Our conceptual framework was inspired by the theory of social conventions and norms,³² which seems to have influenced the practice of infibulation. The practice of infibulation is rooted in gender roles, community influence and a sense of belonging. It is not a religious practice, but a cultural one that reflects shared values concerning family honour, feminine virtues, sexual abstinence and other codes of conduct. The practice of infibulation continues because of custom, belief, social pressure, patriarchy and the assumption that it increases girls’ chances of marriage. However, it is practised to control girls’ bodies and sexuality for the benefit of men, as it is perceived as guaranteeing virginity until marriage. It is passed down from older to younger generations as an obligation accepted by the community. The practice is a symbol of social control over female sexual pleasure and is linked to women’s reproductive roles in society. It is also thought to increase sexual pleasure in men and make women attractive. Social norm theory has become important for understanding why infibulation continues and for designing interventions to end the practice. It is possible to design interventions to accelerate the abandonment of infibulation without undermining traditional values. The present study will investigate the individual and contextual factors (Figure 1) associated with the practice of infibulation among girls aged 0–14 years in Guinea.

This was a secondary analysis of data from the 2018 Demographic and Health Survey (DHS) of Guinea. The primary study was a cross-sectional survey of women of reproductive age with daughters aged from 0 years to 14 years old who had undergone FGM. Data collection took place from March 2018 to June 2018 through face-to-face interviews by trained data collectors to standardise understanding and administration of the questionnaire.

The Republic of Guinea is located in West Africa, between 7° and 12° north latitude and 8° and 15° west longitude. It is bordered by Guinea Bissau to the northwest, Senegal and Mali to the north, Côte d’Ivoire to the east, Liberia and Sierra Leone to the south and the Atlantic Ocean to the west. The population was estimated at 13 million inhabitants in 2018, of which 51.5% were women. Administratively, Guinea is made up of 7 administrative regions, plus the city of Conakry, which has the status of a decentralised community. The country has 33 prefectures, 38 urban communes, including five in Conakry and 334 rural communes. These communities are subdivided into 308 urban neighbourhoods and 1615 rural districts. The public health system is organised into three levels: the first level, comprising 407 rural and urban health centres, 26 district hospitals and 9 communal medical centres; the second level, comprising eight regional hospitals; and the third level, represented by three national hospitals.³³

The study population consisted of all girls aged from 0 years to 14 years old who underwent female genital mutilation or excision (FGM/E), and residing on the national territory at the time of the survey. In total, the primary survey covered 3950 women with at least one daughter who has undergone FGM.

For this study, secondary data from the 2018 Guinea DHS were used. These data are accessible in the DHS programme repository (https://dhsprogram.com/data/available-datasets.cfm)³³ to all registered users and can be downloaded with permission. After reviewing and understanding the details of the data structure, the appropriate data type for the analysis of female infibulation was selected. For this study, the file that contains data for women of reproductive age (15–49 years old) was used. A sample of 3950 women who had excised daughters with infibulation was considered. Data on infibulation and its potential explanatory variables (individual and contextual level variables) were extracted.

Additional recording of variables was carried out as needed for the analysis. The data being processed and cleaned were analysed using Stata Version 17, and QGIS/GIS 3.10 software was used for spatialisation of infibulation data. Out of a sample of 3950 excised girls, 21 (0.53%) were excluded because of missing data at the time of modelling. Model specification was performed using the Hosmer-Lemeshow test while multicollinearity was tested using the variance inflation factor (VIF). Sampling weights were applied to correct and account for the sampling structure, while the SVY command (in Stata is used to analyse complex survey data: stratification, clusters, etc.) was used to address the complex nature of the survey and generalise the results.

The analyses consisted of basic descriptive statistics to describe the data and characteristics of the respondents as well as the calculation of the prevalence of infibulation in girls aged from 0 years to 14 years old. The Chi-square test of independence was used to assess the statistical significance of the association between each of the explanatory variables and infibulation in girls aged from 0 years to 14 years old. Variables that showed significant associations with infibulation at the Chi-square level at a p-value less than 0.20 and variables that were considered to have an association with infibulation through the literature review were used in the logistic regression modelling. The results of the regression analysis were presented as crude odds ratios (bivariate analysis) and adjusted odds ratios with their p-value and 95% confidence interval (CI) indicating precision. For the validation of the final model, the estimation quality measures and influential values were used as post-estimation tests (likelihood ratio test, model discriminatory power, global model fit, model diagnostics, etc.).

The primary study had obtained authorisation from the Health Research Ethics Committee and, in our analyses, we respected the confidentiality of the participants. As the data were not collected by the authors of this study, we requested authorisation from the Demographic and Health Surveys (DHS) Programme to exploit the DHS-2018 data, and access to the data was provided after our intention of the request was assessed and approved. The dataset is freely available at (https://dhsprogram.com/data/available-datasets.cfm).

The study focused on 3950 girls who have undergone FGM residing in all regions of Guinea. Table 1 presents the individual and contextual characteristics of the study population. The average age of the mothers was 35.5 (7.1) years old and that of the girls was 7.5 (13.9) years old. The male gender of the head of household was the most represented in the sample (87.6%). Mothers aged from 35 years to 44 years old represented 42% and more than half (58.5%) of the infibulated girls were aged from 5 years to 9 years old. The majority of mothers were married (95.5%), without any level of education (87.7%). More than a third (34%) of them had five to six living children and 93.8% practised the Muslim religion. More than 11% of mothers were infibulated, while 15% of them did not know their infibulation status. About 46% of mothers lived in a household with a poor wealth index and 74.5% resided in rural areas.

The overall prevalence of infibulation among girls who underwent FGM/C in Guinea was 16% (95% CI: [0.1–0.2]).

This prevalence was 17% (95% CI: [0.1–0.2]) among girls aged from 0 to 4 years old and 14% (95% CI: [0.1–0.1]) and 15% (95% CI: [0.1–0.2]) for the other two age groups.

There is a variation in the distribution of this prevalence according to the place of residence. It was 13% (95% CI: [0.1–0.2]) in urban areas and 16% (95% CI: [0.1–0.2]) in rural areas. As for the distribution of prevalence according to regions of residence (Figure 2), it was 31% (95% CI: [0.2–0.4]) in N’Zerekore; 23% (95% CI: [0.2–0.3]) in Boké; 16% (95% CI: [0.1–0.2]) in Labé; 15% (95% CI: [0.1–0.2]) in Kindia; 14% (95% CI: [0.1–0.2]) in Faranah; 12% (95% CI: [0.1–0.2]) in Conakry; 11% (95% CI: [0.1–0.2]) in Kankan and 8% (95% CI: [0.1–0.2]) in Mamou.

In the bivariate analysis, the variables significantly associated with infibulation among girls who underwent FGM/C included the age of the head of household, age of the mother, number of living children of the mother, current occupation of the mother, religion of the mother, education level of the husband, infibulation status of the mother, wealth index, area of residence and region of residence. In multivariate analysis, four individual factors and one contextual factor were significantly associated with infibulation among girls who underwent FGM/C namely: age of the mother, current occupation of the mother, religion of the mother, infibulation status of the mother and region of residence (Table 2).

Table 2 presents the factors associated with infibulation among girls who underwent FGM/C in multivariate analysis. Girls whose mothers were aged from 25 years to 34 years and 35–44 years old had, respectively, 1.7 (OR = 1.7, 95% CI: [1.1–2.6]) and 1.4 (OR = 1.4, 95% CI: [1.0–2.1]) times higher odds of being infibulated compared to girls whose mothers were 45 years and older. Girls whose mothers were in an income-generating activity were 1.7 (OR = 1.7, 95% CI: [1.3–2.2]) times more likely to undergo infibulation compared to those whose mothers were not in an income-generating activity. By religion, Christian mothers were 2.7 (OR = 2.7, 95% CI: [1.2–5.8]) times more likely to have infibulated daughters compared to Muslim mothers. Daughters whose mothers were infibulated were 22.1 times more likely to undergo infibulation compared to those whose mothers were not infibulated (OR = 22.1, 95% CI: [16.6–29.4]). In contrast, daughters whose mothers did not know their infibulation status compared to those whose mothers were not infibulated were 7.7 (OR = 7.7, 95% CI: [5.8–10.2]) times more likely to be infibulated. Compared to girls residing in the region of Conakry, those living in the region of Boké were 2.8 (OR = 2.8, 95% CI: [1.6–4.8]) times more likely to undergo infibulation.

Some limitations deserve to be highlighted in the context of the realisation of this study. Firstly, the DHS is a cross-sectional study. Although it allows for establishing associations between the practice of infibulation and various explanatory variables, it does not establish a causal link. Secondly, some information may introduce measurement bias, such as the infibulation status of the girl, as the data collection relied solely on the mother’s declaration. Although self-reported data are commonly used and can provide useful information, it is essential to consider the potential biases they introduce. To increase the reliability of results, researchers should consider adopting rigorous validation strategies, combining objective data and employing carefully designed, diversified collection methods. Thirdly, some independent variables such as the mother’s current job could lead to a selection bias. Finally, the lack of qualitative data limits the in-depth understanding and explanation of the models generally characterised by a quantitative analysis.

Despite these limitations, the strengths of the study lie in the use of nationally representative data and a large sample size that guarantees the reliability and generalisability of the results to all girls in Guinea. The study could contribute to improving understanding of factors related to the practice of infibulation among infibulated girls aged from 0 years to 14 years old in Guinea.

In Guinea, infibulated women are the most supportive of the infibulation of girls, and there is also great potential to educate and persuade this group about the importance of eliminating the practice. Given the continued and consistently high prevalence of infibulation over the decades and the controversial cultural mores associated with infibulation practices, it would be useful for the Ministry of Health to ensure the implementation of regulatory texts prohibiting the practice of infibulation in all health institutions. In addition, it would be crucial to adopt policies that strengthen the engagement of communities and religious leaders in the prevention and control of infibulation. Present a precise framework for campaign implementation, with clear stages, necessary resources, monitoring indicators and impact assessment mechanisms. This will help policy-makers understand how to translate recommendations into concrete action. Furthermore, in synergy with other policies, raising awareness and promoting behavioural change among populations could ultimately lead to ending the practice of infibulation in Guinea.