Most of us can gauge another person’s feelings or mood by observing their tone of voice, facial expressions, and overall body language. We then adapt our choice of words and mannerisms based on what we perceive. As long-term relationships take shape, we start to learn other people’s patterns and preferences—enabling increasingly productive and enjoyable interactions.
Patient support program agents attune to individuals’ emotions and needs to effectively answer caller questions and drive single-call resolution. It’s not always easy. Each interaction may be a first-time encounter, and most interactions occur via phone—in other words, without the benefit of visual cues. Additionally, questions are often complex and require agents to probe to understand the caller’s issue so that an answer can be researched and delivered with a high degree of accuracy.
How well are the agents in your program assessing and addressing patients’ needs? It is difficult to answer that question with traditional call center metrics like average speed to answer and abandonment rates. These measures address how a call is handled—not whether the caller’s question is successfully resolved or how the caller feels. Indeed, until recently, patient support programs have not deployed sentiment analysis to help focus agents by assessing and responding to people’s feelings in real time. These insights can help leaders improve program performance via insights focused on whether a problem was resolved to the caller’s satisfaction.
The IQVIA Patient Relationship Management (PRM) platform integrates sentiment analysis to drive a more thorough understanding of patient needs and behaviors. This heightened understanding of a patient during their treatment journey is a tool that supports improved health outcomes. For over a year, we’ve been using this capability to support multiple client programs covering various therapeutic areas—from rare diseases to more common conditions.
Sentiment analysis combines natural language processing and machine learning to review text or speech and then identify and categorize the emotional tone behind the words as “positive,” “negative,” or “neutral.”
In our PRM, sentiment analysis works in the background as an emotional detector. In the moment, it alerts agents if patients are beginning to express negative sentiment so they can course-correct. Over time, it generates insights about how a program may consider adjusting its approach, including specific word choices.
Our early experience indicates that leveraging these behavioral science insights improves the ability to meet patients’ needs. By combining technology with human empathy, we’re creating a more responsive and effective patient support ecosystem. What follows are three reasons we believe every program should be taking advantage of automated sentiment analysis.
Sentiment analysis is helpful in identifying barriers and drivers for individual patients in real time. In many cases, the human agents will still be best positioned to gauge how a caller is responding, but the AI-generated insights can provide validation and/or alert if the conversation is taking a bad turn.
For example, if the sentiment reveals patient frustration when discussing insurance requirements, the agent can quickly shift to a more supportive approach. That may include offering additional explanations or resources to better explain a complex concept involving insurance coverage. Although these adaptations may sound trivial, they are instrumental in preventing escalation, building trust, and resolving the person’s question in the first call.
Our experience affirms that for patients, words matter. Real-time sentiment analysis helps illuminate which words a specific person responds to in a positive and in a negative manner. With that insight, the agent can modify their language to generate more positive sentiments, which aligns to better caller understanding.
For example, even simple adjustments—substituting “please wait” with “just a moment”—leads to a positive feeling from patients. Our analysis has revealed some other specific word choices that produce negative or positive sentiments among patients. More specifically, our analysis found that patients exhibit negative sentiment toward these words and phrases:
We identified positive sentiment from patients when agents used these words and phrases:
Identifying these kinds of nuances in word choice can dramatically alter the tone of a conversation—helping build rapport and a clearer understanding even in short interactions.
Positive patient sentiment is ultimately driven by receiving well-researched answers to specific questions or challenges leading to single call resolution. Sentiment analysis offers a new way to evolve the call center metrics traditionally used to measure the efficacy of patient support services.
IQVIA research shows that patients consistently express positive sentiment when their problem is resolved—no matter how much time it took to get to that resolution. Our early findings suggest that traditional call center metrics such as average speed to answer and hold times are not levers that affect patient understanding, resolution of issues, or satisfaction. This challenges conventional theories regarding patient support program success. Instead, we are learning that there is an opportunity to redefine patient support program metrics to be more closely aligned with patient experience and therefore provide a more meaningful measurement of program efficacy.
Integrating sentiment analysis into patient support is just one example of how we can leverage AI-enabled capabilities to enhance the patient experience. As technologies advance, we anticipate more sophisticated capabilities for understanding subtle emotional cues to better predict patient needs.
Technology is helping us to improve patient support programs by systematically understanding and responding to patient emotions. This helps us to create stronger connections with patients, which we believe will improve health outcomes. As we continue our journey with AI and sentiment analysis, we are combining human empathy with technology to meet people where they are—emotionally and practically—to improve the patient experience and journey.
Want to learn more about how AI is transforming Patient Support? Connect with us to explore innovative strategies and technologies that are revolutionizing patient care to drive optimal health and well-being.
Advanced technologies like AI empower organizations to better identify patient needs and deliver timely, personalized support. In this article, explore how AI is transforming patient support services by enhancing personalization, improving quality and compliance, and evolving performance metrics to improve patient outcomes. Learn about the challenges of adopting AI, such as resistance to change, and the benefits, including real-time sentiment analysis and full-scale case audits.