Clinical Evidence in Indonesia: Class C/D Requirements and Local Clinical Verification

1. Key takeaways

Clinical evidence follows the classification, and the classification is not a matter of opinion. Class C and Class D devices must submit preclinical or clinical evidence; Classes A and B are generally not required to, although reviewers may ask for supplementary clinical evaluation or literature argument for specific products. The judgement always lands on the classification result — never on the manufacturer’s conviction that the product is mature and well established.

For most devices the practical question is not whether clinical evidence exists but which form it may take. A clinical evaluation report built on equivalence and literature is the common route; local clinical verification in Indonesia is the exception, but it is a mandatory exception in defined cases, and the defined cases have grown with the arrival of AI-specific regulation.

• Default rule: Class C and D require preclinical or clinical evidence; A and B generally do not, subject to reviewer discretion for specific products.
• CER route: acceptable where a comparable equivalent device and sufficient literature exist; it stands on the equivalence argument, the literature’s coverage of the submitted intended use, and clear conclusions.
• When CER fails: innovative devices with no similar product on the Indonesian market may need new clinical data or local clinical verification.
• Local verification, case 1: innovative high-risk Class D devices — Kemenkes may require local clinical verification in the Indonesian population.
• Local verification, case 2: imported AI devices — under KMK No. HK.01.07/MENKES/951/2026 (7 September 2026), imported AI products used for diagnosis, screening or clinical decisions must complete local clinical verification in Indonesia and submit the results within 1 year after the licence is granted (see article 29).
• HTA: affects access and procurement, especially public channels — a separate line from marketing authorisation review.

2. Legal basis

The clinical evidence requirement for Class C and D devices is set out in Permenkes No. 11/2025, Appendix 3. The AI-specific local verification obligation comes from KMK No. HK.01.07/MENKES/951/2026 of 7 September 2026. Clinical evidence occupies its own section of the ASEAN CSDT, and the role of health technology assessment reflects public statements by Ditjen Farmalkes. The references below were verified on 26 October 2026.

On HTA, Ditjen Farmalkes has made clear that it uses health technology assessment to ensure the quality and accuracy of devices and to evaluate their benefits. HTA mainly affects the access and procurement level, particularly public channels, and runs on a separate track from marketing authorisation review — but it has a substantive impact on competitiveness in government procurement, which means the commercial case for a device can turn on evidence work that sits outside the registration dossier.

3. Expert insights

Insight 1: the CER route is a three-legged argument, and all three legs must hold. A clinical evaluation report is acceptable where there is a comparable equivalent device and sufficient literature. It stands on three points: the equivalence argument must be sufficiently made, the clinical literature must cover the intended use as submitted, and the conclusions must be clear. Where the product is an innovative device with no similar product in Indonesia, the CER is difficult to sustain, and the file may need new clinical data or local clinical verification. Recognising which situation the product is in — before the submission is designed — is the entire planning question.

Insight 2: the AI local verification rule is the hardest line in current Indonesian clinical policy. Under KMK No. HK.01.07/MENKES/951/2026, effective 7 September 2026, imported AI products used for diagnosis, screening or clinical decisions must complete local clinical verification in Indonesia and submit the results within 1 year of the marketing authorisation being issued. This shapes the entry rhythm for AI imaging, AI-assisted diagnosis and clinical decision support products more than any other single requirement: the licence and the verification now run on parallel clocks that meet one year out.

Insight 3: CE and FDA files buy efficiency, not exemption. A complete CE technical file or FDA 510(k) record reduces the volume of data a reviewer asks to be supplemented and thereby shortens the cycle indirectly, and it markedly improves the reviewer’s ability to judge the product’s maturity and evidence completeness. But it replaces no mandatory requirement. The same principle seen elsewhere in this guide applies here: reference-country approvals strengthen the file; they do not subtract from the Indonesian evidence set.

4. Practical recommendations

• Decide whether local verification is needed at project initiation. Make it a distinct budget and schedule item from day one, not a discovery during review.
• Match the evidence route to the classification, not to internal beliefs about product maturity.
• Test the CER’s three legs before committing to it: equivalence argument, literature coverage of the submitted intended use, and clear conclusions.
• For innovative Class D devices, assume Kemenkes may require Indonesian-population local verification and plan the option accordingly.
• For imported AI products, schedule backwards from the 1-year post-licensing deadline set by KMK No. HK.01.07/MENKES/951/2026.
• Use CE and FDA documentation to pre-empt data requests, while keeping the Indonesian evidence requirements intact in the plan.
• Treat HTA as a commercial workstream where public procurement is part of the market strategy.

The reason for early decisions is arithmetic: the preparation cycle for clinical evidence runs far longer than the review cycle itself. Discovering at the review stage that local verification is needed amounts to rescheduling the project — the review clock stops while a verification programme that should have started months earlier is only being scoped. The cheapest point to answer the local-verification question is before the first rupiah of dossier work is spent.

5. Verification checklist

• □ Classification confirmed and evidence depth matched to it (Class C/D preclinical or clinical evidence per Permenkes No. 11/2025, Appendix 3)
• □ CER route validated: equivalence argument, literature coverage of the submitted intended use, clear conclusions
• □ Innovative-device status assessed against Indonesian market comparators
• □ Class D innovative high-risk products: local verification option planned for Kemenkes discretion
• □ Imported AI products: local clinical verification scheduled against the 1-year post-licensing deadline of KMK No. HK.01.07/MENKES/951/2026
• □ CE/FDA documentation positioned as request-reducing evidence, not as a substitute
• □ HTA implications assessed where public procurement is targeted

How GRZAN can help. GRZAN provides medical device and IVD registration, quality system, authorised representative and local testing support across Southeast Asia, including Indonesia. Our team prepares dossiers to ASEAN CSDT structure, manages AKL/AKL-I submissions with Kemenkes, and supports post-market vigilance obligations. Business enquiries: info@grzanglobal.com