cytological and histological abnormality. The Ayre spatula, Cytobrush, or cotton swab alone generally performed significantly worse than the combination of a spatula and Cytobrush or cotton swab, the extended tip spatula, or the Cervex brush. There were no substantial differences in sensitivity or positive predictive value between the sampling methods examined. The results suggest that a combination of endocervical and ectocervical sampling is appropriate. They support the use of either the extended tip spatula, the combination of any spatula plus Cytobrush or cotton swab, or the Cervex brush for cervical screening.

---

The jury is still out on effectiveness of common treatments for neck pain

Patients with neck complaints are common in clinical practice, and healthcare providers often have to make decisions about which treatment(s) to recommend. Ideally, these decisions are based on good scientific evidence - that is, clinical trials. On p 1291 Aker et al summarise the evidence from randomised trials on conservative management of neck pain, including drug treatments, manual treatments, physical medicine methods, and education. In general, no treatments had been studied in enough detail to make strong recommendations for their use. There is early evidence to support the use of manual treatments in combination with other treatments for short term pain relief. Many current approaches have not been adequately assessed. Further clinical trials are needed to allow healthcare providers, policey makers, educators, and the public to make informed decisions.

---

Population mixing and socioeconomic status are linked with leukaemia in children

A series of studies by Leo Kinlen and colleagues provided epidemiological evidence that childhood leukaemia might be a rare response to an as yet unidentified infection. Hitherto, the relevant studies have concentrated on areas which have experienced extremely high levels of population mixing. Stiller and Boyle (p 1297) have now analysed the incidence of acute lymphoblastic leukaemia in children for the whole of England and Wales in conjunction with interdistrict migration data from the census. They found that incidence in young children increased with the "entropy" attributable to migration into their districts of residence. These results indicate that population mixing even at relatively low levels may be important in the aetiology of childhood leukaemia. Previous findings linking higher incidences with higher socioeconomic status of areas of residence were also confirmed. More affluent areas tend to attract more incomers, and the authors suggest that studies showing an effect of social class on the incidence of leukaemia may in fact have been indirectly observing a population mixing effect.

---

GPs are reluctant to support home births, despite low perinatal mortality and low intervention

Home birth continues to be a subject of debate. The government report Changing Childbirth suggests it should be an option for women. In the first of four papers this week discussing home births, Davies et al (p 1302) report a prospective study of planned home births in the United Kingdom. They examined all known requests for home birth in the Northern region in 1993 and found that home birth is rare but increasing. Women liked home birth because they felt in control, and this strongly positive feeling was shown even by women transferred in labour. However, almost one third of women had their planned place of delivery changed to hospital antenatally. In half the cases there was no obstetric indication for this change. General practitioners often disagreed with the woman's choice and actively discouraged home birth, mainly for fear of complications. Midwives reported a lack of professional support and identified several equipment needs.

A second study, from the Northern Region Perinatal Mortality Survey Coordinating Group (p 1306), examined all perinatal deaths in the region associated with planned or unplanned birth outside hospital between 1981 and 1994. Unplanned home birth was hazardous, but mortality among babies booked for a home delivery (irrespective of where this eventually occurred) was low. The authors believe that death after delivery outside hospital (only 2% of which resulted from planned home birth) warrants further attention.

In the third study, of 1,836 women with low risk pregnancies, Wiegers and colleagues found that the outcome of planned home births was at least as good as that of planned hospital births (p 1309). And in the fourth study of home births Ackermann-Liebrich et al (p 1313) followed up 498 women who at their first antenatal visit decided to have a home confinement and their matched controls who asked for hospital delivery. They found a higher risk for any intervention (induced labour, medication, caesarean section, forceps, vacuum extraction, episiotomy) among women who requested a hospital delivery. Duration of labour, blood loss, and birth complications did not differ between the groups, nor did the clinical condition of the babies. The authors think their findings should encourage all those who wish to improve the circumstances of childbirth.