Access and understand the views and voices of your patient community
Having an efficient, integrated call center (sometimes referred to as the Patient Care Advocate office in hospitals) is a major approach to capture vitally important information from the customer/patient for pharma and healthcare organizations. Call centers are a key interface between healthcare professionals, patients and the company, and organizations need an effective streamlined process to capture, evaluate and route call feeds for appropriate action to be triggered.
Additionally, these voice of the customer (VoC) call feeds can provide a rich source of real world data (RWD). Pharma organizations need to understand how their therapeutics are being used by patients and understand a broader picture of patient reported outcomes. Are patients struggling with the packaging, with the formulation? Are there unexpected safety signals? Are patients switching? What competitor products are being mentioned? What are the trends from key opinion leaders or healthcare professionals? Additionally, regulatory agencies are becoming more focused on involving the patient voice in drug development.
Other sources of patient “voice” data
Secure emails are now a common means of communication between patients and providers, healthcare professionals and pharma organizations. This information is rich in knowledge and the language, vocabulary and grammar is often similar to that found from call center transcripts.
With the increased use of alternative care environments (such as urgent care centers, home care environments, telemedicine) and the adoption of healthcare technology being utilized at home important patient information can easily be missed. For example, telemedicine transcripts may also provide useful information from the patient that may not be directly recorded by the physician within the EHR.
The challenges in analyzing call center and patient voice data
The challenges in gaining benefit from these patient verbatims include:
- Language used is informal and not scientific
- Vocabularies are inconsistent
- Critical information is not easy to analyze. Information needs to be extracted, standardized and categorized so trends can be visualized.
Natural Language Processing-based text mining solution
By integrating the latest technologies with proven business practices, pharmaceutical companies can optimize and streamline their call center communications, which will yield significant returns and competitive advantages. Healthcare organizations will also gain valuable information on their patient population health.
Linguamatics NLP enables flexible search strategies to capture the key facts, add metadata to provide topics and themes, show sentiment, and transform the unstructured call feeds into structured data that can be visualized across global product teams, and population management teams alike.
Use Cases for Transforming Voice of the Customer data with NLP
Pharmaceutical organizations like Novo Nordisk, J&J, AstraZeneca and Pfizer use Linguamatics text mining solution to extract the essential nuggets of valuable information from customer call feeds.
Novo Nordisk has developed a combined integration of Linguamatics NLP, Tableau and AWS to automate analysis of customer call feeds. With the new system, Novo Nordisk has reduced manual work by FTEs, reduced vendor spend, automated the process of generating insights, and significantly broadened access to these insights across a global team.
“Linguamatics NLP is a very big time saver, and also introduces new capabilities for us - things we weren’t able to do before because we didn’t have the manual resources. Linguamatics is very powerful; it offers you a lot of features, a lot of functionality. Frankly, there aren’t other tools that are comparable—and I have seen quite a few tools, especially for unstructured data. Linguamatics NLP is definitely our tool of choice.”
Thierry Breyette, Associate Director, Information Analytics, Novo Nordisk
Johnson & Johnson uses NLP to annotate and categorize “voice of the customer” (VoC) call feeds, to gain insights into the real world use of their drugs. Researchers in the Predictive Analytics group have built an end-to-end workflow to process the call transcripts, using agile text mining to make sense of the unstructured feeds. The calls are categorized and tagged for key metadata such as caller demographics and reason for calling (e.g. complaint, formulation information, side effect, drug–drug interactions).
Smita Mitra, Principal Data Scientist at J&J estimated that to follow the call feeds for a single drug this automation & workflow saved 4 FTE weeks per year. Thus, even just monitoring 12 drugs means that this text analytics workflow would save around 1 FTE-year.
“We have built a great solution to address a real need."
Smita Mitra, Principal Data Scientist at Johnson & Johnson
Pfizer also uses Linguamatics NLP in an automated workflow to take call transcripts and process these unstructured feeds using advanced text analytics to understand trends and build predictive models from real world data from patients, medical assistants, pharmacists and sales representatives.