Individual and contextual factors associated with incidence of ovarian cancer in the United States.

Abstract

e17545 Background: Ovarian cancer has the worst prognosis and highest death rate of all gynecological cancers in the US. Since data on incidence and risk factors are important for strategic program planning, we sought to determine the independent effects of individual-, zip code- and state level factors on the incidence of ovarian cancer in the US. Methods: This retrospective cohort study used de-identified data from the All of Us (AoU) research database to identify women aged 18 years and above with and without ovarian cancer from July 2017 to July 2023. Mixed-effects Cox regression models were used to analyze data on 78,770 individuals, nested within 974 zip codes and within 50 states across the US. Results: Approximately 4 out of every 1,000 patients in this study had ovarian cancer. After all other factors were adjusted for in the final model, individuals aged 50-59 years (adjusted Hazard Ratio [aHR] 2.22; 95% confidence interval [CI] 1.37–3.61) and 60-69 years (aHR 1.76; 95% CI 1.05–2.95) had increased risks of developing ovarian cancer compared to those <50 years. Being a retiree, using hormone replacement therapy, and having annual income of at least $50,000/year increased the risks of ovarian cancer. Individuals residing in zip codes in the Northeast region of the US had increased risks of having ovarian cancer compared to those residing in zip codes in the Midwest region (aHR 1.66; 95% CI 1.09–2.53). There was significant clustering of ovarian cancer at the zip codes and state levels. Conclusions: This study showed that individual and contextual level factors were associated with increased risk of ovarian cancer in the US. Interventions and program designed to lower the risk of ovarian cancer should be directed at both individual and contextual level factors.