The technological revolution being experienced in modern times is shaping the way organisations conduct business to achieve their missions and visions. Specifically, big data drive critical decision-making processes to improve efficiency and quality of products and services. Big data in healthcare is a collective term used to refer to the process of collecting, analysing, leverage, and make sense of complex and immense patient and clinical data in a way that traditional data processing cannot handle such information. In other words, big data could be defined as the vast health data collected from different sources in areas where patients interact with all other stakeholders within the care system continuum.
The major data sources include patient records, government interventions, insurance records, pharmaceutical research, genomic sequencing, and information collected from wearable devices among other similar areas where patient digital footprints can be traced (Ristevski & Chen 2018). There are many definitions of big data, but they all converge at the point of the major characteristics that underline the concept.
For any dataset to be classified as big data, it should have the five Vs – volume, veracity, variety, value, and veracity (Ishwarappaa & Anuradhab 2015). The objective of this paper is to discuss the concept of big data as applied in the healthcare sector, especially in health economics. The SWOT analysis approach is used to assess the strengths, weaknesses, opportunities, and threats of big data in health economics. Specific examples are also given to highlight how companies are using this concept in the quest to improve care delivery processes to populations.
Big Data in Healthcare Sector
The healthcare industry plays a central role in ensuring that populations remain healthy together with maintaining a high quality of life. However, with the increasing population sizes, the traditional ways of collecting and sampling data for clinical decision-making and policy formulation in the sector are becoming ineffective. For example, the conventional way of conducting evidence-based research to avail useful information in healthcare has been randomised clinical trials (RCTs) (Mayo et al. 2017). However, this approach is limited in its scope because it requires a controlled environment and populations.
Additionally, the RCTs are time-consuming, tedious, and expensive because they test one or limited variables at a time. Therefore, the emergence of big data has helped the industry to collect and analyse large datasets using different algorithms to achieve the desired end within a short period. According to Habl et al. (2016), big data in health refers to “large routinely or automatically collected datasets, which are electronically captured and stored…reusable in the sense of multipurpose data and comprises the fusion and connection of existing databases for improving health and health system performance” (p. 11). One of the major problems in health economics is the ever-increasing cost of care services and products.
For instance, in 2017, the US spent around 17 per cent of its GDP on healthcare, which is approximately $10,000 per capita (Organisation for Economic Co-operation and Development (OECD) 2018). In other words, the cost of care is prohibitive, but big data is playing a critical role in ensuring the affordability of services to patients. The quality of care is another major issue affecting care delivery due to the lack of appropriate systems for timely intervention measures, medication errors, and understanding disease co-morbidity among other associated aspects.
Big data contribute to the industry by increasing the quality and effectiveness of treatments through timely disease intervention, minimising medical errors, establishing causalities and cross-linking care providers to mention but a few. It also promotes disease prevention efforts, improves patient safety through pharmacovigilance (Ventola 2018), predicts outcomes, and disseminates knowledge to help professionals in the industry to remain updated on new developments in the sector.
Literature Analysis
Overview
According to Collins (2016), in the past data were mainly used to test hypotheses, but the scenario has now changed with the entry of neural networks and data mining. Under the new regime, datasets are used to generate hypotheses and identify links that could not be noted using logical methods. In most cases, big data involves open datasets meaning that they are shared in the public domain. For instance, the UK government has a program (data.gov.uk), which is a central repository whereby data is availed for public access and use (Department for Business Innovation & Skills 2014).
In the US, the adoption of electronic health records (EHR) allows the fusion of data from all stakeholders in the sector for easy access of information and promotion of evidence-based practice (Menius & Rousculp 2014). Big data has created opportunities for players in the healthcare sector to access large datasets that link systems using evidence-based information to highlight how drugs interact and point out underlying safety issues and side effects.
The beauty of such systems is that they work within a short period to generate large volumes of relevant data, which is useful in health economics and pharmacoepidemiology. The systems also create the possibility of measuring, analysing, and storing complex structured and unstructured information, which could not be achieved through the conventional antiquated information systems that lacked the capacity to communicate with each other.
In the UK, the National Health Service (NHS) has been increasingly using service line reporting to capture health outcomes and measure spending in different programs, especially the national Patient Reported Outcome Measures (PROMs) for effective decision-making (Prodiger & Taylor 2018). In addition, large datasets are being used to monitor consumer behaviour in the sector to determine the role of health insurance in improving care provision and other outcomes. For example, the Oregon health insurance experiment (OHIE) was conducted to assess the effects of expanding Medicaid in the state of Oregon in 2008 (Allen, Wright, & Broffman 2018).
According to Finkelstein, Hendren, and Luttmer (2018), the OHIE results showed that patients are highly likely to use care services and self-report their mental and physical health if they can access affordable health insurance. Such findings could help governments make informed decisions to improve the health and quality of life for their citizens. Without big data, such information could not be available and thus policymakers would continue making theoretical decisions, which might not have a significant impact on the target population.
One of the areas where big data has availed significant benefits in health economics is in bio-monitoring. As people continue to use passive bio-monitoring devices, such as smart-watches and heart rate monitors, such real-time information can be aggregated to give health indicators that are useful in making clinical and policy decisions. Buekers et al. (2018) posit that policymakers use bio-monitoring data to assess whether the “health of people in specific regions or subpopulations is at risk, or whether the body burden of chemical substances (the internal exposure) varies with, for example, time, country, sex, age, or socioeconomic status, need to be answered” (p. 2085).
Such indicators could be used in policy-making to ensure that laws are made according to real-life evidence to protect human beings from exposure to harmful chemicals. Collins (2016) adds that the immense lifestyle data, from activity monitoring through the Internet of Things (IoT) could also be used to make public health policies that promote physical exercising and access to healthy foods. These accomplishments have only been made possible through the application of big data protocols.
Big data has also allowed health economists to generate real-world evidence and promote personalised medicine. For example, drug companies are employing different techniques embedded in big data to trawl clinical data systems to assess the effectiveness of their products before releasing them into the market (Vallance, Freeman & Stewart 2016). Personalised medicine is becoming a reality using big data whereby clinicians can choose “treatments that are most effective and less likely to be discontinued because of side effects, and, in the case of drugs, they can select a dose that is tailored more to an individual, potentially making the healthcare system more efficient” (Collins 2016, p.105).
Ultimately, as argued before, big data will be a central component of ensuring a healthy population by improving decision-making due to the availability of evidence-based information. A SWOT analysis will help to understand the strengths, weaknesses, opportunities, and threats of adopting big data in the healthcare sector based on the current trends and future prospects.
Strengths
In health economics, data plays an important role in decision-making to ensure that the allocation of resources is based on evidence to achieve specific goals. As such, open datasets avail robust and long-term data outcomes for creating economic models that could be applied in different populations. For example, the Cost-Effectiveness Analysis (CEE) registry and the NHS Economic Evaluation Database (EED) derive information from big data to make economic evaluations and create models (Al Kadour, Al Marridi & Al-Badriyeh 2018).
Patients can be tracked from their lifestyle behaviours (using IoT) to the time they seek medical services and how they are attended to including costs incurred. Such aggregated data would be highly useful in tailoring care services to meet patients’ needs for improved care outcomes (Joiya et al. 2017). In addition, open datasets will allow other parties outside the field of research to access, analyse data, and test hypotheses to inject fresh ideas and perspectives towards the creation of transparency in the healthcare sector.
Using information big data repositories, individual-level data allows clinicians to understand the effectiveness of certain drugs to specific population cohorts, which ultimately improves efficiency in clinical decision-making. In policymaking, big data will continue informing policies based on real-life evidence for the benefit of target populations. For instance, in the UK, “the utility scores associated with particular health states in the EQ-5D (which is seen as the gold standard by NICE in calculating quality-adjusted life expectancy) are calculated using time trade-off methods and a general population sample” (Collins 2016, p. 105).
Decision-making supported by real-life evidence is improved to ensure that resources are allocated in areas that will have maximum returns to the targeted populations. Ultimately, with big data, patients are assured of affordable and quality care services informed by evidence-based practices.
Weaknesses
The costs associated with storing and manipulating large volumes of data to make sense out of it could be prohibitive. For example, in the US, the cost of implementing EHR system for five-person practice is around $162,000 in the first year and another annual $85,000 in maintenance costs and the figures could rise to millions of dollars for individual care facilities (Palabindala, Pamarthy & Jonnalagadda 2016). In addition, big data does not achieve the scientific rigor associated with RCTs, which means such traditional methods will continue to be used in clinical research. Additionally, big data is designed to benefit only individuals who are connected digitally, thus it excludes billions of people in developing and underdeveloped countries.
Personal data privacy and security issues are major concerns affecting big data systems. Abouelmehdi, Beni-Hessane and Khaloufi (2018) argue that people might feel that they are being overly monitored, which infringes on their rights to privacy. As such, individuals might turn down the opportunity to participate in trials involving big data systems.
Therefore, in a bid to ensure the security of people’s data that is captured on health big data systems, extra resources have to be deployed for stringent countermeasures, such as installing expensive firewalls and other related steps. This aspect ultimately increases the cost of acquiring and maintaining such systems, thus making them unavailable to most players in the industry. Consequently, patients and other public beneficiaries might not enjoy the gains of big data in healthcare.
Opportunities
The future of big data is unlimited with the evolving technology space being witnessed contemporarily. With the introduction of artificial intelligence, large datasets with the capacity to communicate with each other will be created to execute complex analyses (Combi & Pozzi 2019). Similarly, with large trial registries backed by big data, drugs with similar effects will be matched to assess how best they can be used together for improved efficacy. Medical errors are common occurrences in the process of patient care, but through big datasets, such incidences might be reduced significantly through improved decision-making capacity.
Cirillo and Valencia (2019) posit that big datasets are advancing individualised genetic mapping for people to understand their level of predisposition to certain diseases and take the appropriate preventive measures.
Additionally, people can use information gathered through bio-monitoring to gain the relevant knowledge needed for self-care for better health and quality of life. When people take responsibility for their own care, health outcomes ultimately improve significantly and the pressure imposed on care facilities is reduced, as people do not become sick often. Islam et al. (2018) add that big datasets improve the accuracy with which the relevant bodies predict outbreaks, thus allowing for quick and timely interventions.
For instance, Erikson (2018) highlights the effective use of HealthMap – an Internet-based big data platform that aggregates news from online sources, such as Facebook and Twitter, in detecting and containing Ebola outbreaks in West Africa. Big data will also help in understanding how different cancer-predisposing factors operate in the progression or remission of the disease, thus improving how the health condition is handled.
Threats
Despite the many opportunities and directions that big data can progress moving forward, several threats are associated with such systems. First, understanding risk factors may cause anxiety among people, especially in cases where the underlying disease is incurable or has no known treatment currently. Care provision may also become expensive with insurance companies using such datasets to increase premiums for people who are genetically predisposed to some health conditions. In addition, there are ethical issues associated with excessive genetic screening, which is characteristic of big data systems (Ienca 2018; Knoppers & Thorogood 2017).
Data loss and access by third parties is also a significant threat to health big data systems. Finally, institutions might collude to profiteer from data collected from bio-monitoring and other strategies of data gathering at the expense of patients. For instance, Hunter (2016) is concerned that data “brokers”, such as Pfizer, which spends over $12 million to acquire anonymous health data, will continue to exploit this development for profiteering. Patients’ welfare should precede any form of financial gain in an ethical healthcare environment, but in a capitalistic system that defines the contemporary marketplace, such ethical and moral standards may not be applicable.
Conclusion
Big data has revolutionised how stakeholders in the healthcare sector relate to each other. Policymakers benefit from the system by relying on real-life experiences when making laws on public health and other related matters. Clinicians rely on large datasets to promote evidence-based practice, which ultimately improves patient outcomes and quality of care. The available literature shows that while big data in healthcare has many strengths and opportunities for growth, it also comes with weaknesses and threats.
Players in the industry should focus on maximising the strengths and exploiting the available opportunities while at the same time improving on the weaknesses and minimising the inherent threats. Ultimately, when all factors are considered, big data will continue to shape the healthcare industry based on how policymakers and other related stakeholders formulate health laws, clinicians make care decisions, and patients receive care services.
Reference List
Abouelmehdi, K, Beni-Hessane, A & Khaloufi, H 2018, ‘Big healthcare data: preserving security and privacy’, Journal of Big Data, vol. 5, no. 1, pp. 1-18.
Al Kadour, A, Al Marridi, W & Al-Badriyeh, D 2018, ‘Pharmacoeconomics evaluations of oral anticancer agents: systematic review of characteristics, methodological trends, and reporting quality’, Value in Health Regional Issues, vol. 16, no. 1, pp. 46-60.
Allen, H, Wright, B & Broffman, L 2018, ‘The impacts of Medicaid expansion on rural low-income adults: lessons from the Oregon health insurance experiment’, Medical Care Research and Review, vol. 75, no. 3, pp. 354-383.
Buekers, J et al. 2018, ‘Development of policy relevant human biomonitoring indicators for chemical exposure in the European population’, International Journal of Environmental Research and Public Health, vol. 15, no. 10, pp. 2085-2103.
Cirillo, D & Valencia, F 2019, ‘Big data analytics for personalised medicine’, Current Opinion in Biotechnology, vol. 58, no. 1, pp. 161-167.
Collins, B 2016, ‘Big data and health economics: strengths, weaknesses, opportunities and threats’, PharmacoEconomics, vol. 34, no. 2, pp. 101-106.
Combi, C & Pozzi, G 2019, ‘Clinical information systems and artificial intelligence: recent research trends’, Yearbook of Medical Informatics, vol. 28, no. 1, pp. 83-94.
Department for Business Innovation & Skills 2014, Open data strategy 2014-2016. Web.
Erikson, S 2018, ‘Cell Phones ≠ self and other problems with big data detection and containment during epidemics’, Medical Anthropology Quarterly, vol. 32, no. 3, 315-339.
Finkelstein, A, Hendren, N & Luttmer, E 2018, The value of Medicaid: interpreting results from the Oregon health insurance experiment. Web.
Habl, C, Renner, A-T, Bobek, J & Laschkolnig, A 2016, Study on big data in public health, telemedicine and healthcare, Gesundheit Österreich GmbH, Wien.
Hunter, P 2016, ‘The big health data sale: as the trade of personal health and medical data expands, it becomes necessary to improve legal frameworks for protecting patient anonymity, handling consent and ensuring the quality of data’, EMBO Reports, vol. 17, no. 8, pp. 1103-1105.
Ienca, M, et al. 2018, ‘Considerations for ethics review of big data health research: a scoping review’, PloSone, vol. 13, no. 10, pp. 1-15.
Ishwarappaa, A & Anuradhab, J 2015, ‘A brief introduction on big data 5vs characteristics and Hadoop technology’, Procedia Computer Science, vol. 48, no. 1, pp. 319-324.
Islam, M, et al. 2018, ‘A systematic review on healthcare analytics: application and theoretical perspective of data mining’, Healthcare, vol. 6, no. 2, pp. 54 – 96.
Joiya, J, Liaqat, R, Farooq, A & Rehman, S 2017, ‘Internet of medical things (IOMT): applications, benefits and future challenges in healthcare domain’, Journal of Communications, vol. 12, no. 4, pp. 240-247.
Knoppers, M & Thorogood, A 2017, ‘Ethics and big data in health’, Current Opinion in Systems Biology, vol. 4, no. 1, pp. 53-57.
Mayo, C et al. 2017, ‘Big data in designing clinical trials: opportunities and challenges’, Frontiers in Oncology, vol. 7, no. 187, pp. 1-7.
Menius, J & Rousculp, M 2014, ‘Growth in health care data causing an evolution in the pharmaceutical industry’, North Caroline Medical Journal, vol. 75, no. 3, pp.188-90.
OECD (2019), Health spending (indicator). Web.
Palabindala, V, Pamarthy, A & Jonnalagadda, N 2016, ‘Adoption of electronic health records and barriers’, Journal of Community Hospital Internal Medicine Perspectives, vol. 6, no. 5, pp. 1-6.
Prodiger, B & Taylor, P 2018, ‘Improving quality of care through patient-reported outcome measures (PROMs): expert interviews using the NHS PROMs Programme and the Swedish quality registers for knee and hip arthroplasty as examples’, BMC Health Services Research, vol. 18, no. 87, pp. 1-13.
Ristevski, B & Chen, M 2018, ‘Big data analytics in medicine and healthcare’, Journal of Integrative Bioinformatics, vol. 15, no. 3, pp. 1-5.
Vallance, P, Freeman, A & Stewart, M 2016, ‘Data sharing as part of the normal scientific process: a view from the pharmaceutical industry’, PLoS Medicine, vol. 13, no. 1, pp. 1-3.
Ventola, L 2018, ‘Big data and pharmacovigilance: data mining for adverse drug events and interactions’, Pharmacy and Therapeutics, vol. 43, no. 6, pp. 340-351.