This article was published as a part of the Data Science Blogathon.
Introduction
“Big data in healthcare” refers to much health data collected from many sources, including electronic health records (EHRs), medical imaging, genomic sequencing, wearables, payer records, medical devices, and pharmaceutical research. Its characteristics distinguish it from traditional electronic medical and human health data used for decision-making: It is available in extremely high volume; moves at high speed, and spans the massive digital universe of the healthcare industry; and because it comes from many sources, it has a very variable structure and nature.
What is Big Data in Healthcare?
The main sources of healthcare include payer records, EHRs, smart devices, genetic databases, and the government. Due to the diversity of the format, type, and context, big health data is difficult to merge into conventional databases, making it enormously challenging to process and for industry leaders to capitalize on its significant promise to transform the industry. Despite these challenges, several new technological advancements are making it possible to transform big healthcare data into useful and actionable information.
Despite these challenges, several new technological advancements are making it possible to transform healthcare data into useful and actionable information. With the right software tools, big data informs the movement toward value-based healthcare and opens the door to remarkable advances while reducing costs. With the wealth of information provided by healthcare data analytics, caregivers and administrators can now make better medical and financial decisions while providing ever-increasing quality patient care. However, the adoption of big data analytics in healthcare lags behind other industries due to health information privacy, security, classified data, and budget constraints.
Today, at least two trends encourage the healthcare industry to embrace big data. The first is the shift from a fee-for-service model that financially rewards caregivers for performing services to a value-based model of care that rewards them based on the health of their patient population. Data analytics in healthcare will enable the measurement and monitoring of population health, enabling this transition. The second trend involves using data analytics to provide evidence-based information, increase efficiency over time, and help better understand best practices associated with any illness, injury, or disease.
Using data analytics to provide evidence-based information will increase efficiency and help us better understand best practices associated with any illness, injury, or disease. Embracing big data in healthcare can transform the industry, moving it away from a fee-for-service model. Simply put, it can deliver on the promise of reducing healthcare costs while uncovering ways to deliver superior patient experiences, outcomes, and treatments.
Applications
Staying healthy and avoiding illness and other diseases are at the top of any priority list. Consumer products such as the Apple and Fitbit and Watch activity trackers monitor individuals’ physical activity levels and can also report on specific health-related trends. The resulting data is already sent to cloud servers and provides information to doctors who use it as part of their overall health and wellness programs.
Fitbit has already partnered with United Healthcare, which rewards its policyholders up to $1,500 yearly for regular exercise. Meanwhile, Apple HealthKit, CareKit, and ResearchKit use technology built into Apple mobile devices to help patients manage their conditions and allow researchers to collect data from hundreds of millions of users worldwide.
https://docs.microsoft.com
Healthcare Applications – diagnostics, prevention, Precision Med, research, reduced costs
The Expanding Diagnostic Service
This allows patients better access to specialist care. Mobile apps like Aetna’s Triage advise patients about their health status using aggregated data and can recommend patients seek medical care based on app input.
• In another of its healthcare initiatives, Apple teamed up with Stanford researchers to see if the Apple Watch’s heart sensor could be used to detect atrial fibrillation. This condition kills about 130,000 Americans each year. Apple may alert the wearer to seek medical attention if the device successfully detects an illness.
• Propeller Health uses a Bluetooth-enabled sensor that connects to inhalers and spirometers for people with asthma or COPD. The company monitors environmental conditions at sensor locations and sends messages to patients’ phones so they can understand better the causes of their symptoms and take measures to prevent issues. The company also sends notification reminders about when to take medication. Propeller reports that patients experience 79 per cent fewer asthma attacks and enjoy 50 per cent more symptom-free days.
• Based on the Excellence in Health Innovation report, prescription errors cost approximately $21 billion annually, affecting more than 7 million patients in the U.S. and leading to 7,000 deaths. Israeli startup MedAware is working with healthcare organizations to deploy their decision support tool that uses big data to detect prescription errors before they happen.
Cost Reduction
The greater insight medical data provides to physicians translates into better patient care, shorter hospital stays, and fewer admissions and readmissions.
• The Mayo Clinic is using big data analytics to identify patients with more than one chronic disease (comorbidity) likely to benefit from early interventions in nursing homes, saving them emergency room visits.
• Insights derived from big data analytics provide healthcare providers with clinical insights. It allows them to do necessary treatments and make clinical decisions with greater accuracy, eliminating the guesswork often involved in treatment and leading to lower costs and better patient care.
• Big data analysis in healthcare also contributes to better insight into patient cohorts most at risk of disease, enabling a proactive approach to prevention. In short, big data analytics in healthcare can identify outlier patients who use healthcare services far beyond the norm. It can pinpoint protocols and processes that provide substandard results or whose costs are too high compared to the results. By merging financial and clinical data, it can highlight the effectiveness and efficiency of treatment plans.
Data Lakes
Every researcher prefers to work with sample sizes of millions of values rather than just hundreds; the more information a large data sample contains, the better. The general term “data lake” is often used to describe a collection of big raw data; several developments are underway that promise to build what could be called a “data ocean” brimming with research and analytics opportunities.
https://docs.microsoft.com
For example, the University of Oxford’s Li Ka Shing Center for Health Information and Discovery provides access to the U.K. Biobank and plans to add 50 million electronic patient records.
• The European Medical Information Framework (EMIF) helps improve access to health data from the electronic health records of approximately 50 million Europeans and cohort datasets from participating research communities.
• Built in collaboration with academic and commercial organizations, it enables users to obtain information and make decisions about complex pharmacological issues.
• A division of Dutch multinational NV Philips collected more than 15 petabytes of data obtained from 390 million medical records, patient inputs, and imaging studies. Healthcare staff can access this extensive collection to obtain critical data to inform clinical decision-making.
• The National Institute of Health has established the Big Data to Knowledge (BD2K) program, which aims to bring big biomedical data to researchers, doctors, and others. Initiatives such as these will enable healthcare providers to improve patient care while addressing an unsustainable cost trajectory. It will also provide researchers with rich data and information for disease prevention and treatment.
Challenges for Implementing
Healthcare organizations face data challenges that fall into several main categories, including data aggregation, policies and processes, and governance. Let’s explore them.
Challenges in Data Aggregation
First, financial and patient data are often spread across multiple payers, hospitals, administrative offices, government agencies, servers, and file cabinets. It takes hard work to put this together and get all the data producers working together to create new data in the future. In addition, each participating organization must understand and agree to the types and formats of big data they intend to analyze. Aside from questions about the format in which it is stored (paper, film, traditional database, EHR, etc.), the accuracy and quality of this data must be settled. Has the data been recorded accurately, or have errors crept in, perhaps over many years?
Political and Procedural challenges
Once data is validated and aggregated, various process and policy issues must be addressed. HIPAA regulations require policies and procedures to protect health information. The task is complicated by authentication, access control, security in transit, and other rules. This multifaceted problem has been addressed to some extent by cloud service providers, perhaps most notably Amazon AWS, which offers HIPAA and Protected Health Information (PHI) compliant cloud services.
Managerial Challenges
Finally, realizing the promise of data analytics in healthcare requires organizations to adapt how they do business. Data scientists are likely to be needed along with I.T. staff with the required skills to perform the analysis. Some organizations may struggle with having to “rip and replace” much of their I.T. infrastructure, although cloud service providers alleviate some of these concerns. Clinicians and administrators may need time to believe in the never-before-seen advice that big data can provide.
Future Prospects
Just as executives in the business and industrial sectors claim that their initiatives have been successful and transformative, the outlook for healthcare is even more exciting.
Precision Medicine
The National Institutes of Health has sought to enroll one million people to volunteer their health information to the All of Us research program. This program is part of the NIH Precision Medicine Initiative. According to the NIH, the initiative aims to “understand how a person’s genetics, environment, and lifestyle can help determine the best approach to preventing or treating disease. The goals of the Precision Medicine Initiative focus on bringing precision medicine to all areas of health and healthcare at scale.”
IoT Wearables and Sensors
As mentioned above, have the potential to revolutionize healthcare for many patient populations – and help people stay healthy. A device or sensor may one day provide a direct, real-time feed into a patient’s electronic health record, allowing medical staff to monitor and then consult with the patient, either face-to-face or remotely.
Machine learning, a part of artificial intelligence that relies on big data, is already helping doctors improve patient care. IBM has already partnered with its Watson Health computer system with the Mayo Clinic, CVS Health, Memorial Sloan Kettering Cancer Center, and others. Machine learning, along with healthcare big
Conclusion
Using analytics to provide evidence-based information will increase efficiency and help us better understand best practices associated with any illness, injury, or disease. This can transform the industry, moving it away from a fee-for-service model. It can deliver on the promise of reducing healthcare costs while uncovering ways to deliver superior patient experiences, treatments, and outcomes.
- Data analytics to provide evidence-based information will increase efficiency and help us better understand best practices associated with any illness, injury, or disease. Embracing the use of big data in healthcare can undoubtedly transform the industry.
- It is helping the movement toward healthcare and opening the door to remarkable advances, all while reducing costs with the wealth of information provided by healthcare data analytics.
- Propeller Health uses a Bluetooth-enabled sensor that connects to inhalers and spirometers for people with asthma or COPD. The company monitors environmental conditions at sensor locations and sends messages to patients’ phones so they can better understand the causes of their symptoms and take measures to prevent attacks.
- The task is complicated by authentication, access control, security in transit, and other rules. This multifaceted problem has been addressed to some extent by cloud service providers, perhaps most notably Amazon AWS.
The media shown in this article is not owned by Analytics Vidhya and is used at the Author’s discretion.
