ERA and Deposit Linkage Research

Research question 3: when do remittance and banking references support a reproducible relationship without proving posting correctness?

September 1 scope: freeze the population at a named UTC or local cutoff, define the unit of analysis, and retain the source export. Include ordinary, exception, reopened, access-limited, and unresolved records so the cohort does not describe only convenient completions.

Method: collect stable references, original status wording, event timestamps, source versions, assigned role, access limitations, and observed disposition. For this era and deposit linkage research, keep repeated events linked to one record and distinguish event time from entry, retrieval, review, and decision time.

Classification: use source-confirmed, conflicting source, missing evidence, access blocked, owner interpretation required, reopened, and unresolved at cutoff. Report raw counts and denominators next to rates, medians, or intervals. Explain every exclusion and denominator change.

Evidence boundary: an observed relationship establishes only what the retained records show. Sequence does not prove causation, a matching amount does not prove identity, transmission does not prove receipt or adjudication, and a balanced total does not prove component treatment is correct.

Role boundary: a Philippines-based support specialist may retrieve approved records, preserve chronology, reproduce calculations, classify evidence states, and prepare a neutral owner question. The specialist must not change claims or accounts, select codes, interpret coverage, move money, grant access, or promise outcomes.

Quality review 3: have a second authorized reviewer reproduce a mixed subset from the same references. Compare source selection, linkage, date handling, exclusions, and labels. Disagreement is evidence of ambiguity or transformation and should remain visible.

Limitations for study 3 include incomplete logs, payer-specific semantics, interface changes, access restrictions, non-random sampling, local policy, and evidence created after the observed event. The cohort cannot determine fraud, medical necessity, legal compliance, liability, or the correct financial disposition.

Conclusion: this era and deposit linkage research can improve source capture and owner handoffs when repeated with stable definitions. It cannot automate an authorized decision. Publish unknowns beside resolved cases and preserve superseded evidence for later review.

Sources checked September 1, 2026: CMS Medicare Claims Processing Manual — https://www.cms.gov/regulations-and-guidance/guidance/manuals/internet-only-manuals-ioms-items/cms018912

CAQH CORE Operating Rules — https://www.caqh.org/core/operating-rules

HHS HIPAA Security Rule — https://www.hhs.gov/hipaa/for-professionals/security/index.html

NIST SP 800-66 Revision 2 — https://csrc.nist.gov/pubs/800/66/r2/final