Pharmaceuticals and biotech / Pharmaceuticals
Takeda: Drafting answers to medical information inquiries from healthcare providers
- Company
- Takeda
- Country
- Japan
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
According to a Salesforce customer story, Takeda's medical information teams frequently receive complex questions from healthcare providers about efficacy and safety data, administration and dosage, prescribing guidelines and contraindications. Ainhoa del Romero, Head of Global Medical Information, said the questions are diverse and the team needs to be prepared to answer quickly and compliantly. The same story explained that preparing a response meant searching a large volume of approved medical content, identifying relevant evidence and drafting accurate, compliant answers, which took significant manual effort and time.
Technology and data
Takeda built two agents with Salesforce Agentforce. When an analyst gives the SEIMEI agent the medical content to synthesize in Life Sciences Cloud, the agent extracts and surfaces key findings from PDF-based clinical literature and internal documents, and the analyst reviews the content and sources and directs how the information is compiled. The company extended SEIMEI to summarize transcribed conversations with healthcare providers at medical conventions into structured records for medical teams. The AIMI agent, which supports inquiry responses, finds relevant medical information and prepares a citation-backed draft response when a staff member enters the product and the provider's question. According to the same story, MuleSoft connects Veeva Vault to Data 360, where more than 3,100 regulated documents including clinical study reports, prescribing information and internal medical literature are ingested, and AIMI uses retrieval-augmented generation on top of Mavens Medical Information Cloud from Komodo Health to ground its answers in that content. Medical staff review every response for accuracy and compliance before sending it to the provider, inquiries are logged and auditable, and permissions restrict users to authorized information.
Results
According to the Salesforce customer story, AIMI has supported staff in handling more than 25,000 medical information requests from healthcare providers and 8,000 non-medical requests on topics such as order management, pricing, rebates and patient program eligibility. The results summary of the same story said insights are turned into actionable intelligence in hours instead of weeks. James Hooley, Head of Enterprise Medical Digital Innovation, said the next focus is improving both response times and the responses themselves, and that the team also sees an opportunity to bring in external datasets to augment the content of its responses.
Limits and open questions
The story does not disclose over what period, or in which countries and teams, the request volumes accumulated, nor when the two agents went into operation. The summary about turning insights into intelligence in hours gives no baseline or measurement method, so the size of the effect cannot be confirmed yet.
Sources
- Takeda turns medical complexity into trusted clarity with Agentforcesalesforce.com, Accessed
Compiled from public sources. These are not results from ATF Works customers.