ROUND-UP Research

Cost-effectiveness vs. disease severity for setting health priorities in Uganda

Studies on the setting of health priorities in developed countries have usually shown a strong preference for severity of disease over the cost-effectiveness of an intervention as the leading parameter. However such preferences may differ in resource-poor countries. This study of these criteria in Uganda used a self-administered questionnaire with three case scenarios and asked respondents to prioritise treatments. Just over half of the 413 respondents were health professionals at sub-district level; the remainder included donor organisations, politicians, NGOs, patients and others. The survey found that respondents considered both severity of disease and cost-effectiveness important criteria for setting priorities, with disease severity as the leading principle (83% of those who responded to this question). However, international development partners emphasised cost-effectiveness of interventions as the more important. This discrepancy in attitude requires more investigation and needs to be openly debated to ensure that health priorities are set appropriately in Uganda. ¹

1. Kapiriri L, Arnesen R, Norheim OF. Is cost-effectiveness analysis preferred to severity of disease as the main guiding principle in priority setting in resource poor settings? The case of Uganda. Cost Effectiveness and Resource Allocation 2004;2:1. At: 〈www.resource-allocation.com/content/2/1/1〉.

Improving caesarean section outcomes in Malawi

Between January 1998 and June 2000, over 8,000 caesarean sections were performed in Malawi, 94% of them as emergencies. Obstructed labour was the most common indication (63%), but other indications were fetal distress, antepartum haemorrhage and pre-eclampsia. The most common pre-operative complications were haemorrhagic shock (7.6%), anaemia (6.2%) and ruptured uterus (4.1%). This study of maternal mortality considers events in the first 72 hours after caesarean section in an effort to pinpoint the risk factors that could be most easily addressed. Eighty-five women (1.1%) died, the majority post-operatively on the ward. Higher maternal mortality was associated with a ruptured uterus, low levels of anaesthetic training and the use of general as opposed to spinal anaesthesia. Blood loss requiring transfusion of two or more units of blood was also an important factor; the greater the mismatch between number of units required and number given, the greater the risk of death. These results indicate that more use of spinal anaesthesia, fluid replacement as an alternative to blood transfusion and generally improving care on the post-operative ward might be the best way to decrease maternal mortality related to caesarean sections. ¹

1. Fenton PM, Whitty CJM, Reynolds F. Caesarean section in Malawi: prospective study of early maternal and perinatal mortality. BMJ 2003;327:587–91.

Using HPV tests for preliminary cervical cancer screening

Almost all cases of cervical cancer are caused by a limited number of oncogenic human papillomaviruses (HPVs). Many HPV infections are transitory, so an HPV test is not in itself specific for cervical cancer. This study of over 10,000 women attending for routine cervical screening used a second sample for HPV screening with the aim of finding out whether and how HPV testing could contribute to cervical cancer prevention. HPV testing for high-risk virus types was a more sensitive primary screening technique than cervical cytology. Using HPV testing in addition to cytology could improve the detection rate of high-grade pre-invasive cancer, whilst reducing the need for colposcopy. Rather than introduce an extra test, however, the authors suggest it might be more cost-effective to do HPV screening first and use cytology only for those who test positive. This approach would need further study to confirm its value. ^1,2

1. Cuzick J, Szarewski A, Cubie H, et al. Management of women who test positive for high-risk types of human papillomavirus: the HART study. Lancet 2003:362:1871–76.

2. Franco EL. Are we ready for a paradigm change in cervical cancer screening? [Commentary]. Lancet 2003;362:1866-67.

Cervical cancer screening in low-resource settings–which way forward?

Cervical cancer is the leading cause of cancer-related death in women in developing countries, whereas mortality is considerably reduced in the developed world due to comprehensive screening. Pap smear screening has low sensitivity, requires laboratory personnel with high levels of experience, and involves long waiting times for results and a high level of recalls for re-testing and follow-up. Costs per test are considered high. With few functioning, population-wide screening programmes in the developing world, two “camps” have formed: one arguing that Pap screening is affordable and feasible in low-resource settings, the other promoting alternative low-technology methods. Recent evidence from the few developing countries where Pap smears have been piloted is reviewed here.

In Cameroon a split-sample study comparing results in Cameroon and Switzerland showed a lower atypical rate in the African laboratory (3.6%) than the Swiss laboratory (12.6%) but a higher detection rate of biopsy-confirmed cases (21% vs. 19%) and a lower false negative rate (4.7% vs. 7.8%). ¹ These data can be viewed to mean that Cameroon is not detecting sufficient atypical smears, or that the detection rate of biopsy-confirmed cases is as good as in a western laboratory.²

In South Africa the policy is to encourage all women to have three Pap smears in their lifetime. Studies show that less than two-thirds of rural women had heard about Pap smears and those who had were significantly older, with higher parity and a higher level of education than those who had not. Of those with this knowledge, only 56% had had a smear. ³ Lack of facilities and lack of promotion lead to poor uptake in South Africa. Nevertheless, even the limited screening available does result in reduced risk of cervical cancer, a reduction which would increase with appropriate expansion of services. ⁴

Vietnam believes that Pap tests offer the best way forward, arguing that previous failures have been due to quality limitations rather than technical ones, that quality issues can be addressed, and that it is better to use a known, reliable method than wait for alternatives, which in any case might well require a Pap smear at a later stage. ⁵ Indeed some argue that the cost of Pap smears in resource-poor settings has been over-estimated, ² whilst others point out that there are ways of reducing the costs of consumables by adapting procedures. ⁶

Alternatives include visual inspection using acetic acid, which is simple and inexpensive and gives an immediate result, although it has the disadvantage of giving a high level of false positives. Detection of high-risk HPV has good sensitivity and specificity but is expensive, particularly with respect to consumables, although there are hopes that low-cost, same-day HPV tests will be available in the near future. Coloposcopy has a high sensitivity and specificity, and gives an immediate answer, but it needs highly trained personnel and the colposcope is expensive, though with low running costs. ⁷ A study among 5,564 patients in Costa Rica used HPV-testing for initial screening, following up HPV-positive women with visual inspection. This combination gave a sensitivity of 65% comparable with the Pap smear, but increased specificity of 97%. If low-cost, rapid HPV testing becomes available, this kind of regime might be the most suitable for programmes where women are tested only once or twice in a lifetime, as it would allow many more women to be tested and treated. ⁸

There are those who believe that none of these options is cost-effective and therefore affordable in low-resource settings. ⁷ But far too many women are at risk to wait for cheaper methods to be validated or for prophylactic vaccines to become available. They believe Pap smear screening programmes should be set up without delay, until such time as other alternatives show they can reduce cancer deaths. ⁵

1. Robyr R, Nazeer S, Vassilakos P, et al. Feasibility of cytology-based cervical cancer screening in rural Cameroon. Acta Cytologica 2002;46(6):1110–16.

2. Suba EJ, Raab SS. Feasibility of cytology-based screening in rural Cameroon [Letter]. Acta Cytologica 2003;47(5):948–49.

3. Lartey M, Joubert G, Cronjé HS. Knowledge, attitudes and practices of rural women in South Africa regarding the Pap smear. International Journal of Gynecology and Obstetrics 2003;83:315–16.

4. Hoffman M, Cooper D, Carrara H, et al. Limited Pap screening associated with reduced risk of cervical cancer in South Africa. International Journal of Epidemiology 2003;32:573–77.

5. Suba EJ, Raab SS, Viet/American Cervical Cancer Prevention Project. Pap screening in developing countries: an idea whose time has come. American Journal of Clinical Pathology 2004;121(3):315–20.

6. Laniran TO, Anjorin AS. A low cost Papanicolaou stain protocol for cancer screening [Abstract]. Acta Cytologica 1996;40(5).

7. Cronjé HS. Screening for cervical cancer in developing countries. International Journal of Gynecology and Obstetrics 2004;84:101–8.

8. Jeronimo J, Castle PE, Herrero R. HPV testing and visual inspection for cervical cancer screening in resource-poor regions. International Journal of Gynecology and Obstetrics 2003;83:311–13.

Methodological biases affect coronary heart disease data on HRT

The effect of hormone replacement therapy (HRT) on coronary heart disease is not at all clear. A number of well-conducted retrospective observational studies, supported by animal and basic research, show a protective effect. The recent prospective randomised women's health initiative (WHI) study shows increased risk. The latter type of study is usually considered to have advantages over the former and therefore to produce more reliable data. However, the WHI trial was subject to a considerable degree of unblinding before the trial was stopped, with over 40% of the HRT users and nearly 7% of the placebo users unblinded for medical reasons. This may have resulted in a detection bias in the HRT users, particularly of acute myocardial infarction which can be undetected at the time of occurrence in 22–44% of cases. Factoring this into the WHI data would considerably lower the reported crude rate ratio from 1.28 to 1.02. In observational studies the apparent protective effect of HRT may be biased in the opposite direction by the exclusion of unhealthy women in the cohort, giving a healthy-user bias, and by an inappropriate choice of reference group. The combined crude rate ratio of 0.67 from these studies is increased to 0.82 by using an alternative reference group. Thus the diametrically opposite effects of HRT on acute coronary outcomes found between the observational studies and the WHI study may be a result not only of bias in the observational studies but also of bias in the WHI study. ¹

1. Garbe E, Suissa S. Hormone replacement therapy and acute coronary outcomes: methodological issues between randomised and observational studies. Human Reproduction 2004;19(1):8–13.

Breast cancer risk increased in HRT users

A number of studies have suggested that use of HRT increases the risk of breast cancer, but few have correlated patterns of use of various HRT preparations with breast cancer incidence and mortality. This five-year cohort study of over a million women in the UK showed a relative risk of developing breast cancer for current users of HRT of 1.66 compared to never-users, and an increased risk of dying from breast cancer of 1.22. However, past users of HRT were not at increased risk of incidence or fatal disease. Overall, incidence was significantly increased for current users of oestrogen-progestogen (2.0), tibolone (1.45) and oestrogen only (1.3) preparations, with the risk increasing with increasing total duration of use. Results varied little between specific oestrogens and progestogens or their doses, or between continuous and sequential regimens. These data mean an estimated extra five or six cancers per 1000 women with five years' use. ¹ The evidence is clear. Although there is no need to panic as the absolute risk is still low, HRT prescription should be discouraged and those already taking it should be encouraged to discontinue as soon as possible. ²

1. Million Women Study Collaborators. Breast cancer and hormone-replacement therapy in the Million Women Study. Lancet 2003;362:419–27.

2. Lagro-Janssen T, Rosser WW. Breast cancer and hormone-replacement therapy: up to general practice to pick up the pieces. Lancet 2003;362:414–15.

Belgian doctors continue prescribing HRT

In light of the confusing and sometimes contradictory data on the value and long-term effects of HRT, it is left to the individual physician to decide whether or not to advise use of HRT and which brand. This survey of 1,374 Belgian physicians was based on one case study and asked whether the physician would pursue, discontinue or modify the regimen for that woman. Despite the negative findings of the WHI study, most of the respondents would have continued prescribing one HRT regimen or another. There was a considerable level of disparity in their responses, showing that there is no overall consensus even on a single case. Only 12% would advise discontinuing HRT after two years and less that 25% would spontaneously discontinue prescribing it even after 11 years of use. ¹

1. Ena G, Rozenberg S. Issues to debate on the Women's Health Initiative (WHI) study. Prescription attitudes among Belgian gynaecologists after premature discontinuation of the WHI study. Human Reproduction 2003;18(11):2245–48.

No link between abortion and breast cancer

At least three US states require by law that women considering an abortion should be told it will raise their risk of breast cancer, and anti-abortion groups are lobbying hard for this requirement to be extended to other states. Against this background, the US National Cancer Institute (NCI) convened a workshop of over 100 of the world's leading experts to review all available data. The meeting concluded that many of the studies which had suggested a link were flawed, often because women with breast cancer were more likely to report having had an abortion in their search for an explanation of their disease. However, a Danish study of 1.5 million women, published in 1997, is often considered the definitive study and it found no association between abortion and breast cancer. Although anti-abortionists called the workshop's conclusion “an overstatement”, the NCI clearly states that “having an abortion or miscarriage does not increase a woman's subsequent risk of developing breast cancer”. ^1,2

1. Rubin R. USA Today. 3 March 2003. At: 〈http://www.usatoday.com/news/health/2003-02-26-cancer-usat_x.htm〉.

2. NCI. 30 May 2003. At: 〈http://cis.nci.nih.gov/fact/3_75.htm〉.

Long-term oral contraceptive use doubles cervical cancer risk

Using data from 28 studies involving over 12,000 women, the risk of cervical cancer was found to increase with increasing length of use of oral contraception. Relative risks rose to 1.1, 1.6 and 2.2 in women using oral contraceptives for less than 5, 5–9 and 10+ years, respectively, compared with never-users. For HPV-positive women the risks were 0.9, 1.3, and 2.5 respectively. The results were broadly similar for invasive cancers and cervical cancer in situ, for squamous cell and adenocarcinoma, and in studies which adjusted for HPV status, number of sexual partners, cervical screening, smoking or use of barrier contraceptives. There is some evidence that the risk decreases when oral contraceptives are discontinued, but currently available data are inconclusive. ¹

1. Smith JS, Green J, de Gonzalez AB, et al. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159–67.

Mifepristone-induced abortion does not affect subsequent pregnancies

Pregnancy outcomes in 4,925 Chinese women who had no history of induced abortion were compared with those of 4,931 women who had had one previous surgical abortion and 4,800 women who had had one mifepristone-induced abortion. Women in the mifepristone group were 23% less likely to have a pre-term delivery, compared with women with no abortion history, and the average birthweight of their infants was slightly higher. However, the likelihood of low birthweight, and the average length of pregnancies were similar in the two groups. There were no significant differences in risk of pre-term delivery, frequency of low birthweight or mean infant birthweight between women with mifepristone abortion and those with surgical abortion. This suggests that early abortion induced by mifepristone in nulliparous women has no adverse effects on the outcome of a subsequent pregnancy. ¹

1. Chen A, Yuan W, Meirik O, et al. Mifepristone-induced early abortion and outcome of subsequent wanted pregnancy. American Journal of Epidemiology 2004;160(2):110–17.

Home use of medical abortion safe and preferred in Tunisia

Tunisia is one of the few predominantly Islamic countries where abortion is legally permitted. This study considers the social dimensions of abortion in Tunisia and offers evidence supporting the provision of medical abortion (mifepristone+misoprostol) to specific populations, such as young and unmarried women. Between April 1999 and March 2001, 121 married and 101 unmarried women seeking an abortion were recruited at three clinics in Tunis. All received a regimen consisting of 200mg oral mifepristone followed two days later by 400mg oral misoprostol, administered either at home or in the clinic. The unmarried women (94.8%) were as likely as the married women (94.1%) to have a complete abortion with this regimen. Both groups showed a strong initial preference for home use of the misoprostol (married 80%, unmarried 73%), which grew even stronger after the procedure. They said that home use was desirable because transportation to the clinic was expensive (33%), and it was more confidential (26%) and more convenient (13%). Both groups expressed a high degree of satisfaction with the method. Medical abortion with the option of home use of misoprostol was thus found to be both safe and feasible for married and unmarried women in Tunisia. ¹

1. Blum J, Hajri S, Chélli H, et al. The medical abortion experiences of married and unmarried women in Tunis, Tunisia. Contraception 2004;69(1):63–69.

Sublingual and vaginal misoprostol are equally effective in early medical abortion

There are various modes of administration of misoprostol in a mifepristone + misoprostol regimen. The vaginal route appears more effective than the oral route but women often find the oral route less intrusive. A pilot study comparing sublingual misoprostol (dissolving the tablet under the tongue for a period of 10–15 minutes) with vaginal misoprostol showed complete abortion in 99% (93/96) of women in the sublingual group and in 96% (51/53) of women in the vaginal group. This suggests equivalent efficacy, but is not considered definitive as it was not randomised; the women chose their preferred mode of administration. ¹ A randomised, controlled trial with 224 women gave similar results, however. Complete abortion occurred in 98% of those in the sublingual group and 94% in the vaginal group, with no statistical difference between the two groups. There were three ongoing pregnancies in the vaginal group but none in the sublingual group. Neither group had serious complications though fever, chills, nausea, vomiting and diarrhoea were significantly more common in the sublingual group. The two routes of administration are thus equally effective, but further study is needed to optimise the dose of sublingual misoprostol. ²

1. Hamoda H, Ashok PW, Dow J, et al. A pilot study of mifepristone in combination with sublingual or vaginal misoprostol for medical termination of pregnancy up to 63 days gestation. Contraception 2003;68:335–38.

2. Tang OS, Chan CCW, Ng EHY, et al. A prospective, randomised, placebo-controlled trial on the use of mifepristone with sublingual or vaginal misoprostol for medical abortions of less that 9 weeks gestation. Human Reproduction 2003;18(11):2315–18.

Manual vacuum aspiration safe for early second trimester abortions

Manual vacuum aspiration (MVA) has already been shown to be an effective option for patients seeking surgical abortion during the first trimester. This was a retrospective review of 110 abortions performed in the US at 14–18 weeks of pregnancy, 73 with electric vacuum aspiration (VA) and 37 with MVA. There were no significant differences between the two groups in terms of procedure time or ability to complete the procedure, irrespective of type of cervical preparation and parity. Procedure time increased with length of pregnancy and decreased with physician experience, equally with both types of equipment. Although in small numbers of women, this study suggests that MVA as well as VA is safe and appropriate for pregnancies of 14–18 weeks, making it particularly valuable where resources are limited. ¹

1. Todd CS, Soler ME, Castleman L, et al. Manual vacuum aspiration for second-trimester pregnancy termination. International Journal of Gynecology and Obstetrics 2003;83(1):5–9.

Medical treatment or hysterectomy for menorrhagia?

Two recent studies suggest that about half the women being medically treated for menorrhagia will eventually need surgery. The first study, comparing 236 women treated with either a hormone-releasing intrauterine system or hysterectomy, showed that the two groups did not differ substantially in terms of health-related quality of life or psychosocial well-being at five-year follow-up. Both groups were satisfied with their treatment, although 42% of the women assigned to the group treated with levonorgestrel eventually had a hysterectomy. ¹ A second study of 63 women found that the women assigned to hysterectomy had a higher quality-of-life rating at six months into the study than those assigned to medical treatment, but this difference had been reduced considerably by the two-year follow-up, by which time 17 of the 32 women in the medical treatment group had had a hysterectomy. ² These results can be interpreted in two ways: either surgery is so likely that it should be done as a matter of course to save 50% of women having continuing symptoms, or as 50% of women can be satisfactorily treated medically, medical treatment should be used first for everyone, thus avoiding both the trauma of major surgery and the associated costs for half of them. ³

1. Hurskainen RM, Teperi J, Rissanen P, et al. Clinical outcomes and costs with the levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia: randomized trial five-year follow-up. JAMA 2004;291(12):1456–63.

2. Kuppermann M, Varner RE, Summitt RL Jr, et al. Effect of hysterectomy vs. medical treatment on health-related quality of life and sexual functioning: the medicine or surgery (Ms) randomized trial. JAMA 2004;291(12):1447–55.

3. Gottlieb S. Medical treatment for menorrhagia may only delay hysterectomy [News extra]. BMJ 2004;328(7442):730.

Stem cells isolated from cloned human embryos

Stem cells have recently been isolated from six-day-old cloned human embryos and should provide a new way to study the genetic components of human diseases. As with all current cloning techniques, the method was inefficient in that only 30 of 242 embryos developed normally and only one stable cell line was derived from the 20 embryos used. This new technique will not necessarily entirely replace the use of stem cells derived from aborted fetuses, but as with them, will be of particular value in the study of diseases such as motor neurone disease where the gene(s) involved are not yet known. Eventually cell lines may be used to help in the treatment of disease as well. ¹

1. Wilmut I. Human cells from cloned embryos in research and therapy. BMJ 2004;328:415–16.

Laparoscopic excision of endometriosis improves quality of life

Of 235 women with chronic pelvic pain from endometriosis who participated in a study of laparoscopic excision of endometriosis, complete data on pre-operative, intra-operative and post-operative condition were available with follow-up for 2–5 years for 135. These women showed significant reductions in pain scores, non-menstrual pelvic pain, dyspareunia and dyschesia. Quality of life and sexual function with pleasure were increased and discomfort during sex decreased. About one-third of the women required further surgery during the follow-up period, but returned endometriosis was found in only 68% of them. This confirms previous studies which show that pelvic pain is not always associated with confirmed endometriosis, and that recurrence of pain after endometriosis excision does not necessarily mean recurrence of endometriosis. ¹

1. Abbott JA, Hawe J, Clayton RD, et al. The effects and effectiveness of laparoscopic excision of endometriosis: a prospective study with 2–5 year follow-up. Human Reproduction 2003;18(9):1922–27.

Smoking damages reproductive health

Smoking may be having a far more serious effect on reproductive health than previously thought. Recent figures from the UK suggest it is involved in 120,000 cases of impotence, 1,200 cases of cervical cancer and 3,000–5,000 cases of miscarriage. It also affects fertility, with a 40% reduction in chances of conception in women who smoke, and a poorer response to infertility treatment in couples who smoke. Smoking in pregnancy increases the risks of some fetal malformations and stillbirth, and increases risk of death in newborns. It also triples the chances of having a low birthweight baby and reduces the quality of breastmilk. ¹ More needs to be done both to discourage people contemplating having a baby and pregnant women and their partners from smoking and to protect people from the harmful effects of passive smoking. ²

1. Kmietowicz Z. Smoking is causing impotence, miscarriages, and infertility. BMJ 2004; 328(7436):364.

2. British Medical Association. Smoking and Reproductive Life: The Impact of Smoking on Sexual, Reproductive and Child Health. At: 〈www.bma.org.uk〉.

Fertility rate declines in Central and Eastern Europe after communism

There has been a substantial decline in fertility in all former communist countries of eastern Europe with total fertility rates (TFR) of 1.3 children per woman in 2000, the lowest TFR in the world. The strong attachment to the two-child norm, characteristic of the mid-1980s, was part of a system of early family formation and an early start to childbearing, with short spacing of births. Family size was controlled mostly by traditional contraceptive methods–coitus interruptus and periodic abstinence–and high levels of induced abortion, usually after the second child, rather than with use of modern contraception. In 1989 the TFR was 1.8–2.2 in most of the 16 countries, but as low as 1.5 in Slovenia and as high as 2.8 in Moldova. By 2000 it had dropped in all 16 countries to between 1.1 in Ukraine and 1.4 in Croatia. The decline in the central European countries is mostly associated with a shift towards later childbearing, mimicking the pattern already seen across western Europe. However, in Romania, Bulgaria, the Russian Federation and republics of the former USSR, the decline is mainly due to a trend of one-child families. The difference in causes is reflected in the relative success of the countries post-communism. The more socially and economically developed countries of central Europe have adopted prolonged education, the pursuit of leisure activities and consumerist lifestyles, mirroring western fertility patterns by delaying childbearing, particularly with the use of modern contraceptives. The more eastern countries have had a less successful transition to capitalism and the trend towards one-child families might be due to economic difficulties, although childbearing in these countries has also been delayed. Under the circumstances, it is likely that fertility will decline even further in these states before it stabilises. ¹

1. Sobotka T. Re-emerging diversity: rapid fertility changes in Central and Eastern Europe after the collapse of the communist regimes. Population-E 2003;58(4–5):451–86.

Should obstetrician–gynaecologists provide care for rape victims, asks Mexican study

The obstetrician/gynaecologist is often the first professional that a woman who has been raped may see. This study in Mexico investigated obstetrician–gynaecologists' attitudes to sexual violence and their priorities when faced with a raped woman. 39% of those participating had dealt with at least one rape case in the previous year and 86% had handled at least one at some time, with one physician reporting over 60 cases. Over half did not believe themselves qualified to deal with rape cases, and two-thirds admitted to limitations in how they handled them, in particular as regards legal concerns, and lack of training or knowledge. Only 23% felt they had no limitations in this respect. When questioned about the most immediate problems to be addressed when dealing with a rape case, 56% stated searching for the aggressor, 42% attention to physical damage, 28% prevention of sexually transmitted diseases, 27% care of psychological trauma and 22% prevention of pregnancy. Overall, these professionals were found to be more focused on improving their professional knowledge than on socio-medical considerations, raising questions about how well they prioritised the women's health needs. ¹

1. Uribe-Elias R. Sexual violence and the obstetrician/gynecologist. International Journal of Gynecology and Obstetrics 2003;82:425–33.

Are clinical settings the right place to deal with domestic violence?

Sixty-seven adult women took part in focus group discussions in a US study on services for those who experience intimate partner violence. The women had responded to an advertisement asking for participants in a survey on improving health and community services for women and were not chosen according to any history or experience of domestic violence. A standardised interview guide was used for each of the six focus groups, concentrating on services such as who a woman would turn to for help with domestic violence, what barriers might exist in using services, the role of the health care system in providing services, and suggestions for improving access to and quality of such services. The women identified a range of currently available services including health care providers, police, shelters and churches. They highlighted a number of barriers to addressing the problem of domestic violence including cost, the risk of having children removed by social services, violence being too personal to discuss, and the inability of services and particularly doctors to provide what the victims really needed. The participants emphasised the need for community-based prevention efforts and referral to useful services such as job training and financial support. The women believed that the health care system is not an appropriate setting for dealing with domestic violence and that community-based services would be more valuable. This observation is at odds with the current emphasis on the part of many health care organisations on using clinical settings to address the issue of intimate partner violence. ¹

It is estimated that around a quarter of women in the UK have been hit by a partner at some time. Those affected can suffer from a range of health problems including physical injuries, chronic illnesses, unintended pregnancies, higher rates of sexually transmitted infections and higher rates of depression, anxiety, post-traumatic stress disorder, self-harm and suicide. Although they may receive medical help in the most severe cases, many either do not know how to get help to deal with the underlying problem or are reluctant to do so for social reasons. This has led some health professionals ² to suggest that health settings are an appropriate place for routine enquiry as to whether women are experiencing domestic violence.

Routinely enquiring about domestic violence in such settings could have the advantage of changing perceptions on the acceptability of violence in relationships, making it easier for women to access support services early on and bringing about a change in the knowledge and attitudes of health professionals towards domestic violence. However, there are problems with routine enquiry. One is that women are concerned about breaches of confidentiality. A second is the need for training staff. Time pressures on staff can be a particular problem, not only at the enquiry stage but also in maintaining support and referral. Other health workers ³ believe that routine enquiry is just not appropriate, partly because of professional reluctance to get involved but also because there is little research on its acceptability to women.It is also unclear what happens after disclosure. Still others ⁴ cite studies from primary care facilities which show that many women object to being routinely questioned about domestic violence, though they point out that in antenatal settings routine enquiry does seem more acceptable to the women.

1. Peterson R, Moracco KE, Goldstein KM, et al. Women's perspectives on intimate partner violence services: the hope in Pandora's box. Journal of American Medical Women's Association 2003;58(3):185–90.

2. Taket A, Nurse J, Smith K, et al. Routinely asking women about domestic violence in health settings. BMJ 2003;327:673–76.

3. Duxbury F. Routinely asking women about domestic violence: seeking the causes of disease, not routine inquiry, is good practice [Letters]. BMJ 2003;327:1345.

4. Boyle AA. Routinely asking women about domestic violence: inquiry may be acceptable in different health care environments and to different women [Letters]. BMJ 2003;327:1345.

Homosexuality not a disease to be cured

Homosexuality was only removed from the international classification of diseases in 1992; prior to that it was defined as a mental illness. This definition inevitably led to attempts to “cure” homosexuality. A survey of 31 men treated in the UK between 1960 and 1970 with behavioural aversion therapy, including the use of electric shock, showed an unsurprising lack of success. None of the participants felt they had benefited from the treatment, which they felt was unsophisticated and poorly delivered and followed up. The treatment clearly had a negative impact on their sense of identity and place in society. A companion study on the experience of professionals involved in giving such treatments confirmed that there were no standards, protocols or ethical guidelines in use at the time. Although some professionals had been convinced of the value of such treatments, many others expressed increasing doubts about the efficacy and ethics involved. The authors conclude that it was inappropriate and harmful to use mental health treatments to change human behaviour that is disapproved of on social, political, moral or religious grounds. ^1,2

1. Smith G, Bartlett A, King M. Treatment of homosexuality in Britain since the 1950s–an oral history: the experience of patients. BMJ 2004;328(7437):429–30.

2. King M, Smith G, Bartlett A. Treatment of homosexuality in Britain since the 1950s–an oral history: the experience of professionals. BMJ 2004;328(7437):427–28.

Girl babies in China seven times more likely to die than boys

The outcome of 3,697 pregnancies was followed using civil registration data in a rural area of China. Only three cases were lost to follow-up. Fifteen per cent of the pregnancies were either aborted (312) or ended in miscarriage (240). The perinatal mortality rate (stillbirths and deaths in the first seven days) was 69 per 1000 births, and the early neonatal mortality rate (death in the first seven days after birth) was 46 per 1000 live births. There was a very large difference in this rate between the sexes, 29 for boys and 69 for girls. Although the risk for first-born girls was slightly higher than for first-born boys, the difference was not significant. However, the relative risk of early neonatal mortality for second-born girls was 7.15 compared to second-born boys. The perinatal mortality rate increased notably with parity and was higher in townships with lower per capita income. The rate of stillbirths increased strongly between first and second pregnancies, the opposite of the situation in developed countries. Overall the mortality rate was much higher than previously reported and the differential figures between the sexes provides evidence of the effect of the modified family planning policy which allows those in rural communities to have a second child if the first born is a girl. ¹

1. Wu Z, Viisainen K, Wang Y, et al. Perinatal mortality in rural China: retrospective cohort study. BMJ 2003;327(7427):1319.

Poverty increases risk of STIs and HIV in Haiti

The prevalence of STIs such as syphilis, gonorrhoea and chlamydia as well as HIV is high in urban slums in Haiti, with 47% of women attending antenatal clinics found to have at least one STI. In the early 1990s, it was shown that women who worked as servants in the city of Zanmi Lasante were more likely to have HIV infection than rural market women who governed their own affairs. This new and larger study shows that ten years on, the situation has barely changed. The strongest risk factors for chlamydia and/or gonorrhoea were economic variables, with work as a domestic servant increasing the risk four-fold, whereas work as a market vendor reduced the risk by almost half. Increasing poverty and the collapse of the public health system in Haiti means that most health centres lack basic equipment, and pregnant women are rarely tested for HIV or STIs, as reflected in high rates of congenital syphilis and mother-to-child transmission of HIV. This study supports the thesis that it is the economic situation in Haiti that pushes young women towards sexual unions for financial security. Lacking other employment opportunities and with little schooling, they enter sexual relationships out of economic necessity, and typically do not have the power to demand use of condoms. From this point of view, one of the mechanisms for reducing the risk of STIs and HIV in women would involve increasing their economic opportunities. ¹

1. Smith Fawzi MC, Lambert W, Singler JM, et al. Prevalence and risk factors of STDs in rural Haiti: implications for policy and programming in resource-poor settings. International Journal of STD and AIDS 2003;14:848–53.