Blog

# COFEPRIS, INVIMA & ANVISA approval for medical device trials: a country playbook

Running a medical device clinical trial in Latin America requires authorization from two separate bodies in every country: an ethics committee and the national regulatory authority. Neither approval substitutes for the other, and enrollment cannot begin until both are in hand. That sequencing rule is where most first-time sponsors lose weeks or months, not because the regulators are slow, but because the dossier was assembled in the wrong order or was missing documents that gate the next step.

This playbook covers the exact workflow for COFEPRIS (Mexico), INVIMA (Colombia), and ANVISA (Brazil): what to submit, in what order, what typically triggers a deficiency notice, and how to move through each country’s process without rework.

## The two-track model every sponsor must understand

Every jurisdiction in the region runs the same structural model:

– **Track 1:** Ethics committee (REC/IRB/IEC/CEP) issues a favorable opinion on the protocol, informed consent, and study procedures.
– **Track 2:** National regulator (COFEPRIS, INVIMA, or ANVISA) authorizes the clinical investigation, typically conditional on or concurrent with the ethics approval.

Enrollment cannot begin before both are cleared. Many sponsors try to run these tracks truly in parallel, but in practice, most regulators want at least a pending or favorable ethics opinion before they’ll accept a regulatory dossier as complete. Confirm the specific sequencing requirement for each country before filing.

## Step 1: Confirm your trial type and device classification before filing anything

The documents you’ll need, and the scrutiny you’ll face, depend entirely on what you’re studying and what risk class the device falls into. A first-in-human feasibility study of a novel implantable device triggers a far more detailed dossier than a post-market study of a Class II device with an existing clinical record.

Before drafting any submission, confirm:

– Whether the study qualifies as a clinical investigation, an early feasibility study, or a post-market/confirmatory study (each country treats these differently)
– The device’s local risk classification (Class I/II/III or equivalent) and intended use
– Whether a full device technical file or a lighter-weight summary will satisfy the reviewer

Required sponsor inputs at this stage include a finalized protocol concept, investigator brochure or equivalent device technical documentation, a risk management rationale aligned with ISO 14971, investigational product details (manufacturing, lot/batch controls, sterility), and a labeling and IFU approach.

## Step 2: Ethics committee approval, what the committee must see before enrollment

ISO 14155 (Clinical investigation of medical devices for human subjects) is the organizing standard here. The FDA recognizes ISO 14155 Third Edition (2020-07) as a consensus standard for medical devices, and the CITI Program confirmed ISO 14155:2026 updates that reinforce its global applicability. Designing your study to this standard isn’t optional if you want the resulting data to credibly support any future FDA pathway under [21 CFR 812.28](https://bioaccessla.com/fda-acceptance).

Ethics committees across LATAM consistently require:

– Final protocol (signed, versioned)
– Informed consent and patient information sheet in the local language
– Recruitment and advertising materials
– Safety reporting plan with clear SAE/SUSAR/AE definitions and notification timelines
– Clinical trial insurance or patient compensation evidence
– Investigator CVs and site qualifications
– Risk-benefit narrative aligned with the protocol

The most common reasons for ethics rework are: incomplete consent language (especially around risks, alternatives, and voluntary withdrawal), mismatched risk-to-benefit framing between the protocol and consent, missing or insufficiently described insurance/compensation coverage, and vague SAE definitions. Fix these before submission, ethics committees in the region rarely grant conditional approvals on major items.

## Step 3: Country-specific regulatory submission, what changes by jurisdiction

### COFEPRIS (Mexico)

According to ClinRegs (NIH, October 2025), COFEPRIS requires protocol authorization as a separate step from ethics approval. Sponsors must obtain a favorable opinion from a registered Research Ethics Committee (REC) and approval from a clinical site specifically licensed to conduct investigational studies before COFEPRIS authorization is complete. Regulatory submissions are processed through DIGIPRiS, COFEPRIS’s electronic platform, and the lifecycle includes fees, pharmacovigilance/safety reporting obligations, and investigational product import requirements.

In March 2025, COFEPRIS published a Resolution (Official Gazette, 24/03/2025, reported by Pérez-Llorca on 30/05/2025) introducing a Trusted Regulatory Practices (Reliance) framework that can simplify authorization when prior approvals from recognized authorities exist. Reliance doesn’t eliminate the local ethics or site authorization requirements, but it can reduce duplication in the regulatory dossier.

The typical COFEPRIS dossier for a device clinical investigation includes: Spanish-language protocol, scientific justification, investigator documents, patient-facing consent, investigational device import/handling documentation, and REC favorable opinion. All primary documents must be in Spanish or accompanied by certified translations. See the [Mexico clinical trial regulatory guide](https://bioaccessla.com/regulatory-guide/mexico) for the complete document checklist.

### INVIMA (Colombia)

Colombia uses a strictly sequential model. Ethics committee approval from an INVIMA-registered IEC/IRB must precede INVIMA submission. The total regulatory and ethical review timeline typically spans 120 to 280 days, depending on device risk classification and application completeness.

Required documentation for INVIMA includes: IEC/IRB approval letter, device technical file elements (biocompatibility data, analytical tests, risk analysis), GMP and ISO certificates, clinical trial insurance, and import license documentation. What device teams most often miss: an incomplete technical file (especially biocompatibility and risk analysis sections), insufficient insurance coverage documentation, and absent study registration in a recognized registry. For a structured walkthrough of the [INVIMA clinical trial approval process](https://bioaccessla.com/blog/invima-clinical-trial-approval-process-colombia), review the country-specific submission guide.

### ANVISA (Brazil)

Brazil’s framework has changed materially. Law 14.874/2024 and ANVISA RDC 837/2023 introduced a Clinical Investigation Dossier (DICD) concept for medical devices with risk-based delimitation of ANVISA’s approval scope. Under the modernized framework reported by Med Device Online (July 2025), ethics review is capped at 30 business days from acceptance and ANVISA regulatory analysis may not exceed 90 working days. Both timelines represent meaningful improvements over the prior system.

The DICD consolidates what was previously fragmented across multiple submissions into a single, structured dossier. Ethics governance in Brazil runs through CEP (institutional ethics committee) and CONEP (national ethics council) depending on the study type. Dossier incompleteness and device-class pathway nuances are the primary causes of clock resets in Brazil, sponsors that submit a near-complete package often find the 90-working-day clock paused at first technical review.

## Step 4: Investigational device import, don’t treat this as an afterthought

Every country in the region requires documented authorization to import investigational devices before they can be used in a clinical study. Regulators treat this as a separate gate, and a sponsor who holds ethics and regulatory approval but lacks import authorization cannot legally administer the device to patients.

Typical import documentation includes: import permit or letter of authorization from the regulator, GMP evidence or manufacturer declaration, investigational labeling (often country-specific), chain-of-custody documentation, and a storage, handling, and disposition plan.

This workstream often takes 2-6 weeks after regulatory approval, depending on the country and device type. Planning it in parallel, not sequentially, with site activation is the single biggest schedule optimization available at this stage.

## Step 5: Site activation and start-up readiness

Ethics and regulatory approval get you to the starting line. Site activation gets the trial running. Before the first patient is screened:

– Clinical trial agreements must be fully executed
– Delegation of responsibilities log must be in place
– All study staff must complete protocol and GCP training
– EDC or data capture system must be configured and validated
– Device accountability procedures and storage conditions must be documented and verified
– Monitoring plan must be approved and the first monitoring visit scheduled

Site selection should prioritize institutions with a track record of ISO 14155-aligned device studies, an accredited or registered ethics committee, investigator experience with the specific device category, and the operational infrastructure to handle investigational product controls. Picking sites that look strong on paper but lack device-trial-specific SOPs is a reliable way to generate protocol deviations in the first month.

## Step 6: Safety reporting and documentation during the trial

Sponsor and investigator responsibilities for safety reporting are distinct, and both are auditable. Sponsors are responsible for aggregate safety surveillance, SUSAR/expedited reporting to regulatory authorities, and DSMBs (where required). Investigators are responsible for identifying, documenting, and promptly reporting individual adverse events to the sponsor.

Timelines for regulatory reporting of SUSARs vary by country: typically 7 days for fatal/life-threatening events and 15 days for others, but confirm the specific requirement for each jurisdiction in your safety management plan. Audit readiness requires that source data, signed consent forms, and protocol deviations are documented contemporaneously, not reconstructed. [Clinical trial costs and timelines](https://bioaccessla.com/costs-and-timelines) across key LATAM jurisdictions reflect these operational requirements.

## Pre-submission completeness audit: run this before you file

A deficiency notice resets your timeline by weeks. Before submitting to any country, verify:

– Protocol version matches the version referenced in the informed consent
– Consent language covers all study procedures, risks, alternatives, and compensation terms
– Insurance certificate is valid for the full study duration and covers all enrolled subjects
– SAE/SUSAR/AE definitions are consistent across the protocol, consent, and safety management plan
– Device technical file is complete for the specific device, not a generic corporate summary
– Import documentation package is drafted and in queue, not waiting for regulatory approval
– All local language requirements are met (Spanish for Mexico/Colombia, Portuguese for Brazil)

Run this checklist against every section of your dossier, not just the cover page.

## Common pitfalls that trigger deficiency notices

Across all three countries, the same themes appear in deficiency letters:

– Protocol and consent are different versions or contain conflicting risk language
– Insurance evidence is present but doesn’t name all participating sites or cover all study phases
– Investigational product supply documentation is generic (e.g., a catalog GMP certificate) rather than study-specific
– Safety reporting plan lacks a notification workflow diagram or fails to define “unexpected” in the context of the device’s risk profile
– Technical file doesn’t include biocompatibility evidence appropriate for the intended contact type and duration

Fixing these before submission, not in response to a deficiency, is the difference between a 90-day and a 6-month approval timeline.

## When foreign approvals can help: reliance and streamlining

Reliance mechanisms allow regulators to recognize prior approvals from trusted foreign authorities as part of the local review, reducing the analytical burden on the national reviewer. COFEPRIS’s March 2025 Resolution is the clearest example in the region: sponsors with recognized foreign approvals may qualify for a simplified authorization pathway.

Two guardrails apply everywhere. First, eligibility for reliance depends on trial phase, device class, and the specific regulatory resolution, it’s not a blanket right. Second, reliance never removes the requirement for local ethics approval and local regulatory submission. It may shorten the review, but it doesn’t eliminate the track.

## FAQ

**How do I get COFEPRIS, INVIMA, or ANVISA approval to run a device trial?**
Obtain ethics committee approval first, then submit a complete regulatory dossier with device technical documentation and investigational product import evidence to the national authority.

**Do I need ethics approval before regulator approval?**
Yes, in all three countries. The sequencing is sequential (ethics first), though the specific timing requirements differ by jurisdiction.

**How long does the process take?**
Mexico (COFEPRIS): approximately 8-16 weeks total from a complete submission. Colombia (INVIMA): 120-280 days depending on device risk class and application completeness. Brazil (ANVISA): ethics capped at 30 business days and ANVISA review not to exceed 90 working days under the modernized framework (Law 14.874/2024).

**What documents do I need to submit?**
At minimum: protocol, informed consent, investigator documents, device technical file, GMP/ISO certificates, clinical trial insurance, risk management documentation, and investigational device import/supply plan. Country-specific additions apply.

**Can ISO 14155/GCP-aligned data support FDA planning?**
Yes. The FDA recognizes ISO 14155 as a consensus standard and may accept foreign clinical data under 21 CFR 812.28 when the study is conducted under ISO 14155/ICH-GCP with independent ethics oversight. This is a case-by-case determination, not a guarantee. See the [FDA acceptance of Latin America clinical data](https://bioaccessla.com/blog/fda-acceptance-latin-america-clinical-data-guide) guide for the full criteria.

**Do I need a local representative or importer?**
Yes, in all three countries. A local regulatory representative or importer of record is required for investigational device import authorization and regulatory correspondence. In Brazil, an authorized representative in-country is a structural requirement for ANVISA submissions.

## Get your submission-ready plan in 2-4 weeks

bioaccess® operates as a [first-in-human CRO](https://bioaccessla.com/first-in-human-cro) purpose-built for MedTech startups that need submission-ready human evidence on a predictable timeline. The FIH-12™ Medical Device Program structures every workstream required to go from protocol to clinical study report: FDA-anchored strategy, ISO 14155 protocol architecture, ethics and regulatory submissions in Mexico, Colombia, and Brazil, site selection and activation, investigational device import logistics, patient enrollment, monitoring, data management, and final report delivery.

The program runs under ICH-GCP standards with an ACRP-certified operations team across a network of 50+ pre-qualified clinical trial sites. Clinical data is structured from day one for potential FDA consideration under IDE, 510(k), De Novo, PMA, or HDE pathways.

If you’re planning a device trial in Latin America and want a country-by-country readiness assessment with a realistic timeline deliverable, [contact bioaccess®](https://bioaccessla.com/first-in-human-clinical-trials) to schedule a scoping call. The assessment covers ethics/regulatory sequencing, dossier gap analysis, import logistics planning, and a milestone-anchored submission timeline, typically delivered within 2-4 weeks of engagement.

Read the full newsletter online here.

Two major regulatory shifts landed in Latin America last week alongside a global GCP framework update and a fresh U.S. IND acceleration proposal, giving sponsors and CROs new operational levers to consider before their next protocol filing. This edition covers the developments most likely to affect clinical research planning through Q3 2026.

• Brazil operationalizes continuous submission for clinical trial dossiers under RDC 945/2024
• Peru rescinds its long-standing clinical trial technical-evaluation procedures at the INS
• ICH adopts E6(R3) Annex 2, formalizing GCP for decentralized, pragmatic, and real-world data trials
• Bayer’s Lampit clears ANVISA for pediatric Chagas use down to 2.5 kg body weight
• FDA opens comment window on its Expedited IND Pilot Program through July 22
• U.S. CRO consolidation intensifies as NAMSA and Veranex accelerate early-phase hiring

Take a closer look at the six clinical research signals below that sponsors, CROs, and site networks should factor into their next planning cycle.

Brazil’s ANVISA Operationalizes Continuous Submission for Clinical Trial Dossiers

On July 8, ANVISA’s Diretoria Colegiada approved a new Instrução Normativa operationalizing the continuous submission mechanism established in RDC nº 945/2024. Sponsors can now submit documents composing the Dossiê do Produto sob Investigação — including stability data and analytical method validation — progressively as they are generated, rather than in a single bundle at DDCM submission. The rule also permits references to CADIFA and pharmacopeial monographs for active pharmaceutical ingredient documentation in specific situations. The change complements Lei 14.874/2024’s parallel-review provisions, signaling Brazil’s regulatory culture is converging with EMA-style rolling review conventions.

Why It Matters: Sponsors and CROs should redesign document-readiness sequencing to align with incremental submission windows and prepare API dossier alternatives leveraging CADIFA cross-references. Programs currently pending Brazilian authorization should confirm with regulatory counsel whether the IN’s transitional provisions permit re-filing under continuous submission.

Peru’s INS Dissolves Legacy Clinical Trial Evaluation Procedures

On July 8, Peru’s Instituto Nacional de Salud published Resolución Directoral N.° 291-2026-INS/DI, rescinding POE-DIIS-002 (technical intake and assignment of clinical trial dossiers) and POE-DIIS-003 (technical review of trial authorization applications). Both procedures had structured how DIIS reviewers evaluated investigator qualifications, sponsor documentation, and protocol suitability for the past several years. Sponsors with active or pending INS submissions should confirm with local regulatory counsel which replacement procedure applies to their expediente, as legacy references in existing submission templates must be updated to avoid procedural rejection.

Bottom Line: Peru’s clinical trial pathway is entering a transition window with no published replacement procedures. Sponsors evaluating Peru as an FIH or Phase 2 destination should model both accelerated and delayed scenarios in their country-selection matrices until the INS publishes the replacement framework, expected within thirty days per Peruvian administrative convention.

ICH Adopts E6(R3) Annex 2 for Decentralized, Pragmatic, and RWD Trials

The International Council for Harmonisation adopted Annex 2 to the ICH E6(R3) GCP guideline on July 10, 2026. Annex 2 codifies GCP expectations for decentralized trial elements (home visits, remote patient interactions), pragmatic trial elements (integration with routine clinical care), and real-world data use in interventional studies, covering IRB communication, investigator oversight, sponsor governance, safety monitoring, informed consent, and digital-health-technology validation. ANVISA, COFEPRIS, and ANMAT are ICH Regulatory Members and are expected to implement Annex 2 within their normal transposition timelines.

What to Focus On: Companies running or planning hybrid-decentralized programs across LATAM sites should audit platform vendors’ GCP certifications, refresh investigator training to reflect Annex 2 obligations, and review sponsor oversight documentation before the end of Q3 2026.

Bayer’s Lampit Cleared by ANVISA for Pediatric Chagas Down to 2.5 kg

Through Resolução RE nº 2.631 published in the Diário Oficial da União on July 2, ANVISA approved pediatric use of Bayer’s Lampit (nifurtimox) for Chagas disease in patients from 2.5 kg body weight through age 18. The decision expands treatment access for a disease that remains a public health priority across Brazil, Bolivia, Argentina, Paraguay, and other endemic Latin American countries. Pediatric Chagas programs have historically struggled with low commercial pull and complex enrollment logistics; a Brazil-approved comparator now exists for future Phase 2/3 or long-term extension protocols.

Why It Matters: Sponsors developing complementary Chagas therapies planning LATAM-inclusive Phase 2/3 programs now have a comparator benchmark cleared by ANVISA and should factor Lampit’s labeled pediatric indication into study design and control-arm strategy. Program-level economics for LATAM-focused neglected-tropical-disease work continue to strengthen.

FDA Opens Public Comment on Expedited IND Pilot Program

The U.S. FDA opened a public request for information on its proposed Expedited Investigational New Drug Pilot Program, with comments accepted through July 22, 2026. The initiative proposes a network of qualified research institutions, CROs, and regulatory advisors that would collaborate with sponsors to accelerate the path from drug discovery to first-in-human studies. The pilot is intended to identify structural bottlenecks in current IND workflows and pilot new coordination models across sponsor-CRO-institution triads.

Bottom Line: LATAM-based CROs with FIH execution capability and cross-border regulatory advisory experience should evaluate whether to submit an RFI response — for direct participation in any pilot network and to raise visibility with the sponsor community currently forming its position on the framework. Program-strategy leaders should consider how a U.S. Expedited IND pathway would sequence with LATAM FIH programs already positioned as time-and-cost advantaged.

NAMSA and Veranex Intensify Early-Phase Clinical Hiring

Two U.S.-anchored CROs made notable moves during the week of July 4-10 signaling continued consolidation pressure in early-phase clinical research. NAMSA posted three new roles across five hubs, including a Senior Clinical Research Associate contractor position, an Associate Study Director in Northwood, Ohio, and multi-city laboratory-scientist postings across Irvine, Atlanta, St. Paul, Minneapolis, and Northwood. Veranex separately posted a Senior Clinical Study Manager role explicitly supporting first-in-human and early-feasibility-study execution, alongside a senior business development leader and expanded engineering and quality capacity. Both firms are systematically building integrated preclinical-clinical development platforms via distributed U.S. hubs.

What to Focus On: LATAM-focused CROs should sharpen positioning around country-level regulatory fluency, site relationships, and speed to first-patient-in — the competencies U.S. integrated platforms cannot easily replicate. Sponsors should evaluate NAMSA and Veranex on lifecycle-integrated development economics rather than headline pricing, and LATAM providers on their ability to compress the FIH-to-Phase-2 transition.

Advance Your First-in-Human Trials with Confidence

At bioaccess®, we help MedTech, Biopharma, and Radiopharma innovators accelerate first-in-human trials across Latin America — with the regulatory depth, site relationships, and operational execution that emerging markets demand. From INVIMA to ANVISA to COFEPRIS to ANMAT, our teams work inside the frameworks reshaping the region so sponsors don’t have to. Explore our roadmap.

Key Takeaways

• Brazil’s ANVISA continuous submission Instrução Normativa is a structural change to how sponsors sequence document readiness for DDCM approval.
• Peru’s INS has entered a procedural transition window; sponsors should adjust country-selection modeling until the replacement framework is published.
• ICH E6(R3) Annex 2 codifies decentralized, pragmatic, and RWD trial conduct — LATAM programs should update master protocols and DHT vendor documentation now.
• ANVISA’s pediatric Chagas approval of Bayer’s Lampit down to 2.5 kg strengthens the LATAM neglected-tropical-disease research economics.
• FDA’s Expedited IND Pilot RFI window closes July 22 — a strategic moment for LATAM CROs to signal capability to U.S. sponsors.
• NAMSA and Veranex are aggressively building integrated early-phase platforms; LATAM CROs must sharpen regulatory-fluency and speed positioning.

bioaccess® | Fast-Tracking First-in-Human Trials, Anywhere.

Follow bioaccess® on LinkedIn. | Subscribe to Global Trial Global Trial Accelerators™ · Edition 9 · Clinical-trial intelligenceAccelerators™.

Language: English · Español

Working document · bioaccess® regulatory team · August 4, 2026

Structured text of the 66 observations that bioaccess® submitted to Colombia’s Ministry of Health and Social Protection during the second public consultation on the draft resolution that establishes the requirements for health research involving human beings and partially repeals Resolution 8430 of 1993. Each observation states its regulatory reference, the issue identified, and our summarized recommendation. See the official MinSalud page and our analysis for sponsors.

Executive summary

1. The draft constitutes a substantial and necessary advance over Resolution 8430 of 1993, and several of its components must be expressly preserved. The following are first-order strengths: the adoption of the risk-proportionate regulation approach (Art. 6, item 12; Art. 7, para. 1); the alignment of the consent of adults with disabilities with Law 1996 of 2019 through supports and reasonable accommodations (Art. 17), which corrects a serious deficit of the current regime; the replacement of the requirement of two witnesses with the figure of the impartial witness (Art. 9, para. 2); the technique of dynamic reference to international standards “in their current version” (recitals), which avoids regulatory obsolescence; the reasonable and proportionate limitation of post-study access (Arts. 9, item 25; 34, item 3; 39), notably more viable than the Chilean model; the express exemption from insurance for observational and minimal-risk studies (Art. 35, item 1); the acceptance of global/international policies with local enforceability (Art. 35, item 3); technological neutrality regarding preclinical evidence (Art. 42, final subsection); the transitional recognition of international registrations (Art. 49, para. 5); and the enablement of reliance and mutual recognition mechanisms (Art. 27, item 19).
2. There is a single mandatory deadline in the 50 pages of the draft, and it has no associated legal consequence. Article 27, item 1, sets thirty (30) business days for the opinion of the CEI. There is no deadline whatsoever for: the approval of the protocol by INVIMA (Arts. 43 and 45); the review by the CEIs of each participating center in multicenter studies (Art. 6, item 6); the assessment of amendments; the Good Clinical Practice certification of the centers (Art. 44); the authorization of the importation of supplies (Art. 46); or the prior-consultation certification of the Ministry of the Interior (Art. 6, item 6). In addition, Article 29, Roman I.III delegates to the Standard Operating Procedures of each CEI the setting of “mandatory response times”, which neutralizes the only established deadline. The result is a framework whose total start-up duration is, by design, indeterminate. No reference jurisdiction with which Colombia competes is today in that position.
3. Article 25, Phase 2, point B conditions the conduct of all greater-than-minimal-risk research —that is, of every clinical trial— on the investigator “conclusively demonstrating” that the knowledge derived responds to the national burden of disease, to unmet basic needs, to equity, or to emergency response capacity. This is, in the commentator’s view, the single most consequential provision of the draft for the country’s competitive position. It contradicts Article 32, item 1 itself (which expressly protects studies in orphan diseases and targeted populations) and Article 26, item 3 (which rejects measuring social value by mass population impact). As drafted, it enables the denial of a global clinical development program on grounds of national epidemiological prioritization. It must be reformulated as a requirement of justification of social and scientific value, not as a condition of execution.
4. The multicenter studies model generates cumulative reviews with no time limit or scope delimitation. Article 6, item 6 simultaneously requires the approval of a “reference CEI” and the review by the CEIs of each participating center, without defining what each one reviews, without a deadline for the local reviews, and without a deference rule. This is the design flaw that Regulation (EU) No. 536/2014 resolved through its Article 8, paragraph 2 —a single binding conclusion, with a closed and exhaustive list of three grounds for discrepancy— and that Brazil resolved through Article 14, § 7 of Law 14.874/2024 —review by a single CEP for all national multicenter research—. Without correction, this item alone may add months to the start of national multicenter studies.
5. Ethical review and regulatory review are not articulated as parallel processes, and the duplication of scientific evaluation is expressly provided for. Article 6, item 6 suggests sequentiality; Article 45 assigns to INVIMA the analysis of the product’s benefit-risk balance and to the CEI the verification of “methodological soundness and scientific validity”, enabling non-approval on grounds of methodological shortcomings; and Article 25, paragraph 4 allows INVIMA to reclassify the risk category assigned by the CEI. It is recommended: (i) to expressly authorize simultaneous filing before the CEI and INVIMA; (ii) to delimit non-overlapping scopes; and (iii) to establish that the scientific evaluation of the product carried out by INVIMA is not re-evaluated by the CEI.
6. Verifiable technical defects are identified that must be corrected independently of any policy consideration. Among others: Article 5, item 2 (“QSAR Analysis:”) lacks a definition and is blank; Article 27 contains an empty item 2; there are two chapters numbered “CHAPTER III” (Arts. 25 and 27); Article 16, paragraph 3 refers to “Article 8, paragraph 5”, which does not exist (Article 8 has four paragraphs; the correct reference is to paragraph 3); Article 2, paragraph 3, item 2 exempts systematic reviews and meta-analyses from CEI review, while Article 25, Phase 2, point A classifies them as minimal risk and Article 49, paragraph 2 includes them among the research subject to mandatory registration in the PNRIS; Article 26, item 4, Roman I prohibits compensating “the invasive nature of the procedure”, while Article 36, item 2 allows compensating the “discomforts associated with the protocol procedures”; and Article 19, paragraph requires an insurance policy for all clinical research with intervention involving women of reproductive age, contradicting Article 35, item 1, which limits that requirement to greater-than-minimal risk.
7. Article 8, paragraph 2 orders the deletion or return of the non-anonymized data of the participant who withdraws, which is incompatible with the integrity of the safety database, with the document-retention obligations of the draft itself, and with the medical-record retention regime. The withdrawal of consent must cease the prospective collection of data, not destroy the dataset already incorporated into the analysis and into pharmacovigilance. This observation is classified as a technical error, not as a policy disagreement: as it stands, the provision makes the simultaneous compliance with Resolution 1995 of 1999, Law 2015 of 2020, and the safety-reporting obligations that the draft itself imposes in Articles 37 and 38 unenforceable.
8. The transitional and entry-into-force provisions create a period of legal uncertainty of indefinite duration. Article 53 provides for immediate entry into force upon publication, while at least five substantive obligations depend on instruments that do not yet exist: the National Technical Guide and the Unified Matrix (Art. 26, para. 6), the national CEI accreditation system (Art. 27, para. 1), the expedited ethical-review procedures for emergencies (Art. 27, para. 3, within twelve months), the specific conditions of post-study access (Art. 39, para., within twelve months), and the full operability of the PNRIS (Art. 52), whose certification has no deadline. Article 52 grants eighteen months of adaptation only to the insurance requirements and only with respect to research already approved under Resolution 8430 of 1993. A general deferred entry into force and an express rule of non-enforceability of the obligations dependent on instruments not yet issued are recommended.
9. The Explanatory Memorandum presents two verifiable defects: it asserts the non-existence of operating costs and of economic impact, and it bases the Ministry’s competence on a repealed decree. In its Section 4, the Explanatory Memorandum states verbatim: “With the issuance and implementation of the present administrative act there will be no additional operating costs; therefore, it would not be considered to generate an economic impact.” And in its Section 5: “The draft resolution does not contemplate any budgetary availability.” Both assertions are untenable in the face of the articles, which create a national accreditation system with a public registry and metrics, a national technical guide with a mandatory matrix, a national registration platform, institutional obligations to finance the CEIs including “decent fees” for their members and electronic filing and archiving platforms, new insurance requirements, psychosocial support and periodic assessment of emotional well-being in greater-risk studies, and GCP certification of centers by INVIMA — all of them recurring budgetary burdens on public IPS, public universities, and INVIMA itself (observation C-65). Separately, Section 3.1 of the Memorandum bases the competence on “Item 7 of Article 2 of Decree 4107 of 2011” and on “Article 25 of Decree 4107 of 2011”, a decree that was repealed by Article 63 of Decree 120 of 2026, which is —correctly— the one invoked by the recitals of the draft itself. The Memorandum and the draft are thus based on different norms, one of them repealed (observation C-66). It is recommended to correct both defects and to issue a Regulatory Impact Analysis and a fiscal note.
10. Highest-yield policy recommendation. If the Ministry were to adopt only three of the changes proposed in this document, those with the greatest combined impact on predictability and protection would be: (a) legal maximum deadlines, staggered by phase and risk level, with express rules for tolling the term and with a defined consequence in the event of the authority’s silence, applicable to both the CEI and INVIMA (observations C-33 and C-39); (b) a model of a single binding opinion of the reference CEI with a closed list of grounds for local discrepancy and a term of fifteen business days for local verification, following Article 8, paragraph 2 of Regulation (EU) No. 536/2014 (observation C-31); and (c) a structured reliance route with a shortened term, supported by INVIMA’s status as a Regional Reference National Regulatory Authority Level IV of PAHO and by the 2025-2026 work plan of that network (observation C-48). None of the three reduces ethical standards or participant-protection standards; all three are procedural mechanisms.

The 66 observations

Technical Defects of Drafting, Numbering, and Cross-Reference

1. Blank definition: “QSAR Analysis”

Reference: Article 5, item 2. · Priority: High

Issue. The item is empty. The term is used substantively in Article 6, item 1, where “QSAR analysis” is accepted as prior scientific substantiation alternative to animal experimentation. The absence of a definition leaves without normative content a route of preclinical substantiation that the draft itself promotes in Article 6, item 2 (“the use of alternative methods and prior computational analyses shall be encouraged”).

Recommendation. Computational method for predicting the biological, toxicological, or pharmacokinetic properties of a molecule based on the statistical correlation between its chemical structure and the activity observed in analogous compounds, used as alternative or complementary preclinical evidence, in accordance with the standards…

2. Nonexistent cross-reference: Article 8 has no paragraph 5

Reference: Article 16, paragraph 3. · Priority: High

Issue. Article 8 contains four paragraphs. The conditions for the waiver of informed consent are in paragraph 3. The reference to “paragraph 5” is nonexistent and renders the waiver route inapplicable precisely in the scenario —secondary use of public health registry data for purposes other than the original ones— where it is most needed.

Recommendation. Replace “Article 8, paragraph 5” with “Article 8, paragraph 3”. Additionally verify that the reference in Article 5, item 8 (“ethical and legal requirements established in Article 8 of the present Resolution”) is likewise specified as “Article 8, paragraph 3”.

3. Duplication of chapter numbering and empty item

Reference: Chapter heading preceding Article 25 (“CHAPTER III ON THE IDENTIFICATION AND MANAGEMENT OF RISK…”) and heading preceding Article 27 (“CHAPTER III RESEARCH ETHICS COMMITTEE-CEI”). Additionally, Article 27, Roman I, item 2. · Priority: Medium

Issue. There are two chapters numbered “III”. The general sequence of chapters also does not restart by Title (Title I contains Chapter I; Title II Chapters II and III; Title III Chapter IV), which hinders the precise citation of the act once issued —a relevant practical problem for the references that will later be made by the Standard Operating Procedures of the CEIs, the contracts with sponsors, and the acts of INVIMA.

Recommendation. Renumber the chapters continuously and without duplication (Chapters I to VIII), or restart the numbering within each Title consistently. Delete the empty item 2 of Article 27 and renumber items 3 to 19 accordingly.

4. Duplicated and inconsistent citations of the medical-record framework

Reference: Article 37, item 2, bullet points relating to document custody. · Priority: Medium

Issue. Two bullet points of the same item impose the same obligation with different normative bases that are partially incompatible as to retention periods. In addition, the draft does not set its own retention period for the study archive, unlike Regulation (EU) No. 536/2014, Article 58, which establishes twenty-five (25) years.

Recommendation. Consolidate into a single bullet point: Retain and safeguard, in physical or digital form, the master file of the research for a term of no less than fifteen (15) years counted from the formal conclusion of the study, or the longer term required by the regulations applicable to the investigational product. The…

5. Contradiction regarding systematic reviews and meta-analyses among three articles

Reference: Article 2, paragraph 3, item 2; Article 25, Phase 2, point A, second subsection; Article 49, paragraph 2. · Priority: High

Issue. Three provisions of the same act assign three different regimes to the same class of study: exempt from ethical review; subject to risk categorization by the CEI; and subject to mandatory registration. The contradiction is substantive, not merely formal: it determines whether a meta-analysis carried out by an academic group requires any procedure at all.

Recommendation. — keep without modification, and add the following subsection: “The activities indicated in the present paragraph shall not be subject to risk categorization under Article 25, nor shall they be subject to the mandatory registration provided for in Article 49. The investigator or the institution may, in a…

Scope of Application, Exemptions, and Competence

6. Blank exemption from ethical review and consent by “order of the competent authorities”

Reference: Article 16, first subsection. · Priority: High

Issue. The provision is, as drafted, the most problematic in the draft from the perspective of participant protection, for three concurrent reasons. First, it is circular. Paragraph 1 of the same article establishes that epidemiological surveillance, outbreak control, and public health activities by legal mandate “do not constitute research involving human beings within the meaning of the present resolution”.

Recommendation. Replace the first subsection of Article 16 in its entirety with: Article 16. Public health activities by legal mandate and their delimitation vis-à-vis research. The activities of public health surveillance, mandatory notification, field investigation of outbreaks, epidemiological control, and…

7. “Evaluation of quality of care” as an open exemption

Reference: Article 2, paragraph 3, item 1. · Priority: Medium

Issue. The “evaluation of quality of care” ranges from an internal audit of indicators —correctly exempt— to a prospective process-improvement study with allocation of patients to different modalities of care, which is health services research and is expressly included in the scope by Article 3, item 5 of the draft itself. The exemption does not distinguish.

Recommendation. The actions of public health surveillance, the operation of epidemiological information systems, outbreak control, and the activities of evaluation of the quality of care and of public health programs, provided that they are not designed to produce generalizable knowledge, do not involve…

8. Competence to create the National Accreditation System and assign functions to other entities

Reference: Article 27, paragraph 1. · Priority: High

Issue. A resolution of the Ministry of Health and Social Protection cannot, on its own, impose obligations or assign functions to the Ministry of Science, Technology, and Innovation or to the National Bioethics Council, which is a body created by Law 1374 of 2010 with legally defined functions. The verb used is imperative (“shall create”), which aggravates the defect.

Recommendation. The Ministry of Health and Social Protection, in coordination with the Ministry of Science, Technology, and Innovation and subject to the prior opinion of the National Bioethics Council, shall promote the adoption, by means of the normative instrument of the corresponding rank and within a term of no more than…

9. Sanctions regime and statutory reservation

Reference: Article 51 and its paragraphs. · Priority: Medium

Issue. The article does well in referring to “the sanctions provided for in the current legislation” instead of creating new sanctions —which would be barred to a resolution by statutory reservation—. However, the enumeration of graduation criteria, the statement that “not every irregularity or non-compliance shall be presumed to be willful conduct”, and the reference to the application of disciplinary sanctions may be read as the configuration of an autonomous sanctioning regime.

Recommendation. Non-compliance with the provisions of the present resolution shall be assessed by the competent authorities in the exercise of the powers of inspection, surveillance, control, and sanction attributed to them by law, in particular those provided for in Law 9 of 1979, Law 1751 of 2015, Law 1437 of 2011, and the norms…

Informed Consent, Capacity, and Populations

10. Contradiction between the vulnerability principle and its definition

Reference: Article 4, item 4, vis-à-vis Article 5, item 26, and Article 23, first subsection. · Priority: Medium

Issue. The principle of Article 4, item 4 constitutes, in the commentator’s view, one of the most valuable and technically most solid contributions of the draft: it abandons the general presumption of vulnerability based on socioeconomic condition —which in practice operates as a mechanism of exclusion of the populations that most need access to research— and replaces it with verifiable criteria of lack of protection. It is exactly the correction that the CIOMS 2016 Guidelines introduced with respect to the 2002 version.

Recommendation. Individuals or groups in respect of whom any of the specific criteria of lack of protection indicated in Article 4, item 4 of the present resolution concur, and who therefore present a significantly greater probability of suffering physical, psychological, or social harm, or a real limitation of their capacity…

11. Deletion of data upon withdrawal of consent: incompatibility with data integrity and retention obligations

Reference: Article 8, paragraph 2, second subsection; consistent with Article 9, item 9. · Priority: High

Issue. This provision is, in the commentator’s view, the technical defect of the greatest practical gravity in the draft, because it makes the simultaneous compliance with other obligations that the same act imposes materially impossible.

Recommendation. The inalienable right of the participant to withdraw from the study at any time and without this entailing sanction, retaliation, or loss of the benefits to which they were entitled shall be recognized. The exercise of withdrawal shall produce the following effects: (i) all intervention shall cease immediately and all…

12. Absence of recognition of broad and dynamic consent in Article 8, and contradiction with Article 33

Reference: Article 8, paragraph 3, vis-à-vis Article 33, item 5. · Priority: Medium

Issue. Article 33 recognizes three routes for the secondary use of data —broad consent, dynamic consent, and waiver—, but Article 8, which is the substantive norm on consent, regulates only the last. Neither of the first two figures is defined in Article 5.

Recommendation. Add a new paragraph to Article 8, and incorporate the corresponding definitions into Article 5: Paragraph 5. Broad consent and dynamic consent. For research involving the future use of data or biological samples for health-related purposes not…

13. Absolute standard of comprehension: “fully understood”

Reference: Article 10, final subsection; consistent with Article 12, second subsection. · Priority: Medium

Issue. “Full comprehension” is an absolute and unverifiable standard. No participant fully comprehends the entirety of the information of a Phase III protocol; the international standard is sufficient comprehension to make an informed decision.

Recommendation. No intervention may begin as long as there is no documentary record that the informed consent process was carried out in accordance with the approved protocol and that the participant expressed a sufficient comprehension of the nature, procedures, risks, and alternatives of…

14. Personal contact details of the principal investigator in the consent document

Reference: Article 9, item 2. · Priority: Low

Issue. The requirement is in practice satisfied with personal data of the investigator, whose turnover requires amending the consent document and re-consenting. In addition, a personal channel does not guarantee continuous availability to report an adverse event, which is the critical function.

Recommendation. The identification of the principal investigator, of the institution responsible for the research, and of the responsible sponsor, including a permanent institutional contact channel —email and telephone number attended during the term of the study, with indication of the contact mechanism in…

15. Age ranges of Article 18 vis-à-vis Law 1098 of 2006: risk of normative hierarchy

Reference: Article 18, first subsection, and paragraph 5. · Priority: Medium

Issue. The draft establishes three ranges (under 7 years; from 7 to under 14; from 14 to under 18) that do not coincide with the categories of Law 1098 of 2006 (child from 0 to 12 years; adolescent from 12 to 18). The invocation of the “principle of normative specialty” is not apt to justify that a resolution establish age categories different from those of a law: specialty operates between norms of equal hierarchy.

Recommendation. “The age ranges indicated in the present article constitute guiding criteria of maturity for the individual assessment that corresponds to the Research Ethics Committee and the investigator, and shall be applied in subordination to the categories, the definition of the best interest, and the…

16. Consent of both parents and the veto of one of them

Reference: Article 18, item 1, point a); item 2, point a); paragraph 3. · Priority: Medium

Issue. Three difficulties. First, the requirement of the consent of both parents is stricter for the group of children under 7 years at all risk levels, while for adolescents from 14 to under 18 the consent of one suffices (item 3, point a). The gradation is inverted with respect to vulnerability.

Recommendation. “When both parents exercise parental authority, are identified, locatable, and legally capable, and one of them objects to participation, inclusion shall not proceed in studies without the possibility of direct benefit for the child. When it concerns research…

17. Absence of the category of “minor increase over minimal risk” in pediatric research without direct benefit

Reference: Article 18, paragraph 4. · Priority: Medium

Issue. The threshold is stricter than the international standard and has a counterintuitive consequence: it prevents essential pediatric research —for example, a pharmacokinetic study requiring an additional venipuncture, or an MRI without sedation in a neurodevelopmental cohort— and, in this way, perpetuates the practice of prescribing to children medicines evaluated only in adults, with the burden of risk that this transfers to the pediatric population as a whole.

Recommendation. In research with the possibility of direct benefit for the participant, the risks shall be minimized and proportionate to the prospects of obtaining such benefit. In research without the possibility of direct benefit for the child or adolescent, the research shall be admissible…

18. Consent by an independent person in a subordinate population: proportionality

Reference: Article 24, item 3, and paragraph. · Priority: Medium

Issue. The requirement is unconditional for the entire category of subordinate population, which the article itself defines broadly (“students, employees, members of the armed forces, persons deprived of liberty, institutionalized persons”). In university studies with students —a frequent modality and of typically minimal risk— it entails funding and training an external consent-taker for each project, which in practice discourages formative research.

Recommendation. The informed consent process shall be carried out by a person independent of the research team and outside the hierarchical relationship, when the research is classified as greater-than-minimal risk, when the participant is institutionalized or deprived of liberty, or when…

Risk Classification and Management

19. Conditioning the conduct of all greater-than-minimal-risk research on its alignment with the national burden of disease

Reference: Article 25, Phase 2, point B, final subsection. · Priority: High

Issue. This provision is, in the commentator’s view, the one of greatest individual consequence in the draft for Colombia’s position as a destination for clinical research, and it deserves careful consideration. Real scope. Every interventional clinical trial with an investigational product is classified, by definition of point B, as greater-than-minimal risk.

Recommendation. “The protocol of research classified as ‘Greater-than-Minimal Risk’ shall contain an explicit justification of the social and scientific value of the study, in accordance with Article 7 of the present resolution. Such justification may be supported, among others, by the contribution to the…

20. Elimination of the “no-risk” category and absence of an intermediate low-intervention category

Reference: Article 25, Phase 2 (two categories: Minimal Risk and Greater-than-Minimal Risk), in relation to the repeal of Title II of Resolution 8430 of 1993 (Article 53). · Priority: High

Issue. Resolution 8430 of 1993 contemplated three levels: no-risk research, minimal-risk research, and greater-than-minimal-risk research. The draft reduces the scheme to two.

Recommendation. Restructure Phase 2 of Article 25 into three categories, adding a category of exempt research and one of low intervention level: Phase 2. Categorization of the Risk to the Participant. Once ethical viability is established, research shall be classified, according to the probability and magnitude…

21. Categorical classification of research with artificial intelligence as minimal risk

Reference: Article 25, Phase 2, point A, second subsection. · Priority: High

Issue. The provision classifies the risk based on the state of the input data and not on the risk of the intended use of the model, which is technically incorrect and contradicts Article 3, item 8 of the draft itself, which includes within the scope AI systems “when they process information derived from identified or identifiable persons and may generate consequences on their health, care, or rights“.

Recommendation. Delete from point A the mention of the development, training, and validation of algorithms, and add a specific paragraph to Article 25: Paragraph 7. Categorization of research with artificial intelligence systems and automated analysis. Research related to the…

22. Self-defeating proviso in the definition of minimal risk

Reference: Article 25, Phase 2, point A, first subsection. · Priority: Low

Issue. Every venipuncture alters, by definition, physical integrity. The proviso, read literally, excludes from the minimal-risk category the very procedure that the subsection itself includes in it.

Recommendation. “…and minimally invasive procedures such as the extraction of peripheral venous blood, provided that the volume, frequency, and conditions of the extraction do not exceed the routine clinical parameters defined in the National Technical Guide provided for in Article 26, paragraph 6, taking into account the age and the…

23. “Preliminary rejection” without opportunity for correction, deadline, or appeal, and financial evaluation by the CEI

Reference: Article 25, Phase 1, in relation to Article 28, item 1, point c). · Priority: Medium

Issue. First, the “preliminary rejection” lacks a deadline, an opportunity for cure, a requirement of reasons, and an appeal. A preliminary rejection for a curable documentary deficiency requires restarting the entire procedure, with the complete loss of the thirty-business-day deadline. This is contrary to the principles of administrative procedure, in particular to the duty to require cure before rejecting.

Recommendation. “Before proceeding to categorize the risk, the Research Ethics Committee shall verify compliance with the ethical viability requirements indicated below. When it identifies deficiencies, it shall require the applicant, on a single occasion, to cure them, within five (5) business days…

24. “Quantitative metrics” as a risk-evaluation criterion, without definition

Reference: Article 25, paragraph 1, second subsection. · Priority: Low

Issue. It is not identified which metrics, with what methodology, or with what consequence. The verb is imperative (“shall require”), such that the provision creates an obligation of indeterminate content. Heterogeneity among committees will be generated and, predictably, requests for information that are not comparable among centers of the same multicenter study.

Recommendation. “When the methodological design permits, the Committee may request the quantification of the probability and magnitude of the identified risks, in accordance with the methodology and the template of the Unified Matrix of Risk Identification and Management adopted by the National Technical Guide provided for in Article 26…

25. Reference to INVIMA deadlines not established, regarding the reporting of unexpected risks and harms

Reference: Article 25, paragraph 3. · Priority: Medium

Issue. The obligation is of immediate compliance but its content is referred to guidelines whose existence and content are not identified. At the same time, the draft does not set its own deadlines for the reporting of serious adverse events or of serious unexpected adverse reactions, a matter that Resolution 2378 of 2008 —which survives the partial repeal— does regulate through the adoption of the Good Clinical Practice guide.

Recommendation. “Until INVIMA issues specific guidelines, the notification deadlines established in Resolution 2378 of 2008 and in the Good Clinical Practice guide adopted by that resolution, in its current version, shall apply, it being understood in any case that suspicions of serious adverse reactions and…

26. INVIMA’s power to reclassify the risk category assigned by the CEI, without deadline or criteria

Reference: Article 25, paragraph 4. · Priority: Medium

Issue. The power to suspend a study for safety reasons is legitimate and indisputable, and must be preserved. The difficulty lies in the power to reclassify the risk category already assigned by the CEI, exercised “at its discretion”, without deadline and without criteria. Since the risk category determines the insurance-policy obligation (Art.

Recommendation. In clinical trials that involve research with health technologies, INVIMA, in the exercise of its powers of inspection, surveillance, and control, shall verify the risk category assigned by the Research Ethics Committee within the term available to it to resolve the request for…

27. Environmental and “One Health” obligations applicable to all research, with an absolute standard

Reference: Article 26, item 1; consistent with Article 27, Roman I, item 7. · Priority: Medium

Issue. Three difficulties. First, the obligation is imposed on all health-related research, including survey studies, documentary reviews, and qualitative studies, in which there is no environmental impact to manage. Second, the expression “guaranteeing that the execution of the protocol does not alter the ecological balance” is an absolute and unaccreditable standard; no human activity can guarantee the non-alteration of the ecological balance.

Recommendation. “When the nature of the research so requires —in particular in studies that generate biological, chemical, pharmacological, or device waste; that involve environmental sampling; that involve genetically modified organisms; or that are conducted in territories with ecosystems…

28. Contradiction regarding compensation for discomforts and invasiveness of procedures

Reference: Article 26, item 4, Roman I, vis-à-vis Article 36, item 2, and Article 15, first subsection. · Priority: Medium

Issue. The two provisions are directly contradictory regarding a single fact: whether the discomfort derived from an invasive procedure may be compensated. The objective of Article 26, item 4, Roman I is correct and must be preserved —to prevent the amount from operating as an incentive to accept risk—, but the current wording extends it to compensation for discomfort, which is a distinct and legitimate figure.

Recommendation. The amount or nature of the compensation may not be calculated or presented as consideration for the assumption of clinical risk, nor assessed as a function of the probability or severity of the foreseen adverse events. The foregoing does not prevent the reimbursement of expenses in accordance with the…

29. CEI’s power to deny ethical endorsement due to the existence of “competing research”

Reference: Article 26, paragraph 4, second subsection. · Priority: High

Issue. The underlying concern is legitimate: the real operational capacity of the principal investigator and the competition for the same group of eligible patients may compromise quality and safety. But the criterion chosen to resolve it —the existence of protocols from different sponsors directed at the same indication, population, or mechanism of action— turns a question of capacity into a question of competition among sponsors, with three problematic consequences: 1.

Recommendation. “The Research Ethics Committee shall evaluate and document, by means of objective and verifiable criteria, the operational capacity of the principal investigator and their team to conduct simultaneously the protocols under their charge, considering: the accredited dedication time; the composition and availability of the…

30. National Technical Guide and Unified Matrix without an issuance deadline

Reference: Article 26, paragraph 6. · Priority: Medium

Issue. The paragraph correctly identifies the problem that the guide will resolve: “to avoid the heterogeneous application of the evaluation criteria, to prevent regulatory asymmetries, and to guarantee the legal certainty of the investigators”. But it does not set an issuance deadline, and the transitional rule refers precisely to the heterogeneity that it seeks to correct.

Recommendation. Add to paragraph 6: “The Ministry of Health and Social Protection shall issue such instrument within a term of no more than twelve (12) months counted from the publication of the present resolution, following a public consultation of at least thirty (30) calendar days. Until it is issued, the…

Multicenter Studies, Deadlines, and CEI Governance

31. Multicenter studies model: cumulative reviews without a deference rule, without scope delimitation, and without a deadline

Reference: Article 6, item 6, in relation to Article 27, item 1, and Article 27, item 19. · Priority: High

Issue. The item introduces the figure of the reference CEI —which constitutes an advance— but does not assign it any legal effect vis-à-vis the committees of the participating centers.

Recommendation. Replace the subsection relating to multicenter studies of Article 6, item 6, and add a new article: Article 6, item 6, subsection relating to multicenter studies. “In national multicenter studies, the ethical evaluation shall be carried out in accordance with the Research Ethics Committee procedure…

32. Prior-consultation certification: absence of a deadline and overextension of the enforceability scenario

Reference: Article 6, item 6, final subsection. · Priority: High

Issue. First, the enforceability scenario is overextended. The expression “research involving communities” is broader than the constitutional premise of prior consultation, which is the direct impact on ethnic communities. Under the current wording, a national health survey that includes, through random sampling, persons belonging to ethnic communities would require certification from the Ministry of the Interior.

Recommendation. “When the research may entail direct impact on indigenous, Black, Afro-Colombian, Raizal, Palenquera, or Rom communities —in particular when it is conducted in their territories, when it is directed specifically at their members, when it involves access to their knowledge…

33. The only mandatory deadline of the draft is neutralized, lacks tolling rules, and has no associated consequence

Reference: Article 27, Roman I, item 1, in relation to Article 29, Roman I, item III. · Priority: High

Issue. Four concurrent deficiencies turn the only deadline of the draft into a norm without practical efficacy. 1. Neutralization. Article 29, Roman I.III delegates to the operating procedures of each committee the setting of “mandatory response times”.

Recommendation. “To review and issue an opinion on the research protocol, its amendments, and other relevant documents, verifying the basic scientific validity of the design as ethical support for the study, recognizing the specific nature of qualitative, epidemiological, observational, or clinical designs…

34. The differential evaluation routes are optional and discretionary, not mandatory

Reference: Article 27, paragraph 2; consistent with Article 29, Roman I, item II. · Priority: High

Issue. The paragraph states the correct principle but leaves it entirely to the discretion of each committee. The foreseeable consequence is heterogeneity: some committees will adopt expedited review, others will not, and none will be obligated. For the investigator —particularly the academic one and the one at regional institutions, who is the one who would benefit most— the existence of an abbreviated route will depend on the institution to which they are affiliated, not on the risk of their study.

Recommendation. The evaluation by the Research Ethics Committee shall be carried out in accordance with differentiated routes according to the risk level, in the following terms, which are of mandatory application: 1. Determination of exemption. It applies to the research covered by Article 2…

35. Requirement that the sponsor’s insurance policy cover the professional civil liability of the members of the CEI

Reference: Article 27, Roman III, item 14, vis-à-vis Article 28, item 1, point c), and Article 35, item 2. · Priority: High

Issue. Three concurrent objections, the third of them substantive.

Recommendation. “To verify that the compensations and incentives offered to the participants do not constitute undue inducement, and to verify the existence and validity of the certificate of the insurance policy or insurance mechanism required under Article 35, in the terms of item 4 of that article.” Article 29…

36. Mandatory annual report for all research, regardless of the risk level

Reference: Article 25, paragraph 2, first subsection; consistent with Article 27, Roman II, item 8. · Priority: Medium

Issue. The expression “regardless of its risk category” directly contradicts the proportionality principle of Article 6, item 12, and the second subsection of the same paragraph itself, which reserves reinforced monitoring for greater-than-minimal risk. A retrospective study with pseudonymized data, approved in January and with analysis foreseen for December, does not generate safety information to report.

Recommendation. “The principal investigator of all approved research has the obligation to submit to the Research Ethics Committee progress and safety reports, with the frequency corresponding to the risk category of the study, as follows: (i) exempt and minimal-risk research…

37. CEI obligation to “independently validate the state of the art”

Reference: Article 27, Roman I, item 4. · Priority: Medium

Issue. The independent validation of the state of the art —that is, the autonomous verification of the worldwide scientific literature on the evaluated intervention— is not materially feasible for a committee of five to seven members for each protocol, and duplicates functions of the sponsor, of the investigator, and, in the case of health technologies, of INVIMA under Article 45.

Recommendation. “To verify that the protocol adequately justifies the state of the art in relation to the evaluated intervention, through the review of the scientific substantiation presented and of the references that support it, and to critically assess its sufficiency as ethical support for the…

INVIMA / CEI Competences, Parallelism, and Reliance

38. Absence of express authorization of parallel filing and evaluation before the CEI and INVIMA

Reference: Article 6, item 6; Article 43; Article 45. · Priority: High

Issue. Neither Article 6, item 6 nor Article 45 establishes whether the two evaluations may be carried out simultaneously. In the absence of express authorization, administrative practice will lean toward sequence —INVIMA will require the CEI’s endorsement, or the CEI will await INVIMA’s opinion—, whereby the times add up instead of overlapping.

Recommendation. Add a new article in Title VI and adjust Article 6, item 6: Article [new]. Parallel filing and evaluation. 1. The request for ethical evaluation before the Research Ethics Committee and the request for authorization of the protocol before INVIMA…

39. Total absence of a deadline for the authorization of the protocol by INVIMA

Reference: Articles 43, 45, and 46; consistent with Article 44. · Priority: High

Issue. The draft establishes a deadline of thirty business days for the Research Ethics Committee and none for INVIMA, for the Good Clinical Practice certification of the center, or for the authorization of the importation of supplies. Since the regulatory evaluation of the product is, in most clinical trials, the step of the longest duration, the effect is that the draft regulates the deadline of the fast component and leaves that of the slow component open.

Recommendation. Add the following paragraphs to Article 43: Paragraph 3. Authorization deadlines. INVIMA shall resolve the requests for authorization of research protocols within the following maximum terms, counted in business days from the admission of the request: | Type of request |…

40. Duplication of the scientific evaluation between INVIMA and the CEI

Reference: Article 45, third to fifth subsections, in relation to Article 27, second subsection, and Article 25, paragraph 4. · Priority: High

Issue. Article 27 limits the committee to the “basic scientific aspects” and prohibits it from substituting technical competences; Article 45 entrusts it with verifying the methodological soundness and the statistical methods, and attributes to it the power not to approve on that ground. At the same time, the same Article 45 assigns to INVIMA “the scientific analysis of the benefit-risk balance” and “the classification of the risk level of the study”.

Recommendation. “Delimitation of competences and non-duplication. For the purposes of the differentiated and complementary evaluation provided for in the present article: (i) The evaluation by INVIMA of the pharmaceutical quality, the non-clinical evidence, the pharmacological profile, and…

41. Absence of an appeal against the decisions of the Research Ethics Committee

Reference: Normative gap. Consistent with Article 25, Phase 1 (preliminary rejection), Article 25, paragraph 1 (declaration of unacceptable risk), Article 29, Roman I.VII (denial for omission or falsehood in the conflict-of-interest declaration), and Article 45 (non-approval for methodological shortcomings). · Priority: High

Issue. The draft attributes to the committees powers of preliminary rejection, of declaration of unacceptable risk, of non-approval for methodological shortcomings, of denial for conflict of interest, and of suspension or recommendation of termination of the study. It does not provide for any appeal against any of those decisions.

Recommendation. Article [new]. Reconsideration and second instance. 1. Reconsideration. Against the decisions of non-admission, non-approval, conditional approval, declaration of unacceptable risk, suspension, or termination adopted by a Research Ethics Committee, there shall lie…

42. CEI composition: barriers to entry for regional and smaller institutions

Reference: Article 28, item 1, and item 2. · Priority: High

Issue. Item 2 correctly identifies the risk —”operational barriers in small or regional institutions”— but grants the flexibility exclusively to committees dedicated to social, educational, or observational minimal-risk research, that is, precisely to those where the requirement of profiles is least critical.

Recommendation. “To guarantee deliberative plurality, the profiles indicated in points a), d), and e) of the present item shall be accredited by different persons. The profiles of points b) and c) may be accredited by the same person when that person simultaneously meets both…

43. CEI financing, fees, and safeguarding of independence

Reference: Article 29, Roman III, items 1 and 2; Article 27, Roman III, item 17. · Priority: Medium

Issue. The design is correct in its intention —the financial sustainability of the committee is a condition of its independence and of its capacity to meet deadlines— but presents three gaps. First, “a proportion” is not quantified. Without a floor, the institution may allocate a nominal fraction, and the committee will remain underfunded.

Recommendation. “Resources. To guarantee the financial, physical, and technical sufficiency for the autonomous functioning of the Committee. To this end, no less than seventy percent (70%) of the income received from the charging of protocol-evaluation fees shall be allocated directly to the budget…

44. “Social need” and “non-duplication” as a criterion for approval by the CEI

Reference: Article 7, item 1, second subsection. · Priority: Medium

Issue. The purpose —to avoid the unnecessary exposure of participants to risks when the question is already answered— is correct and corresponds to the standard of the Declaration of Helsinki. The difficulty lies in two elements of the wording.

Recommendation. “The Research Ethics Committee shall verify that the protocol adequately justifies the social and scientific value of the study and the reason why the exposure of participants is required, when relevant prior evidence exists on the research question. For these purposes, it shall not…

Data Protection, Artificial Intelligence, and Cybersecurity

45. International data transfer: the CEI as a body for determining legality, and circular reference to the biobank regime

Reference: Article 33, item 4. · Priority: High

Issue. Three deficiencies. First: circular reference to a nonexistent standard. The parenthesis “(Biobanks)” refers to the biobank regime, with respect to which Article 8, paragraph 4 itself declares that “The specific regulation of Law 2287 of 2023 on biobanks shall be the subject of an independent administrative act”. Equivalence is being required with a standard that the resolution itself acknowledges as not yet issued.

Recommendation. The international transfer of personal data shall be subject entirely to the regime of Statutory Law 1581 of 2012, in particular to its Article 26, and to the provisions issued by the Superintendency of Industry and Commerce in the exercise of its competences, including the declarations of…

46. “Cybersecurity” required without a reference standard, and absence of treatment of the re-identification risk

Reference: Article 25, Phase 3, point A; Article 33, items 2, 3, and 6; Article 6, item 11. · Priority: High

Issue. “Requiring cybersecurity” is not a determinable obligation: it does not identify controls, a reference standard, a level of requirement, or a form of accreditation. An ethics committee cannot verify compliance with an obligation whose content is not defined, and an investigator cannot accredit it. In practice, the provision will be complied with through generic declarations.

Recommendation. Replace the expression “requiring cybersecurity” with: “requiring the adoption and documentation of technical and organizational information-security controls proportionate to the sensitivity of the data, the volume of the information, and the risk level of the study, in accordance with a framework of…

Insurance, Compensation, and Care Costs

47. Prohibition of transferring costs to EPS and PBS: need for a non-denial-of-care clause

Reference: Article 34, item 1. · Priority: High

Issue. The provision is correct and necessary: it prevents the externalization to the public system of the costs derived from private research. Its preservation is recommended. However, as drafted, it generates a foreseeable risk against the participant: the EPS or the IPS may invoke it to deny or delay the care of a participant while it is determined whether or not the event is attributable to the study.

Recommendation. “The prohibition provided for in the present item is directed at the final allocation of the cost among the sponsor, the insurer, and the health system, and in no case constitutes grounds for denying, delaying, fragmenting, or conditioning the provision of the health services that correspond to the…

48. The reliance mechanism is merely declaratory: proposal for a structured route with a shortened term

Reference: Article 27, Roman III, item 19; Article 49, paragraph 5. · Priority: High

Issue. Item 19 constitutes one of the potentially most valuable provisions of the draft, and at the same time the one of the least normative density. It states the correct purpose —”to avoid evaluative reprocessing in the country”— but does not establish: (i) which foreign authorities or committees are eligible; (ii) which documents or determinations may be the object of reliance; (iii) which matters remain subject to full national evaluation; (iv) the invocation procedure; or (v) any deadline benefit.

Recommendation. Replace Article 27, item 19, and add a new article in Title VI: Article 27, Roman III, item 19. “To apply the mechanisms of trust and mutual recognition (reliance) provided for in Article [new] of the present resolution, in order to ensure the unity of…

49. Additional validity of the insurance policy defined as a “reasonable period”

Reference: Article 35, item 5. · Priority: Medium

Issue. “Reasonable” is a concept that each committee will determine differently. Since the extension of the validity is a direct component of the cost of the insurance policy and of its insurability, the indeterminacy translates into the impossibility of pricing the risk before knowing the position of the committee —and, in multicenter studies, of the committees— which delays the contracting and, with it, the filing.

Recommendation. The coverage of the insurance policy or equivalent mechanism shall be in force throughout the entire execution of the study and shall extend, at a minimum, for twenty-four (24) months counted from the last visit of the last participant in the national territory. The protocol may provide for a longer extension, which shall be…

50. Mandatory insurance policy for all research with intervention involving women of reproductive age: contradiction with Article 35 and risk of discouraging the inclusion of women

Reference: Article 19, paragraph, vis-à-vis Article 35, item 1. · Priority: High

Issue. Four concurrent difficulties. First: express normative contradiction. Article 35, item 1 uses the words “solely and exclusively” to limit the insurance-policy requirement to greater-than-minimal risk. Article 19, paragraph extends it to all clinical research with intervention involving women of reproductive age, without reference to the risk level.

Recommendation. “Paragraph. Insurance in research with women of reproductive age. When the protocol contemplates interventions, medicines, investigational products, or procedures with potential teratogenic, mutagenic, or embryotoxic effect, or when the research is…

51. Consent during pregnancy: contradiction between Articles 20 and 22

Reference: Article 20, paragraph, first subsection, vis-à-vis Article 22, paragraph, first subsection. · Priority: Medium

Issue. Article 22, paragraph establishes the correct and constitutionally adequate rule: during pregnancy, consent corresponds only to the pregnant woman, and the intervention of the other parent is activated only once birth has occurred.

Recommendation. “During pregnancy, the informed consent for participation in the research corresponds exclusively to the pregnant woman, in the exercise of her autonomy and of her fundamental rights, in accordance with Article 22, paragraph of the present resolution, even when the…

Responsibilities, Health Technologies, and Registration

52. Absolute prohibition of delegating the analysis of adverse events

Reference: Article 37, item 2, final bullet point. · Priority: Medium

Issue. The responsibility of the principal investigator for the safety of the participants is non-delegable, and that principle must be preserved. But the absolute prohibition of delegating the analysis of adverse events is incompatible with the standard organization of a research team and with the delegation logic of Article 8 of the draft itself.

Recommendation. “The principal investigator is the primary and non-delegable party responsible for the safety of the research participants. Consequently, they shall maintain effective supervision of all delegated activities and shall retain ultimate responsibility for the assessment of the events…

53. Psychosocial support and periodic assessment of emotional well-being as a general obligation in greater risk

Reference: Article 38, items 25, 26, and 27. · Priority: Medium

Issue. Item 26 is correctly conditioned (“in studies that involve significant physical or emotional risk”). Items 25 and 27, by contrast, are unconditional: item 25 applies to all greater-than-minimal-risk research —that is, to every clinical trial— and item 27 does not distinguish any category. The content of the “psychosocial support” is not defined, nor is the frequency, the instrument, or the party responsible for the “periodic assessment of emotional well-being”.

Recommendation. Replace items 25, 26, and 27 with a single item: “25. Proportional psychosocial support. To ensure the availability of psychosocial support and, when appropriate, of mechanisms for the early identification of emotional impact, in the studies in which the Research Ethics Committee…

54. Conceptual confusion between a Phase IV study and a new-indication study

Reference: Article 43, paragraph 1, third subsection; consistent with Article 5, item 9 (definition of Phase IV). · Priority: Low

Issue. A study that evaluates an unauthorized indication is not, by definition, a Phase IV study: Phase IV corresponds, in accordance with Article 5, item 9 of the draft itself, to studies carried out “once the health technology has been authorized for use”, with the objective of expanding the knowledge on safety, effectiveness, and rational use “under real conditions of clinical practice”.

Recommendation. “Studies that evaluate therapeutic indications, populations, doses, routes of administration, or conditions of use not covered by the current marketing authorization of the product, even when they maintain the same dose and presentation authorized for another indication, shall not be considered…

55. Phase I and first-in-human studies: deferred, optional, and deadline-less procedure

Reference: Article 43, paragraph 2. · Priority: High

Issue. The Phase I and first-in-human segment is precisely the one whose capture differentiates a country that hosts sites from a country that hosts programs. It is also the segment of the greatest economic value per participant, of the greatest transfer of technical capacity, and the one that consolidates a center’s position as a regional reference. The draft refers it to future guidelines, with an optional verb (“may”), without an issuance deadline and without a rule applicable in the interim.

Recommendation. “Phase I clinical studies, first-in-human studies, and studies with strategic technologies for national health sovereignty are governed by the general regime of the present article and are authorized in accordance with the deadlines of paragraph 3, with the following rules…

56. Good Clinical Practice certification extended to “any other health technology”, without a deadline

Reference: Article 44, in relation to Article 5, item 31. · Priority: High

Issue. The combination of both provisions produces a result of disproportionate scope.

Recommendation. “Article 44. Good Clinical Practice Certification. Centers that conduct interventional research with medicines, biological products, advanced, gene, or cell therapies, radiopharmaceuticals, or medical devices of classes IIb and III in accordance with…

57. Twelve-month publication deadline and its articulation with the results registration

Reference: Article 48, paragraph 3, in relation to Article 49, paragraph 4. · Priority: Medium

Issue. Two difficulties. First, ambiguity of the starting point: “the conclusion of the study” and “the primary data collection” are different moments and may be separated by months or years in a study with prolonged follow-up; the disjunction “or” does not allow determining which one applies.

Recommendation. “The interested parties shall comply with the following disclosure obligations: (i) Publication of a summary of results in the National Platform of Health Research Registries-PNRIS and in the international registry in which the study is registered, within the…

58. Registration in the PNRIS: excessive scope and need for permanent recognition of international registries

Reference: Article 49, subsections and paragraphs 2, 3, and 5. · Priority: High

Issue. First, the scope is excessive. The obligation encompasses “all research involving human beings”, including undergraduate degree works, minimal-risk surveys, and —in accordance with paragraph 2, and in contradiction with Article 2, paragraph 3— bibliometric reviews and meta-analyses. The resulting volume amply exceeds the management capacity of a platform and dilutes its value as an instrument of transparency: a registry of everything is, in practice, a registry of nothing.

Recommendation. “The registration by the principal investigator of the following research in the National Platform of Health Research Registries-PNRIS, before the recruitment of the first participant or the start of the data analysis, as appropriate, is established as mandatory: (i)…

59. Transition and entry into force: immediate entry into force of obligations dependent on instruments not yet issued

Reference: Articles 52 and 53. · Priority: High

Issue. Six concurrent deficiencies. 1. Immediate entry into force of a regime dependent on future instruments. At least five substantive obligations depend on acts that do not exist: the National Technical Guide and the Unified Matrix (Art.

Recommendation. 1. General deferred entry into force. The present resolution shall enter into force twelve (12) months after the date of its publication, with the exception of the provisions indicated in item 2, which are in force from publication. 2. Provisions of immediate entry into force. In force are…

Matters Absent from the Draft

60. Total absence of regulation of the decentralized elements of clinical trials

Reference: Normative gap. Consistent with Article 8 (modalities of obtaining consent), Article 9, paragraph 1 (electronic, digital, or remote consent), and Article 5, item 32 (impartial witness “through approved technological means”). · Priority: High

Issue. The draft recognizes electronic, digital, and remote informed consent —which constitutes a success and should be highlighted— but does not regulate any of the other decentralized elements that today characterize the design of clinical trials: telemedicine visits, remote evaluation of outcomes, remote monitoring of source data, direct shipment of the investigational product to the participant’s home, obtaining samples at home or in local proximity laboratories, use of portable devices and sensors for data capture, and outcomes reported by the participant through applications.

Recommendation. Article [new]. Decentralized elements and hybrid designs. 1. Admissibility. Health-related research may incorporate decentralized elements, understood as those study activities that are carried out totally or partially outside the center of…

61. Absence of a proportionate regime for academic and non-commercially-sponsored research

Reference: Normative gap. Consistent with Article 27, paragraph 2 (mentions “formative academic research” only for the purposes of accelerated review) and Article 41 (additional benefits in research financed with public resources). · Priority: Medium

Issue. The draft applies the same set of obligations to the multinational clinical trial sponsored by industry and to the trial initiated by an investigator at a public university without commercial sponsorship.

Recommendation. Article [new]. Non-commercially-sponsored research. 1. Definition. Non-commercially-sponsored research is understood as that in which: (a) the sponsor is a higher-education institution, a health services provider institution, a…

62. Absent definitions, inconsistent terminology, and lack of consolidation in Article 5

Reference: Article 5 (definitions), in relation to multiple provisions. · Priority: Medium

Issue. Article 5 contains thirty-five definitions, but several of the concepts of the greatest operational weight in the draft are defined in the body of other articles or are not defined at all. This forces the addressee to reconstruct the meaning from scattered provisions, with the risk of divergent interpretation among committees. Concepts defined outside Article 5: “minimal risk” and “greater-than-minimal risk” (Art.

Recommendation. Move to Article 5, preserving their substantive wording, the definitions of: minimal risk; greater-than-minimal risk; exempt and low-intervention research, in accordance with observation C-20; unacceptable risk; substantial and non-substantial modification; coding, pseudonymization, and anonymization…

63. Independent data and safety monitoring committee: single mention without a regime

Reference: Article 25, paragraph 2, item 1. · Priority: Medium

Issue. The figure is mentioned a single time, as a monitoring strategy that the committee “requires”, without regulating its composition, its independence, its functions, its relationship with the ethics committee and with INVIMA, or the disposition of its recommendations.

Recommendation. “1. Independent Data and Safety Monitoring Committee. Its constitution shall be mandatory in studies that evaluate mortality or major morbidity outcomes, in studies with pre-specified interim analyses that may lead to early termination, in Phase III studies…

64. Articulation with Resolution 2378 of 2008 and formal adoption of ICH E6(R3)

Reference: Article 29, first subsection; recitals; Article 53. · Priority: High

Issue. The draft establishes a complete regime of composition, functions, and operation of the ethics committees (Arts. 27 to 29) and, at the same time, orders that their operating procedures align with the technical annex of Resolution 2378 of 2008, which contains its own regime of ethics committees for institutions certified in Good Clinical Practice.

Recommendation. Add a new article and adjust Article 29: Article [new]. Adoption of the Good Clinical Practice standard and normative harmonization. 1. For all purposes of the present resolution, the applicable Good Clinical Practice standard is the Good…

Observations on the Explanatory Memorandum

65. Untenability of the assertions of absence of economic impact and of budgetary availability

Reference: Explanatory Memorandum. · Priority: High

Issue. Neither of the two assertions is tenable in light of the content of the articles themselves, and their concurrence aggravates the problem: the Memorandum does not merely maintain that the economic impact is low or difficult to quantify, but that there will be no additional operating costs and that the draft does not contemplate any budgetary availability. That is, it is simultaneously asserted that the new obligations do not cost anything and that no source of financing is foreseen for them.

Recommendation. The issuance and implementation of the present administrative act generates additional operating costs, identified and estimated in the Regulatory Impact Analysis that accompanies this draft, as follows: (i) to be borne by the Ministry of Health and Social Protection, the design and operation of the National System of…

66. The Explanatory Memorandum bases the Ministry’s competence on a repealed decree

Reference: Explanatory Memorandum. · Priority: High

Issue. Two of the three competence norms invoked by the Explanatory Memorandum belong to a decree that is repealed. The invoked decree is repealed. Decree 4107 of 2011 —”By which the objectives and structure of the Ministry of Health and Social Protection are determined and the Administrative Sector of Health and Social Protection is integrated”, published in Official Gazette No.

Recommendation. The Ministry of Health and Social Protection is competent to issue the present administrative act on the basis of: (i) item 2 of Article 173 of Law 100 of 1993; (ii) item 7 of Article 2 of Decree 120 of January 30, 2026…

About bioaccess®

bioaccess® is a CRO specialized in first-in-human and early-feasibility studies, with regulatory operations across Latin America. We submitted these 66 observations because the detail of this regulation will shape Colombia’s competitiveness as a clinical-research destination. Talk with bioaccess® about your regulatory strategy →

This document reproduces, in structured and summarized form, technical comments submitted by bioaccess® to a public consultation; it is general information, not legal advice, and does not represent the position of any authority. The final text of the resolution may differ.