A piece published on SuspectFile by security researcher Marco A. De Felice argues that the dominant response to cybersecurity incidents — identifying and attributing the threat actor — consistently draws attention away from a more fundamental problem: organizations accumulate and centralize far more sensitive data than they need, creating conditions where a single successful intrusion produces disproportionate harm. For healthcare organizations subject to HIPAA's minimum necessary standard, the argument maps directly onto longstanding compliance obligations that many covered entities and business associates have not fully internalized.
The structural problem De Felice identifies
De Felice's framing centers on what he describes as structural fragility — a condition in which the architecture of data collection, centralization, and retention is itself the precondition for large-scale exposure, regardless of which threat actor eventually exploits it. The argument is not that attribution is worthless, but that attribution answers the question of who while leaving the question of why so much data was available largely unexamined.
For healthcare practices, this distinction carries regulatory weight. OCR's minimum necessary standard under the HIPAA Privacy Rule requires covered entities to limit the protected health information they collect, use, and disclose to what is reasonably necessary for a given purpose. In practice, many organizations treat that standard as a documentation exercise rather than an architectural discipline applied to how data flows are designed and how long records are retained.
Why centralization amplifies breach impact
The analysis highlights centralization as a specific risk multiplier. When sensitive records are aggregated into a single platform or repository — a common pattern in EHR consolidation, revenue cycle outsourcing, and third-party analytics — a single point of compromise becomes a single point of mass exposure. This dynamic explains why healthcare breach notifications increasingly involve millions of records rather than hundreds: the data was already assembled in a form that made bulk exfiltration straightforward.
De Felice's argument implies that incident response plans built around detection and containment address the breach after structural conditions have already determined its potential scale. Organizations that have not audited what data they hold, where it sits, and whether it is still operationally necessary are essentially managing the consequences of an earlier architectural decision they may not have consciously made.
What this signals for compliance operations
The piece's practical implication is that data minimization and retention scheduling deserve the same operational attention that practices routinely give to access controls and endpoint security. Several areas warrant direct examination:
- Retention schedules in business associate agreements. Many BAAs specify that a business associate will return or destroy PHI upon contract termination but say little about how long data may be retained during the active relationship. Reviewing those terms against actual retention practices closes a gap that often goes unnoticed until an incident occurs.
- Legacy data stores. Archived records from defunct service lines, historical billing data, and de-commissioned application exports frequently remain on accessible infrastructure long after their operational purpose has ended. Periodic inventory of data assets — not just active systems — is the mechanism for identifying and eliminating unnecessary exposure.
- Third-party data flows. Centralized analytics platforms, population health tools, and clearinghouses receive ongoing feeds of PHI. Whether those flows are calibrated to minimum necessary, and whether the receiving party's own retention practices are contractually constrained, are questions that compliance reviews should address explicitly.
Where incident response doctrine falls short
De Felice's broader critique applies to the healthcare sector's incident response culture as much as to any other vertical. Post-incident reviews in healthcare tend to focus on the attack vector, the dwell time, and the notification timeline. Fewer organizations formally assess whether the volume of records exposed was a foreseeable consequence of data practices that could have been structured differently.
Regulators have begun signaling interest in exactly this question. OCR resolution agreements over the past several years have included corrective action plan provisions addressing risk analysis and the scope of PHI maintained by covered entities — language that reflects concern not just with how data was protected but with whether so much data needed to be held at all. Organizations that treat De Felice's structural argument as an abstract intellectual exercise rather than a compliance planning prompt are likely underestimating where regulatory scrutiny is heading.