Data quality assessment in healthcare data: A case study

• Autor(a) principal: Monteiro, Stephanie Cardoso
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/31061
• topic: Healthcare data; Data quality; Data integration and data quality assessment; Dados de saúde; Qualidade de dados; Integração de dados e avaliação da qualidade dos dados

Reliable data is essential for monitoring and evaluating business activities. When critical domains such as Healthcare are involved, data quality has a crucial impact on de- livering more accurate and fast healthcare services. Considering epidemiological scenarios such as the COVID-19 pandemic, data can assume an essential role in supporting social answers carried on by the primary decision-makers. For that, sharing and having an inte- grated view of the data allow for identifying the best approaches and critical signals that could lead to better treatments and diagnoses. Nevertheless, leading with data extraction from several sources is not an easy task and can lead to enormous challenges related to data accessibility, representation, and interpretation. Several data quality problems can occur and, when not adequately addressed, can question the decision-making support. The contribution of this thesis was to perform a data quality assessment from a subset of data from a Portuguese hospital when used in the context of integration to a common shared repository within the scope of a European project. A deep data profiling analysis in the source database was conducted, identifying the main characteristics and, after- wards, the main issues. Each issue was later mapped with its corresponding data quality violation, and rules were defined as guidelines to address these issues and prevent future ones. To classify the quality of the source data, a methodology was proposed to evaluate the data into two levels, quality roles level and data quality dimensions level, calculating the data quality score. The final results are discussed and evaluated in this work.