Cervical cancer diagnosis and treatment delays in the developing world: Evidence from a hospital-based study in Zambia
Description
The dataset presented in this article is secondary data obtained from the registry unit of the only cancer treatment Centre in Zambia, the Cancer Diseases Hospital (CDH) in Lusaka. Information relating to patients with a maiden diagnosis of cervical cancer presenting to CDH is outlined in the data. The data is a set of patient demographics, relevant clinical information and the clinical referral channel obtained from patient case files stored at CDH.
Files
Steps to reproduce
Data processing involved the generation of categorical/factor variables from the raw data set that would then fit into the final model. We specifically generated age category (21-30, 31-40, 41-50, …, 81-90), marital status (single, married, divorced, widowed), HIV status (yes, no), stage of cancer (I-IV), lab location (Lus, CB, CP, other), treatment type (CRT, radiotherapy, palliation, chemotherapy, surgery) among others. Being a count data, Poisson distribution was a natural choice but the assumption of the equality of mean, and variance did not hold. Thus, a negative binomial distribution was used to adjust for overdispersion. Analysis was conducted in R (4.0.2) and RStudio (1.4.1103). All the code and the accompanying data are available upon request.
Institutions
- University of Namibia Faculty of Agriculture and Natural Resources
- University of KwaZulu-Natal - Howard College Campus
- University of Zambia