Form 8-K Item 4.01 Structured Data API

Auditor changes are disclosed under Item 4.01 of Form 8-K. The /form-8k endpoint returns structured data from these disclosures, and every filing that contains a parsed Item 4.01 carries the item4_01 object on the response. To retrieve the full universe of auditor-change announcements, search for filings where the item4_01 object exists at all (any value, including empty).

Results default to descending filedAt order, so the most recent disclosures appear first. The dataset spans from 2004 (when Item 4.01 was introduced) to the present, with over 25,000 disclosures available.

The /form-8k endpoint accepts a ticker filter on the issuer's trading symbol (ticker). To pull a single company's full history of auditor changes, restrict results to filings where the item4_01 object is present and ticker matches the symbol you care about, for example UGRO for urban-gro, Inc. or INFO for IHS Markit.

Sort ascending by filedAt if you want the changes in chronological order from oldest to newest. Note that the ticker stored on each record reflects the trading symbol at the time of indexation, so a company that has changed tickers historically may be more reliably tracked by CIK (see the CIK-based question).

Use the /form-8k endpoint with a date range filter on the filing-acceptance timestamp (filedAt). Restrict results to filings where the item4_01 object exists and where the value of filedAt falls between 2023-01-01 and 2023-12-31.

Because the total may exceed the 50-results-per-page limit, paginate by incrementing the offset until no more results come back or the 10,000 hard cap is reached. If the annual universe exceeds 10,000 records, split the year into smaller windows (months or quarters) and iterate.

Each disclosure carries the issuer's Central Index Key on the response (cik). On the /form-8k endpoint, restrict results to filings where the item4_01 object exists and cik equals the company's identifier — for example 1318605 for Tesla, Inc. or 1706524 for urban-gro, Inc. Leading zeros are stripped from the CIK in the index, so pass the integer form without padding.

If you only know the company by ticker and want the more stable CIK identifier, the /mapping endpoint resolves a ticker to its CIK; from there you can run the CIK-based query against /form-8k.

Every filing has a unique SEC accession number (accessionNo), for example 0001144204-10-033846. On the /form-8k endpoint, filter on that exact value to retrieve the one record. The match must be exact, including the hyphens.

A single accession number resolves to at most one structured Item 4.01 disclosure, so the response will contain either zero or one item under the data array.

The EDGAR form type is recorded on each record (formType). Original auditor-change disclosures carry formType of 8-K, while amendments that revise a previously filed disclosure carry formType of 8-K/A. To list only originals on the /form-8k endpoint, restrict formType to 8-K; to list only amendments, restrict it to 8-K/A.

An amendment typically references the same periodOfReport as the original and updates or clarifies fields such as the reason for the change or attachments referenced in the disclosure.

The reason the prior accountant's engagement ended is recorded as an enum on each filing (item4_01.engagementEndReason). Possible values are resignation, dismissal, dissolution, and declination to stand for reappointment. To find cases where the former auditor resigned on its own initiative, on the /form-8k endpoint restrict item4_01.engagementEndReason to resignation.

For example, a 2026 disclosure by UMeWorld Inc. records that J&S Associate PLT resigned effective December 31, 2025, with WSJ & Partners engaged shortly after as the replacement.

Use the same engagement-end-reason field on the /form-8k endpoint and restrict item4_01.engagementEndReason to dismissal. This isolates cases where the issuer terminated the auditor's engagement rather than the auditor resigning, being dissolved, or declining to stand for reappointment.

Dismissals are the most common reason in the dataset. Examples include urban-gro, Inc. dismissing BF Borgers CPA PC on May 6, 2024, and IHS Markit Ltd. dismissing PricewaterhouseCoopers LLP on July 12, 2016.

Use the engagement-end-reason enum on the /form-8k endpoint and restrict item4_01.engagementEndReason to the exact string declination to stand for reappointment. This separates voluntary non-reappointment by the auditor from outright resignation or dismissal by the issuer.

In declination cases, the audit firm chose not to seek another term despite the existing relationship being available to renew, which is treated differently for disclosure purposes than a mid-engagement resignation.

A boolean flag indicates whether the former accountant expressed substantial doubt about the issuer's ability to continue as a going concern (item4_01.goingConcern). On the /form-8k endpoint, restrict results to filings where item4_01.goingConcern is true. The narrative explanation behind the flag, when present, is in the accompanying details string (item4_01.goingConcernDetail).

Going-concern paragraphs are common in disclosures by small, early-stage, or shell-like companies. For example, AlumiFuel Power Corp's 2011 disclosure flags a going-concern paragraph noting no revenues and significant operating losses.

A boolean flag records whether the disclosure reported material weaknesses in internal controls over financial reporting (item4_01.reportedIcfrWeakness). On the /form-8k endpoint, restrict to filings where item4_01.reportedIcfrWeakness is true. The narrative explaining the weakness is in the accompanying details field (item4_01.icfrWeaknessesDetails).

Material-weakness disclosures often appear together with reportable events under Item 4.01 — for example, an auditor-change filing might note weaknesses related to the accrual of non-routine transactions or inadequate IT general controls impacting financial systems.

A separate boolean flag records whether previously reported ICFR weaknesses have been remediated (item4_01.remediatedIcfrWeakness). On the /form-8k endpoint, restrict to filings where item4_01.remediatedIcfrWeakness is true. The textual description of the remediation steps, when included in the filing, appears in the accompanying details field (item4_01.remediatedIcfrWeaknessDetails).

This flag is set independently from the underlying weakness flag, so a filing can have both item4_01.reportedIcfrWeakness and item4_01.remediatedIcfrWeakness set to true when the issuer is acknowledging a historical weakness while also confirming its remediation.

A boolean flag indicates whether disagreements with the former accountant are mentioned in the disclosure (item4_01.reportedDisagreements). On the /form-8k endpoint, restrict to filings where item4_01.reportedDisagreements is true. The specific disagreements, when itemized, appear in a list field (item4_01.disagreementsList), and a related flag records whether they were resolved before the change took effect (item4_01.resolvedDisagreements).

Most auditor-change disclosures report no disagreements, so this filter narrows the dataset to the small subset of contentious changes that may warrant closer review.

The type of opinion in the prior accountant's last report is recorded as an enum on each filing (item4_01.opinionType). The possible values are unqualified, qualified, and adverse. To isolate non-clean opinions on the /form-8k endpoint, restrict item4_01.opinionType to either qualified or adverse.

Qualified and adverse opinions are uncommon in the dataset relative to unqualified opinions, so this filter typically returns a small, targeted set of filings where the prior accountant flagged material issues with the financial statements before the engagement ended.

A disclaimer of opinion is recorded as a separate boolean flag (item4_01.auditDisclaimer), independent of the unqualified/qualified/adverse enum. On the /form-8k endpoint, restrict to filings where item4_01.auditDisclaimer is true. The narrative explanation, when present, is in the accompanying details field (item4_01.disclaimerDetails).

A disclaimer means the auditor was unable to form or refused to express an opinion, often citing insufficient evidence — a stronger signal than a merely qualified opinion.

The name of the firm being replaced is stored on each filing (item4_01.formerAccountantName). On the /form-8k endpoint, restrict item4_01.formerAccountantName to the firm's exact name as it appears in disclosures, for example BF Borgers CPA PC. The corresponding date the engagement ended is recorded separately (item4_01.formerAccountantDate), which lets you order the resulting list of departing clients chronologically.

This is the natural way to study the fallout from a regulatory action against an accounting firm. For example, after the SEC permanently barred BF Borgers CPA PC from practicing before the Commission in May 2024, many of its clients filed Item 4.01 disclosures shortly afterward — urban-gro, Inc. among them.

The name of the newly engaged firm is stored on each filing (item4_01.newAccountantName). On the /form-8k endpoint, restrict item4_01.newAccountantName to the firm's name (for example Ernst & Young LLP) and the value of filedAt to the date range you care about (for example between 2023-01-01 and 2023-12-31).

The date on which the new engagement took effect is recorded separately (item4_01.newAccountantDate), which you can use to refine the timing further or order the resulting client list chronologically by engagement start.

On the /form-8k endpoint, run one query per calendar year filtering on filings where the item4_01 object exists and the value of filedAt falls inside that year (for example between 2023-01-01 and 2023-12-31). The result's total.value field gives the count of auditor-change disclosures for that year directly, without needing to fetch the matching records.

Repeat for each year of interest to build the time series. Bulk historical analysis can use coarser windows (yearly or quarterly), while operational monitoring typically uses smaller windows (monthly or weekly).

On the /form-8k endpoint, restrict to filings where the item4_01 object exists and filedAt falls inside the period of interest, then paginate through the results and tally the value of item4_01.formerAccountantName on the client side. Sorting those tallies in descending order produces a ranked list of firms by number of departing clients in that window.

The API itself returns one disclosure per record, so the aggregation happens in the caller. If the period yields more than 10,000 records, split the date range into smaller sub-windows and combine the counts.

Use the same approach as for tracking departures, but aggregate on the new-firm side. On the /form-8k endpoint, restrict to filings where the item4_01 object exists and filedAt falls inside the period of interest, then paginate through the results and tally the value of item4_01.newAccountantName on the client side. Sorting those tallies in descending order produces a ranked list of firms by number of newly engaged clients in that window.

Note that some filings report a new firm engagement without naming a former firm (for example, a first-time auditor engagement), so the totals on the gained side will not exactly mirror the totals on the lost side over the same window.

Two date fields on each filing record the relevant timing: the date the former engagement ended (item4_01.formerAccountantDate) and the date the new accountant was engaged (item4_01.newAccountantDate). On the /form-8k endpoint, pull the relevant subset (for example all auditor-change disclosures within a year), then on the client side compute the difference in days between the two dates per filing and average across the population.

In many disclosures both dates are the same day (an immediate handoff), but in others — for example a 2026 UMeWorld Inc. disclosure where the former firm's resignation took effect on December 31, 2025 and the new firm was engaged on February 5, 2026 — there is a measurable gap that this calculation surfaces.

A boolean flag indicates whether the registrant consulted the new accountant before the formal engagement (item4_01.consultedNewAccountant). On the /form-8k endpoint, restrict to filings where item4_01.consultedNewAccountant is true. This identifies cases where the company sought the prospective auditor's views on accounting or auditing matters in advance, which carries specific disclosure obligations under Item 4.01.

In most of the dataset this flag is false, so the filter isolates a relatively narrow subset of filings where pre-engagement consultations were acknowledged.

A boolean flag records whether the change was approved by the board or audit committee (item4_01.approvedChange). On the /form-8k endpoint, restrict to filings where item4_01.approvedChange is false. This isolates the unusual subset of auditor changes that proceeded without board or audit-committee approval, or where the disclosure did not affirm that approval was obtained.

Most disclosures in the dataset record item4_01.approvedChange as true, since SEC rules generally require board or audit-committee involvement in auditor selection, so a false here is worth investigating closely.

Two free-text fields on each filing carry narrative content: a short summary of the disclosure (item4_01.keyComponents) and the detailed reason explaining the engagement's end (item4_01.reasonDetails). On the /form-8k endpoint, run a full-text search across these fields for the topic of interest — for example mentions of Sarbanes-Oxley partner rotation, fraud, or restatement.

Because the relevant phrasing may live in either field depending on the filing, search across both to maximize recall. For example, one disclosure records the reason for the change as inability to rotate partners as required by the Sarbanes-Oxley Act of 2002, which surfaces via a text search of item4_01.reasonDetails.

Every 8-K filing lists the items disclosed in the form (items), as an array of strings prefixed with the item number (for example Item 4.01: Changes in Registrant's Certifying Accountant and Item 5.01: Changes in Control of Registrant). On the /form-8k endpoint, restrict to filings where the item4_01 object exists and items also contains an entry starting with Item 5.01 for changes in control, Item 2.01 for completed acquisitions, or Item 1.01 for material definitive agreements.

A 2008 disclosure by Cornerstone Therapeutics, for example, contains Item 4.01 alongside Items 1.01, 2.01, 5.01, and 5.02 — a merger that triggered both an auditor change and other board and capital-structure changes in the same filing.

Each record carries the date of the underlying event separately from the filing-acceptance timestamp (periodOfReport). The event date represents when the auditor change actually took place, while filedAt represents when the 8-K was accepted by EDGAR (typically a day or more later). On the /form-8k endpoint, restrict the value of periodOfReport to the event window of interest, for example between 2024-05-01 and 2024-05-31.

This is the right filter when you care about market-relevant timing of the event itself — for instance, isolating auditor changes that happened in the days immediately following the SEC's bar of BF Borgers CPA PC in early May 2024.

The /form-8k endpoint defaults to descending order by filing-acceptance timestamp (filedAt), so the most recent filings appear first. To reverse the order, override the sort to ascending on the same field. The earliest disclosures in the dataset go back to 2004, when Item 4.01 was introduced.

If you need event-time ordering rather than filing-time ordering, sort ascending on the event date (periodOfReport) instead. The two orderings will differ slightly because filings are typically accepted by EDGAR a day or more after the event itself.

The /form-8k endpoint returns at most 50 records per call. To walk a larger result set, increment the starting offset by the page size on each successive call — 0, 50, 100, and so on — until the response returns no more results. The starting offset is capped at 10,000, which is also the maximum cumulative size of a single query's result set.

If the total reported in total.value exceeds the 10,000 cap, you cannot reach the remaining records by pagination alone. Split the search universe into smaller sub-ranges (typically using filedAt) and paginate each separately.

The /form-8k endpoint enforces a hard 10,000-record cap per query. To retrieve a larger universe, slice the query along the filing-acceptance timestamp (filedAt) into windows small enough that each individual window stays under the cap — month-by-month works for most periods, and weekly or daily windows handle the densest stretches. For each window, paginate from offset 0 in steps of the page size until you have collected the window's results, then move on to the next window.

If you prefer to slice by event date rather than filing date, the same approach applies to the event-date field (periodOfReport). Either way, combine the per-window result sets in the caller, deduplicating on the accession number (accessionNo) if windows overlap.

The narrative explanation of the reason for the change is captured in a free-text field on each filing (item4_01.reasonDetails). On the /form-8k endpoint, run a full-text search of item4_01.reasonDetails for keywords associated with regulatory action — for example barred, SEC order, PCAOB, or the name of a barred firm. The accompanying summary field (item4_01.keyComponents) often mirrors the same language, so searching across both improves recall.

For example, urban-gro, Inc.'s 2024 disclosure records the reason for dismissing BF Borgers CPA PC as the firm being permanently barred by the SEC from appearing or practicing before the Commission, which surfaces directly via a text search for barred against item4_01.reasonDetails.