Intended for healthcare professionals


Female genital mutilation/cutting

BMJ 2013; 347 doi: (Published 24 September 2013) Cite this as: BMJ 2013;347:f5603
  1. Henrietta L Moore, William Wyse chair of social anthropology
  1. 1Division of Social Anthropology, Cambridge CB2 3RF, UK
  1. Hlm10{at}

Targeted interventions can work, but more remains to be done to change people’s behaviour

In December 2012, the United Nations General Assembly adopted a resolution to intensify global efforts to eliminate female genital mutilation/cutting. As the recent Unicef report argues, evidence played a major part in driving this resolution through.1 But what is the character of the available evidence, and what is known about how to accelerate change to bring about the desired result? Although the tone of the report is resolutely upbeat, the reality on the ground seems more uncertain and fragile.

The report provides the largest ever number of nationally representative surveys from all 29 countries where female genital mutilation/cutting is concentrated, as well as providing comparative data for age cohorts. This allows an assessment not only of how these practices are changing, but also the progress being made in the battle to eliminate them. However, the report also notes that even with the current evidence for declining prevalence rates, if current trends persist, as many as 30 million girls are at risk of being cut over the next decade. Clearly, there is much more to be done, but do we really know what to do and are we currently doing the right things?

It is true that there are marked reductions in prevalence among young women in certain countries, notably Kenya. But, alarmingly, this does not necessarily correlate with expenditure or the implementation of eradication programmes. Kenya has been a recipient of several major campaigns against female genital mutilation and cutting, so we might expect a reduced rate of female genital mutilation and cutting. Indeed, demographic and health survey data show a steady decline in prevalence in Kenya across all age cohorts and analysis in the report suggests that the practice of female genital mutilation/cutting has been steadily declining since the 1970s.

However, leaving aside the colonial campaigns of the 1920s and 1930s, sustained campaigning and investment really began in the 1990s with initiatives to develop alternative rites of passage that would maintain cultural values but encourage individuals and communities to give up traditional circumcision rituals and abandon female genital mutilation/cutting.2 How then should we explain the steady decline from the 1970s, which began before these initiatives, and the apparent lack of improvement in rate of decline during and after these recent campaigns in Kenya? Campaigns may not be the answer. The Central African Republic shows one of the sharpest declines in prevalence,1 but it has not been a major focus for campaigning. In contrast, there has been no significant change in prevalence in Senegal, which is often regarded as doing the most to end these practices.1

A series of factors other than specific projects and national campaigns might influence prevalence rates. There is good evidence from several contexts in Africa that mother’s educational attainment impacts positively, but evidence from Somalia and the Sudan shows higher prevalence rates for girls whose mothers have secondary education.1 Generally, prevalence rates are lower for girls in wealthier households, yet in Sierra Leone and Sudan there is no such association, and in Mali, 78% of girls from wealthier households are cut compared with 70% from poorer households.1 This suggests that the association is not robust or is only robust when certain other (unknown) circumstances prevail, or that there are problems with the aggregated national figures.

Clearly, there are problems with the data—as would be expected with such a sensitive topic and in such challenging research and reporting circumstances. For example, in Mali, surveys were conducted across age cohorts through time, so women who were aged 20-24 years in the 1995-96 survey were 25-29 in 2001, 30-34 in 2006, and 35-39 in 2010. Given that female genital mutilation and cutting usually occurs under the age of 20 years, the prevalence within each of these age cohorts would be expected to be about the same across surveys. Instead the reported prevalence was 94% in 1995-96, 92% in 2001, 84% in 2006, and 90% in 2010. Even assuming under-reporting in the 2006 survey, how robust are these figures in relation to an argument about declining prevalence rates and their causes?1

The truth is, as this report and others show, that targeted interventions do work at the small scale to bring about low falls in prevalence, but their impact and results differ, as do the pathways to change, both within and across countries.3 4 5 This report is very welcome, and is based on much brave community activism and good research, but more remains to be done. As with all public health initiatives, changing people’s behaviour is often a slow process.


Cite this as: BMJ 2013;347:f5603


  • Competing interests: We have read and understood the BMJ policy on declaration of interests and declare the following interests: None.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract