Health insurance companies have a lot to lose if the Genetic Nondiscrimination Act passes. Though most people may view health insurance companies negatively, their potential use of genetic data to adjust the costs of insurance sounds like a solid data driven product, on paper. Ethics and privacy are inseparable from big data and the relationship grows exponentially when we’re discussing the largest collection of data points that distinguish us from other people, the human genome. I chose health insurance lobbyists because I’m a staunch supporter of universal healthcare and I’m challenging myself to see things from the other side. I’d like to play devil’s advocate and play the role of the much-despised health insurance company.
Seeing things from the perspective of a Data Scientist for a health insurance company; there are many ethics questions faces in various parts of the data science life cycle. In the data collection phase, health insurance companies have to take measures to lawfully collect data from customers in order to create customized insurance pricing. One of the main issues concerning the Genetic Nondiscrimination Act, is the use of genetic data from a patient’s research study or other publicly available genomic data. In these such cases it’s possible a patient isn’t aware their genomic data can be accessed by an insurance company. The issue of patients consenting to the release of their genome for public use is a difficult one. Representing insurance companies means supporting making human genomic data easily accessible for public benefit. In order to attribute SNPs with diseases, likelihood for developing a disease, or allergy to medicine, vast amounts of genomic data must be referenced.
This data helps us decide how much to charge a patient for healthcare coverage. After going through the proper measures of data collection, using the statistics in accurate SNP pattern detection to mediate the cost of insurance in fact, lowers insurance rates for generally healthy people.
Health Insurance companies have to take extreme care to prevent data breaches, especially with data as extensive and personal as genomic data. During nearly any point in the data science life cycle a data breach can occur. Although the repercussions of these breaches can be traumatic, amplified by how detailed genomic data is, companies like ourselves take every opportunity to safeguard our customer’s information and strive to mediate the severity the impact of a data breach might have on a customer. Most healthcare information is protected under HIPAA and news laws to protect patients against data misuse isn’t necessary.
A major issue we health insurance companies and lobbyists face is that of data ownership. This is a tricky situation when considering genomes because we haven’t established standards for ownership. Would it be possible that an insurance company owns a person’s genomic data if they collected and paid for the services? It appears to be the clearest definition of ownership, however it’s understandable that patients would feel some rights to information about their own genes. Along with information ownership is the responsibility for information.
If data is owned by an insurance company, there might be some standards for the handling of that information. What types of modifications of uses does that data then provide? I think whoever owns a piece of data should have full, not partial rights to it.
From the perspective of a health insurance company, we feel it’s best to use a person’s genetic makeup to adjust insurance prices based on their probability of developing certain diseases. The opposite perspective, the side I’m personally on, believes genomic data should be treated like every other part of a person’s medical records, protected under HIPAA law. While it may seem like a great idea to base a person’s health insurance on their likelihood to get sick, this takes out of the equation the individual’s life choices. In addition, because someone is predisposed to a condition doesn’t mean they’ll suffer symptoms of the disease. In my knowledge of health insurance companies, I have no faith human genomic data would be used for “good” in their hands. I believe information about one’s self is owned by one’s self and should be respected as property.
In conclusion, I found that there are logical answers to both sides of the fence of what is and isn’t ethical in data science. Concerning genomic information, I found that insurance companies have very valid reasons for accessing a patient’s data in order to set insurance prices. Companies may intend to lower insurance rates for the majority of people, without considering that decisions based on a person’s genomes is discrimination. When erasing the human element of data it’s easy to agree that a genomic-based insurance plan would be logical. However, the detrimental impact of data breaches, illegally acquired data, and improperly controlled information outweigh the benefits of a genomic-based health insurance policy