Big Data

The Use of Big Data in Health Insurance and its Potentially Dangerous Consequences

Health Insurance Data

Cathy O’Neil of Bloomberg has written a very interesting article on big data and how its application in the Health insurance industry could lead to undesirable consequences. The essence of insurance is risk pooling. In health insurance, you need both healthy and sick people to take up insurance for it to be affordable for the sick and profitable for the insurers. If it happens that only the sick take out insurance, they would be forced to pay very high premiums, failing which, the insurer will be operating at losses; and you can bet that the insurers will certainly seek to offset the costs by increasing the premiums.

Understanding the importance of risk pooling is one thing and transforming it into a policy is a different thing altogether. Indeed, Gregory Mankiw, an economist at Harvard notes that there are many obstacles. Firstly, sick people are more likely to take out health insurance and have more information about their health than insurers do. Secondly, coverage encourages people to use more health care thus increasing the expense incurred by the insurer.

Big Data could be used to separate the healthy from the sick. Artificial intelligence and algorithms are capable of identifying patterns of behaviour from which certain outcomes can be predicted. If health insurers use this technique, they would then sort people seeking to be insured according to the expected cost of their health care. Incidentally collected data such as public records, shopping history, demographic data as well as social media data sets can be used to assess peoples’ health. A perfect example is the LexisNexis product known as Socioeconomic Health Score which seeks to predict costs by using such data as criminal records, education and personal finance.

In the U.S., corporate wellness programmes and different startups are gathering information on doctor-patient interactions and ongoing surveillance of patients. It is estimated that healthcare data is increasing by 48 percent every year. Predictive analytics carries the potential to improve healthcare. Artificial intelligent systems will be able to access large volumes of data and based on the symptoms or health status, help doctors to prescribe the most appropriate treatment or tests. As a result, a doctor will spend less time with a patient making it possible to attend to many other patients.

Insurance companies, like any other business, are keen on making profits. If they can predict your healthcare costs, they will definitely increase premiums depending on your risk unless forbidden by law. As a result of targeted increase in premiums, it would become too expensive for those considered riskiest forcing them to drop out. Consequently, only the relatively healthy people would be covered, defeating the whole purpose of insurance, which is risk pooling.

Since the use of big data in the health sector has both pros and cons, the best approach would probably be to limit its application where it is likely to result in undesirable consequences. For instance, the law could prohibit the application of artificial intelligence in assessing health insurance risks. However, the extent to which such a law could be effectively enforced is debatable.

To Top

Pin It on Pinterest

Share This