Tech companies know too much about us. So, it makes sense that the conversation around digital privacy revolves around keeping our personal information encrypted and tucked away, where few can access it. The issue becomes more nuanced, however, when one considers healthcare data. On a moral level, it’s shocking that health and insurance providers control our medical information and thwart our access to it. On a practical level, healthcare data has the power to do a lot of good: If aggregated, anonymized, and properly regulated, the enormous amounts of medical data being recorded and digitized every day could revolutionize treatment plans across all fields of medicine. Although people may balk at the prospect of giving their private healthcare information to tech companies, granting access to patient data for consumer apps and AI research could fundamentally transform the American healthcare industry and deliver power and autonomy back to the patients themselves.
In February, the Department of Health and Human Services (HHS) proposed new rules surrounding its MyHealthEData initiative. These new rules will support seamless access, transfer, and use of patient data through standardized, machine-readable formats for electronic health information (EHI). For the first time, upon patient approval, health providers will be allowed to send patients’ EHI to third-party consumer apps. The Centers for Medicare & Medicaid Services (CMS) have long been committed to increasing accessibility of data for consumers. Now, they are proposing requirements that federal health insurance programs—Medicare, Medicaid, and any plans that participate in federal exchanges—be required to provide their 125 million enrollees electronic access to medical claims and other health information at no cost by 2020.
Not only will these rules expand patient choice by increasing market competition, they will also foster “innovation that promotes patient access to and control over their health information.” Currently, most medical organizations advocate for providers, and it is within providers’ interest to keep patient data hidden from their competitors. However, without access to pricing information and sometimes even their own medical data, patients are unable to make autonomous choices. With the newly proposed rules, healthcare providers and plans would be required to “open data sharing technologies to support transitions of care.” The rules would clear up discrepancies and confusion in the sharing of EHI, allowing all sides of the system—providers, caregivers, and especially patients—to securely access and share health information.
What’s more, the rules aim to provide a suite of new tools that will make healthcare data even more accessible. Standardized application programming interfaces (APIs) will make data available on smartphones, implementation of the 21st Century Cures Act will ensure that the information is free, and feedback sections on pricing information will allow the public to gauge pricing discrepancies when making decisions. Indeed, an interoperable health IT infrastructure across systems is long overdue and will allow all parties—patients, providers, and health IT developers—to obtain immediate access to data when and where it is necessary.
There are many reasons why it’s important for patients to control their own medical information. For one, such control eliminates redundant procedures and testing, thus giving clinicians the time to focus on improving care coordination, and, ultimately, health outcomes. Patients are also able to use their EHI to seamlessly switch to other providers who can provide them with the best standard of care. Indeed, CMS Administrator Seema Verma describes the new policy changes as “the beginning of a digital data revolution that truly empowers American patients.”
Despite these benefits, there are dissenting voices. Groups such as the American Medical Association (AMA) and the American Hospital Association (AHA), are stringently opposed to the newly proposed rules. They argue that because federal privacy protections do not apply to consumer apps, patients could unwittingly sign over sensitive data such as mental health history or prescription drug information. This data could then be sold and disseminated, eventually leading to employment or insurance discrimination. These groups claim that if patients worry about the privacy of their data, they are more likely to withhold sensitive clinical information from their doctors, potentially compromising the physician-patient relationship. For instance, patients tend not to disclose their substance abuse history, which may significantly impact the physician’s treatment plan for the patient.
Although these concerns are valid, the benefits of increased health data accessibility outweigh them. This is especially true in the age of artificial intelligence. Aggregated health care data has the potential to save millions of lives as it allows researchers to develop new medical treatments and improve clinical outcomes. Groundbreaking algorithms that could revolutionize the healthcare industry already exist; however, these technologies require exorbitant amounts of employee training and data testing, and in the current environment, even anonymized data is extremely hard to obtain due to HIPAA and bureaucratic hurdles.
Google has already begun to explore the use of artificial intelligence in diagnosing cancer and preventing blindness. Researchers recently used 45,856 chest CT scans, obtained from an NIH’s research dataset, to develop cutting-edge algorithms. In June, The National Academies published a report detailing a new way to combat cancer in the US by harnessing the power of data from cancer registries, electronic health records, insurance claims, sensor technologies, and even social media. Data from these sources can be used in advanced technologies such as cloud computing, machine learning, and artificial intelligence. This sort of approach can be adapted to all fields of medicine, offering the promise of higher profits, greater efficiency, more accuracy, and better performance.
For niche studies with no large-scale datasets, however, the prospects are currently limited. Small datasets do not suffice to train these technologies. But there is a way to get more data: If patients had more control over their own healthcare information, they could allow researchers to use their data to develop these new programs. Companies can incentivize such data-sharing by promising compensation in the form of money or tech-related health services. Of course, regulation is needed. But done correctly, AI can drastically cut costs and improve outcomes in the healthcare industry.
There’s a successful model for this type of voluntary-data-sharing: Facebook, Amazon, and other tech giants already track health-related consumer information in a process known as “digital phenotyping.” Researchers can now use the information that tech companies collect about our digital behavior to track our mood and cognition, which can, in turn, shed light on our physical and emotional health.
Hysteria about data breaches and privacy violations has taken over the news and our society; yet, considering the groundbreaking role our MRI scans or blood tests could play in medical innovation, perhaps we should rethink our hesitancy toward sharing our data. If my scan could help save someone’s life, I should be allowed to donate it to research.
Illustration by Liam Archibald ’20: instagram.com/liam_archibald_art