Sexual quality of life in women who have undergone female genital mutilation: a case–control study

Authors


Dr SHA Andersson, Department of Anaesthesia, Trafford Hospitals, Central Manchester University Hospitals NHS Foundation Trust, Moorside Road, Davyhulme, Manchester, M41 5SL, UK. Email: stefan.andersson@nhs.net

Abstract

Please cite this paper as: Andersson S, Rymer J, Joyce D, Momoh C, Gayle C. Sexual quality of life in women who have undergone female genital mutilation: a case–control study. BJOG 2012;119:1606–1611.

Objective  To investigate the sexual quality of life of women who have undergone female genital mutilation (FGM) and compare them with a similar group who has not undergone FGM.

Design  Case–control study.

Setting  A large central London teaching hospital.

Population  A total of 73 women who had undergone FGM and 37 control women, who had not undergone FGM but were from a similar cultural background where FGM is practiced.

Methods  The women completed a questionnaire containing the Sexual Quality of Life-Female (SQOL-F) questionnaire.

Main outcome measures  SQOL-F score.

Results  Women who have undergone FGM of any type have a significantly lower (P < 0.001) overall SQOL-F score than control women (mean = 62.44, SD = 27.93 versus mean = 88.84, SD = 13.73). Women who were sexually active and had undergone FGM type III differed the most from sexually active controls (< 0.05) in their SQOL-F score. Women who were sexually inactive but who had undergone FGM reported significantly lower overall SQOL-F scores (P = 0.015) than sexually inactive controls, but were not differentiated by type of FGM.

Conclusion  FGM significantly reduces women’s sexual quality of life, based on the results of the SQOL-F questionnaire.

Introduction

With such a significant expansion of the migrant population in the UK, female genital mutilation (FGM) is becoming an important issue within the practices of many health professionals, particularly, general practitioners, obstetricians and gynaecologists.

Female genital mutilation is defined as any procedure ‘involving partial or total removal of the external female genitalia without medical reasons.’1 It has been classified by the World Health Organization (WHO) as follows:

  •  Type I: Partial or total removal of the clitoris and/or the prepuce (clitoridectomy)
  •  Type II: Partial or total removal of the clitoris and the labia minora, with or without excision of the labia majora (excision)
  •  Type III: Narrowing of the vaginal orifice with creation of a covering seal by cutting and appositioning the labia minora and/or the labia majora, with or without excision of the clitoris (infibulation)
  •  Type IV: All other harmful procedures to the female genitalia for non-medical purposes, for example, pricking, piercing, incising, scraping and cauterization.2

As many as 140 million women and girls worldwide have had the control of their sexuality and the full enjoyment of their rights and liberties taken away from them.3,4 Another 3 million girls are at risk of experiencing the same fate.5 These women and girls are, or will be, the victims of FGM.

Female genital mutilation is always traumatic and has no known health benefits,2,3 and the complications range from bleeding and infection to death. Although the physical complications have previously been reported, the effects on sexual function and satisfaction have been poorly reported.6,7 The British Medical Association states in their leaflet on FGM: ‘Little is documented about the psychosexual and psychological sequelae of female genital mutilation.’8

We hypothesise that women who have undergone FGM will have a low sexual quality of life; the aim of this study was therefore to explore the impact of FGM on sexual quality of life in women who have undergone this procedure.

Methods

An African Well Women’s Clinic was established at Guy’s Hospital in 1997 to serve women who had previously undergone FGM. Initially the clinic mainly attracted women who were pregnant but has expanded to receive women who have had FGM requesting reversal and women who have had FGM and have gynaecological problems. One hundred and twenty-eight women were recruited between March 2006 and September 2010 from a central London Teaching Hospital women’s services directorate that has a special interest in FGM and serves an ethnically diverse area of London. The controls were recruited from the general gynaecology clinic, which has a very diverse ethnic population and also the antenatal, postnatal and gynaecology wards.

Inclusion criteria were: aged 16 years or older; to have undergone FGM, or to be from a cultural background where FGM is frequently practiced but had not undergone FGM. The women had to be able to understand English or be accompanied by an approved interpreter. The women were only asked to participate when a researcher was present in the clinic. A study like this is prone to selection bias as the practice of FGM is deeply rooted in cultural traditions. We tried to address this bias by recruiting controls from cultural backgrounds where FGM is practised.

The women were asked to complete a questionnaire that was divided into two parts: part one asking demographic questions and a second part containing the Sexual Quality of Life-Female (SQOL-F) questionnaire (previously validated for measuring sexual quality of life in women with female sexual dysfunction aged over 18 years).9 Use of the Sexual Quality of Life Assessment Tool was made under licence from Pfizer Ltd, Sandwich, UK.

Written informed consent was obtained from all women and all questionnaires were anonymous.

Demographics measured included, age, country of birth, civil status, number of children, smoking status, alcohol consumption, self-reported history of depression and type of FGM. Women choose their type from a table describing the different types. In some cases, the researcher specified this. The FGM classification was based on the WHO 1995 typology2 of FGM with the alteration of type IV, which was further subdivided into Angurya (scraping of tissue around the vaginal orifice) or Gishiri cuts (cutting of the vagina), and introduction of corrosive substances or herbs into the vagina.

This was a case–control study with a two to one ratio.

Statistical analyses were conducted in MATLAB 2006b (MathWorks Inc., Cambridge, UK), with exploratory analysis (analysis of variance) with independent variables being group (control versus FGM), age, country of origin, marital status, children, sexually active, smoking, alcohol and self-reported depression and the dependent variable being SQOL-F. Further post-hoc multiple comparison analysis, with Bonferroni correction, explored the effect of FGM type on SQOL-F.

Ethics approval was gained from Bromley Local Research Ethics Committee, REC reference number 06/Q0705/3.

Results

A total of 128 women were recruited, of whom two withdrew (one case and one control), 12 did not complete the SQOL-F questionnaire appropriately for a score to be calculated, one control from Zimbabwe and one from Angola were excluded as there are no reports in WHO data2 of FGM being practiced in these countries. Two further controls from the Democratic Republic of the Congo were also excluded because there are only anecdotal reports of FGM being practiced there,2 resulting in 73 women with FGM and 37 control women being included in the analysis. One participant was found to have completed two questionnaires over the years so the second one was voided.

Only women who had a good command of English or who were accompanied by a suitable interpreter were invited to participate when the researcher was present, hence the prolongation of the recruitment period.

Within the FGM group the mean age was 29.16 (n = 70) with a range of 19–39 years and in the control group the average age was 34.03 (n = 35) with a range of 20–59 years.

Of the women who had had FGM, the type of FGM was distributed as follows: FGM type I 27.4% (n = 20), FGM type II 12.33% (n = 9), FGM type III 35.62% (n = 26), Angurya/Gishiri cuts 1.37% (n = 1), Corrosive substances 1.37% (n = 1), unspecified (unknown or not specified) 21.92% (n = 16). (Table 1).

Table 1. Type of FGM (based on WHO classification)
 Cases, n (%)
Type I20 (27.4)
Type II9 (12.33)
Type III26 (35.62)
Angurya/Gischiri cuts1 (1.37)
Corrosive substances1 (1.37)
Unspecified (unknown or not specified)16 (21.92)

Sixty-three (86.3%) of the 73 women who had undergone FGM were from Somalia, Sierra Leone, Nigeria or Eritrea, whereas of the 37 controls, 32 (86.49%) were from Nigeria and Ghana (Table 2). All women were immigrants to the UK.

Table 2. Country of origin of women participating in the study
 Cases, n (%)Controls, n (%)
Cote d’Ivoire1 (1.37)1 (2.7)
Eritrea7 (9.59)0
Ethiopia1 (1.37)0
Gambia2 (2.74)0
Ghana010 (27.03)
Guinea1 (1.37)0
Kenya1 (1.37)1 (2.7)
Liberia2 (2.74)0
Nigeria9 (12.33)22 (59.46)
Sierra Leone13 (17.81)2 (5.41)
Somalia34 (46.58)0
Sudan2 (2.74)0
Uganda01 (2.7)

Exploratory analysis (analysis of variance) examining group (control versus FGM), age, country of origin, marital status, children, sexually active, smoking, alcohol, self-reported depression on the dependent variable SQOL-F showed a main effect of group (F1,71 = 11.74, P = 0.001), and being sexually active, (F1,71 = 17.21, P = 0.0001) with control women reporting higher SQOL-F ratings (mean = 88.84, SD = 13.73) than the FGM group (mean = 62.44, SD = 27.93).

Being sexually active was a potential confounder in reported SQOL-F, where those women who were sexually active had a higher mean SQOL-F (= 92, mean = 76.26, SD = 24.19) than women who were sexually inactive (= 16, mean = 44.13, SD = 25.31). Therefore, further analysis was conducted on two groups separated into those women who were sexually active and those who were not.

Sexually active group

A multiple comparisons procedure (with Bonferroni correction) within the group of sexually active women, explored the effect of FGM type on SQOL-F. This revealed that controls (no FGM; n = 34; mean = 91.20, SE 3.76) had significantly higher SQOL-F scores (at the significance level P < 0.05) than type 3 (n = 20; mean = 63.28, SE = 4.83) and unspecified FGM (n = 12; mean = 68.70, SE = 6.24).

Sexually inactive group

Within the group of women who were not sexually active there were six controls (women with no FGM) and 12 women with FGM. Analysis of variance showed that there was no significant effect of FGM type on reported SQOL-F (F5,15 = 1.56, P = 0.26). However, a two-tailed t-test showed that for sexually inactive women, the mean SQOL-F for the controls was significantly higher (n = 6; mean = 69.43, SD = 28.52) than for women with FGM of any type (n = 12; mean = 35.70, SD = 18.48); t16 = 2.78; P = 0.015.

The effect of type of FGM undergone on SQOL-F score was compared and the data are shown in Figure 1.

Figure 1.

 Mean SQOL-F scores in control and FGM categories. Black filled circle and vertical dotted line: mean SQOL-F in control group with 95% confidence interval; black filled circles and horizontal lines: mean SQOL-F scores for FGM type and controls that differ significantly at the < 0.05 level, with line length indicating 95% confidence interval on the means. Clear circles: FGM categories for which SQOL-F mean scores were not significantly different from control at the 95% confidence level.

Discussion

This study demonstrates that there is a statistically significant difference in sexual quality of life scores between women who have undergone FGM and those who have not with the FGM women scoring lower than control women. In sexually active women this study has established that FGM type III is associated with the lowest sexual quality of life scores compared with controls. Further, in sexually inactive women, FGM still lowers sexual quality of life, but participant numbers limited the extent to which we could determine the effect of each type.

There is a limited amount of data available that explores this issue, which has such a dramatic effect on the lives of women.

Previous studies have shown that FGM adversely affects sexual function. Alsibiani and Rouzi6 demonstrated in their study of 260 women, half of whom had undergone FGM, and half of whom were controls; a statistically significant reduction in the overall Female Sexual Function Index score for the former group as well as some individual domains.

In a study by Catania et al.,10 57 women with type III FGM reported higher scores than controls in several domains of the Female Sexual Function Index, but this study was confounded by the control group consisting mostly of western women (54 of 57). In our case–control study, we tried to eliminate this confounder by recruiting women from similar cultural backgrounds.

Almrot et al.11 previously demonstrated the negative impact that FGM has on fertility, specifically the correlation between anatomical extent of the FGM and fertility. This study demonstrates that in addition to these negative physical effects, FGM has a negative psychological (holistic) impact, as demonstrated in the reduced SQOL-F score thereby adding further to the evidence of the harmful effects of FGM.

The four commonest countries of origin of the participants in this study were Somalia, Sierra Leone, Nigeria and Eritrea. Within the last 10 years, data collected on FGM in these countries shows a prevalence of 97.9%, 94%, 19% and 88.7%, respectively [2]. One point worth noting about the data collection in this study was the difficulty in recruiting controls. As can be seen from the statistics outlined above, the prevalence of FGM is so high, that suitable case controls are rarely available. For example, we were unable to recruit controls of Somalian descent, although the greatest numbers of participants in the FGM group were from this nation, because the reported FGM prevalence is 97.9%.2 This is a potential source of bias. Ideally, the control group should better reflect the nation of origin of the majority of cases, however this is impossible to achieve. Some countries practice particular types of FGM and different cultural factors play a significant role in sexuality and these different factors cannot be elucidated in a study such as this.

Limitations of the study

Studies of human sexuality are prone to bias and have confounding factors because of the wider cultural context as well as psychosocial factors that define this aspect of human behaviour. The choice of questionnaire could be criticised because we did not have evidence that all women had female sexual dysfunction. However, we felt that this questionnaire was the most relevant for our study because it did address sexual quality of life and we felt that it was the best available at the time that had been validated. We were also assuming that women would have some degree of sexual dysfunction having had their genitalia mutilated. Another specific source of bias arising from the use of this questionnaire is that it has been validated in women from the UK and USA9 and therefore could be criticised as it has not been validated within different cultural groups. However, quantitative studies aiming to show between-group differences require a validated instrument. Additionally, our initial analysis of variance showed no effect of country of origin on the SQOL-F score regardless of FGM status.

In terms of recruitment bias, the group of women who have undergone FGM were mainly recruited from the African Well Women’s Clinic where they would have presented for reversal of FGM or a gynaecological complaint. However, given the focus of our study on FGM, recruiting appropriate women for this type of study from a different setting would not have been possible.

This study recruited women from a variety of countries, and subsequently has a control group consisting predominately of women of Nigerian origin, whereas the FGM group consists mainly of women of Somali origin. Our data may therefore be confounded by cultural attitudes to sexual activity and satisfaction, rather than being solely attributable to FGM-specific differences. We have not been able to find good published data relating to attitudes to female sexuality in Somalia versus Nigeria so we are unable to comment as to how this may have influenced our results.

There was a high risk of both response and recall bias in this study, as this study was investigating aspects that many women would consider to be of a very private nature. This meant that women may not have taken part at all or, may have given a falsely high or low score.

Designing and implementing a study like this has been difficult, particularly as the populations being examined are immigrants.

Conclusion

As a result of high rates of migration of women from nations where this practice is considered commonplace, to areas where it is not, it is important to all multidisciplinary practitioners to have an understanding of the impact of this practice on the women for whom we provide care. The effect of a reduced or poor sexual quality of life is likely to have an effect on the general wellbeing of women.

With increasing numbers of women who have undergone FGM integrating into western societies, the sexual function (and its impact on psychological health) of these women is going to become an increasingly important issue. It is important to raise the profile of the damage that is caused with regards to sexual function in the hope that this may go some way to eradicating the procedure for future generations.

This study was conducted mainly on an immigrant population in London. This has enabled us to give a wider picture of the sexual quality of life of immigrant women affected by this practice in the UK.

Disclosure of interest

All authors believe FGM to be harmful and believe its practice should be eradicated.

Contribution to authorship

SA and JR took part in the conception and development of the study; SA took part in acquisition and analysis of data, and in drafting and editing the paper, JR took part in acquisition of data and revising the paper and acted as supervisor. DJ analysed data, drafted the results section and revised the paper. CM took part in acquisition of data and editing/revising the paper and CG in acquisition of data and drafting the paper.

Details of ethics approval

Ethics approval was obtained from Bromley Local Research Ethics Committee, REC reference number 06/Q0705/3.

Funding

No funding sought or obtained.

Commentary on ‘Studying the psychosexual impact of female genital mutilation’

Female genital mutilation (FGM) is still widely practised in many parts of the world despite legislative attempts to eliminate it. FGM describes a range of procedures that involve cutting, excising or otherwise deliberately injuring the external genitalia of female babies, girls and young women. It is firmly rooted in the tradition of those cultures that practise it, where it is believed to protect women from disgrace and promiscuity, or is used as an initiation ritual that celebrates a girl’s rite of passage into womanhood. But tradition cannot justify damaging women’s long-term health and sexuality. Female genital mutilation is condemned by the international medical community as a morally offensive and unacceptable act that causes harm and violates human rights. The physical effects of FGM are well characterised and include both immediate threats (haemorrhage, infection) and long-term consequences (problems with childbirth, pelvic pain, sexual dysfunction). By contrast, comparatively little is known about its psychological sequelae, in particular, the effects of FGM on sexual quality of life.

Immigrants seeking refuge from war, civil unrest and drought in FGM-practising countries such as Somalia and Eritrea are increasingly settling in Western Europe and North America, where, as a result, FGM has become an important public health issue. The psychological impact of FGM on sexual health is poorly understood but is likely to be particularly complex for immigrant women experiencing new attitudes to FGM and different cultural norms of sexual expectation compared with those with which they grew up. An improved understanding of these psychosexual issues will greatly enhance our ability to provide appropriate, specialised care for women who need it. In this study, Andersson et al. explored the impact of FGM on sexual quality of life in women originating from African countries where the practice is endemic. Studies like these are vitally important but challenging to design and conduct. Their design is hampered by the lack of a validated tool for the population of interest; the Sexual Quality of Life-Female (SQOL-F) questionnaire was validated on women from Europe, the USA and Australia (Symonds et al., Journal of Sex and Marital Therapy 2005;31:385–97), all countries with very different expectations of female sexuality. There are also challenges in recruitment to this type of study. Andersson et al. recruited women from a specialised African Well Woman Clinic designed to address obstetric concerns, gynaecological problems and reversal of FGM. In such a cohort, women experiencing sexual difficulties are likely to be over-represented. Participants originated mostly from Somalia, Eritrea and Sierra Leone, where FGM is widely practised, whereas women in the control group originated mainly from Ghana and Nigeria, where it is less common, adding a serious potential source of bias. On top of these difficulties, recruitment was slow, partly because of language barriers but perhaps also because studies like this ask difficult questions about topics that evoke shame or fear of reprisal in prospective participants. In combination, these confounding factors limit interpretation of the data and our ability to draw robust conclusions from them. Even so, studies like these are a brave attempt to explore tough issues in the real world, where bias cannot always be controlled for. Future work is required to substantiate these findings.

Disclosure of interests

I declare no conflicts of interest.

E Crosbie

University of Manchester, St Mary’s Hospital, Manchester, UK

Ancillary