Drug accountability · Medication compliance count · Investigational product tracking
FREE
The free Clinical Research Planner — protocol snapshot with space for visit windows, daily clinical flow, PI signature tracking, and monitor-readiness checklists.
Download the free planner →Drug accountability, formally called Investigational Product (IP) accountability, is one of the most closely scrutinized aspects of any clinical trial. It is the systematic process of tracking every unit of investigational product from the moment it arrives at your site until it is either administered to a participant, returned to the sponsor, or destroyed. Regulators, monitors, and auditors view meticulous IP records as a direct indicator of site quality and data integrity.
The FDA's 21 CFR Part 312.62 and ICH E6(R2) Good Clinical Practice (GCP) guidelines require sites to maintain complete accountability for all investigational product. Failure to do so is one of the most common findings during FDA inspections and sponsor audits. Discrepancies, even small ones — can trigger a full audit, raise questions about participant safety, and in serious cases result in a Form 483 observation or warning letter.
Beyond compliance, IP accountability protects participants. If a patient experiences an unexpected adverse event, the accountability log helps the medical team confirm exactly what dose was dispensed and when — information that can be critical to clinical decision-making. It also prevents diversion of investigational medications, which are often unapproved substances with no commercial equivalent.
Enter the quantity received, dispensed, and returned/destroyed to instantly verify your accountability math. The calculator uses the standard formula: IP Compliance % = (Doses Dispensed / Doses Expected) × 100. Most sponsors require ≥80% compliance, with many oncology trials requiring ≥75% given toxicity-related dose holds.
Reconcile your IP log at every patient visit, not just at the close of the study. Catching a discrepancy immediately, while the patient is still available to clarify — is far easier than trying to resolve it months later during a monitoring visit. This tool is a data-entry aid; always verify your final numbers against your source documents and signed dispensation records.
Investigational product accountability is one of the most consistently cited findings in FDA Bioresearch Monitoring (BIMO) inspections of clinical investigator sites. The reason is straightforward: IP records are objective. Either the numbers reconcile or they don't. Unlike source documentation findings that require judgment about adequacy, an IP discrepancy is a mathematical fact, and inspectors know it.
Understanding not just how to calculate the compliance percentage but what specific errors generate findings, what distinguishes a discrepancy from a deviation, and what regulators actually look for when they review your IP log is what separates sites that sail through monitoring visits from those that accumulate corrective action plans.
IP accountability requirements are codified at 21 CFR 312.62, which requires investigators to maintain adequate records of the disposition of the drug, including dates, quantities, and use by subjects. This is an investigator obligation, not just a sponsor policy requirement. ICH E6(R3), Section 5.14, extends this framework to the sponsor side, requiring that the sponsor implement a system for IP accountability that ensures complete traceability from manufacture to subject to return or destruction.
Together, these requirements mean that at any point during a study, you should be able to account for every unit of IP: where it came from, where it went, and what happened to whatever wasn't used. Gaps in that chain are the source of findings.
The standard IP compliance formula is:
IP Compliance % = (Doses Dispensed ÷ Doses Expected) × 100
But "doses expected" requires clarification, and this is where errors originate. In most studies, doses expected is not a fixed number: it is calculated from the participant's dosing schedule over the accountability period. A participant who was on study for 28 days at once-daily dosing has 28 expected doses. If they had a protocol-specified dose hold for 5 days due to toxicity, the expected dose count drops to 23. If the dose hold is not correctly reflected in the IP log, the expected dose count will be wrong, and the compliance percentage will be wrong.
This is the most common source of IP compliance calculation errors: failing to document dose modifications, holds, and missed doses as distinct events with their own entries rather than simply noting a lower returned quantity at the next visit. Every departure from the standard dosing schedule needs its own log entry with date, reason, and authorization.
To see how the calculation comes together — and why the same record can yield two slightly different percentages — work through a single participant. Suppose a participant is dispensed 30 tablets of a once-daily oral IP on 1 March, returns to the clinic on 29 March, and brings back 3 unused tablets, with none lost or damaged.
Note: the participant and figures below are a hypothetical illustration chosen for clarity — not data from an actual study.
The compliance percentage is built in four steps:
Now the part that matters most. That 28-day window assumes the participant did not take a dose on the return-visit day — that the bottle was handed back before the 29 March dose. If the protocol instead expects a dose on the return day (for example, the participant takes it at home before traveling to the clinic), the return day counts as a dosing day and the window becomes 29 days. The same record then reads:
Neither figure is wrong. The roughly three-point spread between 96.4% and 93.1% comes entirely from a single day, and which one is correct depends on what the protocol says about dosing on the return-visit day. This is a factor that must be taken into consideration when reconciling an accountability log, and it is worth confirming the convention before a discrepancy is ever raised. Near an acceptability threshold — a sponsor minimum of 80%, say — that one day can be the difference between a result that passes and one that prompts a query, so the convention is not a rounding detail.
The calculator on this page exposes this directly through the "Dose taken on the return-visit day" option, which adds the return day (+1) to the window. Whichever convention your protocol uses, the four steps above let you reproduce any result by hand and confirm it against your source documents before finalizing the accountability log.
Not all IP discrepancies are created equal. Understanding the spectrum helps you prioritize what needs immediate action:
A quantity discrepancy occurs when the math doesn't work: received minus dispensed minus returned/destroyed does not equal zero (or whatever balance is expected). Small discrepancies sometimes resolve with documentation, a missed entry for a returned partial blister pack, for example. Unresolvable discrepancies, where no documentation supports reconciliation — are major findings. They raise the question of whether IP was diverted, taken by a non-enrolled individual, or administered outside the protocol.
In blinded randomized studies, IP kits are assigned specific randomization numbers. The kit number dispensed to the participant must match exactly what is recorded in both the IP log and the CRF. A transposed digit, kit 1023 recorded as kit 1032, is a major documentation error in a blinded study because it breaks the traceability chain. Even if the clinical impact is negligible, the finding cannot be resolved without a formal explanation and CAPA.
Most investigational products have defined storage requirements, a specific temperature range that must be maintained continuously from receipt to dispensation. When a temperature logger indicates that IP was stored outside its specified range, that IP must be quarantined and the excursion reported to the sponsor before any decision is made about whether the affected units can still be used. IP that was exposed to an excursion and then dispensed without sponsor notification is a serious finding, because the safety of administering out-of-spec IP to a participant cannot be assumed. The accountability log must reflect the quarantine, the sponsor notification, and the sponsor's disposition decision — use, return, or destroy.
When IP is destroyed on-site, whether because it expired, was returned used, or was involved in a temperature excursion, the destruction record must include: the date of destruction, the method, the name of the person performing the destruction, and a witness signature. An IP log that simply shows units "destroyed" with no supporting documentation treats destruction as equivalent to disappearance. Inspectors and monitors cannot verify what happened to those units.
IP dispensed to a participant after their documented discontinuation date is a serious finding with potential safety implications. If a participant withdrew on Day 42 and your IP log shows a dispensation on Day 45, the log is either wrong (a documentation error) or IP was actually administered after discontinuation (a protocol deviation with possible safety significance). Both require investigation, documentation, and likely IRB and sponsor notification.
These terms are often used interchangeably in practice but they are distinct. A discrepancy is an inconsistency within the IP records themselves, the numbers don't reconcile, or documentation is missing. A protocol deviation is a departure from the protocol's requirements — dispensing the wrong dose, dispensing outside a specified window, or dispensing to a participant who did not meet continued eligibility criteria.
A deviation always requires a deviation report. A discrepancy may resolve with documentation without rising to the level of a formal deviation, depending on its nature and the sponsor's requirements. When in doubt, discuss with the CRA or sponsor's clinical team before deciding whether a discrepancy warrants a deviation report, but don't delay that conversation.
At final closeout, every unit of IP must be accounted for across the study's full lifecycle. Under FDA's risk-based monitoring guidance, monitors typically perform a line-by-line review comparing the IP accountability log against CRF dispensation records, visit notes, and participant-signed drug accountability forms (where applicable). The final reconciliation must demonstrate that: all received IP has been either dispensed, returned to sponsor, or destroyed with documentation; all dispensations are supported by signed visit records; all dose modifications are documented; and the final compliance percentage is within the sponsor's acceptable range for each participant who completed the study.
Prepare for closeout by performing a self-audit of your IP log before the monitor arrives. Identify any gaps, discrepancies, or missing entries now: it is far less problematic to raise a question to your monitor before a monitoring visit than to have the monitor discover it independently.