Physician–patient communication in decision-making about Caesarean sections in eight district hospitals in Bangladesh: a mixed-method study

BACKGROUND: Caesarean sections (CS) in Bangladesh have risen eight-fold in the last 15 years. Few studies have explored why. Anecdotally, physicians suggest maternal request for CS is a reason. Women and families suggest physicians influence their decision-making. The aim of this research was to understand more about the decision-making process surrounding CS by exploring physician–patient communication leading to informed-consent for the operation. METHODS: We conducted a mixed-method study using structured observations with the Option Grid Collaborative’s OPTION5 tool and interviews with physicians and women between July and December 2018. Study participants were recruited from eight district public-sector hospitals. Eligibility criteria for facilities was ≥ 80 births every month; and for physicians, was that they had performed CSs. Women aged ≥ 18 years, providing consent, and delivering at a facility were included in the observation component; primigravid women delivering by CS were selected for the in-depth interviews. Quantitative data from observations were analysed using descriptive statistics. Following transcription and translation, a preliminary coding framework was devised for the qualitative data analysis. We combined both inductive and deductive approaches in our thematic analysis. RESULTS: In total, 306 labour situations were observed, and interviews were conducted with 16 physicians and 32 women who delivered by CS (16 emergency CS; 16 elective CS). In 92.5% of observations of physician–patient communication in the context of labour situations, the OPTION5 mean scores were low (5–25 out of 100) for presenting options, patient partnership, describing pros/cons, eliciting patient preferences and integrating patient preferences. Interviews found that non-clinical factors prime both physicians and patients in favour of CS prior to the clinical encounter in which the decision to perform a CS is documented. These interactions were both minimal in content and limited in purpose, with consent being an artefact of a process involving little communication. CONCLUSIONS: Insufficient communication between physicians and patients is one of many factors driving increasing rates of caesarean section in Bangladesh. While this single clinical encounter provides an opportunity for practice improvement, interventions are unlikley to impact rates of CS without simultaneoulsy addressing physician, patient and health system contextual factors too.