The Use and Impacts of a Corporate Data Warehouse: The Case of the National Health Laboratory Service

Abstract

The promotion of population health is a key concern of government authorities and various health institutions. With the prevalence in South Africa of communicable diseases like AIDS and Tuberculosis, the increase in the incidence of non-communicable diseases like Cancer, Cardiovascular disease and Diabetes, as well as health care access, management and insurance issues, it is clear that the area of population health in South Africa is in need of urgent attention. Decisions about interventions that may impact on the health of the population tend to be far reaching and must be made with care. Basing these decisions on facts drawn from relevant data collected on various aspects of population health would be highly advantageous.

This study explores the potential and application of Business Intelligence and Data Warehouse technology for the population health space. Specifically, the National Health Laboratory Services (NHLS) was identified as the context for the study. The National Health Laboratory Services (NHLS) is responsible for laboratory tests done at public health institutions throughout South Africa, and as such keep records for 80% of the population, all of which are accessible from the NHLS Corporate Data Warehouse (CDW) as a BI tool. The full impact of the NHLS CDW cannot be realised by simply providing the technology and hoping that it is well used. The CDW can only make a difference if the information it contains is utilised by people who base their decisions on it. An understanding of the issues and challenges experienced by these people in obtaining information from the CDW would facilitate positive outcomes which would in turn promote a positive impact on population health.

Drawing on Activity Theory, this study explored and described the ways in which the NHLS CDW is being used, and the impact that this use is perceived to be having on population health in South Africa. To this end, the following research questions were addressed:

RQ1: What are the current applications of the NHLS CDW? RQ2: What are the issues and challenges in the use of the NHLS CDW? RQ3: What are the individual outcomes for the NHLS CDW user and data requestor? RQ4: What are the perceived benefits (impacts) of the NHLS CDW use for the NHLS as an institution and for population health in general?

After a preliminary session with the staff of the CDW department, ten interviews were conducted with members of staff from four separate parts of the NHLS – the National Institute for Communicable Diseases, the National Cancer Registry, the National Priority Program and operational management. In relation to RQ1, the NHLS was found to be an exceptionally large organisation involved in many areas of population health. Consequently, a broad range of applications of the NHLS CDW were identified in areas as diverse as population health tracking, medical research and administration of laboratory performance. In relation to RQ2, issues and challenges were identified in terms of difficulties obtaining a patient-oriented view of the data, data quality, staffing issues, time delays, restrictions due to privacy and security and the need to access external data sources. In relation to RQ3, it was found that the NHLS CDW was considered by all participants interviewed as an essential and relevant tool in the accomplishment of their jobs – most are highly appreciative of it. In relation to RQ4, the general perception is that the CDW is invaluable to the NHLS as an organisation and for the improvement in population health in general. Evidence stored within the CDW clearly shows the impact on population health of NHLS population health programs that focus HIV/AIDS and Tuberculosis. These programs depend heavily on the CDW for their effective operation. There was, however, a definite sentiment amongst those interviewed that however much is already being done with the CDW, there is still potential for so much more.This research has contributed to the body of knowledge and the body of evidence on data warehouses, particularly on data warehouses implemented in South Africa and/or in the population health context. The extension of the use of Activity Theory into the study of data warehouses as information systems, and the demonstration of Activity Theory use in elucidating issues and challenges in CDW use, also constitute a theoretical contribution. Contributions to practice include possible NHLS use of study findings to resolve identified issues and challenges, or utilisation by other organisations of research results as input to decisions about existing or potential datawarehouse applications.