Staying Ahead of Potential Breachers
Proactive Measures and Security Layers
Another highly recommended action is to proactively make updates to the system’s security and infrastructure, like securing the actual data when it's sitting at rest. This involves assurance that information is distributed across multiple different layers while the minimum required data is being shared at each layer.
When the information is effectively viewed by the user, it's literally only shown from a single endpoint. The user interface itself only has access to the data in a very minimized format that is only specifically available to that authenticated user. Essentially, scope is limited to the extent that each authenticated user only gets access to data they’re authorized by their username and password to see.
Going up the chain, it’s meaningful to consider that the information isn’t saved anywhere in the user interface. It's neither cached nor stored in the user interface and the application itself is not saving sensitive data. What the application would potentially save is a key value that would practically leverage tokenization. You can read more about this in our other articles on medical billing portals.
The tokenization would consistently provide validation that a user was authenticated, without tying any PHI data to that user. Then that token would go back to the API which would essentially be in another layer, potentially behind a physical firewall. Those APIs are only accessible via a call from the application, which is inside the physical server as opposed to a user's browser. The application server can be whitelisted, which prevents someone outside of that whitelist from accessing the API data. The big idea is that multiple layers are present in the system and they collectively provide redundancy.
The actual application data that's hosting the PHI via HIPAA-compliant web hosting might be behind a separate physical firewall, which keeps data and the application itself encrypted at rest. When data is transferred, everything is sent securely over SSL. This set of multiple layers ensures nobody can get information unless they're properly authenticated and authorized. That constitutes a profoundly robust set of randomized strings that pose a great challenge to replication attempts. Penetration testing is meant to complete a running stream of validation that none of those different layers can be penetrated or exploited easily (and ideally, not exploited at all).
The infrastructure is set up with the intent to minimize exposure through every step of the process. Ultimately, if the above procedures are ongoing and continuously performed, then we can remain assured that we're -at minimum- following industry best practices. In other words, we're effectively removing or significantly reducing the risk of any security breach that could threaten the application.