As a male obstetrician and gynaecologist, I am fascinated by childbirth, even though I have never gone through it personally. In many ways (although not all), I regret this. However, I do not regret the fact that I am unlikely ever to experience stress urinary incontinence. Millions of women worldwide suffer true incontinence from obstetric vesicovaginal fistula, over a million in northern Nigeria alone. But, many more suffer the consequences of a short urethra stretched beyond adequacy by childbirth. On page 700, Wesnes et al. report their studies of the link between urinary incontinence during pregnancy, mode of delivery, and incontinence postpartum. They studied more than 12 500 women who were having their first pregnancy and were previously continent of urine. Six months postpartum, almost one in three had developed urinary incontinence, about half being stress incontinence. One in 20 reported leaking on a daily basis. Women who reported having leaked urine during pregnancy were 2.3 times more likely to have a problem at 6 months postpartum than those who did not. Compared with women delivering by elective caesarean section, the relative risk of leakage was 4.0 for forceps and 3.2 for spontaneous delivery or ventouse extraction. Elective caesarean section was proportionately equally protective for women with or without incontinence during pregnancy, despite the difference in absolute risk. Despite the high proportion of women affected, stress urinary incontinence receives relatively little attention in the mass media. It is almost ‘the condition that dare not speak its name’. The Greek fisherman in the movie ‘Shirley Valentine’ admonishes Shirley for being embarrassed about her striae gravidarum—‘they are the badge of motherhood, you should wear them with pride’ he declaims. I doubt that he would have said the same thing about stress urinary incontinence. What are the barriers that prevent women talking freely about this condition and seeking appropriate treatment? This question is addressed by Basu and Duckett on page 726. They use the term ‘iceberg effect’ to describe the fact that many more women live with their symptoms than seek treatment. They studied 49 women who had had an operation for genital prolapse, using qualitative techniques in which transcripts of interviews were analysed using ‘Grounded Theory’, as described by Glaser and Strauss (if like me you have never previously heard of this approach, Basu and Duckett explain how it works in their paper. Basically, it is a sequential approach to identifying themes that emerge from each woman’s description of her experience). A major theme that emerged is that many women regard stress urinary incontinence as ‘just a woman’s lot, especially as she gets older’. Many were unaware of the potential for effective treatment over and above the operation they had undergone for prolapse. Interestingly, fatalism emerged as a stronger factor than embarrassment. So what is important to women in relation to stress incontinence? This question is addressed by Ternent et al. on page 719. Papers from urogynaecologists tend to emphasise aspects that are readily measurable quantitatively. However, these are not necessarily the aspects that are most important to women themselves. Ternent et al. asked 188 women contacted through the Bladder and Bowel Foundation to select, weight, and rate the importance of particular aspects of their problem. One hundred and five women responded, although only 61 completed their questionnaires with total accuracy. The most frequently mentioned problem was a restriction in socialisation and ‘going out’ but almost as many were concerned about personal hygiene, and the way the condition interfered with their sleep. Interference with shopping and travel were also major concerns. What would women with stress incontinence like to see investigated further? This is addressed in the paper by Herbison et al. on page 713. Although participants were asked to list their priorities for research, in the event their comments were largely related to problems with service delivery. This is perhaps not surprising as few would have had any training in medical research, and most would have been unaware that operational research is difficult to fund. Accordingly, they asked primarily that seeking assistance for their problems be made easier (again, overcoming the embarrassment factor is probably important) and that techniques be developed to make their day-to-day life more manageable. They also thought that it is important to discover the true cost to society of the effects of stress urinary incontinence, to learn more about the underlying causes, and the effects of lifestyle modification (including the effects of fluid intake, exercise, and complementary therapies). The aspect of their investigation that was new to me was the use of ‘citizens’ juries’. Participants were recruited by placing advertisements in local newspapers, offering an honorarium of 150 New Zealand dollars for taking part in a workshop, which ran from a Thursday evening to the following Saturday morning. Following introductory lectures, juries were asked to deliberate on the question ‘what can researchers study to make your life better?’ Qualitative techniques were then used to identify the major themes. Perhaps, using such juries to highlight patients’ priorities could be used to influence funders? The last paper in our urinary theme for this month is from Touboul et al. on page 708. They investigated the vascular anatomy of the obturator region in relation to the placement of mesh needles during cystocele mesh repair. The posterior needle passed within 1–2 mm of the posterior division of the obturator vessels on each side, and they highlighted damage to these vessels as a potential risk during this procedure. We continue to learn about both the potential and the hazards of these new approaches. Reliable figures on maternal mortality were first published in the UK in 1860, and the rate remained stubbornly above 400 per 100 000 women (1 in 250, a rate similar to that in India today) for the next 60 years. However, the rate fell dramatically over the next 60 years, reaching a low of less than 10 per 100 000 women (1 in 10 000) in 1980. The improvements were due largely to the introduction of antibiotics, safe blood transfusion, and improved anaesthesia and surgery. In the 1980s, we used to ask ourselves whether we could get the rate even lower. The answer was unfortunately no, and now, we are faced with rising mortality rates. Why is this? A major factor has been the recrudescence of cardiac disease, highlighted in a commentary on page 609 by Gelson et al. The heart is the final common pathway of many problems related to diet, particularly obesity, hyperlipidaemia, and diabetes, which when added to a sedentary lifestyle adds up to ischaemic heart disease. Acquired heart disease from rheumatic fever, largely abolished in developed countries by penicillin, is resurgent as developed countries with low birth rates have encouraged major increases in immigration to rebalance the age pyramid. And while techniques of managing congenital heart disease in pregnancy have improved greatly, so has the success of corrective surgery in infancy, resulting in many more women with impaired cardiac function surviving to reproductive age (and heart disease is one of the most common congenital abnormalities, occurring in about 0.7% of all births). In my view, we need more investment to improve the diet of our children (this should surely be a major health priority, not left only to activists such as the TV chef Jamie Oliver in the UK and Morgan Spurlock [of ‘Super Size Me’ fame] in the USA). We also need to improve preconception counselling and health evaluation of teenagers, especially in women with cardiac disease and in immigrants. An interesting feature of the past 10 years has been the growth of ‘research into other people’s research’, in other words, meta-analyses and systematic reviews. Cynics might say that they have become popular because they can be carried out from the comfort of one’s desk, do not require a year-long dialogue with the ethics committee, and require relatively little funding (apart of course from the salary of those doing the analysis and a good computer). Inevitably, we are now seeing systematic reviews of systematic reviews, which often point out that not all systematic reviews are equally valid or useful. One that we did think would be useful to our readers is the ‘best evidence’ review of induction of labour by Mozurkewich et al. on page 626. Much of my research in the 1970s concerned induction of labour, and in the UK, at that time, rates had risen to almost 50% because of concern about the problem of the ‘mature stillbirth of unknown cause’. Mozurkewich et al. confirm that the evidence for induction of labour at or beyond 41 weeks of gestation is strong. However, they do not comment on the fact that most studies are based on women of white European origin, and there is growing evidence that gestational length varies by racial group (Balchin et al., BMJ 2007;334:833–5) and that, for example, women of black African or south Asian origin can appropriately be induced 1 week earlier (Balchin et al., Obstet Gynecol 2008;111:659–6). Another of Mozurkewich’s conclusions is that induction of labour within 12 hours following prelabour rupture of membranes reduces the risk of infectious complications compared with expectant management, without increasing the operative delivery rate. This may be particularly important in women of black African origin, who have higher infectious morbidity rates, and challenges the expectant policy adopted by many advocates of ‘natural childbirth’. A further conclusion is that there is no convincing evidence of an optimal timing for induction of labour following preterm prelabour rupture of the membranes. Policies should therefore be tailored to the local availability of neonatal intensive care and individual patient factors. There are many logistical advantages to having a consistent gestation for preterm induction in any particular unit, but this can be chosen according to local circumstances. There is a growing fashion among obstetricians, midwives, and the women themselves to request early induction of labour for suspected macrosomia because ‘it is obvious that a smaller baby will be born more easily’. Unfortunately, what is gained on the swings of reduced size is lost on the roundabout of impaired labour efficiency, and Mozurkewich et al. highlight that although weak, the evidence suggests that such inductions are not beneficial (with the possible exception of diabetic pregnancy). For almost all other indications, including mild pre-eclampsia and fetal growth restriction, the evidence is weak or equivocal (e.g. although induction for fetal growth restriction reduces stillbirth, it simply passes the problem to the neonatologist who often cannot compensate for established damage). As always, we need to be humble and recognise that many of the things we think we know, we should in fact not be sure of (this is a variation on the famous quote by Donald Rumsfeld on 12 February 2002 that ‘there are also unknown unknowns. There are things we do not know we don’t know’). This month’s BJOG is full of other articles of potential practical value to readers. For example, on page 648, Eller et al. describe the techniques for managing placenta accreta that they find particularly effective (including, for example, leaving the placenta in situ and using ureteric stents). Should you use gonadotrophin-releasing hormone analogues, triple tourniquets, or both at myomectomy? Read the paper by Magos et al. on page 681 for their views. Want information on how to manage your methadone addicts in pregnancy? Learn from the experience of Dryden et al. on page 665. And guidance on predicting post-traumatic stress disorder and depression following pregnancy is provided by Soderquist et al. on page 672. Bill Gates has written ‘Every now and then I like to pick up a copy of Time magazine and read every article from beginning to end, not just the articles that interest me most. That way you can be certain to learn something you didn’t know previously’. Why not try that with BJOG?