Skip to main content
Promotional banner for the pentest readiness checklist
Snooping Alerts: Which Monitoring Model Fits Your Trust?Supervisory Authorities & Enforcement
5 min readFor Data Protection Officers (DPOs)

Snooping Alerts: Which Monitoring Model Fits Your Trust?

The ICO has contacted over 200 senior health leaders after receiving more than 300 reports of medical record snooping between January and mid-October 2026. These leaders must respond by November 27. Chief executive Paul Arnold seeks personal assurance on three points: who can access patient records, whether access is monitored, and what actions are taken when unauthorized access occurs.

If you're the DPO, you're likely drafting or reviewing that response. The tougher question is: what monitoring model can you honestly assure? This guide will help you decide.

The Decision You Are Facing

You need to determine how your organization detects and responds to inappropriate access by staff with legitimate credentials. This isn't about hacking. The risk involves staff like nurses or receptionists accessing records of acquaintances, ex-partners, or public figures.

Article 5(1)(f) of the UK GDPR requires appropriate security, including protection against unauthorized processing. Article 32 mandates suitable technical and organizational measures. Both apply to insider access. Trusting staff to behave isn't enough.

Key Factors Affecting Your Choice

Consider these factors before choosing a path.

1. What can your record system actually do? Some electronic patient record systems can flag suspicious access in real time. Others can't. Start with a factual inventory of your audit logging capabilities.

2. When did you last review safeguards? The ICO's letter uses the six-month mark as a benchmark. If you haven't reviewed safeguards recently, or you're dissatisfied with the results, you have three more months to address them. Your honest assessment determines your timeline.

3. Can you evidence who has access and why? Role-based access that aligns with job functions is essential. If broad access is granted by default, monitoring will be overwhelmed.

4. Do you have a working sanctions process? Detection without consequences fails the third question in the ICO's letter. Prompt investigation, deciding if patients must be informed, and taking action against staff are all expected.

5. What is your supervisory authority exposure? Unauthorized access can be a criminal offense under the Data Protection Act 2018. The ICO has prosecuted 16 healthcare workers over the past decade and has signaled enforcement against organizations with inadequate protections.

Path A: Real-Time Alerting with Proactive Audit

Choose this if: your record system supports real-time flagging, and you can assign someone to manage alerts.

What it involves:

  • Alert rules for known risk patterns, such as access to colleagues' records or high-profile cases.
  • Defined triage ownership, ensuring alerts reach a designated person promptly.
  • Regular sampling audits in addition to alerts, so you're not solely relying on pre-set rules.

Requirements driving this path: Article 32(1) expects measures proportionate to risk, and health data is special category data under Article 9. For large trusts with high-risk records, this model is defensible to a supervisory authority.

Watch for: alert fatigue. An unreviewed alert queue is worse than no alerts, as it shows awareness without action. Tune thresholds before implementation and document the process.

Path B: Retrospective Audit with Strong Access Governance

Choose this if: your systems log access but can't alert in real time, or your team can't support live triage yet.

What it involves:

  • Scheduled log reviews, risk-weighted rather than random, such as staff accessing records outside their service area.
  • Tight role-based access, minimizing who can open any given record.
  • Patient-facing routes to request access history, handled through your data subject access request (DSAR) process, giving patients a detection channel.
  • A written plan and timeline for transitioning to Path A when systems allow.

Requirements driving this path: accountability under Article 5(2) means you must demonstrate compliance. A documented, risk-based retrospective model with a remediation roadmap is defensible. Silence about the gap is not.

Watch for: detection lag. Retrospective review finds snooping after the harm. Your documentation should explain why this is proportionate now and what compensates for it, like stricter access controls and a visible deterrent message to staff.

Path C: Treat Your Current Position as Non-Compliant and Remediate First

Choose this if: you can't currently say who has access, logs are incomplete, or no one reviews them.

What it involves:

  • A candid gap assessment before providing assurance to the ICO.
  • Immediate access clean-up: remove dormant accounts, narrow broad roles, close shared logins.
  • A funded plan with owners and dates, aligned to the three-month window the ICO has offered for organizations dissatisfied with their safeguards.
  • A review of any open personal data breach assessments, including whether affected patients must be informed.

Requirements driving this path: if you provide assurance you can't evidence, you've created a new problem. Accuracy in your response is more important than speed.

Watch for: the temptation to overstate. An honest remediation plan is better received by a supervisory authority than a confident claim that doesn't hold up under scrutiny.

Balancing Access Needs with Privacy

Clinicians need fast access, especially in emergencies. Two controls can reduce friction without sacrificing protection. First, "break the glass" access: records outside a user's normal scope open only after stating a reason, which is logged and reviewed. Second, role design based on care relationships, not job title. Both preserve emergency care while making casual browsing visible.

Summary Matrix

Factor Path A: Real-Time Alerting Path B: Retrospective Audit Path C: Remediate First
System capability Supports live flagging Logs access, no live alerts Logging gaps or unknown
Resourcing Dedicated triage owner Scheduled review capacity Needs investment first
Detection speed Immediate Delayed Not reliable yet
Main legal anchor Art. 32, Art. 9 risk Art. 5(2) accountability Art. 5(1)(f), Art. 32
Key risk Alert fatigue Detection lag Overstating compliance
ICO response stance Confirm safeguards work Confirm plus roadmap Disclose gaps, use extra window
Sanctions process needed Yes Yes Build now

Whichever path you choose, the sanctions process and the honesty of your assurance are non-negotiable. The ICO has asked whether safeguards work in practice. Your answer needs to be one you can prove.

a promotional banner asking how ready are you for PCI DSS 4.0? With a call-to-action to get the checklist now.

You Might Also Like