The impact of surgical treatment on the mental health of women with obstetric fistula


Dr A Browning, Barhirdar Hamlin Fistula Centre, PO Box 1739, Barhirdar, Ethiopia. Email


Obstetric fistula is estimated to affect 2 million women worldwide and has considerable social and psychological effects on affected individuals. In this prospective study, 51 consecutive women with obstetric fistula admitted to the Barhirdar Hamlin Fistula Centre in the north of Ethiopia were screened using the General Health Questionnaire (GHQ-28) for potential mental health disorder before and 2 weeks after fistula surgery. Prior to surgery, all women screened positive. By 2 weeks after, this had dropped to 36% (P = 0.005). 27% of the 45 women who were cured of their incontinence screened positive, while all 6 of those with severe residual incontinence continued to screen positive. We conclude that surgical treatment of obstetric fistula results in marked improvements in mental heath.


Obstetric fistula represents a significant global health problem. It has been estimated that there are currently 2 million women with obstetric fistula waiting for treatment and approximately 50 000 to 100 000 new cases occurring each year, predominantly in the developing world.1

A woman with obstetric fistula suffers a great deal. She has frequently delivered a stillborn child (up to 98% in some surveys),2 has to cope with being completely incontinent of urine and/or faeces with resulting poor hygiene and is often divorced and isolated from her community due to her injuries.3 There is little research into the mental health status of these women, but suicides and suicide attempts have been reported. A recent observational study by Goh et al.4 questioned women with obstetric fistula from Bangladesh and Ethiopia with a shortened version of the General Health Questionnaire (GHQ-28) for psychiatric disorders and found that 97% of women screened positive for potential mental health dysfunction.

The GHQ-28 is a validated screening tool aimed to detect current psychiatric dysfunction that has been successfully used in developing world settings.5,6 It contains questions covering four areas: social dysfunction, somatic symptoms, anxiety/insomnia and depression. There are 7 questions in each of the 4 domains, giving a total of 28 questions. Answering five or more questions positively indicates a probable case of mental health disorder,5 and it is estimated that 24–40% of women screening positive will have major depression.7 Combining this data with that from the Goh survey,4 it is estimated that between 23 and 39% of women with fistula have major depression. This is in contrast to controls surveyed during the same study in Bangladesh and Ethiopia where 32% screened positive with the same questionnaire, equating to a major depression prevalence of 8–13%. This rate is similar to the general population prevalence in Western countries.7

Women with obstetric fistula usually live in the developing world where mental health services are few or nonexistent. It has often been stated that women cured from obstetric fistula are incredibly grateful, rejoin their families, remarry and sometimes have successful pregnancies. The hospital environment, being one of care and fellowship among women with fistula, will undoubtedly go a long way to relieve mental suffering (it has been described as the ‘sisterhood of suffering’ by Dr Catherine Hamlin of the Addis Ababa Fistula Hospital). But until now no one has described the impact on mental health by the treatment of their condition.

This paper describes a cohort of women with obstetric fistula in the Barhirdar Hamlin Fistula Centre, Barhirdar, northern Ethiopia, a dedicated obstetric fistula unit. Screening was conducted using the shortened version of the GHQ-28 and was performed on both admission and just prior to discharge after treatment. As there is no institutional ethics committee in Barhirdar, institutional approval was gained from the Addis Ababa Fistula Hospital research and ethics committee (of which Barhirdar is a subsidiary).


All women admitted to the Barhirdar Hamlin Fistula Centre in northern Ethiopia between February and April 2005 were asked if they would like to partake in the study and informed of what was involved. The 51 women (who were all illiterate) gave verbal consent. They consisted of 49 women with obstetric vesicovaginal fistulae and 2 women who had previously had a fistula repair but had severe residual urinary incontinence. The latter two patients’ urinary incontinence was so severe that they were not voiding any urine normally and they were continually wet, thus having the same symptoms as a woman with obstetric fistula.

As all women were illiterate, a qualified nurse (W.F.) collected details of the patients’ personal and medical history and administered the presurgery GHQ-28 through interview. The principal investigator (A.B.) instructed her in the use of the questionnaire, and the initial interviews were performed under his supervision. All women were subsequently treated surgically and had a 2-week postoperative recovery period with free bladder drainage through an indwelling Foley catheter. At the end of the 2 weeks, the catheter was removed and the clinical outcome was recorded (classified as: cured with no incontinence, fistula closed but mild residual incontinence on exertion, fistula closed but sustaining severe urethral incontinence on walking and/or lying in bed or fistula repair failed). None of the women received any formal counselling during their admission as this was not available in Barhirdar. During discharge, the GHQ-28 was again administered, but this time by a nursing student who was blinded to the answers to the first questionnaire. She had been trained by the nurse (W.F.) and doctor (A.B.), and the initial interviews were performed under their supervision. All interviews were undertaken in Amharic (the regional language of Barhirdar), before being translated to English, transferred onto a formatted sheet and entered into a database. The results of the GHQ-28 were analysed using the chi-square and Mann–Whitney U tests.


All women with obstetric fistula admitted during the time period of the study agreed to partake in the survey. The mean age of the women was 27 years, and the mean time from antecedent event to presentation of fistula was 3 years. The mean parity was 3.1, and the mean number of living children was 1.2. There was only one live child from the causative deliveries (98% stillbirths), and 42% of women were divorced as a result of their injuries.

Following surgery, 3 of the 51 repairs were considered to have failed (5.9%). Of the 48 remaining women, 3 had severe continuing urinary incontinence per urethra (6.2%) and 6 had mild stress incontinence with activity (12.5%).

On admission all 51 women screened positive on the GHQ-28 questionnaire with a mean score of 24.8, median 27 (Table 1). On discharge the same questionnaire was administered to all women. This time only 18 of the 51 women screened positive (36%), and the overall mean score of the 51 women was only 4.7 with a median of 2. The difference in prevalence of screen-positive women was statistically significant (chi-square test, P = 0.003) as was the difference in median score (Mann–Whitney U test, P < 0.05).

Table 1.  Results of GHQ-28 screening scores for women with vesicovaginal fistula before and after treatment
 nNumber (%) screening positive (GHQ score >4)Median (IQR)
  1. IQR, interquartile range.

Admission (before treatment)5151 (100)27 (23–28)
Discharge (after treatment)5118 (35)2 (0–6)
Failed repairs33 (100)23
Successfully closed fistula4815 (31)7 (6–15)
 Closed and continent4513 (29)7 (5.5–7)
 Closed, but severe incontinent33 (100)20

The postoperative score was highly dependent on the success of the treatment. All six women for whom the operation was not curative (three failed repairs and three with severe residual incontinence) scored positive with median scores of 23 and 20, respectively (Table 1). The median score of those with failed repairs was significantly higher than for those with a successfully closed fistula (P < 0.05).


This study has demonstrated that women who undergo successful operative repair of obstetric vesicovaginal fistulae improve markedly in mental health scores. This occurs despite the lack of any formal psychological or psychiatric input.

The results of this survey may be open to bias as the women in this cultural context are disempowered and often give answers that they think will be pleasing to the health professional and not the honest answer. Furthermore, women may be feeling better from the experience of simply being cared for in the hospital for the previous 3 weeks. However, the experience of those women with failed treatments who continued to score highly seems to suggest that neither of these are major confounding factors.

The use of different people to conduct the interviews before and after surgery could also have biased the results. This effect was minimised by each of the interviewees undergoing the same instruction and initial supervision.

Those women who do not improve from the operation either because the repair failed or because they remained with severe urethral incontinence, still tested positive for potential mental dysfunction. In the event of operative failure, it is the usual practice in fistula centres to ask the patient to return home and return after 3 months for another attempt at operative repair. In the case of severe continuing incontinence, the women are taught pelvic floor exercises and asked to go home and return after 6 months for a stress incontinence operation. The results of this study suggest that these women continue to be under severe mental strain and should therefore be offered a great deal of psychological support. It may be appropriate to provide a waiting area close to the fistula unit where they can stay until they are operated and maybe to offer the second operation much sooner. At admission, the woman should also be realistically informed of the likelihood of success of the operation and of the steps that might be taken to help if the operation does not succeed.


Women who undergo successful obstetric fistula repair have markedly improved mental health scores on the GHQ-28 screening test. In contrast, those women with unsuccessful repairs continue to have high levels of psychiatric morbidity. Greater attention needs to be paid to the welfare of these women with continuing incontinence from not only a medical/surgical management viewpoint but also a psychiatric viewpoint.


The authors would like to thank Dr Catherine Hamlin and the Addis Ababa Fistula Hospital for their support in writing this paper. The authors also thank Prof Allan Chang, Director, Mater research Support Centre, Mater Misericordiae Hospital, Brisbane Australia for assistance with the statistical analysis.