Illuminating the path to success: Using AI to develop a strategic plan for differentiated patient engagement

Patient engagement continues to be an underutilized potential driver of pharmaceutical product adoption and growth

We have observed increasing interest from pharmaceutical companies in moving beyond the “physician-as-customer” model to engage patients and drive informed patient decision-making. Studies have suggested that patients are more satisfied if more involved in their treatment decision-making.^1,2 Greater patient involvement can also lead to improved outcomes and lower costs³ and, in competitive spaces, allow patients to advocate for the product that is best for them.

However, only ~half of providers are providing patients as much information as desired, a proportion which has barely changed in many years.⁴ This is despite providers consistently being rated as a highly trusted information source, both across the breadth of Putnam’s client work as well as in third-party studies.⁵

Patients also have a tendency to overestimate their own understanding of their disease and treatment. For example, one study demonstrated that patients with cancer over-estimated their mortality risk by four-fold or more.⁵

All of this suggests that there is a persistent and profound knowledge gap at the patient level, and manufacturers have an opportunity to differentiate by helping fill this gap. Recently, Putnam has been using generative AI to analyze competitor patient engagement strategies to identify untapped market opportunities for new entrants. By employing AI-powered tools, pharmaceutical companies can gain a competitive edge, enhance their marketing strategies, and improve patient knowledge and engagement.

AI-powered analysis of competitor patient-facing product websites can identify content gaps

One valuable resource for understanding competitor messaging strategies and market gaps in the US is patient-facing websites. While on the surface many look the same and seem to offer similar content, there are frequently valuable insights into how competitors position their products, interact with patients, and attempt to address their needs that can have deep subconscious impacts. However, manually analyzing these websites can be time-consuming and prone to human biases. AI offers a solution for automating this process and extracting actionable insights in a way that may reduce the influence of subconscious human bias (although it is important to remember AI can have biases of its own).

Putnam has recently used AI-powered tools to analyze patient-facing websites to reveal competitor messaging strategies, extracting text content to identify recurring themes and messages. Specifically, natural language processing algorithms can extract and interpret information from unstructured text contained on a website and machine learning algorithms can identify patterns and trends within this data. This analysis provides our clients with rich understanding of how competitors are communicating with patients, allowing them to refine their own messaging strategies and differentiate themselves. By comparing competitor strategies with other sources of information (e.g., primary market research with physicians and patients), Putnam has identified gaps in the market where patient needs are not adequately addressed.

For example, in a recent engagement in a highly competitive oncology disease, many competitors were focusing their efforts on educating patients about mechanistic differences between their products. However, separate market research identified that, in fact, patients tended to have more questions about the safety and tolerability profile, particularly when newly diagnosed and facing their first line of therapy.

This suggests that patient messaging relating to our client’s differentiated mechanism will be most successful if it can be linked to what side effects the patient will actually experience in the short term during their first weeks of treatment. In this case, it created an opportunity for education on how a more selective mechanism with fewer off-target effects is a likely driver of the lower rate of side effects seen vs more traditional therapies. This allowed our client to tailor their launch marketing campaign to effectively communicate their unique value proposition.

AI can aid in crafting messages with differentiated tonality that strongly resonate with patients

AI algorithms can also help identify tonality and sentiment in patient-facing websites and other engagement materials. Putnam works across a vast swath of disease and therapeutic areas, but patients are virtually universally experiencing significant emotion relating to their condition.

Different disease spaces have unique characteristics, and the tone on patient-facing materials should reflect these differences. For instance, materials for patients with cancer are more likely to need to adopt compassionate and empathetic tone, acknowledging the emotional challenges these patients are facing. However, the needs may also be different for those newly diagnosed, where there may be a greater need for basic disease education, vs those in later lines of treatment, where there may be a greater desire for detailed information on specific mechanisms and why this mechanism will work when others have failed, creating a need for hopeful and educational tonality. Emotional needs are also likely to be different for a patient with cancer and a patient with psoriasis, both of which can have profound impacts on patients but in different ways.

Patient emotional needs can also vary considerably from patient to patient, even in the same disease space. Identifying any mismatches between competitor tonality and actual patient emotional needs can help create patient-facing content that more strongly resonates with the target audience.

In a recent engagement relating to a disease that is, for most, a devasting diagnosis, Putnam analyzed competitor tonality on patient-facing product websites. We found that most competitors adopted informative, supportive and educational tones. However, hopeful, empathetic, or empowering tonality was lacking. Tone and content also tended to be relatively abstract. A minority of competitors also used directive, professional tones that have the risk of feeling impersonal to a patient facing highly personal, emotional decisions. Collectively, these findings were used to develop suggested potential tonality approaches our client can use to stand out in a crowded competitive space.

We also found that patients have a variety of emotional needs—not all patients have exactly the same experiences or are going to react in the same way. However, many of the manufacturers in this space were taking a “one-size-fits-all” approach to patient engagement. We were able to recommend a suite of different patient-facing offerings that would meet their various needs (e.g., having materials available with a variety of depths of information and different tonalities based on the type of information being provided, while still maintaining consistency with the overall brand strategy and target positioning).

AI can ensure continued branding consistency across messages and channels

As messages are developed, AI can be used to pressure-test tonality and suggest revisions and refinements to messaging to create differentiated tonality that meets patients where they are at in their emotional journey.

Patient-facing content across different platforms (e.g., websites, social media, television and radio advertisements, etc.) can also be analyzed for consistency. Maintaining consistency across platforms and over time (beyond that required from a regulatory standpoint) can help to establish trust and credibility with patients. AI can scan for this consistency and identify opportunities to strengthen.

Lastly, patient needs and expectations are not static and neither are markets or competitor strategies. With regular application, examining both self and competitor materials, AI analysis can establish a feedback loop and monitor for opportunities to improve content and tonality of patient-facing materials. This can help companies understand the impact of tone and make necessary adjustments to improve patient engagement and satisfaction.

With over 30 years of experience, Putnam has been instrumental in helping manufacturers shape messaging strategies for patients as well as physicians and payers. Now, leveraging the power of AI, we empower our clients to maximize their differentiation in patient engagement strategies, revolutionizing the way they connect with patients and drive better outcomes for all.

References:

1. Birkeland S, Bismark M, Barry MJ, et al. (2022). Is greater patient involvement associated with higher satisfaction? Experimental evidence from a vignette survey. BMJ Quality & Safety 31:86-93.
2. Osborn R, Squires D. (2012) International perspectives on patient engagement: results from the 2011 Commonwealth Fund Survey. J Ambul Care Manage. 35(2):118-28.
3. Greene J, Hibbard JH, Sacks R, Overton V, Parrotta CD. (2015) When patient activation levels change, health outcomes and costs change, too. Health Aff (Millwood). 34(3):431-7.
4. Office of Disease Prevention and Health Promotion. (2023) Healthy People 2030. US Department of Health and Human Services, National Institutes of Health. https://health.gov/healthypeople
5. Hoffman RM, Lewis CL, Pignone MP, Couper MP, Barry MJ, Elmore JG, Levin CA, Van Hoewyk J, Zikmund-Fisher BJ. (2010) Decision-making processes for breast, colorectal, and prostate cancer screening: the DECISIONS survey. Med Decis Making. 30(5 Suppl):53S-64S.