Posted: January 18th, 2022

Big Data in the Healthcare Industry

Big Data in the Healthcare Industry

Introduction

Big Data describes large volumes of structured and unstructured data that inundate a business in its daily activities. However, the amount of data does not matter in the context of big data; what an organization does with the data matters.

Big data is becoming common in the current world, whereby organizations employ different strategies, from analyzing data to competing, innovating, and capturing value.

Big data helps an organization create new growth opportunities, and entirely new categories of companies can combine to analyze existing data.

Although big data is relatively new in the market, organizations such as healthcare have been collecting and storing vast amounts of information for analysis. Therefore, data collection and storage, the first step in big data analysis, has only been taking a new trajectory in the millennial error.

The concept of big data gained momentum at the beginning of the 21^st century when industry analyst Doug Laney articulated the now-mainstream of big data as the three Vs. However, according to Barnes (2013), and Manning (2013), big data did not emerge at this time, nor was the invention of Google in 1998 or Apple in 1976.

One is to reconstruct the history of big data over a long time. It has only been made possible due to intertwining different elements with their history and technology to form big data. Therefore, big data dates back to decades ago when various forms of data could be collected and stored. It has only been enhanced by technology in the 21^st century.   

Industry influencers, academicians, and other stakeholders agree that big data has become a game-changer in modern industries. According to Upadhyaya and Kynclova (2017), most organizations have put in place to adopt big data projects, with the primary objective being enhancing customer experience.

Other goals for adopting big data include cost reduction, better marketing strategies, and increased efficiency of the existing processes.

Industries adopting big data include banking and securities, communication, media, the entertainment industry, healthcare providers, education, manufacturing, natural resources, various governments and their agencies, insurance, and the transportation industry.

Therefore, big data is gaining momentum across multiple sectors, enhanced by technology and numerous stakeholders’ goals in each sector.

Artificial Intelligence (AI) is like a magnet to big data whereby organizations collect large amounts of information with promises that valuable insight will result from them.

AI is tearing down information processing barriers whereby it changes the paradigm of supercomputers and mimics human intelligence to programs that can respond to changes in information keyed in and make adjustments in the final output (IDG, 2016).

Therefore, AI makes analyzing big data easier than in earlier times when it depended on supercomputers and more intelligent humans to perform tasks.

A generic overview of Big Data in the healthcare industry

Big Data in healthcare refers to the extensive data amassed from various sources, including Electronic Health Records (EHRs), medical imaging, genomic sequencing, payor records, pharmaceutical research, wearables, and medical devices.

Big data sources in the healthcare industry include patient portals, government agencies, research studies, payer records, search engine data, public records, and generic databases (NEJM Catalyst, 2018). Healthcare organizations use Big Data mainly to improve the service provided to patients.

The process is mainly done through communication from one department to another as well as communicating the process of treatment to patients. Therefore, Big Data in the Healthcare industry helps health professionals deliver quality services to patients by making appropriate decisions using the patient’s available information.

Global healthcare big data continues to rise steadily due to increased investment of management tools, practice management solutions, and EHR systems.

According to Pramanik, Pal, and Mukhopadhyay (2018), big data in healthcare has existed since hospitals kept patient records in hard copy before digitization processes started in the late 20^th and early 21^st century.

Big data was fully incorporated in the healthcare industry from 2011 onwards, especially in developed countries such as the USA and the United Kingdom. Other countries are also introducing policies that fully incorporate big data ideas into their health systems.

The most widely used big data system in the healthcare system is electronic health records, whereby individual health organizations and national levels have adopted it. EHR is used within an organization to share patient data from one department to another.

It is also used to share information from one hospital to another. Therefore, Big Data in healthcare is mainly helping in transferring information from one department and from one hospital to another, thus making the treatment process more effective.

Patients predictions

Big data is used to predict patients’ readmission within a specific period accurately. According to Parslow (2014), in the United States of America, big data has been used to predict patient readmission for 30 days. The analysis also helps in suggesting the actions required for the patients.

Predictive intelligence also has a vast potential for the NHS in the United Kingdom, whereby medics can tell patients their highly accurate expectations. The patient’s more precise treatment is offered by predicting patient conditions, courtesy of big data analysis.

With the world of the internet, devices like Fitbit and Apple Watch can track a person’s physical movements with the capability of sending data to physicians; thus, the progress can be monitored. Therefore, Big data helps predict the patient’s future through accurate analysis and constant monitoring, from which appropriate actions can be taken.

Several errors in the prescription process have been reported to patients worldwide. According to Landau (2019), more than 7 million people are affected by wrong prescriptions in the United States, causing about 7 thousand deaths a year.

However, through Big Data, programs such as MedAware, an Israeli startup, are combating the trend by monitoring errors and taking necessary actions before they occur. Thus, the effects that could have happened to patients are interrupted before they occur.

Electronic Health Record

An electronic Health Record (HER) is an electronic version of a patient’s medical history maintained by the health care provider over time. The record may include administrative and clinical data relevant to the patient. The program automates access to information and can streamline clinicians’ workflow.

The EHR has also been involved in other care-related activities, directly or indirectly, through various interfaces, including evidence-based decision support, quality management, and outcome reporting.

Through EHR, patient care can be improved by reducing medical errors likely to occur due to inaccuracy and unclarity of medical records, availability of patient health information reduced duplication in tests and delays in treatment, and the patient is well informed making thus making better decisions.

Therefore, EHR has improved healthcare delivery by making more patient information available to all medical providers, thus enabling them to make accurate decisions.

In the United States of America, patients’ information is collected through the National Institute of Health (NIH), which any medical provider can later access. The EHR is among the tools used with several advantages, including all healthcare professionals having access to the entire medical history of a patient (Boonstra, Versluis, Vos, 2014).

The EHR information includes diagnoses, prescriptions, data related to known allergies, demographics, clinical narratives, and results from various laboratory tests. Therefore, the treatment process becomes more efficient due to a reduction in the lag time of previous information.

Healthcare organizations have widely adopted EHR and is more advanced in the United States of America, the United Kingdom, Denmark, and the Netherlands. Thus, it is recommended that other countries, especially developing ones, adopt the system to improve the accessibility of information by various healthcare professionals.

Therefore, EHR provides information to all healthcare professionals, thus improving patient treatment.

Real-time alerting

Healthcare professionals require real-time information to carry on the treatment process effectively. Thus, wearable devices such as fitness trackers and wristbands are essential to the healthcare system due to their volume in providing doctors with information. The data from all sensors can be analyzed instantly.

If something goes wrong, the doctor will automatically alert the patient, and a specialist will be sent to the patient to give further instructions on what should be done. The real-time alert requires creating various software programs that will provide solutions from the Big Data solutions.

Real-time alerting is also used in hospitals; Clinical Decision Support (CDS) software analyzes medical data on the spot, providing healthcare professionals with advice on the prescription steps. Doctors also want patients to receive home care treatment to avoid costly in-house treatment; thus, personal analytical devices become crucial in such situations.

Therefore, real-time alerting is mainly used by wearable devices that allow doctors to monitor patient reactions, thus recommending appropriate actions if anything goes wrong.

Other healthcare professions can access the information acquired through real-time monitoring, including other factors such as socioeconomic context. This allows the doctor to modify delivery strategies according to the patient’s status, using the most efficient method to benefit the patient.

An example of a real-time alerting tool used is the Asthmapolis, whereby inhalers are fitted with GPS-enabled trackers to identify asthma trends in individuals and the broader population. Data acquired from this tool is used in conjunction with CDC data to develop better treatments for asthmatic cases.

Other areas such as blood pressure monitoring among people experiencing hypertension, are done using wristbands. Therefore, real-time alerting enables immediate healthcare responses from providers due to the patient’s availability of real-time data.

Opioids

Opioids, or narcotics, include a healthy prescription of pain relievers such as oxycodone, hydrocodone, fentanyl, and tramadol. The illegal drug heroin is also classified as an opioid. Some opioids are made from the opium plant and other synthetic artificial materials.

In Big Data analytics, behavioral health providers are leveraging big data to make a difference in drug consumption rates.

The process of pharmaceutical practices is changing through the incorporation of big data. According to Grape (2019), abuse of opioids is common among patients, and often the wrong prescription is given by healthcare providers due to a lack of enough patient information.

However, pharmacies’ adoption of big data provides the solution whereby the pharmaceutical facet is getting full of data, thus offering the right prescription to patients.

Big data has been used to reveal the abuse of drugs, which has become a particular interest for healthcare providers in light of the current opioid epidemic in the United States.

According to the Department of Health and Human Services, more than 3900 people use nonmedical prescription opioid drugs each day in the U. S., resulting in at least 78 people dying due to inappropriate use of opioid medications.

Therefore, through big data incorporation in the healthcare industry, incorrect prescription and use of opioid drugs are monitored and necessary steps taken, especially by the pharmaceutical sector.

The opioid epidemic is currently ravaging the world, whereby the overuse of drugs is increasing rapidly. However, the solution to the problem is through collaboration between different stakeholders through Big Data analytics to help identify and evaluate entities that contribute to the risk.

According to Grape (2019), traditional data analysis is often missed when tackling the problem, which could offer unique information about the trends of opioid abuse worldwide.

Suppose a step-by-step analysis is taken on the available data. In that case, the risks associated with manufacturing, prescription, and over-use of opioid drugs will be unmasked, helping curb the expanding epidemic. Therefore, a closer look into big data will help identify a solution to overdose using opioid medications.

Challenges of full adaptation of big data analytics in the healthcare industry

Big data management and analysis methods are being developed, especially those focusing on real-time data streaming, capture, aggregation, analytics, and visualization solutions.

Due to increasing information, challenges such as storage, certified analytical tools, data cleaning, unified format, accuracy, image pre-processing, security, meta-data, querying, visualization, and data sharing are adopting big data in the healthcare industry.

According to Dash et al. (2019), the high rate of changes in the healthcare industry leads to interoperability and data-sharing difficulties.

For instance, several EHR programs are available in the United States but are not certified since each contains different clinical terminologies, technical specifications, and functional capabilities.

While the healthcare industry is entering the post-EMR deployment phase, the main objective should be to focus on actionable insights from the vast amount of collected information. Therefore, big data integration in the healthcare industry faces various challenges as more information is gathered and new technology is implemented.

Increased information available in healthcare records leads to the primary challenge of storage. Although several organizations are comfortable with data storage on their facilities, the site server network manager is more expensive and difficult to maintain.

Therefore, some organizations opt for cloud-based services, which need to consider a company with the knowhow of patient information and comply with healthcare standards and security issues. Stored data often undergoes cleaning to ensure accuracy, correctness, consistency, relevancy, and purity after the acquisition.

However, the process may derail big data projects if the right and precise tools are not used. Besides, the traditional EHR format cannot capture the ever-increasing patient data, thus creating a need to codify relevant information surfaced for claims, billing purposes, and clinical analytics.

Patient data is shared between various healthcare professionals, whereby accuracy is needed to avoid wrong prescriptions and diagnoses, a common trend in healthcare organizations.

Therefore, the ever-increasing patient information in healthcare organizations faces various challenges that, if not handled correctly, will lead to inconveniences in the treatment process.

Case Studies

Case 1

The New York-Presbyterian Innovation Center plays an essential role in developing and applying internal and external innovation processes at the New York-Presbyterian Hospital.

According to Fleischut (2016), the innovation center’s primary focus is patient engagement and provider communication, linked with information from one entity to another.

Through the program, doctors are expected to have real-time, streamlined data, which will help them share insights with their peers and make life-saving decisions quickly and accurately.

From the patient’s side, they and their families are expected to know their diagnosis and the treatment plan, which will place them in a better position to participate in the recovery process. The hospital automatically updates patients and their families about operating and emergency room visits.

The work is guided by innovation principles centered on rapid, sustainable, scalable, mobile, and measurable driving big data analytics.

Therefore, the New York-Presbyterian Hospital employs significant data analytics principles to form a communication platform and patient engagement in this mandate to provide quality healthcare services.

Hospitals in the U.S. are investing in Big data and IT development through innovation centers. The focus of these centers is tech innovation that will improve healthcare services.

At the New York-Presbyterian innovation center, the hospital is being reimaged to what it can do to strengthen the world-class care of the hospital patients.

The hospital’s first startup sector was Blueprint Health Accelerator, which was used to bring technologies into use in 2015. The space shows more than 50 companies shared the blueprint, which was targeted at integrating various data into one system that will serve all patients.

Therefore, the New York-Presbyterian Hospital uses technology to introduce big data analytics in its design that will improve patient engagement and communication between patients and medics.

Case 2

Cabell Huntington Hospital (CHH), a 313-bed acute care hospital, serves patients in West Virginia, eastern Kentucky, and Southern Ohio. CHH developed a close working relationship with the School of Medicine at Marshall University to create a data warehouse with analytic tools.

The first step in forming a big data solution is creating an informatics technology that comprises partners from both collaborators who are CHH and Marshall’s University, through the chief Medical information officer. The research was carried out, and two approaches were identified to follow in their project.

The first was to develop a warehouse and purchase a pre-packaged data warehouse and analytic solution. The two entities decided to go for the first option, which faced various challenges in the implementation stage. Therefore, in-depth research must be carried out to determine the best approach to adopt a big data solution.

The relationship between CHH and Marshall faced the challenge of lacking the know-how to work with big data analytics. The two collaborators started the process without a skill set, expertise, or other experience when carrying out such a project.

This left the two teams on the learning curve until they gained enough experience and acquired the required skills to handle the data warehouse. The first thing that was to be done in the project was to hire a data architect who would be left responsible for building a data warehouse.

However, CHH hired two data analysts who work on the day-to-day duties of managing the data. So far, the staff evaluated different data sources, built data dictionaries, and normalized the data to put the data into one place for an analytical report to be produced.

Despite the challenges, the project aims to accrue benefits internally by improving internal operations, efficiencies, working denials, and improving work quality and safety.

The project also focuses on external services such as population health. Therefore, the project could only be a little more time before full benefits are realized due to the challenges faced at the beginning of the project.

Case 3: Diabetic patients

Medical care for patients with chronic diseases is a somewhat challenging process because many checks are required to be carried out frequently. The patients must also pay attention to various aspects such as diet, sports activity, medical analysis, and blood glucose levels.

Diabetic patients must measure blood pressure several times to ascertain that they are in the collect groups. Simultaneously, the patients will not always be present within the doctor’s vicinity to treat them if normal blood pressure measurements are observed.

Therefore, according to research by El Aboudi and Benhlima (2018), a real-time monitoring tool is required to monitor diabetic patients’ conditions. The device will read directly from all incoming data provided by every sensor.

Healthcare measures are then compared with user-defined thresholds to decide if the situation is abnormal or normal. The process is done by using a prototype system that detects the dangers of patients.

The course contains Spark streaming and MongoDB, which implement the emergency detection module. Spark reads data from MongoDB in a batch layer, which runs regularly, thus providing real-time data.

The proposal’s effectiveness was evaluated by carrying out several experiments. A cluster formed through three nodes in an identical setting was configured with an Intel Core i7-4470 processor to visualize patient medical parameters.

An open-source EHR generator was used to evaluate the proposal’s scalability, whereby medical patient data was obtained and loaded to MongoDB as an ISON file.

The request gives two-fold solutions to real-time monitoring of patients with diabetic diseases; one proposed a generic big data architecture for medical-based computing and stream computing, which simultaneously provides accurate predictions and online patient dashboards.

The implementation of the prototype solution followed this through the MongoDB and Spark programs. Therefore, advancement in the future in the proposal will help have real-time monitoring of patients with diabetes and an extension to other chronic diseases.

Case 4

Dignity Health, the fifth-largest health system in the United States and the largest hospital provider in California, uses Big Data and advanced analytic tools to predict potential early-stage sepsis cases. Identifying the disease at this stage, appropriate interventions will be taken, which are helpful.

Through the use of the Sepsis Bio-Surveillance Program, Dignity Health monitors over 120000 people every month in about 34 hospitals, managing more than 7500 patients with Sepsis’s potential.

The program collects data from all patients’ electronic records within its facilities, followed by using natural language processing and a rules engine to monitor factors that indicate a sepsis infection continually.

Since implementing the program, Dignity Health has significantly improved mortality and ICU length of stay for Sepsis patients.

At least 28 of the Dignity Health hospitals have recorded an average of 5 percent decrease in mortality rates because patients stayed for a lower time at the ICU. Therefore, the program is cost-saving in addition to being a life-saving program.

Conclusion

Big Data involves all available data disposable at a particular organization. The data include information gathered from the history of the organization.

Big Data is shaping various industries in the 21^st century, with most of them adopting the idea and employing different tools to analyze the data. Big data has gained momentum in the past two decades in the healthcare industry, with various organizations adapting the trend to improve healthcare delivery.

Most healthcare system information is mainly a patient’s information, which is chare among different healthcare professionals to track patient health history.

Big data is used through electronic health records, patient prediction devices, real-time alerting, and drug use monitoring (opioids). Various organizations, such as the New York-Presbyterian hospital, have formed innovation bodies to integrate technology with service delivery, thus advancing the Big data idea.

Other organizations are collaborating with inventing a big data warehouse, which will help improve service delivery. Lastly, hospitals are developing programs to help improve healthcare delivery to patients. Therefore, big data improve healthcare delivery by making all patient information to health care professionals.

References

Barnes, J. T. (2013). Big little history. Dialogues in Human Geography. 3(3) 297-302. https://doi.org/10.1177/2043820613514323

Beall, A. Big data in health care. SAS. https://www.sas.com/en_us/insights/articles/big-data/big-data-in-healthcare.html

Boonstra, A., Versluis, A., & Vos, J. F. (2014). Implementing electronic health records in hospitals: a systematic literature review. BMC health services research, 14, 370. https://doi.org/10.1186/1472-6963-14-370

Dash, S., Shakyawar, S.K., Sharma, M. Kuashik, S. (2019). Big data in healthcare: management, analysis, and prospects. Journal of Big Data 6(54). https://doi.org/10.1186/s40537-019-0217-0

EL aboudi, N., Benhlima, L. (2018). Big Data management for healthcare systems: Architecture, requirements and implementation. Admissions in Bioinformatics. https://www.researchgate.net/deref/https%3A%2F%2Fdoi.org%2F10.1155%2F2018%2F4059018

Fleischut, M. P. (2016). Big data and the future of the care continuum. Health Manager 16(1). https://healthmanagement.org/c/healthmanagement/issuearticle/big-data-and-the-future-of-the-care-continuum

Grape, R. (2019). Pinpointing patterns in opioid abuse and using data to fight fraud. Medcity News. https://medcitynews.com/2019/02/pinpointing-patterns-in-opioid-abuse-and-using-data-to-fight-fraud/?rf=1

Jain, D. Kumar, V. Khanduja, D. Sharma, Bateja, R. (2019). A detailed study of Big Data in Healthcare: Case study of Brenda and IBM Watson. International journal of Technology and Engineering 7(5). https://www.ijrte.org/wp-content/uploads/papers/v7i5/E1798017519.pdf

Landau, J. (2019). 4 Areas where big data is transforming healthcare right now. HIT Consultant. https://hitconsultant.net/2019/09/09/4-areas-big-data-transforming-healthcare/#.X7MkvWgzbIU

Manning, P. (2013). Big Data in history. Palgrave Macmillan. https://books.google.co.ke/books?hl=en&lr=&id=KtEeAgAAQBAJ&oi=fnd&pg=PP1&dq=big+data+history&ots=s85g6TFKjA&sig=-XysXKRzDTXCp0xTVa048dDLwuY&redir_esc=y#v=onepage&q=big%20data%20history&f=true

NEJM Catalyst, (2018). Healthcare Big Data and the promise of value base care. https://catalyst.nejm.org/doi/full/10.1056/CAT.18.0290#:~:text=What%20Is%20Big%20Data%20in,devices%2C%20to%20name%20a%20few.

Parslow, W. (2014). How big data could be used to predict a patient’s future—the Guardian. https://www.theguardian.com/healthcare-network/2014/jan/17/big-data-nhs-predict-illness#:~:text=In%20the%20US%2C%20big%20data,actions%20needed%20for%20each%20patient.

Pramanik, P. K. Pal, S. Mukhopahyay, M. (2018). Healthcare Big Data: A comprehensive overview. In Intelligent Systems for Healthcare Management and Delivery. IGI Global. https://www.researchgate.net/publication/327845528_Healthcare_Big_Data_A_Comprehensive_Overview

Schaeffer, C., Haque, A., Booton, L., Halleck, J. & Coustasse, A. (2016, April). “Big Data Management in United States Hospitals:Benefits and Barriers.” In J. Sanchez (Ed.), Proceedings of the Business and Health Administration Association Annual Conference, Chicago, IL. https://mds.marshall.edu/cgi/viewcontent.cgi?referer=https://www.google.com/&httpsredir=1&article=1154&context=mgmt_faculty