The authors have declared that no competing interests exist.
Conceived and designed the experiments: AB JM. Performed the experiments: AB SK AS. Analyzed the data: AB HT JM. Contributed reagents/materials/analysis tools: AB SK AS JM. Wrote the manuscript: AB HT JM. Translated initial French document to English language: AB SK.
Obstetric fistula is a sequela of complicated labour, which, if untreated, leaves women handicapped and socially excluded. In Burkina Faso, incidence of obstetric fistula is 6/10,000 cases amongst gynaecological patients, with more patients affected in rural areas. This study aims to evaluate knowledge on obstetric fistula among young women in a health district of Burkina Faso, comparing rural and urban communities. This cross-sectional study employed multi-stage sampling to include 121 women aged 18-20 years residing in urban and rural communities of Boromo health district. Descriptive statistics and multiple logistic regression analysis were used to compare differences between the groups and to identify predictors of observed knowledge levels. Rural women were more likely to be married (
Obstetric Fistula is defined as a direct communication between the vagina and the bladder (vesico-vaginal fistula) and/or between the vagina and the rectum (recto-vaginal fistula) [
In sub-Saharan Africa and South Asia, obstetric fistula is very common, as access to and use of emergency obstetric care is limited [
In Burkina Faso, obstetric fistula remains a huge problem [
Burkina Faso is a francophone West Africa country with an economy that is mainly based on agriculture and livestock [
Currently, there is an on-going 4-year program, covering the period, 2011 - 2015, to eradicate obstetric fistula from Burkina Faso [
Ethics approval for this study was obtained from the ethics committee of the School of Health and Related Research (ScHARR), University of Sheffield, in the UK, as well as Ecole des Hautes Études en Santé Publique (EHESP), in France, and the National Centre for Scientific and Technological Research (CNRST), and the National Health Research Ethics Committee (CERS), in Burkina Faso. Participants’ written informed consent was obtained using an informed consent form, which had been reviewed and approved by all four ethics committees. For those participants who could not read and write, the enumerators read out the research information sheet and the informed consent form, and their thumbprint was taken as proof of consent.
This is a descriptive cross-sectional study, which assesses and compares knowledge of young women (aged 18-20 years) on obstetric fistula in urban and rural areas of Boromo health district. The survey was conducted between May 15, 2013 and May 19, 2013. We employed a cluster sampling method, which included two groups: the only urban area and 3 out of the 9 rural areas in Boromo. The sample size was estimated using STATA SE 12.1, targeting a power of 80% and an α error of 0.05 for the comparison of population level knowledge in both groups (rural Boromo and urban Boromo). Ideally, the knowledge should be 100% widespread, but in this study, we made an assumption that 80% knowledge prevalence was "sufficient" to effectively propagate any piece of information that needs to be disseminated. The sample size was calculated with the intention of being able to describe a 20% significant difference in knowledge, that is, an observed prevalence of 60% or lower. These two proportions were used to generate the sample size, with a design effect (Deff) of 2. The assumption made to estimate "ρ" clusters was similar within each sample, to bring the value to "1". The computation under these assumptions prescribed a sample size of 24 for each group (urban and rural), which adds to 48 for the whole study. Subsequently, factoring in the design effect led to an estimation of a sample size of 96. The method used in populating the groups had been utilized previously in studies of prevalence of meningococcal carriage in rural and urban communities of Bobo-Dioulasso, Burkina Faso [22). The multi-stage sampling entailed that streets or departments would be selected purposely while compounds or villages would be identified by systematic random sampling method. This sampling technique was preferred as it was found to be more feasible and as it allowed enumerators to work more efficiently [
For urban Boromo,
First stage: In the urban area, 3 streets were selected randomly beginning with a randomly identified starting point. This was done after reviewing all streets on the cadastral plans.
Second stage: Identification of compounds to be included in the survey was done by spinning a pen on a notebook and progressing in the direction indicated by the tip and entering every second compound that was encountered during the walk. If no eligible young woman, who met the inclusion criteria, was found within the compound, the investigators continued to the next compound. In each compound, the investigators included only one person, who was in turn identified randomly.
For rural Boromo,
First stage: To start with, 3 of the 9 administrative departments were identified randomly.
Second stage: From the principal departments, 50 hamlets were identified through systematic random sampling. The villages to be included in the survey were identified by spinning a pen on a notebook and moving in the direction indicated by the tip of the pen, entering every second village encountered. One eligible woman was included per hamlets. A hamlet in this setting typically represents small settlements of around 7-10 households’.
In each compound (urban Boromo) and hamlet (rural Boromo), a resident young woman of Burkinabe origin, who is between 18 and 20 years, who had not suffered or is currently not suffering from obstetric fistula, was randomly selected and recruited into the study. The eligible age range was set between 18 and 20 years following the rationale that assessing the prevalence of knowledge in a young group would have important ramifications, as they constitute the most affected age group and as they present with opportunities for early intervention with prevention messages. However, we limited the lower age bracket to 18 years as going below that would entail including children in this sensitive subject area, which would in turn lead to quite strenuous and time consuming processes in terms of obtaining ethics approval from the ethics approval boards of the different institutions involved in the research, as well as the host country.
Female interviewers administered the designed semi-structured questionnaires to assess the knowledge of participants on obstetric fistula. The analysis compared knowledge in rural and urban areas of Boromo. A pretesting of the questionnaire was conducted prior to the main study using 5 volunteers. These volunteers were excluded from the main study. Following the pilot study, the necessary modifications were made to the instrument to simplify the language and facilitate comprehension on the part of participants. The questionnaire was written in French, and the most common local languages Mossi and Dioula. All versions were tested as part of the pre testing. Female research assistants who were trained on the administration of the instrument and who had experience in previous researches were involved in administering the questionnaire.
For the construction of the questionnaire, we consulted previous attempts at assessing knowledge in the area [
The third section of the questionnaire was used to assess participants' knowledge. "Knowledge" in this study was defined as "the knowledge of risk factors and symptoms of obstetric fistula, normal duration of labour and possible sources of emergency obstetric care” (
- Early pregnancy | 0.125 |
- Home delivery | 0.125 |
- Female Genital Mutilation | 0.125 |
- Evil spirits | 0.125 |
- Breach of a prohibited act | 0.125 |
- Prolonged labour | 0.125 |
- Bad luck | 0.125 |
- Malnutrition of the mother | 0.125 |
- Urinary incontinence | 0.2 |
- Faecal incontinence | 0.2 |
- Continuous sleeping | 0.2 |
- Stomach ache | 0.2 |
- Vulva irritation | 0.2 |
|
We used standard statistical methods for the analyses, including accounting for design effect. Participant characteristics were compared between the rural and urban group using Fisher’s exact test for categorical and Wilcoxon ranksum test for continuous variables. Odds ratios for the association between rural residency and awareness or knowledge about obstetrical fistula were calculated using logistic regression. To evaluate whether characteristics differing between rural and urban participants (educational status, previous pregnancy and marital status) could explain or had masked an association between rural residency and the outcomes, we calculated adjusted odds ratios. To evaluate whether educational status or previous pregnancy influenced the association between rural residency and outcomes, we also calculated stratified odds ratios. All analyses were performed using STATA ™ SE 12.1.
In total, 126 contacts were made, whereby 5 declined to participate, making the total number of respondents to be 121. Of the 5 who declined to participate, 3 were from the urban group and 2 from the rural group. The median age of participants was 19 years old. Around half of participants (50.4%) in the total sample had not received school education, 45.5% were married and 45.5% had been pregnant in the past (regardless of outcome) (
Age | Mean (SD) | 19.0 (0.9) | 19.1 (0.9) | 18.9 (0.9) | 0.352 |
Level of education | Non-educated | 61 (50.4%) | 36 (62.0%) | 25 (39.7%) | 0.004 |
Primary | 21 (17.4%) | 12 (20.7%) | 9 (14.3%) | ||
Secondary | 39 (32.2%) | 10 (17.2%) | 29 (46.0%) | ||
Marital status | Married | 55 (45.5%) | 38 (65.5%) | 17 (27.0%) | 0.000 |
Single | 66 (54.5%) | 20 (34.5%) | 46 (73.0%) | ||
Previous pregnancy | Yes | 55 (45.5%) | 34 (58.6%) | 21 (33.3%) | 0.006 |
Age at first pregnancy (years) | Median (min, max) | 18.0 (15, 20) | 17.5 (15, 20) | 18.0 (15, 20) | 0.894 |
Number of pregnancies | Median (min, max) | 1 (1, 3) | 1 (1, 3) | (1, 3) | 0.479 |
Women aged 18 to 20 years in Boromo district, Burkina Faso, 2013.
* Fisher exact or Wilcoxon ranksum test
Only a third of participants were aware of obstetric fistula, with marginal difference between rural (37.9%) and urban residents (34.9%), which was not biased by educational status, experience of pregnancy [
Awareness of obstetric fistula | Yes | 44 (36.4%) | 22 (37.9%) | 22 (34.9%) | 1.14 (0.47, 2.76) | 1.69 (0.61, 4.70) | 1.06 (0.43, 2.62) |
Source of awareness | Word of mouth | 18 (40.9) | 15 (68.2) | 3 (13.6) | |||
School | 6 (13.6) | 2 (9.1) | 4 (18.2) | ||||
Media | 20 (45.5) | 5 (22.7) | 15 (68.2) | 0.14 (0.04, 0.46) * | 0.12 (0.02, 0.81) | 0.13 (0.03, 0.61) | |
Knowledge of obstetric fistula | Informed | 44 (36.4%) | 14 (24.1%) | 30 (47.6%) | 0.35 (0.16, 0.79) | 0.41 (0.18, 0.93) | 0.27 (0.09, 0.79) |
Awareness of risk of complications during delivery | Yes | 13 (10.7%) | 10 (17.2%) | 3 (4.8%) | 4.17 (1.63, 10.66) | 3.95 (1.41, 11.07) | 4.11 (1.18, 14.31) |
Knowledge of transport means in case of emergency | Ambulance | 19 (15.7%) | 7 (12.1%) | 12 (19.0%) | 0.58 (0.24, 1.44) ** | 0.74 (0.31, 1.72) | 0.87 (0.32, 1.41) |
Motorcycle | 87 (71.9%) | 41 (71.7%) | 46 (73.0%) | ||||
Donkey | 1 (0.82%) | 1 (1.7%) | 0 (0.0%) | ||||
Foot | 14 (11.6%) | 9 (15.5%) | 5 (7.9%) |
Women aged 18 to 20 years in Boromo district, Burkina Faso, 2013.
Figures present N (%); OR, odds ratio
* OR for having had information on obstetric fistula by media, vs. word of mouth or school
** OR for mentioning ambulance as the means of transport in case of emergency, vs. other means.
Regarding sources of information, most of the women who were aware of obstetric fistula got the information through the media (45.5%) or attributed the information to family and friends (word of mouth, (41.0%), with predominance of word of mouth among rural women (68.2%). Compared to urban women, and irrespectively of education status, previous pregnancy or marital status, rural women were about 8 times less likely to have received information through the media.
More participants from urban Boromo were classified as ‘sufficiently informed’ on obstetric fistula (47.6% vs. 24.1%)[
In stratified analyses using the adjusted model (
|
||
---|---|---|
Overall (N=121) | 0.35 (0.16, 0.79) | 0.31 (0.10, 0.92) |
Non educated (N=61) | 0.26 (0.11, 0.64) | 0.17 (0.05, 0.55) |
Primary education (N=21) | 0.67 (0.15, 2.89) | 0.69 (0.20, 2.43) |
Secondary education (N=39) | 0.62 (0.08, 4.94) | 0.57 (0.05, 5.96) |
Previous pregnancy (N=55) | 0.18 (0.05, 0.67) | 0.20 (0.05, 0.71) |
No previous pregnancy (N=66) | 0.43 (0.11, 1.68) | 0.51 (0.13, 2.06) |
Women aged 18 to 20 years in Boromo district, Burkina Faso, 2013.
All 121 participants identified the hospital as a site where they could access emergency obstetric care, but only 11.0% of women thought they could have pregnancy complications that would require emergency treatment (
Most women (71.7%) identified the motorcycle and only 15.7% the ambulance as the means for transport in case of emergency, with no significant association with rural residency (
The results of this survey among young women in Burkina Faso suggest that the prevalence of awareness of obstetric fistula and knowledge on prevention are low in both urban and rural areas. From the results, women in rural areas were almost three times less likely to have preventive knowledge on obstetric fistula compared to urban women, and these differences could not be explained by differences in education, experience of previous pregnancy or marital status. A similar cross-sectional study on knowledge of women on antenatal care conducted in Alexandria, Egypt, also showed that urban women had a higher mean total score for antenatal care knowledge than their rural counterparts, with a statistically significant difference (11.23 +/- 2.91 and 6.59 +/- 4.14, respectively and Z = 9.73, P < 0.001) [
Higher level of education amongst urban young girls could explain their higher health literacy, as more young urban women speak and understand the French language, even though there are other factors that could contribute to this rural-urban literacy gap [
It is now widely accepted that keeping girls in schools, especially, ensuring that they complete at least primary education, contributes to women empowerment, curtails harmful traditional practices such as child marriage, promotes gender equality and reduces incidences of maternal morbidity and mortality, including obstetric fistula [
Those young women who were ‘aware’ of obstetric fistula in rural areas mostly attributed their awareness to their family and friends, that is, through word of mouth. In this regard, it is difficult to ascertain the authenticity and quality of the information they receive, in contrast to structured and targeted messages from health professionals, counsellors or through the media. A recent systematic review and meta-analysis demonstrated the positive effects of face-to-face tailored messages from health workers on health behaviours of participants [
There has been a rise in annual rate of institutional deliveries in Burkina Faso, with the most significant increment of 27.3% occurring a year after the introduction of the subsidy for emergency care in 2007 [
Almost half of the females were already married, which is very similar to figures cited by UNFPA [
While it was evident that the motorcycle was the most frequently used mode of transportation in the district, as it was relatively cheaper than a vehicle and faster than a donkey. The concern was for the women who believed an ambulance was the best mode of transport in an emergency. Only the district hospital provides ambulance services [
One of the limitations of our study includes the relatively small sample size. Also, due to cluster sampling, participants in our study may be more homogeneous than the population as a whole [
Although this research has focused on the prevention dimension of obstetric fistula programs, we note the importance of the other two components that make up an effective fistula response, that is, treatment and reinsertion (social and economic). Slinger [
It is also crucial to strengthen research efforts that focus on identifying information gaps and that explore the most effective methods for disseminating information for behaviour change in health. This research has provided essential information for framing policies and designing programs to prevent obstetric fistula. Strategies must be adapted to local settings, whether urban or rural, in order to be more effective. Efforts geared towards tackling obstetric fistula should be made explicit and strengthened within the context of wider developmental objectives and global targets such as the millennium development goals whereby multi-sectoral approaches (including education, health and gender) are deployed, while harnessing the respective contributions of all stakeholders. Indeed, the eradication of obstetric fistula constitutes a human rights issue; a societal ill that has got inequality written all over it. It embodies gender inequality as well as inequality within the same feminine gender as women in rural communities are more disadvantaged compared to those in urban areas. Indeed, “in an unequal world, these women are the most unequal among unequals” [
Direction des Relations Internationales, Ecole des Hautes Études en Santé Publique for providing travel support to the research site.
Barro Emmanuel, Bicata Félicité, Konaté-Isako Insita, Ouedraogo Catherine and Kouanda Rihamata, who worked as field workers in this study.