Big Data to Improve Public Health in Low- and Middle-Income Countries: Big Public Health Data in LMICs

Introduction

The term “big data” refers to the emerging use of rapidly collected, complex data in large quantities (Ward and Barker, 2013). Health care delivery is a complex endeavor at both an individual and population level. However, public health can be improved with big data as it can provide insights into the causes and outcomes of diseases, identify better drug targets for precision medicine, and enhance disease prediction and prevention (Khoury and Ioannidis, 2014). At a clinical level, the tailored provision of care to individuals is guided, in part, by medical history, physical examinations, vital signs, and evidence. At the population level, traditional health data include information from vital statistics registries and hospital admission statistics (Wyber et al., 2015). Advancements in information technologies represent a milestone in facilitating the development of learning systems for care that enables us to improve public health (Kache and Seuring, 2017). Thus, the era of major data transformation has spurred the use of advanced analytics and big data technologies in public health care (Veyane et al., 2018). Electronic medical records, mobile phone applications, geographical positioning systems, and social media are some of the popular big data sources in public health (Lou et al., 2016).

In addition to the transformation potential of big data in health care, there are potential risks involving this approach (Nambiar et al., 2013). Thus, the high quality collection and analysis of data in health care is critical to improving the effectiveness and efficiency of health care delivery, with the added benefit of a concomitant reduction in health costs (Murdoch and Detsky, 2013). A substantial fraction of the waste in public health care policies and expenditure ensues from not knowing what works for particular patients in particular clinical contexts. Consequently, big data has huge potential to improve public health and achieve better health outcomes (Ellaway et al., 2014).

Big data analytics in public health have already led to better outcomes in many countries, by analyzing patient characteristics and the cost and outcomes of care, permitting caregivers better capacity to identify the most clinically and cost-effective treatments, along with offering analysis and tools, thereby influencing provider behavior by applying advanced analytics to patient profiles (Raghupathi and Raghupathi, 2014). The public health system can be transformed by incorporation of family history, genetic data, lab results, and symptoms into a digital patient data storage system and by providing a data-based computational aid system for physicians to use in assessing their patients’ disease risk. Moreover, understanding hospital readmissions and costly illnesses will help health professionals provide a better balance between capacity and costs (Chawla and Davis 2013). In other words, big data analytical systems provide a significant benefit by allowing health professionals to identify patterns of care and control costs using a broader view of evidence-based clinical care (Wang et al. 2018).