Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Dr. Rose Ndumia

Second Supervisor/Advisor

Prof. Stanley Luchters


Imaging and Diagnostic Radiology (East Africa)


Background: The Breast Imaging Reporting and Data System 3 (BI-RADS 3) category is assigned to lesions which are deemed to be probably benign on breast imaging with a low likelihood of malignancy (≤2%). With regular imaging follow-up, this category is intended to reduce the rate of invasive biopsies while enabling early detection of breast cancer. However, imaging follow-up has been reported to be challenging to both radiologists and referring clinicians in terms of its appropriate use on the various imaging modalities and understanding management of the patients in this category. Furthermore, it has been reported to induce patient anxiety and have variable levels of compliance in different studies with various factors postulated or shown to have an association with compliance. A local study in Kenya, on the level of compliance as well as assessment and exploration of the factors associated with compliance could increase our understanding of the utility of this BI-RADS category in our setting.

Objectives: The primary objective was to determine the proportion of women, diagnosed with BI-RADS 3 on breast imaging, who were compliant to follow-up recommendations at 6, 12 and 24 months following diagnosis. Secondary objectives of the study were to assess and explore the factors associated with compliance and to determine the outcomes of the BI-RADS 3 lesions during the study period.

Methodology: A mixed methods study was conducted with the quantitative cross-sectional study using short telephone structured questionnaires, and the qualitative component using in-depth interviews by telephone. The study participants were patients identified from the Aga Khan University Hospital, Nairobi (AKUHN) Radiology Department’s PACS system to have a BI-RADS 3 lesion during the period of January 2016 to December 2018. Data collected from the quantitative component was anonymized, compiled and made available for analysis using R statistical software (version 3.6.0). For the qualitative component, a thematic framework was constructed and applied manually to the data to enable construction of themes and sub themes as per the preset research objectives.

Results: Out of the 168 participants enrolled in the quantitative study, almost two thirds (n=107; 64%) were compliant of whom 60(60/107, 63%) were followed up with imaging alone and 47(47/107; 37%) had tissue diagnosis. Among the patients who were compliant with their v follow up imaging, 38% (23/60) had the imaging done at a facility other than AKUHN, while 51% (24/47) of the patients who had tissue diagnosis through either biopsy or surgery had it done at other facilities. In-depth exploration in the qualitative inquiries elicited perspectives on enablers of compliance such as the quality of care provided at AKUHN, diagnosis of breast cancer in family members/friends, having a medical cover and presence of family support as well as self-efficacy and patients’ concern for their own health. Barriers to follow up mentioned included cost of imaging, desire to seek a second opinion, distance and access to the hospital, resolution of symptoms and patients’ perception of the BI- RADS 3 report or their understanding of what was communicated by their referring clinician. Quantification confirmed the main reason reported for non-compliance was, “the doctor told me it was nothing to worry about.” Factors independently associated with good compliance included being divorced/widowed/separated (Adjusted Odds Ratio (AOR) =6.54, 95%CI: 1.36-49.93) and whether the participant reported being made aware of the need for follow (AOR=6.29, 95%CI: 2.53-17.15). Finally, the malignancy rate in this sample of patients was 0.6% (1/168) which was within the expected ≤2% for BI-RADS 3 lesions.

Conclusion: The findings of this study demonstrated that approximately two-thirds of patients diagnosed with BI-RADS 3 diagnosis are compliant with follow-up recommendations, which is comparable to some prior studies with the malignancy rate within the expected range. As identified in the qualitative component, and subsequently confirmed quantitatively, the most common and independently associated correlate for non-compliance was the referring clinician informing the patient that there was nothing to worry about. Non-compliance was also noted to be more likely among married women when controlled for potential confounders. Patients being made aware of the need of follow up may improve patient compliance, especially among married women. Additionally, increased awareness to clinicians of the BI-RADS 3 category so that they may emphasize on the need for follow up may also increase compliance in this category.

Included in

Radiology Commons