Obstetric fistula is the presence of a connection either between a woman’s genital tract and her urinary system (vesico-vaginal) or between the genital tract and the rectum (recto-vaginal fistula) due to untreated obstructive labor, precipitous delivery, abortive maneuver, or an iatrogenic cause (instrumental delivery, cesarean section) in the absence of appropriate obstetric care [1] - [3] .

The WHO estimates that approximately 50,000 to 100,000 women develop an obstetric fistula each year, with at least 33,000 of these cases occurring in sub-Saharan Africa [4] [5] . Obstetric fistulas represent a major public health issue in developing countries. It is a true social and psychological tragedy; approximately 800 women die every day from complications related to pregnancy or childbirth worldwide. For every woman who dies from maternal-related causes, it is estimated that at least 20 suffer from maternal morbidity, one of the most severe forms being obstetric fistula [6] .

The conditions conducive to the occurrence of obstetric fistulas are based on the three delays described by Maine and adopted by WHO: delay in deciding to seek care, delay in reaching a care facility once the decision is made, and finally, delay in receiving appropriate care once at the hospital [7] . All these factors can be distributed across three levels of responsibility incumbent on the individual, the community, and the health system. Several strategies are implemented to address this, with one of the main approaches being the prediction of delivery in the 8th month and precautions during childbirth, with the midwife as the key player [8] . Thus, the fight against obstetric fistulas is a major concern for midwives.

It is with this perspective that we proposed conducting a study on the Knowledge, Attitudes, and Practices (KAP) of midwives in Brazzaville by analyzing their knowledge, attitudes, and practices concerning obstetric fistulas in Brazzaville in 2024.

This was an analytical KAP (Knowledge, Attitudes, Practices) study conducted from May 10 to September 10, 2024 (four months), in the Gynecology and Obstetrics departments of hospitals in the city of Brazzaville, Congo. The study took place at the following locations: Brazzaville University Hospital Center, Talangaï Referral Hospital, Makélékélé Base Hospital, Blanche GOMES Specialized Mother and Child Hospital, Bacongo Base Hospital, and Pierre Mobengo Central Military Hospital.

The target population consisted of all midwives who participated in childbirth activities and outpatient consultations at the hospitals selected for the study. Excluded were midwives who did not answer all the questions, those with less than 12 months of experience, and those who participated in the pre-test.

A non-probability sampling method was used, with a total sample size of N = 200 midwives.

The variables studied were related to:

For the successful implementation of the survey, a pre-test was conducted to assess the midwives’ understanding of the questionnaire. This pre-test was carried out with 20 midwives randomly selected from a health center that was not included in the study. This phase allowed for the adjustment of certain questions and the adaptation of the questionnaire to the midwives’ level of comprehension.

We performed univariate and bivariate analyses using the software CS Pro 7.4 and EXCEL 2019.

We also conducted an analysis of the degree of certainty of responses using the spectrum of response qualities for Instruction 1 by Dieudonné Leclercq. This method categorizes knowledge-based questions into a spectrum where the surveyor marks the given answer and circles the degree of certainty or probability percentage among the six following options: 0%, 20%, 40%, 60%, 80%, 100%. If the respondent had no knowledge of the question, the surveyor was required to indicate a certainty level of 0%.

Attitudes were assessed using the Likert scale, employing a predefined scale with the following response options: Strongly agree - Agree - Disagree - Strongly disagree practice was considered good when it aligned with medical protocols and international recommendations aimed at preventing and treating obstetric fistulas, and poor when it did not meet these standards.

We used the Chi-square test to assess the homogeneity of the distribution of the study populations. We then obtained the odds ratio (OR), the 95% confidence interval, and conducted Fisher and Pearson tests.

In this study, we used a non-probabilistic sampling method based on the availability of respondents and the accessibility of data. While this approach allowed us to collect relevant information, it has certain limitations in terms of the representativeness of the target population.

The midwives were aged between 22 and 60 years, with an average age of 36.5 ± 9.1. TEBEU et al., in a 2019 KAP study on obstetric fistulas among healthcare professionals in Cameroon, reported an age range of 23 to 62 years with an average age of 39.2 ± 7.8 [9] . More than half of the midwives had professional experience of 10 years or less (76.5%), whereas in the study by TEBEU et al., 141 providers, or 45.6%, had 1 to 9 years of experience.

The analysis of the level of certainty to identify partial knowledge showed that among the 200 midwives who had heard about obstetric fistulas and answered the questionnaire, 81.8% of responses were correct compared to 18.2% incorrect responses. Specifically, 45.9% of the knowledge was correct with 100% certainty; 70.6% was usable, 15.7% was unusable. Additionally, 13.7% of the knowledge was harmful; 6.0% was incorrect with 100% certainty; 11.6% corresponded to acknowledged ignorance.

The ideal distribution of incorrect responses should be J-shaped, which is beneficial because, all else being equal, the more the distribution of incorrect responses resembles a J-curve (shifted to the right), the better: if one must answer incorrectly, it is preferable to do so with the lowest certainty (closest to 20%) possible [10] . In our study, the distribution of incorrect responses formed a sharply left-skewed J-curve, which is concerning because 6% of the 18.2% incorrect responses from midwives were given with maximum certainty (100%), and 15.7% of the knowledge was unusable with 13.7% being harmful. This harmful knowledge, concerning symptoms and prevention, requires corrective intervention for the concerned midwives by organizing continuous training modules on obstetric fistulas; as this situation is due to some midwives only, it is important to identify which ones.

The recklessness or average certainty accompanying incorrect responses is 67.8%, which is concerning as it is too high.

The distribution on the hemisphere of spectral qualities of correct responses ideally has its peak as far to the right as possible (J-curve, again). Here, it is quite steeply J-shaped (a large majority of the correct responses are with 100% certainty). We observe that 70.6% of the knowledge is usable, with 62.9% of confident responses (correct at 80% or 100%), of which 45.9% are perfect responses (correct with 100% certainty). We also see 9.8% uncertain responses (correct with 20% or 40% certainty) and 11.6% admitted ignorance (incorrect responses with 50%).

The confidence or average certainty accompanying correct responses is 79.2%, which is satisfactory. Confidence and recklessness are impact indices that will serve as references during the evaluation of knowledge after competency-based training strategies for midwives.

Among the 146 midwives with satisfactory knowledge, only 55/146 had good practices. Similarly, of the 189/200 midwives with adequate attitudes, only 79/189 had good practices. Knowledge and attitudes significantly influenced practices (OR = 0.35; 95% CI: [0.186 - 0.678]; p = 0.002; OR = 13.9; 95% CI: [1.75 - 110.99]; p = 0.001). In the literature, there is no data on the influence of healthcare providers’ knowledge and attitudes toward obstetric fistulas on their practices. However, these negative practices justify the need for better involvement of healthcare personnel concerning obstetric fistulas.

Nevertheless, the influence of knowledge and attitudes on the practices of midwives in Brazzaville might be partly related to the fact that they are not frequently in contact with obstetric fistulas. For several decades, Congolese women have predominantly delivered in maternity wards, which is not always the case in some sub-Saharan countries where socio-cultural prejudices outweigh the use of health facilities [3] [11] - [13] . Additionally, the frequency of obstetric fistulas is relatively low, as found in a study conducted at the Brazzaville University Hospital in 2002 by Bouya and colleagues, which managed 39 cases of obstetric fistulas in the largest urology department in the country [14] . In contrast, Teguete in Mali found 1096 confirmed cases of fistulas over four years [15] .

The midwives in Brazzaville are mostly young, with a secondary education level and more than 5 years of experience. Overall, they had sufficient knowledge levels and adequate attitudes, but there were significant disparities in the certainty of their responses. Their practices were poor, influenced by both inadequate attitudes and a harmful lack of knowledge. Therefore, a training strategy based on a competency approach regarding obstetric fistulas could help improve their practices.