Label Language: Is English Enough, and Is Bengali Required

1. Key takeaways

§7.2 of the DGDA medical device registration guideline does not require labelling in Bengali. The provision requires marking in accordance with ISO standards and GHTF/SG1/N70:2011, and symbols in accordance with ISO 15223-1:2007, and says nothing about language. Practitioners interviewed for this series are equally clear: English labelling is acceptable, and even household devices do not require a local language. Some sources state that bilingual English and Bengali labelling is required; that claim conflicts with the other sources and is recorded as pending verification.

The distinction that matters is between what the text requires and what it permits. An absence of a language requirement is not the same as an express exemption from one, and the way the question is usually answered in practice, through a conversation with the reviewing officer, reflects that ambiguity rather than resolving it.

• Text: §7.2 sets no language requirement.
• Practice: English labels accepted for registration, including for home-use devices.
• Translations: documents in other languages are filed with an English translation.
• Conflict: claims of mandatory English and Bengali labelling are pending verification.
• Risk area: household and retail packaging, where registration-stage acceptance and retail expectations may differ.

2. Legal basis

Section 7.2 deals with marking content, not with language. It refers to international and ISO standards, names GHTF/SG1/N70:2011 for the marking requirements and ISO 15223-1:2007 for symbols, and leaves the linguistic question outside its scope. The practitioner statements below are recorded as practical estimates rather than as provisions, and the conflict between sources is recorded openly rather than resolved by preference, because an applicant who relies on the wrong reading carries the cost.

Two observations follow. First, the interviewing evidence is unusually direct for this market, since the statements address the language question in terms rather than by implication. Second, direct evidence from practitioners is still evidence about practice, and practice can vary between officers and between product categories, which is why the written confirmation described below is worth obtaining.

3. Expert insights

Insight 1: the absence of a language requirement is itself an answer, but an implied answer rather than an express exemption. Section 7.2 discusses standards and symbols and never mentions language. That can reasonably be read as meaning that Bengali is not mandatory. It cannot be read as preventing DGDA from requiring Bengali in an individual case. The difference between a provision that is silent and a provision that grants a freedom is the difference between an inference and a guarantee, and in our documents we never present the silence as a guarantee. The practical formulation is that English labelling is the working assumption, supported by the text and by practice, and that the assumption is confirmed where the product makes the question material.

Insight 2: the conflict between sources comes from conflating the registration stage with the retail stage. The practitioners interviewed were describing registration submission and label review. The sources that mention bilingual labelling appear to be describing the retail pack. Both can be true at once: English may pass at registration while the packaging that reaches an ordinary consumer is subject to a separate expectation. Household devices, meaning glucose meters, blood pressure monitors and thermometers, are where this divergence is most likely to appear, because they are the products a consumer reads without professional support. Treating the two stages as one question produces an argument that cannot be won, since each side is describing a different moment in the product’s life.

Insight 3: paper submission makes the language question depend on communication rather than on system design. In an online system, field language is fixed by the form and the answer is the same for every applicant. In a paper submission, an officer can raise the point case by case, and the outcome depends on who reads the file and what the local partner has established with the authority. That makes written confirmation more reliable than inference. A one-page letter from the local partner recording the authority’s position is worth more than a careful reading of a silent provision, because it addresses the actual decision-maker rather than the text.

Insight 4: translation obligations are separate from labelling language. The requirement that non-English documents be accompanied by an English translation concerns the dossier, not the pack. Applicants sometimes conclude from the translation rule that Bengali labelling must be required, which reverses the direction of the obligation.

4. Practical recommendations

GRZAN handles the question in three layers. At registration, labels are submitted in English, which both the text and practitioner practice support. For household and retail products, the local partner obtains written confirmation from DGDA before packaging is produced at volume, and the reply is kept on file. Where no written reply can be obtained, we use the conservative option: the main information stays in English, and the product name and warnings are added as a Bengali supplementary label. The cost is low and it removes the disagreement. We do not recommend producing retail packaging at volume on the strength of third-party experience that English is definitely acceptable, because the cost of reprinting a production run is far higher than the cost of one written confirmation.

• File in English, and say so. Record the basis for English labelling in the dossier note.
• Ask in writing for household and retail products. Address the question to the authority through the local partner.
• Keep the reply. A written answer is the only durable form of this evidence.
• Cost the conservative option before deciding against it. A supplementary Bengali label is cheap relative to a reprint.
• Translate the dossier, not the pack. Keep the two obligations distinct in the file index.

5. Verification checklist

• □ Confirmed that guideline §7.2 does not mandate Bengali, with the textual basis retained on file
• □ Registration labels submitted in English
• □ Non-English source documents accompanied by English translations
• □ Written DGDA confirmation arranged before mass production for household or retail products; conservative option used if no reply
• □ The conclusion that English is sufficient recorded as a practical estimate, not as a legal guarantee
• □ Cost of the conservative option, English body plus Bengali supplement, included in the packaging budget

Regulatory information verified as of 27 October 2026.

How GRZAN can help. GRZAN provides medical device and IVD registration, quality system, authorised representative and local testing support across South Asia, including Bangladesh. Our team prepares dossiers to DGDA expectations, manages submissions and deficiency responses, and supports post-market vigilance obligations. Business enquiries: info@grzanglobal.com