What Are the Limitations of Big Data in Healthcare?
There are some limitations of Big data in healthcare that companies need to be aware of
We have previously discussed some of the ways that big data has impacted the healthcare industry. Research shows that it can help healthcare providers deliver care more effectively and at a lower cost. However, there are some limitations of Big data in healthcare that companies need to be aware of.
How will big data affect healthcare?
WullianallurRaghupathi, Professor of the Graduate School of Business at Fordham University, has stated that big data is shaping the healthcare industry in promising ways. However, there are also some major limitations that need to be addressed. According to recent research from MapR, Big data is improving healthcare outcomes and reducing costs. Healthcare providers are able to access many different types of data on their patients, including:
- Medical claims
- Electronic health/medical record data
- Pharmaceutical research, such as clinical trials and genomic data
- Medical device information
Here are some ways this data is helping the healthcare industry:
- Reducing fraud. Fraud is responsible for about 10% of all annual healthcare expenditures. Data has helped healthcare providers improve their fraud detection models. The Centers for Medicare and Medicaid studies reportedly saved $210 million in one year by using big data predictive analytics models.
- Reducing readmission rates. According to the OECD, American healthcare providers need to improve treatment of chronic diseases. Patients with chronic conditions often need to make numerous trips to the emergency room, because their primary healthcare provider does a poor job with preventive care. Big data can help healthcare providers identify high-risk patients and lifestyle factors that need to be addressed.
- Minimizing overhead. The infrastructure of the healthcare industry is very expensive. It could be a lot cheaper if healthcare providers found ways to eliminate waste. Big data has made it much easier for them to tackle this problem.
- Improving healthcare product design. The future of the healthcare industry is also dependent on the companies that produce the tools they depend on. The FDA Medical Device Network has started using big data to help improve R&D. Even companies that sell ergonomic office chairs are starting to use big data to improve their designs.
Raghupathi states that more healthcare companies will invest in big data in the coming years.
What are the limitations of big data?
The benefits of big data are indisputable, but there are also some limitations that need to be discussed as well. Weighted Averages for Actuarial Models Insurance companies, hospitals and other healthcare organizations all depend on sound actuarial models for risk-management. Big data can help them improve their actuarial models to a point. However, there are some limitations. Tod Emerick and David Toomey of Insurance Thought Leadership points out that unstructured healthcare data is not normally distributed. This means that actuaries can’t use weighted averages in their models. Instead, they need to look at data sets for different subgroups. “In most discussions today, employers evaluate the average cost of employees with specific conditions, e.g., diabetes or high blood pressure. This is a flawed approach because spending by employees with various chronic conditions is skewed, thus not really “averageable.” For example, assume 90% of an employee population with diabetes is spending $10,000/year and 10% is spending $250,000/year; the average will be a meaningless $34,000/year. All too often, a wild goose chase ensues, when in fact the focus should be on the $250,000 cohort to understand why they were so much more expensive,” Emerick explains. Unfortunately, many big data extraction tools are still not equipped to analyze data on such a granular level. Difficulty Assessing Doctor Performance Insurance companies, the Center for Medicaid and Medicare Services and other actuaries carefully track the performance of different doctors in their networks. Unfortunately, their methodologies are very imprecise. The problem is that doctors often refer their highest risk patients to their colleagues to improve their own success rate. The best doctors often have subpar performance ratings, because they are responsible for all of the sickest patients. Data Availability and Reliability Big data healthcare models require reliable and detailed data sets. This means healthcare providers need access to as much data on their patients as possible. They also need to vet it carefully, because inaccuracies can destabilize their entire healthcare models. Social media, in particular is often unreliable, because patients are less likely to double check what they post on their profiles. They may even intentionally post inaccurate information to embellish their resumes or look more appealing to their friends.
Big Data is the Future of Healthcare – But Challenges Remain
Big data is changing the future of healthcare in many unprecedented ways. However, there are still limitations that healthcare providers need to overcome. Healthcare providers need to invest more in big data, but they must also be realistic about the limitations. Fortunately, many of these challenges will be addressed in the near future.
