Skip to the stories
AI & ML

Medical Data Breach Exposed Records Tied to 13.3 Million People

  • John Martinez
  • 959 reads

A breach at Perry Johnson & Associates exposed medical and identity data tied to more than 13.3 million people. The companies must pay $300,000 and strengthen security.

Medical Data Breach Exposed Records Tied to 13.3 Million People

The breach at medical transcription provider Perry Johnson & Associates (PJ&A) shows how gaps in basic security controls can turn a network intrusion into a large exposure of sensitive records. Files contained information tied to more than 13.3 million people, including about 4.2 million New York residents. PJ&A and four related companies agreed to pay $300,000 and make security changes; another $9.7 million could become payable if they fail to meet the agreement’s terms.

Digital Footprint Scan

The incident also exposes a difficult question for breach response: how can an organization determine the full scope of access when it hasn’t reviewed every potentially affected file? Investigators found that attackers accessed shared network storage containing personal and health information. The companies did not examine those files to identify whether they involved additional people, so the final number of affected individuals could be higher.

What the exposed files contained

The records included direct identifiers such as names, dates of birth, home addresses and Social Security numbers. They also contained medical record and hospital account numbers, insurance details, admission diagnoses, treatment dates, and clinical information from transcriptions. That clinical material could include test results, medications, treatment facilities and healthcare providers.

This combination matters because medical information can be useful in more than one kind of misuse. Identity details may support attempts to impersonate someone, while treatment and insurance information can make those attempts appear more credible. Health records also carry a practical risk: inaccurate information introduced through someone else’s use of an identity can complicate a person’s records and care.

The number in the agreement documents is larger than an earlier estimate of about 8.95 million affected patients. The newer count covers records associated with more than 13.3 million people. The additional files found in shared storage leave open whether even that figure captures the full exposure.

How the intrusion went unnoticed

The described failures were layered. Remote access did not require multifactor authentication, which adds a second verification step beyond a password. Passwords sent by email were nearly identical, sensitive files lacked file-level encryption, and security software detected malicious activity without alerting security staff.

Each weakness reduces the value of another safeguard. Similar passwords can make access easier to guess or reuse. Without multifactor authentication, a stolen or guessed password may be enough to enter remotely. Encryption at the file level can limit the usefulness of data taken from storage, but it was absent here. And detection software has limited practical value if its alerts do not reach people who can investigate and respond.

The companies discovered the attack after the attackers contacted them on May 2, 2023, and demanded payment. That sequence points to a gap between activity on the network and the organization’s awareness of it. For security teams, the operational lesson is that monitoring needs a clear path from detection to a human response, with named responsibilities and procedures for escalation.

Notification and required changes

PJ&A began telling its clients about the stolen database files in July 2023. Those clients then notified affected individuals from October 31, 2023, through February 8, 2024. The delay was found to violate New York requirements for timely breach disclosure. The payments are owed to the states; the agreement does not announce compensation for patients.

The companies must add multifactor authentication to individual accounts, administrator accounts and remote network access. They must also improve password management and limit employee access to the information needed for each person’s work. These controls address how users enter a system and what they can reach after they get in.

Other required changes focus on data protection and detection. The companies must encrypt personal and protected health information, configure monitoring to alert security staff automatically to suspicious activity, and use network segmentation to limit movement between parts of a network. They must also put measures in place to detect and prevent unauthorized data transfers.

The agreement calls for annual risk assessments and penetration testing, along with written incident response procedures. It also requires clearer rules for keeping sensitive information and securely disposing of it when it is no longer needed. Together, these measures cover prevention, discovery, containment and data handling—the stages that failed or proved incomplete in this incident.

What affected people can watch for

A breach does not establish that someone has misused a person’s records. Still, people who may be affected can review medical bills and insurance statements for unfamiliar appointments, tests, prescriptions or services. Unexpected entries are worth checking with the provider or insurer that issued the document.

People can also examine credit reports for accounts or debts they do not recognize. Messages about the incident deserve caution: a convincing reference to a breach does not prove that a sender is legitimate. Contacting a healthcare provider or insurer through contact information already known to the recipient can help verify a request before sharing information or making a payment.

For organizations handling clinical records, the lasting question is whether these controls operate as a connected system. Authentication, encryption and monitoring each help, but gaps between them can leave sensitive data exposed and delay a response. The additional files that were never reviewed make the quality of incident scoping another point to watch: a credible account of a breach depends on knowing which data was accessed and who it concerns.

Source: John Martinez · www.malwarebytes.com

From the same set