Introduction
Every day, organizations worldwide are generating an enormous volume of data, largely driven by widespread computerization. The International Data Corporation (IDC) approximated the size of the digital universe in 2005 to be 130 exabytes (EB) and has recently projected a staggering expansion to 291,000 EB in 2027 1. To comprehend the scale of this data, it is equivalent to watching the entire Netflix catalogue a mind boggling 663 million times 2. The utilization of big data has transformed the qualitative approach in many sectors into a quantitative approach relying on statistics and models, and propelled by meticulously executed big data analysis. In healthcare, big data plays a crucial role in addressing significant healthcare issues, including public health, quality of care, transparency, health democracy, efficiencies within health systems, rising healthcare expenditures, and innovation in areas such as personalized medicine and information technologies 3. Traditionally, the healthcare industry predominantly relied on personal experience and anecdotes for decision-making. However, over the last decade, there has been a dramatic revolution, with the development of artificial intelligence technologies capable of meticulously executing big data analysis. Health data therefore represent a unique tool and a potential to create value, with the realization of that potential depending on the ability of the healthcare sector to organize the development of an ecosystem that facilitates the use of the data while guaranteeing compliance with privacy requirements and the confidentiality of personal data 3. This paper is the result of a collegial discussion of experts from a range of different private and public professional backgrounds to reflect on the applicability of big data analytics in the healthcare industry.
Patient access to care
Traditionally, patient scheduling has been a time-consuming and error-prone task. The integration of big data analytics, along with continued enhancements in artificial intelligence, has transformed this process, significantly enhancing efficiency as well as the patient experience. By analysing large datasets, healthcare facilities can identify patterns and optimize appointment scheduling based on factors such as preferences, medical history, and provider availability 4. Additionally, big data combined with predictive analytics can help forecast patient demand to improve the overall scheduling process. In one study, big data was applied to analyse patient wait times. Through analysis, changes were made in template design as well as patient scheduling workflows to reduce wait times and increase capacity without the hiring of additional staff, acquisition of new machines or changes in infrastructure 5. Efficient patient scheduling through the application of big data not only benefits healthcare operations, but also enhances the overall patient experience.
Healthcare facilities can personalize appointment scheduling to match individual patient needs. By considering factors like travel distance and preferred time slots, patients experience greater convenience and reduced wait times. The application of big data in healthcare does not need to be at the sacrifice of a personalized experience. When applied thoughtfully and comprehensively, the use of big data with artificial intelligence can increase access to care and provide optimal outcomes for patients, placing them at the centre of their healthcare journey. The integration of big data analytics and automation of patient scheduling has revolutionized the healthcare industry. The ability to process vast amounts of data and extract meaningful insights have improved efficiency and enhanced the patient experience. As technology continues to advance, the impact of big data on automating patient access to care will undoubtedly continue to evolve, leading to further improvements in healthcare delivery and patient care.
Acute care delivery at home
Many factors are contributing to the increasing need for healthcare services, such as the ageing population and the increasing prevalence of chronic diseases. Faced with capacity constraints, healthcare systems are seeking alternative sites of care to meet the growing demand. A patient’s home is now seen by many as the safest and most effective option for certain patients and conditions. Hospital-at-home allows for medium-acuity patients who need hospital-level care, but are considered stable enough, to be treated from their homes 6.
This model of care is seeing increased adoption and broader support, especially in the USA. Big data can be a key enabler for acute care delivery at home, including identifying suitable patients, empowering clinicians with real-time information, and facilitating proactive patient management through big data analysis. To identify the appropriate clinical conditions for hospital-at-home, health systems can analyse the electronic medical records and other relevant datasets to understand care patterns and resources required during a hospital stay and prioritize the types of diseases that can be safely and effectively treated at home. In addition, advanced analytics can be used to build models that incorporate patients’ individual-level data, such as medical history, living conditions, and available family support, to identify the appropriate candidates to ensure optimal clinical outcomes.
Another application of data in acute care delivery at home is to leverage data to support clinical decision-making and proactive patient management. Hospital-at-home programmes often rely on comprehensive remote patient monitoring technologies, such as continuous vital sign monitoring. Through these platforms, real-time data is communicated to the centralized command centre and the patient’s care team. Clinical decision support tools can be integrated into the workflow, ideally within the main electronic medical record system, to enhance evidence-based and effective decision-making and mitigate data overload. The large amount of data collected through sensors, monitors, and electronic medical record systems can then be analysed to predict patient deterioration, enabling proactive clinical management and reducing avoidable exacerbations.
Standard medical practice
Another focus in the global health industry is to enhance the quality of healthcare and to alleviate the financial burden on the healthcare system 7. Many institutions have leveraged big data analytics in this setting and developed customized dashboards to monitor their key performance indicators related to operations, performance, and clinical outcomes. These dashboards have positively influenced hospital workflows, the cost of care, patient safety, and overall healthcare quality 8–11.
The Performance and Clinical Excellence (PaCE) programme was launched in the Emirates Health Services, UAE, almost a decade ago to enhance chronic disease management by adopting evidence-based and patient-centered care approaches in international chronic care models. The project mapped clinical data elements to a quality domain framework, gathered input from clinicians and electronic medical record experts, and considered existing data and literature to establish benchmarks. Over 150 clinical performance indicators related to disease of interest and evidence-informed benchmarks were set. Notably, the project resulted in a reduction of stroke clinical indicator “door to needle” time from 94 minutes to 49 minutes, a critical factor in promoting quality and patient safety. EHS facilities achieved compliance rates of over 85% with stroke measures set by the World Stroke Organization. Regarding diabetes care, an evaluation of 30 million patient encounters showed substantial improvements, with diabetes control rates rising from 68% to 76% since the programme’s inception. Maternal clinical quality outcomes also improved, including a significant decrease in the number of C-section operations performed (from 36% to 30%), addressing staffing shortages, and reviewing complications to prevent adverse outcomes for both mothers and babies.
Medical research
The development of drugs is a complex and costly endeavour that entails searching for therapies, undergoing multiple testing phases, and evaluating them against established standards of care. The intricate nature of diseases further complicates this process. In the field of oncology, the substantial expenses and limited success rates in research and development have sparked concerns, as pharmaceutical companies invest billions of dollars with a mere 3% chance of success for individual projects, adversely affecting patients who have limited treatment alternatives 12. Historically, research focused on a limited aspect of cancer at the molecular and clinical level. However, the advent of big data, including initiatives like the Human Genome Project, UK Biobank, and Real World Datasets, has facilitated the collection of vast clinical data and biological specimens from large cohorts 13–15.
The abundance of large-scale data has generated valuable insights and driven advancements by integrating different modalities, aggregating data from multiple cohorts, and reusing existing datasets. Significant progress is being made in generating and analysing this data, leading to remarkable advancements in the field in our comprehension of cancer biology as well as its practical application in clinical settings. For instance, the Cancer Genome Atlas (TCGA) dataset serves as a prominent example of big data in cancer research. With a vast storage size of 2.5 petabytes, equivalent to 2,500 times the capacity of a modern laptop in 2022, TCGA‘s profound impact is evident through over 10,242 article citations and 11,054 NIH grant mentions, highlighting TCGA as a transformative community resource that has greatly propelled cancer research 16.
In addition, big data offers an opportunity to reduce the number of concurrent control subjects required in developing medicines for certain diseases mainly in the late-phase confirmatory clinical trials 17. A notable example of this is demonstrated in the randomized trials of COVID-19 vaccines, where among the 12 trials, there were 22,578 individuals who received a placebo whereas the control arm could have been shared between trials 18. It is important to note that such an approach does not propose abandoning well-established principles of data generation, analysis, and interpretation. However, numerous areas of medical research can potentially benefit from the analysis of big data, and therefore, warrant further diligent investigation into the methodological aspects involved 19,20.
Revenue cycle management
Revenue cycle management refers to the process of managing and optimizing the financial aspects of a healthcare organization’s revenue generation. It involves the entire lifecycle of a patient account, starting from appointment scheduling and registration, through the provision of healthcare services, to the final payment collection 21. Thus, navigating the complexities of the revenue cycle management requires expertise in healthcare operations, billing and coding, technology, and financial management. To optimize revenue and minimize claim denials, healthcare facilities can leverage big data analysis to extract the clinical context from unstructured physician notes, automatically code each encounter with guaranteed accuracy, and provide real-time patient feedback.
Big data analysis can overcome many of the complexities in revenue cycle management arising at the various stages in patient registration and insurance verification, coding and documentation, claims submission and processing, denials and appeals management, and patient billing and collections 22. By automating repetitive manual tasks, organizations can free up staff resources to focus on other valuable tasks. There are several strategies that can be considered to harness the power of data and enhance operational efficiency while optimizing revenue generation. For instance, algorithms can be trained using millions of medical records to emulate human decision-making in dynamic healthcare environments, ingest and learn clinical and financial data, understand nuances and challenges, and accurately categorize and code the data. Big data analysis can also streamline the coding and documentation process, ensuring compliance with the correct coding initiative regulations and global coding standard.
Challenges and considerations
Over the last decade, big data has enabled healthcare providers to enhance patient access and improve care quality by identifying patterns, trends, and patient preferences. It has also helped allocate resources effectively, reducing wait times and ensuring timely and appropriate care. Big data also transformed healthcare delivery by enabling remote monitoring and management of patients’ conditions, reducing hospital readmissions, and lowering costs. Moreover, it drove medical research advancements through the analysis of large-scale clinical data, leading to personalized medicine approaches. Maintaining a profitable revenue cycle is essential, achieved by optimizing revenue streams, controlling costs, and improving billing processes. This allows sustainable investment in infrastructure, technology, and research while delivering high-quality care. Embracing the potential of big data analytics empowers the healthcare industry to achieve significant advancements in improving patient outcomes and enhancing the efficiency of medical and administrative processes. This not only ensures the industry’s long-term viability but also equips it to thrive in a constantly evolving healthcare landscape.
Applying big data analytics in the aforementioned healthcare domains is a transformative endeavour, yet it poses multifaceted challenges. Patient access to care hinges on navigating data security intricacies and integrating disparate sources, while ensuring seamless interoperability. For acute care at home, reliable remote monitoring systems must be established alongside robust data transmission protocols. Integrating data-driven practices into standard medical procedures requires overcoming inertia within the medical community and validating clinically sound insights. In medical research, the ethical imperative of data sharing coexists with the intricacies of heterogeneous data types. Meanwhile, revenue cycle management grapples with billing precision, fraud detection, and intricate regulatory compliance. Tackling these challenges demands interdisciplinary collaboration between healthcare professionals, data scientists, cybersecurity experts, and technology specialists to extract pertinent insights from the massive amounts of healthcare data available while safeguarding patient privacy and data accuracy.
References
IDC Global DataSphere Forecast, 2023-2027, #US50554523 April 2023.
Allen Hasham, C. It’s time for a new interconnection model for the growing amount of data in motion. https://www.ciena.com/insights/articles/2023/its-time-for-a-new-interconnection-model-for-the-growing-amount-of-data-in-motion.
White Paper ‘Big Data & Prevention: from Prediction to Demonstration’. Healthcare Data Institute https://healthcaredatainstitute.com/2017/01/10/white-paper-big-data-prevention-from-prediction-to-demonstration/ (2017).
Ala, A. & Chen, F. Appointment Scheduling Problem in Complexity Systems of the Healthcare Services: A Comprehensive Review. Journal of Healthcare Engineering 2022, e5819813 (2022).
Kern, C. et al. Big data simulations for capacity improvement in a general ophthalmology clinic. Graefes Arch Clin Exp Ophthalmol 259, 1289–1296 (2021).
The Value Initiative Issue Brief: Creating Value by Bringing Hospital Care Home | AHA. https://www.aha.org/issue-brief/2020-12-03-value-initiative-issue-brief-creating-value-bringing-hospital-care-home.
Adler, P. S. et al. Performance Improvement Capability: Keys to Accelerating Performance Improvement in Hospitals. SSRN Scholarly Paper at https://papers.ssrn.com/abstract=979048 (2009).
Jack Weiner, null, Balijepally, V. & Tanniru, M. Integrating Strategic and Operational Decision Making Using Data-Driven Dashboards: The Case of St. Joseph Mercy Oakland Hospital. J Healthc Manag 60, 319–330 (2015).
Egan, M. Clinical Dashboards: Impact on Workflow, Care Quality, and Patient Safety. Critical Care Nursing Quarterly 29, 354 (2006).
Raban, M. S., Bamford, C., Joolay, Y. & Harrison, M. C. Impact of an educational intervention and clinical performance dashboard on neonatal bloodstream infections. S Afr Med J 105, 564–566 (2015).
Chandraharan, E. Clinical dashboards: do they actually work in practice? Three-year experience with the Maternity Dashboard. Clinical Risk 16, 176–182 (2010).
Pursuing breakthroughs in cancer-drug development | McKinsey. https://www.mckinsey.com/industries/life-sciences/our-insights/pursuing-breakthroughs-in-cancer-drug-development.
The Human Genome Project. Genome.gov https://www.genome.gov/human-genome-project (2022).
UK Biobank. https://www.ukbiobank.ac.uk/.
Surveillance, Epidemiology, and End Results Program. SEER https://seer.cancer.gov/index.html.
Jiang, P. et al. Big data in basic and translational cancer research. Nat Rev Cancer 22, 625–639 (2022).
Lim, J. et al. Minimizing Patient Burden Through the Use of Historical Subject-Level Data in Innovative Confirmatory Clinical Trials: Review of Methods and Opportunities. Ther Innov Regul Sci 52, 546–559 (2018).
Haas, J. W. et al. Frequency of Adverse Events in the Placebo Arms of COVID-19 Vaccine Trials: A Systematic Review and Meta-analysis. JAMA Network Open 5, e2143955 (2022).
Caliebe, A., Leverkus, F., Antes, G. & Krawczak, M. Does big data require a methodological change in medical research? BMC Medical Research Methodology 19, 125 (2019).
Rassy, E. & Andre, F. A forgotten dimension of big data in drug repositioning. European Journal of Cancer In press, (2023). https://doi.org/10.1016/j.ejca.2023.113277
Jewell, B. Revenue Cycle Management. in The Business Basics of Building and Managing a Healthcare Practice (eds. Baum, N. & Kahn, M. J.) 19–29 (Springer International Publishing, 2020). doi:10.1007/978-3-030-27776-5_4.
Schouten, P. Big data in health care: solving provider revenue leakage with advanced analytics. Healthcare Financial Management 67, 40–43 (2013).
