Breast health awareness campaign and screening pilot in a Kenyan County: Findings and lessons

Abstract Background and Aim Breast cancer is the leading cancer in terms of incidence in Kenya. We conducted a breast cancer awareness and screening pilot to assess feasibility of rolling out a national screening program in Kenya. Methods Conducted in Nyeri County during October–November 2019, the pilot had three phases; awareness creation, screening (clinical breast examination and/or imaging) and final evaluation (post‐screening exit interviews and retrospective screening data review). Descriptive statistics on awareness, screening process and outputs were derived. Results During the pilot, 1813 CBE, 217 breast ultrasounds and 600 mammograms were performed. Mammography equipment utilization increased from 11% to 83%. Of 49 women with suspicious lesions on mammography, only 22 (44.9%) had been linked to care 4 months after the campaign. Of 532 exit interview respondents; 95% (505/532) were ≥35 years of age; 80% (426/532) had been reached by the awareness campaign. Majority (75% [399/532]) had received information from community health volunteers; 68% through social groups. Majority (79% [420/532]) felt the campaign had changed their behavior on breast health. Although 77% (407/532) had knowledge on self breast examination (SBE); only 13% practiced monthly SBE. More than half (58% [306/532]) had previously undertaken a CBE. Approximately 70% (375/528) were unaware of mammography before the pilot; 86% (459/532) had never previously undertaken a mammogram. Fifty‐five percent (293/532) of respondents had screening waiting times of >120 min. Conclusion Community health workers can create breast cancer screening demand sustainably. Adequate personnel and effective follow‐up are crucial before national roll‐out of a breast cancer screening program.


| BACKGROUND
Breast cancer is the most common cancer among women, affecting over two million women globally and resulting in over 600 000 deaths in 2018. 1 For effective breast cancer control programs, five key approaches have been described; integration of breast cancer into national cancer control strategic planning by policymakers, development of diagnosis and management guidelines, review of evidencebased practices by clinicians, identification of priority breast cancer control opportunities by advocates and implementation research training and mentorship. 2 Breast cancer is the leading cause of cancer morbidity in Kenya, constituting approximately 13% of all cancer cases; and the third leading cause of cancer deaths with approximately 2600 deaths in 2018. 3 Even in tertiary facilities, about a third of breast cancer cases are diagnosed in stage four, with metastases to bone, brain, lung or liver. 4 This is associated with high costs of treatment and low overall survival rates. In Kenya, breast cancer occurs earlier in women between ages 35 and 45 years which is 10-15 years earlier than the peak incidence in developed countries. 5 Knowledge on approaches for early detection of breast cancer is low, especially in the rural areas. 6 Kenya does not have a mass breast cancer screening program at the population level yet; screening is currently opportunistic and individual-based. The Kenya National Cancer Screening Guidelines 2018 identify breast cancer as one of the cancers planned for population-based screening. 7 The World Health Organization (WHO) recommends conduction of a pilot before launch of a cancer screening program, to guide implementation. 8,9 In 2016, through the Managed Equipment Service (MES) project, the Ministry of Health availed mammography equipment in all the 47 counties in Kenya. However, there were concerns about the low utilization of the equipment for breast cancer screening and early diagnosis. Therefore, the National Cancer Control Program conducted a breast cancer awareness and screening pilot, to assess the feasibility of utilizing mammography equipment available at county referral facilities to support a national, populationbased breast cancer screening program.

| METHODS
The breast health awareness and screening pilot was a 2 month interven-

| Campaign awareness approaches
To reach the target audience, multiple channels were used; including community leaders, community health volunteers (CHVs), mass media (radio and television), social media (Whatsapp and Facebook), advertising materials (leaflets, printed t-shirts, lesos and posters), health facility information activities and the campaign launch event itself, that brought together key stakeholders.

| Length of stay during screening visits
The average length of stay (LOS) for 36% of the clients was more than 180 min (3 h) while 27% and 19% of clients spent 61-120 min and 121-180 min, respectively ( Figure 2). Only 18% of the patients indicated that they had spent less than 60 min at the screening health facility. This is compared to a report from healthcare workers during the baseline survey conducted in April 2019, which estimated the LOS at 60 min.

| Mammography equipment utilization
Based on interviews at the facility, it was established that a maximum of 18 mammograms could be undertaken per day (three per hour for 6 h in a day). On average, two mammograms were performed per day before the pilot; therefore, equipment utilization was calculated as: During the pilot, an average of 16 mammograms was performed per day; therefore the equipment utilization was: Therefore, mammography equipment utilization increased from 11% before the pilot to 83% during the pilot period. However, this was not sustained after the pilot, since the average number of mammograms performed per day begun to rapidly decline post-campaign. In the week following the campaign's end, on average six mammograms were performed per day (EQ = 33.3%).
While in October and November 2018 diagnostic mammograms were more than screening mammograms, in 2019, screening mammograms were more than diagnostic mammograms ( Figure 3). 3.5 | Post-screening exit interview/survey

| Socio-demographic characteristics of survey respondents
The exit interview had 532 respondents. Majority of the respondents (76%) were between the ages 30-59 years age (Table 1). Majority (70%) were married and 76% had at least attained primary education and above. Over half (58%) of the respondents were farmers.

| Mammography
Around 71% were not aware or did not know what mammography is before the campaign. Majority of the respondents (87%) had never previously undertaken a mammogram. Majority (78%) of the mammograms undertaken pre-campaign by the respondents were diagnostic mammograms. Around 73% of the respondents reported that they had not undertaken a mammogram because they did not have a problem with their breast and therefore did not know that they needed one. Other reasons included that a doctor or other health worker had not ordered one (12%) and issues of cost (3%

| Screening process experience
Ninety-nine percent (527/532) indicated that they were satisfied with mammography services at the screening facility and would go back for a subsequent mammogram. Around 87% (463/532) of the respondents indicated they had received respectful care by the person performing their mammogram while 79% (420/532) termed the screening process as comfortable.
3.10 | Health system readiness evaluation The health system structures to support breast cancer screening were evaluated based on the WHO building blocks for health systems. The findings are presented in Table 2. Gaps were noted in human resources, health system capacity to support mammography-based breast cancer screening and an efficient health information system that can track clients through the entire continuum.

| Summary of findings
During the breast cancer awareness and screening pilot, the average number of mammograms performed increased and equipment utilization peaked; however, this was not sustained after the pilot period ended. Mammography screening for every woman invited for screening was found to overwhelm the screening facility and contributing to lengthy waiting times before undergoing screening and delayed reporting of results; a triage with CBE was deemed more feasible.
Limited availability of human resources for mammography screening was a major impediment to optimal equipment utilization and T A B L E 2 Health system readiness evaluation to support mammography based breast cancer screening, Nyeri County, Kenya, 2019

Health information systems
A large number of mammogram reports were uncollected 4 months after the pilot. Data capture was fragmented and findings not fully linked with information captured in the screening registers

Medical products and technologies
The two main supplies that were inadequate at the facility during the campaign period were mammography films and thermoluminescent dosimeters (TLDs) for the radiology staff. The shortages worsened as the number of mammograms performed increased during the pilot Human resources for health Before the pilot, the facility had one radiologist and two radiographers competent enough to perform mammograms. During the pilot, one radiologist and two radiographers were added; however, the workload was still more than this team could handle efficiently Service delivery Awareness campaign created immediate demand. However, since mammography is available only at the county referral facility, women had to endure significant distances and long-waiting times to access screening. The approach had to be adapted during the pilot to use CBE as a triaging for the women prioritized for mammography Health financing Most of the mammograms performed were free, since Nyeri county was also piloting UHC. However, after the end of the UHC pilot, this has not been sustained

Leadership and governance
The pilot involved ensuring existence of policy frameworks combined with effective oversight, coalition building, regulation, attention to system design and accountability. A close-out forum with all the implementing agencies provided a model for stakeholders to adopt in the future planning, implementation and review of campaigns. Screening was guided by the National Cancer Screening Guidelines reporting during this pilot. Approximately 45% of women with suspicious lesions for breast cancer and 40% of those with suspected benign conditions were lost to follow-up before diagnostic work-up could be done. Community strategy utilizing community health workers was the most effective awareness creation and community mobilization approach; this can also be utilized to track screened women and reduce loss to follow.

| Screening process and client experience
A co-test strategy of CBE and mammography (initial use of CBE and risk stratification before mammography) had to be adopted due to overwhelming demand following the awareness campaign. Majority of the clients had screening turn-around times exceeding 1 h for the screening mammography procedure. This was longer than what was reported by healthcare personnel before the pilot, despite the fact that more personnel were deployed to meet the expected increase in demand due to the awareness campaign. There was also an additional period of at least 2 weeks for reporting of the mammograms. This could lead to loss to follow-up and ineffective linkage to further evaluation and/or management. Time factor alone is a key factor in breast cancer screening uptake and adherence to follow-up. 10

| Strengths and limitations
A particular strength of the approach employed during this pilot was the combination of screening process data and feedback from screened clients. This would further inform the planned national rollout of breast cancer screening program. However, the pilot study had some limitations. First, the pilot intervention period was rather short. Therefore, it was not possible to evaluate the medium term effects of the awareness creation. Second, a component of formative research would have given more insights into the target population attributes that would impact breast cancer screening uptake.

| CONCLUSION
A community awareness and provision of information on breast cancer screening can create demand; however, the healthcare system needs to be well prepared to offer the screening and linkage to care to all women seeking screening services. All the pillars of healthcare systems strengthening must be improved to support an effective breast cancer screening program. A CBE-based screening, with linkage to imaging may be the most feasible approach as breast cancer screening is introduced at the population level in Kenya even as we focus on increasing staff training and availability to provide screening services.