How the Nursing Home Violation Registry Is Built
This page documents the method behind the companion registry: the public records we draw on, how violation categories are mapped to governing statutes, the time windows and inspection types covered, the quality controls that gate each row, and the limits of what any registry of this kind can claim. I publish it so any Minnesota family, reporter, or fellow practitioner can audit the sourcing before relying on the data.
Scope note: This is a methodology document. It does not list facilities, findings, or outcomes. Specific facility records live on the MDH and CMS public portals. Statutory framing applies statewide under Minn. Stat. Ch. 144A and Minn. Stat. §§ 144G.90-.91.
Methodology Overview
Two Regulatory Systems, One Registry
Minnesota nursing home oversight runs on two overlapping surveillance systems. The state layer, operated by the Minnesota Department of Health Office of Health Facility Complaints (OHFC), investigates complaints and conducts state licensing surveys under Minn. Stat. Ch. 144A. The federal layer, run through CMS and delegated to MDH surveyors for Medicare- and Medicaid-enrolled facilities, produces the Statement of Deficiencies (CMS Form 2567) that feeds the public Care Compare file. Our registry pulls from both layers and reconciles them against two secondary streams: Ramsey County adult protection public filings under Minn. Stat. § 626.557, and published MDH oversight and corrections-of-deficiency reports.
The method below is written for Saint Paul and Ramsey County facilities. The framing applies statewide to any Minnesota nursing home, assisted-living setting under Minn. Stat. §§ 144G.90-.91, or home-care provider under Minn. Stat. § 144A.44.
Compilation Process
How a Category Row Becomes a Registry Entry
Each registry row starts as a category recognized by MDH or CMS, is tied to its governing statute, and is mapped to the clinical or operational signature the category produces in a medical record. The five-step sequence below runs every time a category or framework is added.
Identify the governing framework
Every category traces to a named statute, rule, or federal F-tag. Vulnerable-adult abuse and neglect: Minn. Stat. § 626.557. Licensing and complaint procedure: Minn. Stat. Ch. 144A. Resident dignity and restraint: Minn. Stat. § 144.651. Federal pressure-injury oversight, F-tag 686; staffing, F-tag 725; infection control, F-tag 880. A category with no statutory anchor does not enter the registry.
Source from the public record
Four public-record streams feed the registry: MDH OHFC investigative reports, CMS Care Compare Statement-of-Deficiencies files for Ramsey County facilities, Ramsey County adult protection filings, and MDH oversight reports. Unpublished anecdotes, private complaints, and attorney-work-product materials are excluded. A family’s own record request runs privately under Minn. Stat. §§ 144.292-.293, not through this registry.
Map to a clinical or operational signature
Each category is paired with the observable signature it produces in a medical or operational record. Pressure-injury categories map to ICD-10 L89 staging, Braden Scale documentation, and admission-versus-in-facility acquisition records. Medication-administration categories map to MAR gaps and adverse drug events. Staffing categories map to missed-medication-pass patterns and delayed call-light response logs. This column is the Pillar II bridge — it tells a family which records to request.
Apply time-window and inspection-type controls
The registry covers three inspection types: complaint investigations (opened on specific allegations), standard recertification surveys (the roughly nine-to-fifteen-month federal cycle), and revisit surveys (verifying a facility’s plan of correction). Coverage begins with the 1980 enactment of Minn. Stat. Ch. 144A and extends to the most recent published report. Records with missing inspection-type coding or undated custodian stamps are held until the gap is resolved.
Verify before publication
Every row is checked against the source document it cites. Statutory references are cross-read against the Minnesota Revisor’s published text. Federal F-tag numbers are cross-read against the current CMS State Operations Manual Appendix PP. Categories that fail any check do not publish. Amendments and F-tag renumberings trigger a re-verification pass and a version bump in the freshness footer.
Data Architecture
Inspection Types, Time Windows, and Source Files
The three inspection types behind the registry are distinct in trigger and scope. Reading them together produces a fair picture; reading any one in isolation is where misunderstandings start.
| Inspection Type | Trigger | Authority | Published Output |
|---|---|---|---|
| Complaint Investigation | Specific allegation filed with MDH OHFC or routed from county adult protection under Minn. Stat. § 626.557. | MDH Office of Health Facility Complaints | Investigative memorandum with substantiation finding. |
| Standard Recertification Survey | Scheduled on a roughly 9–15 month federal cycle for any Medicare- or Medicaid-certified facility. | CMS, delegated to MDH state surveyors | CMS Form 2567 Statement of Deficiencies. |
| Revisit / Follow-up Survey | Verification that a facility implemented its plan of correction after a prior deficiency. | CMS / MDH | Follow-up Form 2567 amendment. |
Source files are cited inline in the companion registry. Category labels there mirror the labels MDH and CMS use, so any row can be traced back to the originating public file without translation.
Quality Controls
Checks That Run Before a Row Publishes
Four verification checks gate every row. None are optional. The goal is a registry a reporter or opposing counsel can audit without finding a fabricated citation, a misrouted F-tag, or a category that drifted from its anchor.
- Statutory pinning: every category is locked to a named Minnesota statute or federal F-tag. A category with no anchor is held.
- Source-file traceability: each row cites the public-record stream (MDH, CMS, Ramsey County) and the document class (investigative memo, Form 2567, revisit amendment, adult protection filing).
- Clinical-signature review: every category maps to a documentable signature in the medical or operational record. Generic language (“poor care”) is rejected in favor of specific findings (“ICD-10 L89 stage-III pressure injury acquired in-facility”).
- Version control: statutory amendments and federal F-tag renumberings trigger a re-verification pass. The freshness footer below records every update date so readers can tell which version they are looking at.
Known Limitations
What This Methodology Cannot Do
The registry is a map of what the regulatory record contains, not a map of what happens inside every facility. Four limitations are real and readers should know them.
- Under-reporting.
- Peer-reviewed literature on elder maltreatment has documented for decades that regulatory filings capture a fraction of actual incidents. A facility with zero substantiated findings is not the same as a facility with zero problems — it is a facility whose problems have not yielded a public filing.
- Self-reporting bias.
- Several violation categories, including certain medication-administration and fall-prevention findings, rely on facility self-reporting inside the inspection process. Facilities with weaker internal reporting cultures produce less paper, not necessarily better care.
- Jurisdictional gaps.
- Assisted-living facilities operated under a much thinner framework before the Assisted Living Bill of Rights took effect on . Pre-2021 coverage is intentionally conservative; we flag rows where the record is incomplete rather than extrapolate.
- No facility-level rank.
- The registry does not rank or score specific facilities. Facility-level findings are available directly at the MDH and CMS portals; we point readers there rather than publish a scoreboard we do not own.
Transparency note: This is a living document. When a limitation above is narrowed by new data (for example, when the full post-2021 assisted-living dataset matures), the change is recorded in the freshness footer and an archived version is retained.
Why We Publish the Methodology
Method Before Verdict
Institutional-care harm cases collapse when the underlying record is treated as a mystery. The registry exists because the clinical-correlation work on every one of these files starts with knowing what the regulator tracks, where it lives, and which category a family’s concern fits into. That is the Pillar II contribution — converting a subjective observation into an objective, documentable pathway. The sourcing and chain-of-custody discipline behind the data is the Pillar I contribution.
This page is part of the evidence-based methodology behind our personal injury practice. The companion data product is the registry of institutional care oversight findings for Saint Paul and Ramsey County. Adjacent methodology resources include the Regions Hospital trauma intake family checklist for hospital-side record capture and the Ramsey County liability decision tree for injury cases for upstream causation logic.
Frequently Asked Questions
Common Questions About This Methodology
Why not publish facility-level rankings? +
A ranking requires weighting assumptions we do not have authority to make on behalf of every family. The underlying records live at the MDH and CMS portals, and readers can evaluate specific facilities there against the categories we document. A ranking layered over the regulator’s own data is a second opinion presented as the first one.
How do you handle under-reporting? +
We name it explicitly in the limitations section. When a category is particularly prone to under-reporting — for example, financial exploitation inside an institutional setting — we flag that context on the row in the companion registry. A clean regulatory record is a starting point, not an endpoint.
What happens when a statute is amended? +
Every affected row is re-verified against the amended text from the Minnesota Revisor, and the registry version is incremented in the freshness footer. F-tag renumberings trigger the same pass against the current CMS State Operations Manual Appendix PP. Rows whose citations cannot be re-confirmed are pulled until the confirmation clears.