Objective: To examine temporal trends in caesarean section (CS) rates and identify clinical and system- level factors driving observed changes in a metropolitan referral hospital in Ghana. Design: Mixed methods study combining retrospective longitudinal analysis and qualitative inquiry. Setting: Tema General Hospital, a major urban referral facility in the Greater Accra Region, Ghana. Population: All recorded deliveries (N=16,517) between January 2020 and December 2024. Methods: A descriptive-analytical approach was employed. Quantitative data from routine hospital records were analysed to estimate annual CS rates and temporal patterns. Associations between year of delivery and mode of delivery were assessed using chi-square tests and Poisson regression models with robust standard errors were used to estimate adjusted risk ratios (aRRs). Semi-structured interviews with 13 key informants were analysed thematically to contextualise observed trends. Main outcome measures: Annual CS rates, relative risk of CS by year and qualitative drivers of CS utilisation. Results: CS rates increased from 38.0% in 2020 to 50.3% in 2024 despite a decline in total deliveries. Compared with 2020, deliveries in 2024 had a 60% higher adjusted risk of CS (aRR=1.603; 95% CI: 1.495-1.719). Qualitative findings identified increasing maternal preference, improved surgical capacity, defensive clinical practice and institutional constraints as key drivers of rising CS use. Conclusions: CS rates at this urban referral hospital substantially exceed WHO recommendations and continue to rise. The findings suggest that increasing CS use reflects complex interactions between clinical need, provider behaviour and health system dynamics. Strengthening clinical governance, audit systems and antenatal counselling is essential to optimise CS use.