Report: 30 Percent of Healthcare Databases Exposed Online

Hackers are using the Dark Web to buy and sell personally identifiable information (PII) stolen from healthcare organizations, and exposed databases are a vulnerable attack surface for healthcare organizations, according to a new cybersecurity research report.

A research report from IntSights, “Chronic [Cyber] Pain: Exposed & Misconfigured Databases in the Healthcare Industry,” gives an account of how hackers are tracking down healthcare personally identifiable information (PII) data on the Dark Web and where in the attack surface healthcare organizations are most vulnerable.

The report explores a key area of the healthcare attack surface, which is often the easiest to avoid—exposed databases. It’s not only old or outdated databases that get breached, but also newly established platforms that are vulnerable due to misconfiguration and/or open access, the report authors note.

Healthcare organizations have been increasingly targeted by threat actors over the past few years and their most sought-after asset is their data. As healthcare organizations attempt to move data online and increase accessibility for authorized users, they’ve dramatically increased their attack surface, providing cybercriminals with new vectors to steal personally identifiable information (PII), according to the report. Yet, these organizations have not prioritized investments in cybersecurity tools or procedures.

Healthcare budgets are tight, the report authors note, and if there’s an opportunity to purchase a new MRI machine versus make a new IT or cybersecurity hire, the new MRI machine often wins out. Healthcare organizations need to carefully balance accessibility and protection.

In this report, cyber researchers set out to show that the healthcare industry as a whole is vulnerable, not due to a specific product or system, but due to lack of process, training and cybersecurity best practices. “While many other industries suffer from similar deficiencies, healthcare organizations are particularly at risk because of the sensitivity of PII and medical data,” the report states.

The researchers chose a couple of popular technologies for handling medical records, including known and widely used commercial databases, legacy services still in use today, and new sites or protocols that try to mitigate some of the vulnerabilities of past methods. The purpose of the research was to demonstrate that hackers can easily find access to sensitive data in each state: at rest, in transit or in use.

The researchers note that the tactics used were pretty simple: Google searches, reading technical documentation of the aforementioned technologies, subdomain enumeration, and some educated guessing about the combination of sites, systems and data. “All of the examples presented here were freely accessible, and required no intrusive methods to obtain. Simply knowing where to look (like the IP address, name or protocol of the service used) was often enough to access the data,” the report authors wrote.

The researchers spent 90 hours researching and evaluated 50 database. Among the findings outlined in the report, 15 databases were found exposed, so the researchers estimate about 30 percent of databases are exposed. The researchers found 1.5 million patient records exposed, at a rate of about 16,687 medical records discovered per hour.

The estimated black-market price per medical record is $1 per record. The researchers concluded that hackers can find a large number of records in just a few hours of work, and this data can be used to make money in a variety of ways. If a hacker can find records at a rate of 16,687 per hour and works 40 hours a week, that hacker can make an annual salary of $33 million, according to the researchers.

“It’s also important to note that PII and medical data is harder to make money with compared to other data, like credit card info. Cybercriminals tend to be lazy, and it’s much quicker to try using a stolen credit card to make a fraudulent purchase than to buy PII data and run a phishing or extortion campaign. This may lessen the value of PII data in the eyes of some cybercriminals; however, PII data has a longer shelf-life and can be used for more sophisticated and more successful campaigns,” IntSights security researcher and report author Ariel Ainhoren wrote.

The researchers used an example of hospital using a FTP server. “FTP is a very old and known way to share files across the Internet. It is also a scarcely protected protocol that has no encryption built in, and only asks you for a username and password combination, which can be brute forced or sniffed

by network scanners very easily,” Ainhoren wrote. “Here we found a hospital in the U.S. that has its FTP server exposed. FTP’s usually hold records and backup data, and are kept open to enable backup to a remote site. It could be a neglected backup procedure left open by IT that the hospital doesn’t even know exists.”

According to the report, hackers have three main motivations for targeting healthcare organizations and medical data:

State-Sponsored APTs Targeting Critical Infrastructure: APTs are more sophisticated and are usually more difficult to stop. They will attempt to infiltrate a network to test tools and techniques to set the stage for a larger, future attack, or to obtain information on a specific individual’s medical condition.
Attackers Seeking Personal Data: Attackers seeking personal data can use it in multiple ways. They can create and sell PII lists, they can blackmail individuals or organizations in exchange for the data, or they can use it as a basis for further fraud, like phishing, Smishing, or scam calls.
Attackers Taking Control of Medical Devices for Ransom: Attackers targeting vulnerable infrastructure won’t usually target healthcare databases, but will target medical IT equipment and infrastructure to spread malware that exploits specific vulnerabilities and demands a ransom to release the infected devices. Since medical devices tend to be updated infrequently (or not at all), this provides a relatively easy target for hackers to take control.

The report also offers a few general best practices for evaluating if a healthcare organization’s data is exposed and/or at risk:

Use Multi-Factor Authentication for Web Applications: If you’re using a system that only needs a username and password to login, you’re making it significantly easier to access. Make sure you have MFA setup to reduce unauthorized access.
Tighter Access Control to Resources: Limit the number of credentials to each party accessing the database. Additionally, limit specific parties’ access to only the information they need. This will minimize your chance of being exploited through a 3rd party, and if you are, will limit the damage of that breach.
Monitor for Big or Unusual Database Reads: These may be an indication that a hacker or unauthorized party is stealing information. It’s a good idea to setup limits on database reads and make sure requests for big database reads involve some sort of manual review or confirmation.
Limit Database Access to Specific IP Ranges: Mapping out the organizations that need access to your data is not an easy task. But it will give you tighter control on who’s accessing your data and enable you to track and identify anomalous activity. You can even tie specific credentials to specific IP ranges to further limit access and track strange behavior more closely.