A cross-sectional mixed-methods design combined quantitative and qualitative approaches to capture both measurable and experiential aspects of care. Conducted from January to May 2025 at HGY and HGOPY who are amongst the principal tertiary hospitals for adult and gynaecologic oncology [6][7]. HGY provides surgery, medical oncology, and radiotherapy; HGOPY offers multidisciplinary gynaeco-oncologic services. Ethical clearance was obtained from the FMBS-UYI Ethics Committee, and administrative authorization from both hospitals. Participation was voluntary and confidential.

Inclusion:

Women aged ≥ 18 years with a confirmed diagnosis of breast cancer, attending either hospital for treatment or follow-up, who provided informed consent and had received breast cancer care within the past five years.

Exclusion: Women who were critically ill, sedated, unable to participate in interviews, or who declined consent were excluded. Patients with severe psychiatric disorders impairing communication or comprehension were also excluded to ensure the reliability of self-reported responses. Women with unstable severe comorbidities requiring emergency management at the time of recruitment were not enrolled because their acute medical condition could substantially interfere with participation and perception assessment.

Sampling:

A mixed-methods sampling strategy was employed. For the quantitative survey, women aged ≥ 18 years with a confirmed diagnosis of breast cancer attending the study sites for treatment or follow-up were eligible. The minimum sample size was calculated using the single population proportion formula:

N = z² × p (1 − p)/d²

where z = 1.96 for a 95% confidence level, p = 0.50, and d = 0.05, yielding a minimum required sample size of 385. To account for non-response, 500 eligible women were selected using simple random sampling from hospital records. A total of 389 participants consented and completed the survey and were included in the analysis.

For the qualitative component, purposive sampling was used to select 40 participants from the survey respondents, ensuring variation by hospital, treatment stage, and sociodemographic characteristics. Four focus group discussions (FGDs) were conducted (two per hospital, 10 participants each), with sampling guided by the principle of thematic saturation.

Quantitative: Data were collected using a structured, interviewer-administered questionnaire adapted from the EORTC QLQ-C30/BR23 and FACT-B instruments. The questionnaire was pretested among 20 breast cancer patients prior to the main study. It comprised sections on sociodemographic characteristics, clinical information, perceived quality of care (including communication, timeliness, empathy, information, and continuity of care), and psychosocial and financial dimensions. Internal consistency of the adapted satisfaction scale demonstrated acceptable reliability with a Cronbach’s alpha coefficient of 0.82. Domains retained from the EORTC QLQ-C30/BR23 and FACT-B instruments included communication, emotional support, treatment burden, continuity of care, and psychosocial functioning.

Responses were recorded on a five-point Likert scale ranging from “very dissatisfied” to “very satisfied.” The primary outcome, overall satisfaction with care, was calculated as a composite score obtained from all satisfaction-related items. Scores above the median composite value were categorised as “satisfied,” whereas scores below the median were classified as “not satisfied” for regression analyses.

Qualitative: Semi-structured FGDs (60 - 90 min) explored patient journeys, interactions, barriers, emotions, and expectations. Discussions were conducted in French, English, and local languages according to participant preference. Sessions were audio-recorded, transcribed verbatim, and translated into English where necessary by trained bilingual research assistants. Thematic coding was performed independently by two researchers, and disagreements were resolved through consensus discussion.

Trained assistants administered tools in private rooms after consent. FGDs were conducted after quantitative completion, moderated bilingually, and anonymised. The principal investigator supervised all procedures and daily data checks.

Data were analysed in SPSS version 26. Descriptive statistics summarised participant characteristics and satisfaction domains. Chi-square tests examined associations between independent variables and overall satisfaction. Variables considered clinically relevant a priori (age, hospital, stage at diagnosis, household income, communication quality, psychosocial support, and waiting time) were entered into the multivariable logistic regression model. Additional variables with p < 0.20 in bivariate analysis were also considered for inclusion. Missing data represented less than 5% of observations and were handled using complete-case analysis. Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were reported. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUC), and calibration was evaluated using the Hosmer-Lemeshow goodness-of-fit test. A p-value < 0.05 was considered statistically significant.

Qualitative data underwent thematic analysis by two independent coders; themes were triangulated with quantitative results for validity.

Of the 500 women identified, 423 met the eligibility criteria, and 389 completed the quantitative survey, yielding a response rate of 92%. The mean age of participants was 45.2 ± 13 years (range: 24 - 78 years). Most participants were married (62.7%), had a low monthly income below 200,000 CFA francs (74.8%), and were diagnosed at an advanced stage of breast cancer (stage III - IV; 71.2%). Only 18% of participants reported having received psychosocial support during their care (Figure 1 and Table 1).

Figure 1.Recruitment flowchart of study participants.

Table 1. Sociodemographic and clinical characteristics of participants (n = 389).

*Not married includes single, divorced, widowed, and cohabiting.

Overall satisfaction with breast cancer care was reported by 65.4% of participants. Satisfaction was highest for perceived technical competence (78.2%) and provider courtesy (71.6%), while the lowest satisfaction scores were observed for information provision (42.3%) and waiting time (39.8%). There was no statistically significant difference in overall satisfaction between the two hospitals (67.1% vs. 63.3%; p = 0.21).

Qualitative findings reinforced these results, with participants acknowledging provider competence but expressing frustration with prolonged waiting times and limited attention to emotional needs (Figure 2 and Table 2).

Figure 2.Profile of perceived quality of breast cancer care.

Table 2. Satisfaction scores by quality dimension and hospital.

NS = not significant.

Regarding communication, 72% of participants rated physician attitude as good, and 65% reported feeling listened to during consultations. However, only 42% stated that they received complete information about their condition and treatment, and 38% reported understanding treatment side effects. Qualitative analysis revealed themes of information deficits, emotional neglect, and the role of empathy in fostering trust. In quantitative analysis, good communication was a strong predictor of patient satisfaction (aOR = 3.03; p < 0.001).

Only 36.8% of participants reported continuous follow-up with the same physician, while 54% experienced difficulties navigating care between departments. Participants described fragmented care pathways, characterised by repeated referrals and lack of coordination. Poor coordination was significantly associated with lower satisfaction (p = 0.004).

A large proportion of participants (81%) reported experiencing financial hardship related to breast cancer care. The mean delay from diagnosis to treatment was 6.4 months. Satisfaction was significantly higher among women treated within three months (74.6%) compared with those treated after more than six months (49.3%; p < 0.001). Waiting time remained an independent predictor of satisfaction in multivariable analysis (aOR = 1.89).

In multivariable analysis, factors independently associated with higher satisfaction were good communication (aOR = 3.03), receipt of psychosocial support (aOR = 2.07), shorter waiting time (aOR = 1.89), and higher income (aOR = 2.79), with good model fit. Differences between hospitals were small and not statistically significant (Table 3 & Table 4).

Table 3. Multivariable logistic regression of factors associated with satisfaction.

aORs adjusted for hospital, age group, clinical stage, household income, communication quality, waiting time, and psychosocial support. CIs are 95% Wald CIs. Interpretations refer to odds, not risk. Model fit: Hosmer-Lemeshow p = 0.68; AUC = 0.81.

Table 4. Alignment of quantitative and qualitative findings.

Participants emphasised the need for improved communication and information, integration of psychosocial services, reduction of treatment delays through decentralised care, financial support mechanisms, and improved coordination and continuity of care.

Box 1.Representative PatientQuotes by Theme

“Doctors are competent,but we wait too long.”“We are treated for the body,but no one asks about our fears.”“Each department works alone; we go back and forth.”“We sold land to buy drugs.”

Box 2. Practical Recommendations for Clinicians

Improve communication and information delivery.Integrate psychosocial support into routine care.Strengthen coordination and continuity of care.Reduce delays and improve accessibility.Consider patients’ financial constraints in care planning.