Cancer staging information in Hospital Cancer Registries (HCR) is essential for cancer care quality evaluations. This study aimed to analyze the completeness of cervical cancer staging in Brazilian HCR and identify individual and contextual factors associated with unknown staging.
The outcome analyzed was missing or unknown staging (Malignant Tumor Classification System and/or International Federation of Gynecology and Obstetrics) in 2006–2015. Individual data on cancer cases were collected from the HCR Integrator. Contextual variables were collected from the Atlas of Human Development in Brazil, the National Registry of Health Facilities, and the Outpatient Information System. The random intercept multilevel Poisson regression model was performed to identify the factors associated with the outcome.
The prevalence of unknown staging data was 32.4% (95% confidence interval [CI], 32.1–32.7). Women aged 18–29 years (prevalence ratio [PR], 1.48; 95% CI, 1.42–1.54), referred by the public health system (PR, 1.16; 95% CI, 1.11–1.21), living in states with a low density of oncologists (PR, 1.70; 95% CI, 1.62–1.79), and with a low cytopathological testing rate (PR, 1.69; 95% CI, 1.57–1.82) showed a higher prevalence of unknown tumor staging data. A lower level of education (PR, 0.91; 95% CI, 0.84–0.98) was associated with complete staging data.
Individual and contextual factors were associated with missing staging data. It is necessary to improve information on cancer in the HCRs by improving the awareness and training of Brazilian cancer care professionals.
- • Individual factors are associated with loss of cervical cancer staging data.
- • Contextual factors are also associated with loss of cervical cancer staging data.
- • The proportion of missing staging data is higher in areas with a low density of oncologists.
- • Low rates of cytopathological testing are associated with missing staging data.
- • Clear and appropriate inclusion of staging data in medical records is necessary.
Cervical cancer has the fourth highest incidence and mortality among the female population worldwide. In middle- and low-income countries, it ranks second in incidence and mortality in women after breast cancer. Approximately 121,000 new cases of cervical cancer are estimated to occur in 2030 worldwide . In Brazil, an estimated 16,590 new cases are expected each year in the triennium 2020–2022, with variations among regions .
Cervical cancer can be prevented and controlled through actions such as vaccination, screening, early diagnosis, and timely access to health services . Cancer registries play a key role in the initiative to control and eliminate cervical cancer .
Hospital cancer registries (HCR) systematically collect, store, analyze, and interpret data on cancer patients assisted in hospital units. The information produced by the HCR allows the monitoring of oncologic care, particularly clinical function. It becomes a tool for evaluating hospital service quality with emphasis on cancer treatment results . It is an essential part of the production of health information fundamental to understanding cancer distribution and its determinants .
In Brazil, HCR were created in the early 1990 s to gather standardized information about the sociodemographic characteristics of cancer patients, clinical characteristics of tumors, and hospital service activity . HCR are located in the Units and Centers for High Complexity Care in Oncology, which operate and feed the data .
HCR data quality is a key factor requiring evaluation. In addition to being an important predictor of cancer survival, clinical tumor staging is an indicator of a health system’s supply capacity and access to diagnostic technologies . Complete and high-quality staging generates important information that increases our understanding of cancer outcomes and enables the monitoring of primary prevention strategies with the goal of the early diagnosis of cervical cancer .
Missing cancer staging information at the time of diagnosis may be due to an incomplete staging assessment or registration system failure and data coding errors. Understanding the mechanisms of missing data and the characteristics of patients for whom staging data are missing is essential to unraveling possible biases and reaching sensible conclusions from HCR results analysis and interpretation .
This study aimed to analyze the patterns of missing data for cervical cancer staging at diagnosis and identify possible mechanisms involved therein as well as associated factors that can improve data completeness.