# DenQAI Governed Owner Case-Series Pilot

Protocol version: 0.3.0-draft
Status: Intake closed; no recruitment, enrollment, submission, or case publication  
Reviewed: July 24, 2026

## Purpose

The proposed pilot would examine mechanisms that shape independent dental-practice ownership decisions and early owner survival. It is intended to improve evidence requests, decision tools, and future research design. It is not designed to estimate national prevalence, prove causation, adjudicate professional misconduct, rank named organizations, or evaluate patient care.

## Research questions

1. Which mechanisms create first-48-month practice and household cash strain?
2. How do guarantees, delay, capital calls, and unpaid management time change the ownership-versus-employment tradeoff?
3. Which advisor incentives or information boundaries changed a decision, and what contemporaneous evidence supports that account?
4. Which facts caused an owner to proceed, redesign, delay, renegotiate, sell, close, or walk away?
5. Which plausible alternative explanations survive a within-case evidence review?

## Proposed pilot boundary

- Population: adult U.S. current or former dental-practice owners with firsthand involvement in the decision being described.
- Initial ceiling: no more than 20 voluntarily enrolled cases in a maximum-variation pilot.
- Design: structured baseline and evidence index, with proposed 6- and 12-month follow-up.
- Extension: 24- or 48-month follow-up requires a revised protocol and renewed governance review.
- Recruitment: neutral and non-coercive; no lender, employer, broker, DSO, school, payer, advisor, or transaction counterparty controls participation or learns who declined.
- Compensation: not yet determined. Any future payment must be fixed, disclosed, modest, and independent of outcome or viewpoint.

## Explicit exclusions

The pilot will not accept:

- patient names, dates, record numbers, chart content, images, clinical notes, or other patient-level information;
- direct owner, practice, employee, advisor, seller, buyer, lender, vendor, or family identifiers in the analytic case file;
- exact addresses, ZIP codes, license/NPI/tax IDs, account numbers, or credentials;
- allegations presented as findings;
- source-document uploads before an approved secure environment and data-flow review exist;
- clinical-necessity, malpractice, coding, legal, tax, lending, valuation, or employment adjudication.

## Regulatory determination gate

The proposed activity is intended to develop knowledge beyond one owner’s decision. Under 45 C.F.R. § 46.102, a systematic investigation designed to develop or contribute to generalizable knowledge can be research. Interaction with living owners or use of identifiable private information can raise human-subject questions.

Before any recruitment or enrollment, DenQAI requires a dated written determination from a qualified independent reviewer or appropriate IRB/institutional official addressing:

- whether the activity is research;
- whether it involves human subjects;
- whether 45 C.F.R. Part 46 applies through funding, conduct, institutional assurance, or another basis;
- whether an exemption or other review path is available;
- whether state, local, professional, contractual, or publication rules add requirements;
- who owns continuing review and protocol deviations.

The protocol will not self-label the activity as exempt, not-human-subjects research, or HIPAA compliant.

Primary references:

- 45 C.F.R. Part 46: https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46
- OHRP 2018 decision charts: https://www.hhs.gov/ohrp/regulations-and-policy/decision-charts-2018/index.html
- OHRP quality-improvement/research FAQ: https://www.hhs.gov/ohrp/regulations-and-policy/guidance/faq/quality-improvement-activities/index.html
- Belmont Report: https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/read-the-belmont-report/index.html

## Independent evidence axes

Each material proposition records these separately:

- **Source kind:** direct record; reconciled summary; firsthand account; or inference/unverified report.
- **Support relationship:** directly supports; partially supports; context only; contradicts; or does not support.
- **Verification state:** verified; partially verified; unverified; or disputed.
- **Period and recency:** the broad period covered and the review date.
- **Review and limitation:** reviewer status, independent control, contradiction, material limitation, and public-use status.

A direct record is not automatically verified or sufficient for every proposition. A firsthand account can be important while remaining unverified. Missing evidence cannot be averaged away, and there is no composite evidence-completeness score.

## Analysis plan

1. Build a within-case chronology that separates what was known at the time from hindsight.
2. Tie turning points to dated evidence and owner-defined decision gates.
3. Record evidence against the preferred explanation.
4. Compare mechanisms across cases using a structured matrix.
5. Search for negative cases: similar exposure without the same outcome, and similar outcome through a different mechanism.
6. Preserve participant and reviewer disagreement.
7. Report bounded observations only. A purposive pilot cannot establish population prevalence or causation.

## Privacy and release design

- The future identity/contact key must be stored separately from coded case content.
- The analytic case file uses broad regions, quarters or years, and financial/time bands.
- Public release requires a separate transformation, rare-combination review, small-cell suppression rule, and qualified re-identification assessment.
- Removing names alone is not de-identification.
- Patient-level information is prohibited even if a contributor believes it has been de-identified.
- Retention, deletion, backup, logging, access, incident, and vendor controls must be approved and tested before intake.

HIPAA references:

- HHS de-identification guidance: https://www.hhs.gov/hipaa/for-professionals/special-topics/de-identification/index.html
- HHS research and PHI guidance: https://www.hhs.gov/hipaa/for-professionals/special-topics/research/index.html

## Participant control

The final consent process must cover:

- voluntary participation and the right to decline without penalty;
- purpose, procedures, time, follow-up, and foreseeable risks;
- compensation and conflicts;
- what data are collected, who can access them, and how they may be reused;
- participant fact-check of the structured case and attributed statements;
- the withdrawal cutoff and what cannot practicably be removed after aggregation or publication;
- complaint, correction, incident, and withdrawal contacts;
- new-information and protocol-change procedures.

## Launch gates

Enrollment remains closed until the public record shows dated evidence for:

1. independent regulatory determination;
2. frozen protocol, eligibility, analysis plan, and data dictionary;
3. approved consent and withdrawal process;
4. approved privacy, de-identification, and state-law review;
5. tested secure collection, access, retention, deletion, backup, and incident controls;
6. approved recruitment, compensation, conflict, and anti-retaliation rules;
7. approved participant review, publication, correction, small-cell, and withdrawal rules.

## Current product behavior

The public DenQAI case builder processes values in the active browser tab and exports a local draft to the user’s device. It does not upload, submit, enroll, consent, or approve a case. Users should not email a draft or source records to DenQAI.