How Do We Find Out Why the Same Findings Keep Repeating?
Reviewed October 2026
In IHS's reading, find out by comparing the failed work across surveys, testing why the previous correction failed, and checking whether the new correction changes actual performance. In IHS's reading, a team with access to its records and authority to change the work can run this five-day diagnostic itself; bring in counsel when the finding carries legal exposure, and a consultant when the team wants its causal reasoning challenged.
Repeated findings call for a causal investigation before another round of training. The same citation does not establish the same cause. The five-day diagnostic below is IHS's proposed adaptation of published root-cause guidance. It is a work plan for investigating a bounded problem, not a regulatory deadline or a promise of correction within a week.
What should we group together on the first day?
Group findings by the work step that failed, while keeping each original citation attached. That is IHS's proposed first-day method. Record the survey date, applicable requirement and version, observed failure, prior correction, and evidence used to close it. Keep unlike failures separate even when they share a citation.
Then compare the scope and sample behind each finding. Ask whether the same task failed again, whether the review reached a different location, or whether the requirement changed. Choose one recurring problem for the week's investigation, prioritizing potential harm; address any immediate risk through your existing response process. CMS's Five Whys tips advise including people with personal knowledge of the processes and systems involved in the problem; in IHS's reading, the week's team should therefore include the people who do the work, and not only those who manage it.
What should we reconstruct on the second day?
Reconstruct what people actually did, including the point where the previous correction should have worked. In IHS's proposed second-day review, compare a failed instance with a successful instance of the same task. Trace the handoff, responsible role, available information, workload, and record created at each step.
The AHRQ-hosted PSNet primer describes reconstruction using records and participant interviews before multidisciplinary analysis. It discusses serious adverse events; applying that approach to recurring survey findings is IHS's inference.
When should we stop asking why?
In IHS's reading, stop when the proposed explanation survives a recurrence test, not when the team reaches its fifth question. CMS's Five Whys tool notes that it can take more than five questions and warns that the method may miss the root cause. It suggests fishbone analysis to consider broader categories. CMS also distinguishes a contributing factor from a root cause by asking whether correction would still leave recurrence likely.
For day three, IHS proposes testing each answer against the reconstructed work. If correcting the stated cause would leave another route to failure, investigate that route too.
Is the cause process, training, staffing, or evidence?
Use those four categories as competing explanations, not labels to assign by instinct. In IHS's diagnostic, a process explanation needs an identifiable failure in ownership, sequence, or handoff. A training explanation needs evidence that the person lacked the knowledge or skill needed for the task.
A staffing explanation needs a connection between available capacity and the missed work; compare workload and coverage when the task succeeded and failed. An evidence explanation needs a distinction between work performed but not retrievable and work never performed. The PSNet primer cautions that interacting system failures can be obscured by concentrating on isolated causes.
What should change on days four and five?
Select a correction for each supported cause on day four, then define how to test it on day five. CMS's RCA guidance ranks training among weaker actions and system changes among stronger actions. It also calls for a responsible person and timing for each action. The same guidance places a new procedure, memorandum, or policy among the weaker actions, so, in IHS's reading, a policy revision alone is a weak answer to a repeat finding.
In IHS's method, test the suspected mechanism: clarify an unowned handoff, demonstrate a missing skill, adjust constrained coverage, or make an existing record retrievable. Have the responsible clinicians approve changes affecting clinical work. Define the eligible work to be checked, the failure being counted, the review interval, and who acts on another failure. Record failures out of opportunities checked, including the period covered. A quiet week with little eligible work cannot establish sustained correction.
What evidence should we hand the surveyor or accreditor?
In IHS's reading, a response showing the problems were corrected documents what was done, by whom, and by when, while a plan to prevent repeat findings also ties each action to a tested cause and adds monitoring results showing the correction held. Prepare a record connecting the finding to its cause, the implemented correction, and subsequent performance. This is IHS's proposed evidence packet, not an acceptance checklist. Include the original finding and requirement, factual reconstruction, supported causal explanation, approved action with owner and completion date, and dated evidence that the change occurred.
Add monitoring results with the sampling method, eligible population, review dates, exceptions, and follow-up decisions. CMS's guidance evaluates implementation, adherence, and effects over time; it does not establish a universal submission packet, and it tells facilities accredited by the Joint Commission, or in states with regulations governing RCAs, to refer to those requirements. In IHS's reading, the client should check the recipient's applicable instructions before sending the relevant evidence. These sources are silent on whether a particular surveyor or accreditor will accept that packet.
What this is NOT
This is a diagnostic for recurring findings, not an emergency response protocol or a substitute for a required investigation. It does not guarantee that a surveyor or accreditor will accept a correction.
This is not legal advice. Seek counsel's review when legal exposure, reporting obligations, or disclosure of investigative material needs a legal determination.
The CMS materials are QAPI tools; the RCA guidance uses resident and facility examples. CMS expressly says its Five Whys tool is optional and completing it does not establish regulatory compliance. This page does not extend these tools to other programs as a requirement.
Frequently asked questions
Do we need a consultant to run this review?
You may not. In IHS's reading, an internal team with access to records, process knowledge, and authority to change the work can run this diagnostic. A consultant can challenge unsupported explanations and help map evidence to corrections. IHS drafts analysis and readiness materials for client review; the client sends them. IHS never contacts or represents a client before an accreditor, regulator, or payer.
Does another training session close the finding?
In IHS's reading, training alone does not demonstrate sustained correction. Use it when a demonstrated knowledge or skill gap explains the failure, then check performance afterward. A signed attendance record documents attendance; it does not test whether the work changed.
About the author
Thomas G. Goddard, JD, PhD, has been Chief Executive Officer of Integral Healthcare Solutions since 2002. He was Chief Operating Officer and General Counsel of URAC from 1998 to 2000.
Integral Healthcare Solutions: accreditation consulting, compliance services, and program development.
Sources
Five Whys Tool for Root Cause Analysis. Centers for Medicare & Medicaid Services. https://www.cms.gov/medicare/provider-enrollment-and-certification/qapi/downloads/fivewhys.pdf
Guidance for Performing Root Cause Analysis (RCA) with Performance Improvement Projects (PIPs). Centers for Medicare & Medicaid Services. https://www.cms.gov/medicare/provider-enrollment-and-certification/qapi/downloads/guidanceforrca.pdf
Root Cause Analysis. PSNet, hosted by the Agency for Healthcare Research and Quality; editorial primer reviewed in 2024, not an official agency policy statement. https://psnet.ahrq.gov/primer/root-cause-analysis
Last updated: October 31, 2026. Sources were retrieved on October 30 and again on October 31, 2026 for the sentences revised that day. Requirements change and must be verified with the issuing body.