The effect of knowledge on uptake of breast cancer prevention modalities among women in Kyadondo County, Uganda

A household survey of women in Kyadondo County was conducted during the months of June, 2014 to August, 2015. Kyadondo County is a part of Kampala city with a proportion of the urban, peri-urban and rural set up [11]. Kyadondo County was chosen for the study because it is the area from which the Kampala cancer registry captures comprehensive data.

Cluster sampling method was used. Kyadondo County was stratified into 20 parishes [clusters] as the primary stratification variable. At this stage, each parish was further clustered into villages. Each parish comprises 10–20 villages and in each village 50–100 households and the 20 parishes were used as the large clusters from which 10 villages were selected. At the household level, 40 households from each village and one respondent each household was selected randomly giving 400 respondents. Each household had 1–4 women who were aged 18 years of age and above. To minimize bias, 1 woman aged 18 years and above was selected from each household. The age of 18 years was chosen because it is the age of consent and that the person has insight to develop attitudes based on knowledge that may influence her practices.

Inclusion criteria: Women 18 years and above who were able to understand and communicate were included in the study. Only subjects of a sound mind, who were able to hear and speak English and Luganda (a local language used in the area).

Exclusion criteria: participants unable to understand and communicate in English or Luganda or those who declined to consent.

Data was captured using a survey questionnaire that was developed from the existing literature on the topic. This was composed in English and then translated to vernacular (Luganda) after being piloted by research assistants. The questionnaire was pre-tested by research assistants among thirty women in Kisugu, Nakawa division to ensure that the most accurate information was obtained and where necessary changes were made. To ensure internal validity, only completed questionnaires were adopted. The questionnaire was interviewer administered.

Study variables: Dependent variable was breast cancer prevention modalities (adopted SBE, clinic breast examination and ultra sound) and the independent variables were background characteristics, knowledge based factors and lifestyle factors.

Knowledge of breast cancer was assessed based on the questions that sought to assess women awareness of breast cancer, knew someone that had breast cancer, risk factors of breast cancer, what to do in order to reduce the risk of acquiring breast cancer, what to do to discover breast cancer early, how to check their own breasts for lumps, and those able to recognize breast cancer signs like discharge from the breast nipple, swelling of the breast among others. The quantification was done as a proportion of the total sample size.

Data was entered and analyzed using SPSS IBM version 20. A completely edited dataset was adopted and double checked for errors and corrections done. We considered breast cancer prevention modalities (adopted SBE, CBE and ultra sound) as the dependent variable and the independent variables were background characteristics, knowledge based factors and lifestyle factors. The data was analyzed at univariate, bivariate and multivariate. Variable at univariate and bivariate with a p value of less than 0.05 were considered for multivariate analysis and p values less than 0.05 were considered statistically significant. Only extract of the bivariate and multivariate were used for this article.

The study was approved by research ethics committee of International Health Sciences University. All respondents provided written informed consent after receiving detailed description of the study. Eligible participants were consented in privacy and no incentives were given. Anonymity of the respondents was ensured at all stages of data analysis.

Only 414 subjects were adopted in the study, (that is 85.5% of those who had consent and had complete questionnaires). The results show that 55.3% were below 30 years, 35.5% were 31–50 years and only 9.2% were above 50 years. The religious affiliation revealed that 25.5% were Muslims, 15.5% were Seventh Day Adventists and atheists, 29.5% were Anglican and Roman Catholic respectively. 74.4% of these women were urban dwellers, 20.8% were in semi urban areas and only 4.8% were in rural areas of Kyadondo County. However, the study findings revealed that, 36.2% compared to 63.8% had adopted breast cancer prevention modalities in Kyadondo Count. Among the background factors that were found to be significantly associated with adoption of breast cancer prevention modalities among women in Kyadondo county included: age (p-value = 0.049), religious affiliation (p-value = 0.054), level of education (p-value = 0.000) and place of residence (p-value = 0.001). Others background characteristics of the women in Kyadondo County such as distance to the health facility, type of health facility and source of breast cancer information were not significantly associated with the adoption of breast cancer prevention modalities. Only the results that were found to be significantly associated with the adoption of breast cancer prevention modalities are documented in Table 1.

The study findings indicate that 91.8% of the women were aware of breast cancer, 44.2% knew someone that had breast cancer, only 30.4% knew what increases the chances of breast cancer, 30.4% of the women knew what to do in order to reduce the risk of acquiring breast cancer, more than half 57.5% knew what to do to discover breast cancer early (that is by going for regular check-ups), 35.2% knew how to check their own breasts for lumps, 32.9% of the women had been examined by a health worker and 2.4% had done an ultra sound or had a mammography done prior to the study. In addition, more than half (56.0%) were able to recognize breast cancer signs like discharge from the breast nipple, swelling of the breast among others. The study revealed that 65.7% of the women knew where to obtain breast cancer treatment services in Kampala.

Table 2 indicates the knowledge based factors associated with the uptake of breast cancer prevention modalities among women in Kyadondo County. These include: awareness of breast cancer (p-value = 0.001), knowing someone who had suffered from breast cancer (p-value =0.000), obtaining information through health talk shows on radio/television (TV) programs (p = 0.004), obtaining information from the health facility (p = 0.018), knowing that as one grew older it increased the chances of breast cancer (p = 0.000), having a family member that had breast cancer (p = 0.003), being obese (p = 0.000), eating lots of fatty foods (p = 0.000), not giving birth before 35 years (0.005), knowing what would increase one’s chance of acquiring breast cancer (p-value = 0.000), knowing how to prevent breast cancer (p-value = 0.000), breast cancer preventions methods like doing exercises daily to control weight (p = 0.009), reducing weight if one is obese (p = 0.031), eating lots of fruits (p = 0.000), reducing alcohol consumption (p = 0.000) avoiding exposure to radiation (p = 0.001), knowing how to check for signs and symptoms of breast cancer (p-value = 0.000), knowing what to do (for instance a clinical breast exam) (p = 0.006), having a breast ultra sound done (p = 0.004), having a mammography done (p = 0.004), ability to tell the different signs of cancer (p-value = 0.000) which included whole breast swelling (p = 0.002), swelling in the arm pit (p = 0.000) and having an abnormal discharge from the nipple (p = 0.000), and knowing where to seek breast cancer treatment services (p = 0.000) and at the health care facility (p = 0.002).

In order to determine factors that jointly influenced uptake of breast cancer prevention modalities, a binary logistic regression model was fitted. The dependent variable was regressed with a set of independent variables including: socio-demographic characteristics and knowledge base factors. The dependent variable was dichotomous and grouped as 1-adopted and 2-did not adopt. The results in Table 3 indicate that subjects who had primary level of education were 5 times and those with secondary education were 3 times more likely to adopt breast cancer prevention modalities respectively. The respondents who were aware of breast cancer were 1.5 times more likely to adopt breast cancer prevention modalities than their counterparts. The study findings revealed that the following increased the chances of acquiring breast cancer: ageing (OR 2.2, CI .53–9.59), having someone in the family with a history of cancer (OR 1.20, CI .31–5.07), eating fatty foods (OR 0.25, CI .06–1.09), drinking alcohol (OR 1.1, CI .33–4.08). It was also established that knowing what to do reduced the risk of breast cancer (OR 2.16, CI .92–5.07). When asked to identify what actually reduces risk of acquiring breast cancer, the women in Kyadondo identified the following: eating plenty of fruits (OR 0.10, CI .01–.85), reducing alcoholic consumption (OR 0.19, CI .05–.78) and avoiding exposure to radiation (OR 0.47, CI .05–4.39). The respondents who knew what to do to reduce their chances of acquiring breast cancer were 2.15 times more likely to adopt breast cancer prevention modalities.

A woman’s knowledge on the signs of breast cancer was analyzed in relation to adoption to breast cancer prevention modalities. The breast cancer signs identified include: small swelling on the breast (OR 0.35, 95% CI .15–.82), swollen breast (OR 1.39, 95% CI 48–4.01) and abnormal nipple discharge (OR 0.27, 95% CI .07–1.01), However it was established that the women in Kyadondo who were able to identify the signs of breast cancer were 0.61 times less likely to adopt breast cancer prevention modalities than their counterparts. When asked what can be done to discover breast cancer early: the women were able to identify the following: self-breast examination (OR 0.24, 95% CI .07–.90), clinical breast examination (OR 0.70, 95% CI .22–2.18), doing a biopsy (OR 0.62, 95% CI .17–2.26), and those who did not know what to do (OR 0.73, 95% CI .21–2.48). Finally, our respondents indicated that breast cancer treatment facilities were likely to adopt prevention modalities (OR 1.92, 95% CI .74–4.99). The results revealed that women who acknowledged that there was breast cancer treatment at the health facility were 0.73 times less likely to adopt breast cancer prevention modalities than their counterparts.