How Clinics Should Respond to a Compromised Microsoft 365 Account

Kevin Henry

Incident Response

Published Updated

8 minutes read
Share this article
How Clinics Should Respond to a Compromised Microsoft 365 Account

How should a clinic contain access to a compromised Microsoft 365 account?

Act quickly, but coordinate the response so you contain the account without disrupting patient care. Notify your IT or security lead and privacy officer through a trusted channel, confirm the affected user’s identity independently, and record when the clinic discovered the compromise.

If an attacker may still be signed in, block new sign-ins and revoke the user’s sessions. Microsoft’s emergency access guidance describes how to disable an account and revoke sessions in Microsoft Entra ID. In an Entra-synchronized environment, coordinate changes with on-premises Active Directory so they are not overwritten or left incomplete.

Whether to disable the account depends on the risk and the user’s role. Disabling it can be appropriate to stop further access, but plan for essential work such as patient scheduling, referrals, and secure clinical communications. Arrange an approved alternate workflow or authorized coverage before disabling the account when circumstances allow.

Revoking sessions is important, but it may not immediately end every application session: access can depend on token expiration and how each application handles revocation. Treat containment as a set of steps, check for remaining access, and preserve relevant logs and evidence before cleanup when doing so will not prolong an active threat.

How should the clinic reset credentials and remove unrecognized authentication methods?

After blocking access, reset the password from a trusted, uncompromised device. Give the user a unique password through a secure channel, and require a change at next sign-in if your account process supports it. Do not send a temporary password to the affected mailbox or rely on contact details supplied in a suspicious message.

Review the user’s multifactor authentication methods in Microsoft Entra ID. Remove any phone number, authenticator app, security key, or other method the user does not recognize. Microsoft’s authentication-method guidance explains how administrators can reset a password, require MFA re-registration, revoke sessions, and manage methods. Re-registration can remove registered methods, so coordinate a secure way for the verified user to enroll again.

Check for and delete unrecognized app passwords as well. These can allow some older applications to authenticate separately from the user’s normal sign-in flow; removing them can disrupt applications that rely on them, so check for legitimate dependencies and restore access only after confirming the need and user.

How can staff check for unauthorized mailbox forwarding and inbox rules?

Inspect both mailbox-level forwarding settings and inbox rules. Attackers may use mailbox forwarding rules to send copies of messages elsewhere, or create rules that move, delete, or hide incoming mail. In Exchange Online, check the mailbox’s forwarding configuration and all enabled and disabled inbox rules, including rules with unfamiliar names or actions.

Microsoft documents how to review current rules with Get-InboxRule and search the Microsoft Purview audit log for rule changes in its guide to identifying who modified mailbox rules. Also review mail-flow or transport rules that could affect the account’s messages.

Record suspicious rule settings and available audit details before removing them, when it is safe to do so. Then disable or remove unauthorized forwarding and rules, and verify that legitimate messages are arriving normally. Microsoft warns that automatic external forwarding can increase takeover risk in its guidance on Exchange Online email forwarding.

Ready to simplify HIPAA compliance?

Join thousands of organizations that trust Accountable to manage their compliance needs.

Get Started Free Today

How should investigators review sign-in logs, audit activity, and message traces?

Set a time window that begins before the suspected compromise and continues through containment. Save relevant records and note timestamps and time zones. Review Microsoft Entra ID sign-in logs for successful and failed sign-ins, including interactive and non-interactive activity. Compare IP addresses, devices, applications, authentication details, and timing with the user’s expected activity. Location information is a clue to investigate, not definitive proof of where a person was.

Search the Microsoft Purview audit log for account and mailbox activity around the same period, including rule changes and relevant mailbox actions. Microsoft explains how to search for mailbox activities in specific mailboxes. Export or otherwise preserve useful results; audit availability depends on the organization’s configuration, license, and retention policies.

Use Exchange Online message trace to investigate message delivery and related events. Microsoft’s message trace guidance explains how to check whether a message was received, rejected, deferred, or delivered. Message trace data is retained for 90 days by default, so start searches promptly and save relevant results.

Build a timeline that brings sign-ins, audit events, mailbox changes, and message activity together. A missing log entry does not by itself prove that no activity occurred; record gaps and limitations and seek additional evidence where possible.

How can the clinic determine whether suspicious emails were sent from the account?

Search message trace for outbound messages during the suspected access period. Narrow the search by sender, recipient, and time, then review delivery status and event details. Compare the results with the user’s Sent Items, drafts, and available audit records; do not rely on Sent Items alone, because messages or other evidence may have been deleted.

For every suspicious message, record the recipients, time, subject, attachments, and available message identifiers. Determine whether it contained malicious links, requests for credentials or payment, or patient information. A trace can help establish message flow and delivery status, but it does not by itself establish who used the account or whether a recipient viewed the message.

If messages reached recipients, coordinate with your security and communications leads to contain or remove them where possible and warn recipients through a verified channel. Do not reply to the suspicious message or use contact details in it to confirm the incident.

How should clinics assess possible exposure of protected health information and maintain patient care?

Work with the privacy officer, security lead, and appropriate legal or compliance support to determine whether protected health information (PHI) may have been accessed, used, or disclosed. Review relevant mailbox contents, attachments, forwarding destinations, and sent messages, while considering what the logs can and cannot show about access or viewing. Document the evidence reviewed, the timeline, containment actions, and any remaining uncertainty.

A compromised account alone does not automatically establish a reportable HIPAA breach. Under HHS’s Breach Notification Rule guidance, an impermissible use or disclosure of PHI is generally presumed to be a breach unless the organization demonstrates a low probability that PHI was compromised through a risk assessment. That assessment considers the nature and extent of the PHI, who received or used it, whether it was actually acquired or viewed, and how the risk was mitigated.

If unsecured PHI was involved and the incident is determined to be a breach, follow the applicable notification requirements and timelines. HHS states that affected individuals must be notified without unreasonable delay and no later than 60 days after discovery of a breach. Document the assessment and decisions, including the basis for any determination that notification is not required.

Maintain patient care using approved alternatives while access is restricted—for example, an authorized colleague or another secure communication workflow. Confirm that essential messages and appointments are covered, and restore the user’s access only after credentials and authentication methods are secured and the clinic’s response lead approves it.

In short, contain access, secure credentials and authentication methods, investigate mailbox and sign-in activity, trace suspicious messages, and document a fact-based PHI risk assessment while keeping clinical workflows running.

FAQs.

Should a clinic disable a compromised Microsoft 365 account during an investigation?

Disable it when necessary to stop ongoing access, especially if an attacker may still be active. Consider the user’s clinical responsibilities and arrange an approved coverage workflow so disabling the account does not interrupt patient care. Blocking new sign-ins and revoking sessions are key containment steps, though existing access may not end instantly across every application.

How can a clinic revoke active Microsoft 365 sessions?

An administrator can select the user in Microsoft Entra ID and use Revoke sessions, which invalidates refresh tokens and prompts reauthentication. If the threat warrants it, also block new sign-ins and reset the password. Microsoft notes that the time for access to end can vary by application and token behavior; see its guidance on revoking user access.

What should a clinic do if the account sent suspicious emails?

Contain the account, preserve evidence, and use Exchange Online message trace and available mailbox records to identify messages and recipients. Remove or contain malicious messages where possible, and warn recipients through a verified channel. Assess whether any messages included PHI and record the actions taken.

How should a clinic assess whether protected health information was exposed?

Review relevant messages, attachments, forwarding destinations, and available sign-in and audit evidence to determine what PHI may have been accessed or disclosed and by whom. The clinic’s privacy officer should document the HIPAA breach risk assessment, considering the nature and extent of the PHI, the recipient, whether it was acquired or viewed, and mitigation. A compromised account by itself does not decide whether notification is required; apply the organization’s breach-response process and HHS guidance.

Share this article

Ready to simplify HIPAA compliance?

Join thousands of organizations that trust Accountable to manage their compliance needs.

Get Started Free Today

Related Articles