The feasibility of applying individual patient data to assess inequity in cancer treatment retention in Northern Ghana - early results
More details
Hide details
School of Health and Related Research, University of Sheffield, United Kingdom
Ghana Regional Health Directorate, Ghana
School of Public Health, University of Ghana, Ghana
Tamale Teaching Hospital, Ghana
Publication date: 2023-04-27
Popul. Med. 2023;5(Supplement):A1525
Background: Cancer poses an increasing burden in the Africa, associated with an epidemiological transition.  Despite the high contribution of preventable cancers, survival rates are low, reflecting limitations in screening, diagnosis, resources and treatment access. One reason for poor cancer outcomes in the Ghana is poor engagement with treatment.  We previously conducted a critical interpretive synthesis of literature on access to cancer treatment in Ghana and found barriers across the social ecological system (Tuck et al.,BMJ Open_ 2022).  However, there was a gap in understanding what influenced treatment completion in the northern region. Objectives: 1. To assess the feasibility of using digital patient records to assess cancer treatment completion in Northern Ghana. 2. To apply the data to understand social, economic and demographic characteristics influencing completion of cancer treatment. Methods: Secondary data analysis of routinely collected cancer treatment data, retrieved from the cancer registry of the Oncology Department of Tamale Teaching Hospital (TTH) (the largest referral hospital in the northern Ghana) will be undertaken. Variables required to perform analysis to address the study objectives include outcome variables: completion of chemotherapy and radiotherapy and explanatory variable: demographics and social economic status. These variables are currently being extracted from anonymised medical records of patients with cancer. Data will be cleaned and sorted in R. Descriptive analysis will estimate the proportions of patients completing chemotherapy and radiotherapy and logit regression conducted to identify characteristics associated with incompletion. Finally, the feasibility of applying multi-level modelling approaches for the intersectionality of social characteristics will be assessed. Results: The results will be presented as summary statistics for the single level and multiple level logit regression models to assess the discriminatory accuracy and intersectionality of characteristics on treatment completion. Conclusions: The key findings and feasibility of using the clinical dataset in future equity research will be discussed.