Awareness, Attitudes and Practices of women related to breast cancer in Niamey and Zinder

Background: In Niger, breast cancer is the rst cancer related morbidity and mortality within the female population. While Breast cancer awareness can contribute to early diagnosis and disease mortality reduction, Niger women’s knowledge of breast cancer is not well documented. The objective of this study was to assess the knowledge, attitudes and practices of women related to breast cancer and have a look on the factors associated this knowledge. Methods: The study was conducted thought a cross sectional survey in women population in Zinder and Niamey regions. A random sampling was used to select women in households within health districts. We dened a breast cancer knowledge score and used a generalized linear model to assess factors associated with breast cancer knowledge. Results: A total of 675 women were included in the survey. Average age of women was 39.2 years (38.2-40.2) and 48.4% (44.7-52.2) of them were not educated. Overall women’s knowledge of breast cancer is relatively low, only 41.2% (37.5-45.0) were aware of breast cancer. An area where women demonstrated an awareness was breast cancer symptoms knowledge with 65.0% (61.3-68.7). Insucient level of knowledge was observed on knowledge of risk and protection factors with only 27.4 % (24.0-31.0) level of awareness and breast cancer good practice with 16.9% (14.2-19.7). Younger age OR=0.98 (0.96-0.99) practicing clinical breast examination OR=2.25 (1.31–3.16), breastfeeding 3.34 (2.12-5.26), not having a history of participation to breast cancer 0.53 (0.34-0.82) and living in rural and peri-urban areas 0.34 (0.20–0.44) were factors related to awareness of breast cancer. Conclusion: Niger women had overall low level of breast cancer awareness. While they were not acquainted with breast self-examination and clinical breast examination, their knowledge of breast cancer symptoms was acceptable. Clinical breast examination, area of residence, age, breastfeeding and history of participation to breast cancer screening were found to be associated with breast cancer knowledge. Awareness programs can promote


Background
Breast cancer remains a public health concern worldwide including sub-Saharan African countries (Brinton et al.;Ferlay et al.;2008). In Niger, breast cancer is identi ed as the rst cause of cancer related morbidity and mortality standing for twenty-eight percent (27.7%) of the overall cancer mortality within the female population, (WHO; Cancer Country Pro les 2014; Globocan; 2012, H. M. Zaki et al.;2013, S Mamoudou et al.;. Studies found that if women adopt early risk-reduction behaviours like physical activity, healthy diets, breastfeeding, non-extensive alcoholic beverage consumption to mention but a few, part of the breast cancer burden can be prevented through primary prevention 2014, Graham and Kari;2015). This is even more important in countries like Niger where mammographic screening facilities are not widely available and research literature on breast cancer epidemiology and awareness is not extensive. Soliman, A. S., et al.; 2015 outlined younger age and late diagnosis are common features of breast cancer in Niger.
Knowledge of breast cancer among women is not generally high in sub Saharan Africa and not very well documented in Niger (Jennifer N. al.;2017). Breast cancer awareness can signi cantly contribute to early diagnosis and disease mortality reduction (Anderson B & Jakesz R 2008).
To the best of our knowledge, no recent studies were conducted in the analysis of knowledge, attitudes and practices of women related to breast cancer in Niger. The country does not yet have an operational cancer policy/strategy/action plan (WHO, Cancer Country Pro les, 2014). In this paper, we used and adapted Breast Cancer Awareness Measure, (BCAM) to asses knowledge and attitudes of women in Niamey and Zinder.

2.1: Study design, sampling and data collection:
The study targeted women aged from 25 and above in Niamey and Zinder regions to participate to the survey. The sampling frame was composed by a list of villages of the health districts and number of households in the two regions with a similar population of women at this age. A two-stage clustered random sampling was used to first select villages in health districts, then households within villages using random walk technique. Finally, women were randomly selected in these households as primary sampling units.
The required sample size is calculated based on a 95% confidence interval, assuming that 50% of women know basics of breast cancer with a correction of design effect for cluster.
We assume an average number of 7 persons per household (national institute of statistics reference from the last 2012 population Census). This gives us about 630 women we adjusted with a 5% increase as a security margin to account for non-response. The collected data was analyzed using R version 3.4.0 Epidata analysis v2.2.2.183 and IBM SPSS Statistics 24. All confidence intervals were computed with 5% error margin. The questionnaires were conducted using face to face interviews with women randomly selected in their households.
The aim of the study was to assess the knowledge, attitudes and practices of women related to breast cancer and have a look on the factors associated this knowledge. A Weighting of the knowledge factors was done to substantiate and allow some leverage for important risk factors generally accepted (WHO) but also accounting for context specific factors. Globally, correct answers were given points (see table 1 below) and a wrong answer was given zero. Mammography is not that much common in Niger context (as of January 2018, the data collection period, only one mammography machine is available for the whole country and it was located in Niamey). It was accessible only to few rich women or women referred under specific conditions. In Niger, screening is done using clinical breast examination. Hormone replacement therapy is not also common in Niger. As showed in the descriptive results, only 0.4% (0.2-1.3) of Niger Women smoke and 0.7% (0.3-1.7) of them drink alcohol, so even if these are known established risk factors, they don't seem to be very relevant in Niger context, so we did not weigh them. Breastfeeding, maternity, physical activity, breast self-examination, clinical breast examination and age related factors were weighted to account for their relative importance. Some important potential breast cancer symptoms were also weighted (see table below).
In general, women should have half of the overall knowledge score (36) to be labeled aware of breast cancer.

3.1: Description of the sample:
The questionnaire was administered by 675 women in the two regions historically hosting national reference hospitals, Niamey and Zinder. Niamey is the capital city and Zinder the most populated region in Niger. Households were selected within health districts in 'commune 2", "commune 3" and "commune 4" in Niamey region and Goure, Miriah and Zinder in Zinder region (see sampling frame in list of annexes). do not practice contraception at all. Finally, 28.3 % (24.9-31.9) had an abortion history. simply agree which is a total of only 29.7% of women that recognizes these risk factors.

Overall knowledge of breast cancer
Only 27.3% agree that physical inactivity and obesity could be a risk factor for breast cancer. We can see from  Only 36.2% of the interviewed women are convinced that long term survival (longer than 5 years) is common when breast cancer is treated in an early stage.
3.4 Women's knowledge of breast cancer warning signs of women can be classified as being aware of breast cancer good practice.
Factors associated with knowledge of breast cancer: Table 7 shows the results of modeling of odds of being breast cancer aware given socio demographic and specific predictors. We fitted both OLR and GEE models. within health center clustering, we found the following results. Niger women that consult in the same health center are more likely to see the same health professionals and receive the same awareness programs. Clinical breast examination, age and zone were found to be associated with breast cancer knowledge. The odds of a woman that practices clinical breast examination is 2.25 higher than of one who is not. Urban women were less likely to be aware of breast cancer as compared to rural women. Younger women were more likely to be aware of breast cancer. We observed an association between knowledge of breast cancer good practice and number of children, women with more children were more likely to be aware of breast cancer good practice.

Discussion
Although the knowledge score was constructed somehow like a composite multidimensional score, overall women's knowledge of breast cancer in Niger was found to be low. Dimensions where women showed speci c low levels are knowledge of breast cancer risk/protective factors and knowledge of ''good practice'' with respectively 27.4% (24.0-31.0) and 16.9% (14.2-19.7). In fact, only 23.6% agree that smoking is a risk factor and 37.8% see alcohol as a breast cancer risk factor. These proportions could be explained by the insigni cant proportion of women who smoke or drink in the sample survey which is representative of the Niger community in general due to religious and personal beliefs. Only 39.3% and 42.2% respectively see physical inactivity and nulliparity as a breast cancer risk factor.
These ndings correspond with those of Azubuike SO; 2017 who recently found a generally poor knowledge of breast cancer among women in Benin city, Nigeria, whose knowledge concerning breast cancer risk factors was also worse than concerning signs and symptoms. Okobia MN et al.;2006  We found women living in urban and peri urban areas to be less likely to be aware of breast cancer compared to those living in rural areas. Niamey being the capital city, almost all women consult a doctor in an urban health center though these are administratively conceived to be rural contrasting the recent urbanization phenomena. Ann Muthoni and Ann Neville Miller; 2010 revealed a difference between rural and urban Kenyan women's knowledge and attitudes regarding breast cancer and breast cancer early detection measures in a qualitative study. In fact, urban middle-income women were more informed about breast cancer risk factors and early detection measures as compared to other groups including urban low income and rural low-income women. Women with history of participation to breast cancer screening are half likely to be aware of breast cancer compared to those who don't participate, this is surprising, may be women who know less on breast cancer are the ones more motivated to participate to screening.
Educational level was not found to be associated with breast cancer awareness. It should be noted that in the general women population, educational level in Niger is not that much heterogenous, in our sample only 5.5% have higher education. Lydia A et al; 2017 could not nd an association nor between educational level neither between age on the one hand and breast cancer awareness on the other hand in a study in Limpopo, South Africa. Nevertheless, some studies established an association between breast cancer awareness and educational level in sub Saharan African context in Nigeria (Okobia et al;2006) and in Ghana (Y. Opoku et al;. We found age to be associated with breast cancer awareness with younger women having more awareness than older ones, in line with ndings by Allam and Abd Elaziz 2012. It should be noted that in our survey, only women from 25 years and older were selected to participate to the survey. Limitations: The limitation of this study could nd its root in the fact that it was conducted in the two regions of Zinder and Niamey, which historically host national reference hospitals. The study could therefore contain bias in terms of reproducibility in the whole country. Results need to be interpreted within the regions framework. Lack of wide access to equipment of mammography could introduce a bias in the proportion of women reporting having undergone a mammography, as during the period of the survey (January 2018), only one mammography machine was available in the whole country which is located in the capital city.
Policy recommendations: Focus breast cancer awareness programs on knowledge of breast cancer risk/protective factors and breast cancer good practice as these are areas of which women know the least.
Support the implementation of cancer national policy in the country with support of WHO Awareness programs should target different age classes but focus on older women Design a strategy to shift from a clincial breast to a mammography screening program in Niger Provide and decentralise mammography units and build mutilateral partnerships to support women to have nancial access to breast cancer screening.
Design a stategy to encourage women to participate in breast cancer screening

Conclusion
Niger women were found to have a relatively low level of breast cancer awareness. Knowledge of risk/protective factors and breast cancer good practice were main factors downscaling this knowledge. Meanwhile women's knowledge of breast cancer symptoms appeared to be acceptable. Clinical breast examination, age, area of residence, breastfeeding and history of participation in breast cancer screening were found to be associated with breast cancer knowledge using OLR. Using GEE, only clinical breast examination, age and area of residence were found to be associated with overall breast cancer knowledge.
This study reported an overview of Niger's women knowledge of breast cancer and can contribute in guiding awareness program for screening/early detection. Awareness programs in Niger should promote clinical breast examination as mammography is not common and target older women both in rural and urban areas with focus on breast cancer risk/protection factors. Participation to breast cancer screening should be encouraged.  -12-2017). The results are expected to contribute in strengthening the disease knowledge and scienti c literature in the country. Each respondent was asked to decide whether she wants to participate to the survey after an explanation and oral informed consent was obtained before proceeding.

Consent for publication : Not applicable
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests Funding This study was funded by myself as PhD student while working for Medecins Sans Frontieres, Doctors without Borders as country director for the NGO in Democratic Republic of Congo. I funded the data