PM360 asked experts in data analytics how the digital transformation of healthcare is changing data needs and the best ways to overcome data gaps. Specifically, we wanted to know:
- How are the data needs and analytical capabilities of life sciences companies evolving following the digital acceleration of healthcare during the pandemic? What other factors or trends are impacting data and analytics?
- How can companies and/or marketers better identify their biggest data gaps? What are the leading reasons behind these data gaps? What can they do to improve their data acquisition strategies or extract better insights from the data they have?
The COVID-19 pandemic is accelerating the data analytics evolution that pre-pandemic forces sparked in pharma, such as virtual detailing and the fact that omnichannel communication is now a necessity to improve customer engagement.
Pharma leverages analytics to find patients, support them through the therapy lifecycle, allocate resources, and communicate with HCPs. This harnesses the explosion in customer and marketing data, analytics, and artificial intelligence and machine learning (AI/ML) on the cloud. During the pandemic, some companies made significant progress in this journey. New marketing data hubs and AI/ML-enabled omnichannel next-best-action processes enable the ability to deliver the right message, to the right customer, through the right channel, at the right time.
However, given the severe and ever-expanding talent shortage in the data science industry, success will require productization. The entire analytics value chain—from the data to insights, strategy, planning, to tactical direction—needs productization and integration on a single technology stack.
Such productization—analytics software verticalized for pharma—will increase effectiveness, accuracy, efficiency, agility, and democratization. Companies that do not leverage enterprise-grade pharma-ready analytics software are unlikely to maintain a sustainable analytics capability and competitive advantage.
One trend that will significantly affect data and analytics is the eventual disappearance of third-party cookies. For years, agencies have been using them not only to track website visitors, but also to serve hyper-targeted ads to specific audiences. Third-party cookies allow marketers to execute personalized, one-to-one marketing successfully. But that will disappear when Google and other browsers phase them out by 2022. Some predict Google may push the deadline further out, but as privacy concerns continue to intensify, it’s unlikely to happen.
Rather than waiting until this occurs, which our industry tends to do, digital agencies must begin planning and testing new, privacy-forward solutions now. Some examples would be to use first-party data that customers have already consented to provide. Another is to unify all data into one hub—a holistic data environment—enabling data that has already been collected to speak with one another, unleashing compelling insights.
Throughout this phase-out period, brand marketers must now—if they haven’t done so yet—start to build customer trust. Transparency will reinforce trust, which opens the door for expanded access of personal data. Incidentally, personal data is still relevant. It can still be used to better understand the target audience.
The pandemic forced the industry to quickly adopt innovative digital capabilities on a broader scale, which changed how we gather and examine information that we collect and analyze. We now need more data than ever before. And for it to be actionable in any meaningful way, it has to be available for scrutiny much earlier than it was previously. Qualitative data, such as how HCP offices changed their workflow to see patients safely, also became key. As a result, every discipline across the healthcare ecosystem sped up its evolution and adoption of new processes. Telehealth is a prime example.
Additionally, a new narrative has begun that focuses on patient outcomes and the value treatments have on patients’ lives instead of solely concentrating on script lift or new patient starts. While the latter will always be valuable, adopting new measurement practices is vital to patient betterment.
Another area of advancement making a significant impact is in geolocation and data set technologies. Geofencing has become increasingly accurate, and innovative approaches have been created to identify an individual’s exposure to a campaign. New technologies have allowed for the compilation and comparison of health data sets, enhancing our ability to measure patient outcomes.
The ability to recognize data gaps requires a willingness to move beyond the silo of a single disease state. Patients, providers, and payers are not experiencing healthcare in a vacuum. They will all interact with each other across various disease states that do not universally overlap with manufacturer portfolios, and, more importantly, they will engage in activities outside of healthcare. A holistic perspective is required to truly understand, and effectively communicate to, any healthcare audience.
Thankfully, data and technology advancements have empowered marketers to see the forest through the trees, provided they are willing to look. Robust medical and pharmacy claims data sets can be integrated with consumer behaviors to provide a comprehensive source of understanding for almost any audience. Advancements in machine learning and artificial intelligence have made it easier to move from insight to action; effectively bridging the gap between data science and customer engagement.
Marketers should promote a holistic understanding of their audience by using data and technology to adopt their perspective and engage with them on their terms. This will enable a deeper connection to your audience and a richer understanding of how to drive micro or macro behavioral changes.
My experience is that people often have a myopic view based on the data they have on hand. They’ve shaped the narrative for so long on the handful of data sets at their disposal that it’s difficult to embrace new data that may change existing hypotheses or expectations.
Historical data silos based on function have expounded the issue. For instance, medical/real-world evidence teams having sole access to claims/EMR data, clinical teams having sole access to clinical data, and commercial teams having sole access to downstream data sets such as Rx/sales data. Three ways to change that cycle are:
1. Break down the silos and truly embrace the notion of data democratization. Build a core center of excellence team across functions that shares high-level strategies and data needs.
2. Set expectations that you likely need more than one data provider. Complex disease categories often require different data providers based on specific business questions.
3. Rely on a trusted third-party analytics provider, that doesn’t sell data, to make sure you have the right data sets to answer the business questions at hand. All data providers will say they have the right data for you.
Pharmaceutical brands must address the lag time and disparate data sets across the healthcare ecosystem for marketers to better understand the underlying decision-making process of their target audience that impacts their medication decisions. Various factors such as educational background, specialty, and content preferences can alter the types of messages that will resonate with HCPs, but without those data sets, marketers won’t be able to go beyond prescribing behaviors and empathize with their intended audience.
To maintain a patient’s privacy, brands must ensure the individual’s identity is anonymized by employing multiple data sources. In order to do so, marketers can leverage synthetic data sets that are significant enough to represent a real-world population and develop predictive algorithms instead of relying on fractured data sources. This will give marketers access to answers that an organization would have gained via a first-party data set. To close the data gap, marketers can run tests to determine whether an algorithm will reduce the gap in the brand’s data points by drawing some inferences learned from running the model on the synthetic data set. With the additional insights, marketers can deliver more effective messages that resonate with an individual’s previously unknown behaviors to optimize a campaign.