By Justin Yamashita, MSc. Benchtop, site, CRO: three levels of basic and clinical research, explained without spin.

From Open Label Media, the team of one behind Root to Rx, comes a quiz to find out your type of skepticism, plus four free interactive games for discussions about claims made online and telling what’s true.

All four play in your browser in about two minutes at openlabelmedia.com, and I’d love to see your results on social media.

Issue 014 was about disclosure, the information your clinician needs and often does not get. Issue 013 took the drug commercial apart on screen. This issue follows the information in the other direction: how the risk data on a drug label is generated, who verifies it, and what happens to that system when it is underfunded.

Broadcast direct-to-consumer prescription drug advertising is permitted broadly in only two countries, the United States and New Zealand. In the US, advertising rules require fair balance: an ad that states a drug's benefits must also present its major risks. The rapid adverse-event list is not voluntary candor. It is a legal requirement, and it is drawn from the drug's approved labeling.

Every line on that side-effect list is the fingerprint of a specific job. Read the trial from first dose to final label and you can name who was watching at each step.

Before anyone enrolls, the medical monitors, physicians and pharmacovigilance teams across both CRO and sponsor set the safety rules for the study: what counts as an adverse event, what counts as serious, and the conditions that stop the trial. Every later judgment traces back to these definitions.

At the site, the investigator, clinical trial coordinator, trial nurses, data coordinators are all involved when a participant reports anything, a headache, a rash, a night in the hospital, the site logs it, whether or not anyone believes the drug caused it. The investigator assesses each event; the coordinator records it, and sometimes a separate data coordinator enters it into the trial’s electronic data capture (EDC) software for collecting and managing patient data digitally.

Checking the site, the clinical research associate (CRA). A monitor compares the record against the source, the real charts and lab reports, to confirm each event actually happened and was reported correctly.

Building the record, the data manager. Data management loads every entry into the study database, runs automated edit checks, and sends queries back to the site until the numbers reconcile.

Naming it, the medical coder. Free-text descriptions are mapped to a standard dictionary (MedDRA) so the same event is counted the same way across thousands of patients.

Judging cause and pattern, the pharmacovigilance safety analyst. Drug-safety scientists assess causality on each serious event, aggregate events across the whole study, and hunt for signals that separate a drug effect from ordinary life. Serious, unexpected, related events trigger expedited reports to regulators while the trial is still running.

The independent check, the Data Safety Monitoring Board. On many trials an outside board with no stake reviews the unblinded safety data on a schedule and can pause or halt the study if harm outweighs benefit.

The math, the biostatistician. Statisticians build the safety tables, each event's rate on drug versus placebo, that decide what is real enough to reach the label.

Writing it down, the medical writer and regulatory affairs. What survives all of that becomes the approved labeling, and the labeling is what the fair-balance rule then requires the ad to read out loud.

After approval it does not stop, pharmacovigilance continues. Once the drug is on the market, safety teams and the FDA keep collecting reports (through FAERS) and can update the label, add a warning, or pull the drug.

What could not be attributed is still reported as observed. The label, and therefore the ad, reflects what all of those people found, not what was convenient.

Oversight runs on several tracks at once: on-site monitoring, centralized or risk-based monitoring that flags statistical anomalies across sites, and for-cause audits when something looks wrong. Above the sponsor and CRO sits the FDA's Bioresearch Monitoring program, which can run a for-cause inspection, issue a Form 483 of observations, escalate to a Warning Letter, begin disqualification through a formal notice, and refer criminal conduct to the Department of Justice.

These are real, and their legal status differs, which matters.

·        A&R Research Group (Pembroke Pines, Florida): the owners pleaded guilty to conspiracy to commit wire fraud for enrolling unqualified participants and falsifying spirometry and echocardiogram data in two asthma-drug trials. The trial’s clinical investigator separately pleaded guilty to lying to an FDA inspector. Guilty pleas entered; sentencing pending. Established fact.

·        Tellus Clinical Research (Miami): the clinic owner, the clinical investigator, and multiple study coordinators were convicted and sentenced to federal prison (terms up to 71 months) for fabricating records and falsely representing that subjects took the study drug. Established fact.

·        FDA Warning Letters to individual clinical investigators continued through 2025 and 2026, citing failures such as enrolling ineligible subjects and protocol deviations. One March 2026 letter faulted an investigator for treating ineligible subjects, which the agency said raised concerns about the reliability of the site’s data. Official regulatory findings.

·        Pines Care Research Center (Pembroke Pines): a physician and staff were charged in June 2026 (indictment unsealed June 10, 2026) with fabricating test data and using real people’s identities to invent enrolled subjects. These are allegations only; the defendants are presumed innocent and have not been convicted.

When misconduct is confirmed, the affected data is excluded from the analysis. The integrity of the approved dataset is protected by removing the corrupted inputs.

Here is the part that rarely gets said out loud. All of this costs money, and that cost is real. Every role in the pipeline is a paid person or a system someone paid for: the monitors who travel to sites, the CRAs who check records line by line, the data managers who chase every discrepancy, the coders, the statisticians, the independent safety boards, the FDA inspectors. Verification, monitoring, and data work are a meaningful share of what a trial costs, and a large trial costs a great deal. When people ask why drugs are expensive, this apparatus is one of the honest answers. You are not only paying for the molecule. You are paying for the proof that it does what the label claims, and does not do what the label warns against.

That cost buys one specific thing: a label you can trust. Cut the verification and you do not get a cheaper drug that is just as reliable. You get the same drug with less certainty behind every number on the insert. The cases above were caught because someone was funded to look. Thin out the monitoring, the audits, and the inspections, and the gap between a problem and its discovery widens. That gap is exactly the window in which fabricated data reaches a database and stays there. Underfunding oversight does not make the field cleaner or the drug safer. It makes the same field harder to verify, and it moves the risk onto the patient who swallows the pill. That is the trade worth arguing about: not whether oversight is expensive, but what you are buying when you refuse to pay for it.

If you work in clinical research or regulatory affairs: the adverse-event list is the visible end of pharmacovigilance and labeling. The same monitoring and audit functions that generate trustworthy safety data are the ones most exposed when budgets tighten. Resourcing detection is a data-integrity issue, not an overhead line.

If you are a patient or patient advocate: the long list on a drug ad reflects a system that was required to look. The absence of a list on a supplement reflects a system that was not. Apply the same questions to both, and keep a written record of everything you take.

If you are a general reader: fraud in clinical research happens, and it is caught by a structured net of people and rules. That net is the reason the medicine that reaches you is not built on the data that got pulled. It is worth understanding, and worth funding.

·        FDA, Basics of Drug Ads (fair balance: a product-claim ad must state the drug’s major risks alongside its benefits). https://www.fda.gov/drugs/prescription-drug-advertising/basics-drug-ads

·        FDA, Prescription Drug Advertising: Questions and Answers (the broadcast “major statement” that makes ads read the risks out loud). https://www.fda.gov/drugs/prescription-drug-advertising/prescription-drug-advertising-questions-and-answers

·        Direct-to-consumer prescription drug advertising is permitted broadly in only two countries, the United States and New Zealand. Wisconsin Watch (2025): https://wisconsinwatch.org/2025/05/prescription-drug-direct-advertising-us-new-zealand-pharmacy/ | The Conversation (2024): https://theconversation.com/most-high-income-countries-ban-direct-advertising-of-prescription-drugs-why-does-nz-still-allow-it-231688

·        FDA, Bioresearch Monitoring (BIMO) Program Information (on-site inspections and data audits across sponsors, CROs, monitors, investigators, and IRBs). https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/fda-bioresearch-monitoring-information/bioresearch-monitoring-program-information

·        FDA, Clinical Investigators – Disqualification Proceedings (the administrative path that can bar an investigator from receiving investigational products). https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/compliance-actions-and-activities/clinical-investigators-disqualification-proceedings

·        FDA, Warning Letters (searchable index of issued letters). https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/compliance-actions-and-activities/warning-letters — Example, Ehsan Sadri, M.D. (03/27/2026), cited for treating ineligible subjects: https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/warning-letters/ehsan-sadri-md-726342-03272026

·        U.S. Department of Justice, Office of Public Affairs: A&R Research Group — medical clinic owners and clinical investigator plead guilty in connection with fraudulent clinical drug trials (guilty pleas; sentencing pending). https://www.justice.gov/opa/pr/medical-clinic-owners-and-clinical-investigator-plead-guilty-connection-fraudulent-clinical

·        U.S. Department of Justice, Office of Public Affairs: Tellus Clinical Research — study coordinators and clinical investigator sentenced for falsifying clinical trial data (convictions and prison sentences). https://www.justice.gov/opa/pr/florida-study-coordinator-sentenced-scheme-falsify-clinical-drug-trial-data

·        U.S. Department of Justice, Office of Public Affairs: Pines Care Research Center — doctor and staff charged with falsifying data in clinical drug trials (indictment unsealed June 10, 2026; allegations only, defendants presumed innocent). https://www.justice.gov/opa/pr/doctor-and-staff-charged-falsifying-data-clinical-drug-trials

·        FDA Adverse Event Reporting System (FAERS): how post-market reports are collected after approval. https://www.fda.gov/drugs/drug-approvals-and-databases/fda-adverse-event-reporting-system-faers-database

·        FDA, Information for Consumers on Using Dietary Supplements (under DSHEA 1994, supplements are not FDA-approved for safety or effectiveness before sale). https://www.fda.gov/food/dietary-supplements/information-consumers-using-dietary-supplements