A blank continuation row is not documented treatment failure
When a specialty pharmacy, regional health system infusion suite, or patient support hub submits a prior-authorization reauthorization packet for an ongoing specialty therapeutic, the continuation file may include a clinical review grid. Within that submission, managed care pharmacists and medical directors may encounter clinical outcome cells captioned “No response recorded,” “Evaluation omitted,” or left completely blank within the clinical response field. Treating that absence of evidence as evidence of absence—reclassifying the unrecorded row as an affirmative lack of efficacy—skips the identity check this worksheet exists to perform.
The fundamental operational reality for specialty-drug renewal reviewers is unambiguous: a blank continuation row is never documented treatment failure. When an incoming reauthorization packet arrives with an unrecorded response field, the evidence reviewer must classify the row across four distinct, mutually exclusive evidentiary states before anyone issues an administrative clarification request or codes the submission as clinical futility:
Missing scheduled assessment: An evaluation was intended under the therapeutic protocol or clinical encounter schedule, but the visit was omitted, canceled, or delayed, with missingness properly documented as an unperformed evaluation rather than therapeutic failure.
Present but noncomparable measurement: An assessment is documented in the submitted medical chart, but it utilizes an alternate rating scale, discordant observation window, different rater, or skip-logic pathway that cannot be mathematically or clinically compared against the baseline continuation criterion.
Documented nonresponse: A source record actually documents nonresponse or inadequate response on the named measure at a named date. A caption of “no response recorded” is not that record, and this worksheet does not invent a continuation cutoff.
Inaccessible source record: An evaluation was performed by an external health system, independent practice, or reference laboratory, but the primary record is not maintained within the designated record set of the submitting provider or specialty pharmacy.
To establish operational context, TrialEndpoints is a publication that covers clinical trial biostatistics, registrational study designs, estimand frameworks, and clinical outcome assessment (COA) measurement operations. It is not a contract research organization (CRO), biostatistical consultancy, commercial health plan, pharmacy benefit manager (PBM), or hub services vendor, and it does not maintain patient-level clinical records. In prospective clinical trials, biostatisticians develop formal frameworks to differentiate missing data from intercurrent events, as examined in TrialEndpoints' operating guide on missing data and sensitivity analysis under ICH E9(R1). That page is confirmatory-trial sensitivity analysis under ICH E9(R1). It is not this specialty-drug renewal worksheet. Clinical trial sensitivity models and statistical estimands do not dictate health insurance coverage criteria. An access reviewer cannot impute a response into a patient's record or recode a blank row as treatment failure.
This evidence-intake worksheet must also be kept distinct from downstream reauthorization appeals and trial-to-payer translation frameworks. Access leaders analyzing how registrational endpoints are structured for early formulary dossiers should review PharmaDossier's guide on translating patient-reported outcome endpoints into payer evidence, while teams confronting formal denials should refer to the GLP-1 reauthorization denial appeal workflow. Whereas an appeal workflow operates reactively after an adverse determination has been issued, this classification worksheet operates proactively at intake, ensuring that incomplete documentation is triaged under appropriate evidence-gathering rules rather than converted into premature treatment terminations.
Keep this receiving-side identity worksheet distinct from adjacent PharmaDossier jobs. GLP-1 reauthorization after BMI drops below the original threshold is continuation when a BMI number falls below an eligibility figure. Renewal denial after stable disease is how to answer a denial after stable disease. FHIR prior authorization fields that cause specialty-drug denials maps denial-triggering data fields. Treatment-regimen estimands in FDA labels is estimand language in labels and dossiers, not this four-row packet check. Digital health endpoints in FDA trials and the payer handoff gap is FDA DHT qualification versus payer HEOR. Biomarker documentation mismatch in oncology prior authorization is label versus NCCN versus payer biomarker proof. Step therapy exception evidence for specialty drugs is exception grounds after fail-first rules. A formulary change notice is not a patient-level coverage decision explains why a formulary notice is not a coverage determination. ChinaMedGlobal: checking access assumptions in a licensed-drug data room separates supplied licensor evidence from speculative U.S. payer assumptions. None of those pages classifies a missing assessment, a noncomparable measurement, documented nonresponse, and an inaccessible source before anyone codes treatment failure.
Record a missing assessment as not done, not as a numeric nonresponse
The regulatory distinction between an unperformed assessment and clinical failure is grounded in foundational federal guidance. In the FDA’s December 2009 guidance Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims (Docket FDA-2006-D-0362; HTML landing page content current as of 2019-10-17) is nonbinding guidance for medical-product labeling claims. The agency addresses the reality that trial participants sometimes fail to report for scheduled study visits, fail to complete questionnaires, or withdraw from a clinical trial before its planned completion. The resulting missing data can introduce bias and interfere with the ability to compare effects in the test group with the control group because only a subset of the initial randomized population contributes, and these patient groups may no longer be comparable. The guidance does not set specialty-drug continuation criteria.
The 2009 guidance does not instruct a reviewer to recode an omitted questionnaire or missed visit as treatment failure, lack of efficacy, or numeric nonresponse. Instead, the clinical trial protocol should describe how missing data will be handled in the analysis and can increase the likelihood that a trial will still be informative by establishing backup plans for gathering all treatment-related reasons for patients failing to report at scheduled times or withdrawing from a treatment or the clinical trial. Patients should remain in the clinical trial, even if they have discontinued treatment, and should continue to provide PRO data. In statistical analysis plans, missing items within a domain and missing entire domains or entire measurements are analysis-handling questions, including instrument-specific rules for how many items can be missing before a domain is treated as missing. Those SAP rules belong to medical-product trials used to support labeling claims. They do not create a specialty-drug continuation threshold, do not authorize filling a blank payer packet with an imputed response, and do not convert an absent scheduled assessment into documented treatment failure.
This principle was formalized into data exchange standards in the FDA’s November 2023 technical specification Submitting Patient-Reported Outcome Data in Cancer Clinical Trials (Version 1.0, November 2023; Docket FDA-2018-D-1216; HTML landing page content current as of 2023-11-03) is a nonbinding technical specification for submitting PRO data from cancer clinical trials. Section 3.1.1.2 recommends representing missing PRO data in the Study Data Tabulation Model (SDTM) Questionnaire (QS) domain. When an individual item response or a source-data summary score that cannot be calculated per the scoring algorithm is missing, the specification says the row should include QSSTAT = 'NOT DONE'. If the operational reason for non-administration is known, it must be recorded in the reason variable QSREASND; otherwise, QSREASND remains null or empty.
Those reason codes are examples, not a closed payer taxonomy. When the patient was not administered the PRO measure at an onsite visit or at a planned offsite PRO timepoint, examples include, but are not limited to, the patient was physically unable to complete the PRO measure due to an adverse event; patient refusal; patient did not provide; study site failed to administer or other site staff error; or technological problems with a PRO administered electronically. Table A2 shows a missed onsite visit due to hospitalization as QSSTAT = 'NOT DONE', QSREASND = 'HOSPITALIZATION', and a null completion timestamp (QSDTC). A blank or absent numeric score is therefore a documented missing record, not a numeric nonresponse and not a treatment-failure value. QSSTAT and QSREASND are an identity analogy for a not-done record. They are not a prior-authorization form field and not a continuation threshold.
When applied to specialty-drug reauthorization, these standards provide a vital framework. When an electronic prior-authorization submission or clinical chart contains a blank response field, the reviewer is looking at a not-done record—an assessment that was not administered or not transcribed—rather than a numeric nonresponse. Converting that omission into an affirmative finding of therapeutic futility recodes a blank as treatment failure. The 2023 specification does not instruct that recoding.
Separate a noncomparable measurement from a missing one
The second evidentiary challenge occurs when a measurement is documented in the chart but is not comparable to the named continuation measure. A different instrument, time window, rater, or skip-logic path that never asked the named item is not the same as an absent scheduled assessment. Reviewers can confuse a noncomparable assessment with a missing one, issuing clarification requests that ask for data the submitting clinic believes it already supplied, or coding failure when the named measure was never asked.
Noncomparability arises across four primary operational vectors:
Instrument discordance: If the named continuation measure is one instrument, a different scale in the same chart is not that measure. A Psoriasis Area and Severity Index (PASI) baseline cannot be subtracted from a later Physician Global Assessment, and a Crohn’s Disease Activity Index cannot be swapped for a rheumatoid arthritis Clinical Disease Activity Index even though both are abbreviated CDAI. Unknown comparability stays unknown. This worksheet does not invent a minimal clinically important difference or a payer continuation cutoff.
Assessment window discordance: If the named continuation measure is defined for a stated observation window, a submitted note from a different week is not that measure. An undated note leaves the window unknown. This worksheet does not invent a typical week-12-to-16 rule.
Assessor discordance: If the named continuation measure is a clinician-rated scale, a patient self-rating is not that measure. If it is a patient-reported scale, a clinician estimate is not that measure. Unknown rater identity stays unknown.
Conditional branching and skip logic: Section 3.1.1.3 of the FDA November 2023 specification treats PRO data not collected because of skip logic or computerized adaptive testing as separate from missing data. A question the instrument never asked is not a blank that can be recoded as nonresponse. In a renewal packet, a skip-logic path that never elicited the named continuation measure is a comparability identity question, not documented treatment failure.
This distinction also reflects the International Council for Harmonisation (ICH) E9(R1) addendum on estimands and sensitivity analysis in clinical trials (Step 4, 20 November 2019; adopted as FDA guidance in May 2021, Docket FDA-2017-D-6113; landing page content current as of 2021-05-11). ICH E9(R1) states that failure to collect relevant data should not be confused with the choice not to collect, or to collect and not to use, data made irrelevant by an intercurrent event. When an access reviewer identifies a noncomparable measurement, the worksheet classifies it as “Present but Noncomparable,” A targeted clarification can ask whether the named instrument, window, or rater exists. It does not recode the submitted different measure as treatment failure, and it does not invent a continuation cutoff.
Documented nonresponse still needs a measure, date, and source
Documented nonresponse is a separate identity from missing or noncomparable data. ICH E9(R1) allows an informative reason such as treatment discontinuation due to lack of efficacy to be handled as an intercurrent event rather than as missing data, but only when that reason is recorded. A caption of “no response recorded” is not that record. To keep documented nonresponse from collapsing into a blank, the worksheet still needs a named measure, a date, and a source statement—not an invented percent-change cutoff:
Instrument fidelity: The source must speak to the same named measure used for comparison, not a different instrument or an unasked skip-logic item.
Temporal validity: The source must carry an assessment date. If the date is absent, the window is unknown. This worksheet does not invent a typical maintenance window.
Objective non-improvement: The source must actually document nonresponse or inadequate response on that named measure at that date. Do not invent a continuation threshold, and do not recode a missing visit as that finding.
Under the ICH E9(R1) framework, an intercurrent event is defined as an event occurring after treatment initiation that affects either the interpretation or the existence of the measurements associated with the clinical question of interest. ICH E9(R1) notes that a prospective plan to collect informative reasons for uncollected data helps distinguish intercurrent events from true missing data. Specifically, what is casually recorded as “loss to follow-up” may in fact represent “treatment discontinuation due to lack of efficacy.” Where treatment discontinuation for lack of efficacy is confirmed, it is classified and analyzed as an intercurrent event under the prespecified estimand strategy, rather than treated as a missing-data calculation.
Crucially, ICH E9(R1) differentiates between discontinuation of therapy (an intercurrent event) and complete study withdrawal (which produces missing data). Furthermore, under composite estimand strategies in registrational trials, statisticians may assign predefined failure scores when an intercurrent event occurs, provided those scores meaningfully reflect the patient’s lack of clinical benefit. However, access reviewers must recognize that composite scoring is an analytical tool for clinical trial research. In the administration of specialty pharmacy benefits for individual patients, a reviewer cannot impute failure scores onto a live patient’s reauthorization request based on silence or missing visits. Documented nonresponse requires an explicit, dated, verifiable medical record entry confirming lack of efficacy.
An inaccessible source is a records problem; missing evidence is a clock, not a failure finding
When a reauthorization dossier arrives without the named clinical measure, the gap may be a records-location problem rather than a clinical omission. 45 CFR 164.524 is an individual right of access to protected health information in a designated record set. 42 CFR 422.568 and 423.568 treat additional evidence and supporting statements as clock questions. Neither converts an inaccessible file into documented nonresponse.
Under the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule, 45 CFR 164.524 gives an individual a right of access to inspect and obtain a copy of protected health information about the individual in a designated record set, with limited exceptions including psychotherapy notes and information compiled in reasonable anticipation of legal proceedings. Title 45 was displayed on eCFR as of 2026-09-03 (last amended 2026-08-31; this section reports no eCFR changes after 2017-01-03). Under §164.524(d)(3), if the covered entity does not maintain the protected health information that is the subject of the request and knows where the requested information is maintained, it must inform the individual where to direct the request for access. Under §164.524(c)(2)(iii), a covered entity may provide a summary or explanation in lieu of access only if the individual agrees in advance to that summary or explanation and to any fees. This section does not require a pharmacy, hub, or manufacturer access team to hold the designated record set, and it does not authorize reconstructing a missing clinic note as nonresponse.
A dispensing specialty pharmacy or hub may not maintain the primary clinical record of an external hospital or academic health system. When a reauthorization packet submitted by that holder lacks a physician clinic note or laboratory panel, the identity question is whether the current holder maintains the source, knows where to direct the request, or lacks both. An inaccessible source is a records-access problem, not proof that no assessment occurred and not proof that treatment failed.
Medicare Advantage and Part D clocks likewise treat missing evidence as a timing and records question, not as documented treatment failure. These are program-specific clocks. They do not create commercial-plan clocks and they do not convert silence on a clinical measure into lack of efficacy:
Medicare Advantage item and service determinations (42 CFR 422.568): Displayed on eCFR as of 2026-09-04 (Title 42 last amended 2026-08-13), paragraph (b)(2)(i)(B) allows an MA organization to extend an item-or-service organization-determination timeframe by up to 14 calendar days when the extension is justified and in the enrollee’s interest due to the need for additional medical evidence from a noncontract provider that may change a decision to deny an item or service. An incomplete source packet is therefore a reason to request evidence, not a documented nonresponse. For a service or item not subject to the prior-authorization rules in §422.122, the standard clock in (b)(1)(i) is 14 calendar days. Beginning on or after 1 January 2026, a service or item subject to §422.122 has a 7-calendar-day standard clock under (b)(1)(ii). Do not quote the older 14-day standard clock as if it still applied to those PA-subject items.
Strict non-extendable clock for Part B drugs (42 CFR 422.568(b)(3)): Paragraph (b)(3) requires notice as expeditiously as the enrollee’s health condition requires, but no later than 72 hours after receipt of the request, and that 72-hour period may not be extended under (b)(2). Do not apply the item-or-service 14-day additional-evidence extension to a Part B drug request.
Medicare Part D coverage determinations (42 CFR 423.568): Displayed on eCFR as of 2026-09-08, when a party makes a request for a drug benefit, the Part D plan sponsor must notify the enrollee as expeditiously as the enrollee’s health condition requires, but no later than 72 hours after receipt of the request. For an exceptions request, that 72-hour determination is clocked from receipt of the physician’s or other prescriber’s supporting statement. If a supporting statement is not received by the end of 14 calendar days from receipt of the exceptions request, the sponsor must notify the enrollee as expeditiously as the enrollee’s health condition requires, but no later than 72 hours from the end of those 14 calendar days. A missing supporting statement is therefore a clock and evidence-identity problem, not proof that treatment failed.
These 42 CFR 422.568 and 423.568 clocks are Medicare Advantage organization-determination and Part D coverage-determination rules. They do not create commercial-payer clocks, typical wait times, or a finding of clinical nonresponse. They also do not authorize imputing a score into a blank continuation row.
Evidence-gap classification worksheet and labeled fictional packets
The original decision asset is an evidence-gap classification worksheet plus labeled fictional packets that cannot be mistaken for an actual patient, coverage determination, or clinical finding. Copy as separate columns the named measure, the baseline used for comparison, the assessment date, the source-record location, whether the submitted measurement is comparable to that named measure, the missingness or not-done reason if any, the request owner, and a neutral clarification sentence that does not recode the row. The four rows below are hypothetical packets. Leave unknown fields unknown. Do not impute a score, invent a response cutoff, or write an automatic coverage conclusion.
| Measure | Baseline Used | Assessment Date | Source Record | Comparability Status | Missingness / Reason | Request Owner | Neutral Clarification Text |
|---|---|---|---|---|---|---|---|
| Hypothetical packet A: Clinical Disease Activity Index (CDAI) | Hypothetical Week 0 baseline recorded as 32.4 (packet fact, not a continuation cutoff) | Named Week 24 window; score blank / unrecorded | Electronic Prior Authorization Portal Form | Scheduled assessment not present | Missing assessment; not done. Packet note: visit delayed due to acute viral infection. FDA 2023 QSSTAT analogy only; not a PA field; not failure. | Clinic Prior-Authorization Coordinator | Please confirm whether a Week 24 CDAI was administered, or provide the rescheduled encounter date. Do not recode this blank as treatment failure. |
| Hypothetical packet B: Physician Global Impression of Severity (PGIS) | Hypothetical Week 0 PASI recorded as 18.6 (packet fact; named continuation measure is PASI, not PGIS) | Month 6 Clinical Follow-up | Outpatient Clinical Progress Note | Present but not comparable: 5-point PGIS is not the named PASI | Not missing; different instrument. Unknown whether a PASI exists on the named date. | Specialty Pharmacy Clinical Navigator | Submitted record contains a 5-point PGIS. Please confirm whether a Month 6 PASI exists. This worksheet does not invent a continuation cutoff. |
| Hypothetical packet C: Crohn’s Disease Activity Index (CDAI) | Hypothetical Week 0 Crohn’s CDAI recorded as 310 (packet fact, not a continuation cutoff) | Week 16 Maintenance Milestone | Audited Gastroenterology EHR Chart Encounter | Comparable on the named Crohn’s CDAI at the dated encounter | Documented nonresponse: Week 16 note states inadequate response on the named Crohn’s CDAI (packet score 318). No numeric continuation cutoff is invented here. | Medical Director / Prescribing Physician | Week 16 source documents inadequate response on the named Crohn’s CDAI at that date. Classify as documented nonresponse, not a missing assessment. No coverage conclusion or treatment change is written from this worksheet. |
| Hypothetical packet D: absolute neutrophil count (ANC) monitoring | Hypothetical pretreatment ANC recorded as 2,100 /µL (packet fact) | Month 9 Maintenance Monitoring Window | Dispensing Specialty Pharmacy Prescription File | Comparability unknown: current holder does not maintain the source | Inaccessible source; 45 CFR 164.524(d)(3) records-direction identity, not a failure coding | Health System HIM Records Liaison | Dispensing pharmacy does not maintain the external laboratory chart. If it knows where the designated record set is kept, tell the individual where to direct the request. This is not proof that monitoring did not occur. |
The same four identities can be walked as labeled fictional packets. They are internally consistent examples, not real cases, not coverage determinations, and not clinical findings:
The featured TrialEndpoints page remains adjacent measurement-identity reading on why a blank cell is not automatically an intercurrent event in confirmatory trials, as outlined in TrialEndpoints' analysis of missing-data assumptions across estimand frameworks. That article is not this specialty-drug renewal worksheet and is not a document.sources authority for insurance rules, coverage, response cutoffs, or clinical treatment changes.
Sources
The regulatory frameworks, technical data standards, federal administrative statutes, and methodological principles analyzed in this guide are drawn from verified primary source records and official agency publications:
U.S. Food and Drug Administration: Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims (Nonbinding guidance for industry, December 2009; Docket FDA-2006-D-0362; addressing missing visits, incomplete questionnaires, reason collection, and retaining discontinued patients in PRO data collection).
U.S. Food and Drug Administration: Submitting Patient-Reported Outcome Data in Cancer Clinical Trials (Technical specifications document, Version 1.0, November 2023; Docket FDA-2018-D-1216; defining SDTM QSSTAT = 'NOT DONE', QSREASND reason taxonomy, and distinguishing skip-logic and CAT routing from missing data).
International Council for Harmonisation: ICH E9(R1) Addendum on Estimands and Sensitivity Analysis in Clinical Trials (Step 4 guideline, 20 November 2019; defining intercurrent events, distinguishing treatment discontinuation from study withdrawal, and defining missing data within estimand strategies).
U.S. Food and Drug Administration: E9(R1) Statistical Principles for Clinical Trials: Addendum: Estimands and Sensitivity Analysis in Clinical Trials (Guidance for industry, May 2021; Docket FDA-2017-D-6113; 86 FR 26047; adopting ICH E9(R1) Step 4 standards into FDA regulatory oversight).
Electronic Code of Federal Regulations: 42 CFR 422.568 Standard timeframes and notice requirements for organization determinations (Centers for Medicare & Medicaid Services; governing 14-day additional medical evidence extensions under (b)(2), the 72-hour non-extendable Part B drug clock under (b)(3), and standard prior-authorization timeframes under §422.122).
Electronic Code of Federal Regulations: 42 CFR 423.568 Standard timeframe and notice requirements for coverage determinations (Centers for Medicare & Medicaid Services; establishing the 72-hour Part D determination timeline, prescriber supporting statement requirements, and 14-day exception evidence windows).
Electronic Code of Federal Regulations: 45 CFR 164.524 Access of individuals to protected health information (Office for Civil Rights, Department of Health and Human Services; establishing individual rights of access to PHI in designated record sets and redirection duties when entities do not maintain records).
