Search for a command to run...
Cervical cancer is the fourth most common cancer among women with significant global disparities in disease burden. In lower-resource settings, where routine screening for cervical cancer is uncommon, higher incidence of advanced-stage disease contributes to increased morbidity and mortality. Understanding care delays may inform strategies to decrease overall time to treatment, and could potentially improve outcomes. We sought to characterize the cervical cancer care cascade and identify factors associated with time to care cascade completion within a Ugandan cohort. We collected sociodemographic, reproductive health and care journey data from 268 Ugandan women newly diagnosed with cervical cancer at Mulago National Referral Hospital and the Uganda Cancer Institute. We characterized time from symptoms to presentation (patient interval), time from presentation to diagnosis (diagnostic interval) and time from diagnosis to treatment (treatment interval) and estimated the influence of patient, health provider, system, and disease factors on length of each interval using survival analysis. Median patient, diagnostic and treatment intervals were 74 days (IQR 26–238), 83 days (IQR 34–229), and 34 days (IQR 18–58), respectively. Patient interval was prolonged by the belief that symptoms would resolve spontaneously (aHR 0.37, 95% CI 0.24–0.57), confusion about where to seek care (aHR 0.64, 95% CI 0.47–0.88), and utilization of traditional care (aHR 0.70, 95% CI 0.51–0.96). Patient interval facilitators included perceiving symptoms as serious (aHR 2.14, 95% CI 1.43–3.19) and suspecting cancer (aHR 1.82, 95% CI 1.12–2.97). Diagnostic interval delays included symptomatic bleeding (aHR 055, 95% CI 0.35–0.85) and visiting > 2 clinics (aHR 0.69, 95% CI 0.49–0.97); facilitators included early-stage disease (aHR 1.41, 95% CI 1.03–1.95) and direct tertiary care presentation (aHR 2.13, 95% CI 1.20–3.79). Treatment interval delays included anticipating long waits (aHR 0.68, 95% CI 0.46–1.02) and requiring blood transfusions (aHR 0.63, 95% CI 0.37–1.07); no facilitators were identified. We identified potentially modifiable barriers and facilitators along the cervical cancer care cascade. Interventions targeting these factors may reduce delays, but are unlikely to significantly improve morbidity or mortality given advanced-stage disease at symptom acknowledgement and relatively timely treatment initiation. Expanding cervical cancer screening and vaccination are of utmost importance.