Data Acquisition: Building the Foundation for Measurable Patient Support Programs [Whitepaper]
This series discusses practical strategies for Patient Support Programs (PSP) optimization, covering data collection, storage, and analysis. By implementing the best practices outlined, organizations can build more effective, measurable, and patient-centric support programs.
Patient Support Programs (PSPs) serve as a vital link between pharmaceutical companies and patients, ensuring access, affordability, and adherence to prescribed therapies. However, despite their importance, many PSPs struggle to demonstrate their true impact due to fragmented and inconsistent data collection and analysis. Without a clear data strategy, organizations risk basing critical decisions on incomplete or misleading information.
Measuring the performance and impact of a PSP involves much more than simply defining the metrics that matter and how frequently you’d like these metrics to be reported to you – that’s just the beginning. Data acquisition (be it directly collected or purchased) needs to be directly informed by the final measurement objectives in mind in order for there to be the correct pieces of data to analyze! This paper explores how to establish a robust data collection framework that not only measures success but also informs strategic decision-making.
Defining the Right Metrics: What Should Be Measured?
The foundation of any data strategy is a clear understanding of what should be measured. Metrics should directly align with the core objectives of a PSP, which often include:
Patient Access
How effectively do patients navigate reimbursement and financial assistance hurdles?
What do we consider “effective” or “successful” and, conversely, what is the threshold for what we would define as “ineffective” and needing improvement?
Given our target patient population, what is the expected mix of insurance coverage patients will have and how does that compare to the actual mix of insurance coverage that we process for our patients?
Affordability
To what extent are patients utilizing offered financial assistance programs? (e.g. How many copay cards downloaded are applied and used, and for how many times?)
How are patients finding out about our copay assistance offering?
How many patients receive copay assistance, and what impact does it have on adherence?
Adherence & Persistence
How long do patients stay on therapy?
How does participation in PSP offerings affect patients’ stay on therapy?
What is the soonest we can reliably measure this and attribute effect to engagement with PSP offerings?
What counts as participation in or engagement with PSP offerings? (e.g. Do we count those who just downloaded the copay card and never filled a script? Do we count those who just filled a script once?)
Does the impact of PSP offerings on patients’ stay on therapy meet our business objectives? (What if it does not?)
Operational Efficiency
Are service-level agreements (SLAs) being met by third-party vendors (e.g. the hub, the specialty pharmacy, etc.)?
How does turnaround time of certain milestones in the patient journey affect patient’s subsequent start of therapy or stay on therapy?
Are there bottlenecks in the process that can be eliminated or improved?
In thinking about what to measure (and therefore what data is needed), we must recognize how data feeds into the four levels of analytics, ensuring insights evolve from basic reporting to strategic decision-making:
1 — Descriptive Analytics – Understanding What Happened
At the base level, every program has “standard” metrics provide a retrospective look at PSP performance, offering basic but essential insights into program operations. These include metrics such as patient enrollment rates, case resolution times, turnaround times, therapy adherence rates. These are outcomes; they tell us factually what happened and are usually not very informative until we add some context (e.g. industry benchmarks, or internally-set targets) to evaluate them through a specific lens.
2 — Diagnostic Analytics – Understanding Why It Happened
This level delves deeper into root causes, allowing us to identify patterns and drivers behind program outcomes. We often want to “slice” or “segment” the data in different ways, wishing to observe any differences as we test hypotheses as to why we see the results that we do in our descriptive analytics. We might find, upon segmentation, that patient demographics play a larger-than-previously-expected role in therapy start rates – and this gives us a direction for further investigation. Or perhaps we may realize that average turnaround times for Benefits Investigations are really amplified by one specific payer type – now we can think about ways to improve processing for that payer type, or decide it’s a better reflection of how our PSP team is doing by separating out that payer type for turnaround time-related metrics.
3 — Predictive Analytics – Anticipating What Might Happen
With the help of your Insights and Analytics team, you can plan for leveraging historical data to do advanced modeling, forecasting trends and potential outcomes. The more historical data (what is used in descriptive and diagnostic analytics) available to be used as the basis of building predictive models, the more reliable the predictive models will be. (Barring seismic shifts in market dynamics such as significant changes in policy, treatment paradigm, etc.) Predictive analytics can help you proactively identify patients at high risk of discontinuation, for example, based on what you’ve already observed about patients’ journeys and therapeutic outcomes.
Predictive analytics is very exciting and of course can be very useful, but it’s a longer-term project that requires upfront planning and continued invested efforts. Talk with your Insights and Analytics partners to identify the problems you’d like to solve, what you’d like to get predictions on, how you’d utilize those predictions and what would make the predictions meaningful to you, and also what data needs to be collected, how much of it, and how that data needs to be organized, cleansed and prepared in order for it to be valuable in the creation of a quality predictive model
4 — Prescriptive Analytics – Determining the Best Course of Action
The “dream” of analytics is getting data-backed recommendations through analytics. Prescriptive analytics transforms insights into actionable intelligence, guiding you to reshape PSP strategy and coaching you on specific interventions to achieve your desired patient outcomes and PSP goals.
Today, you may have a digital campaign that attempts to re-engage patients with a series of pre-designed emails, triggered by defined business rules. With prescriptive analytics, you could identify what kind of message may resonate the most with your patients at risk of churning, personalized based on their previous interactions, their individual concerns and circumstances. These predictions need to be driven by a lot of data, and in the early stages of deployment, will likely need continued refinement. Getting to reliable prescriptive analytics is a long-term investment that requires dedicated resources every step of the way.
However, even if we do not yet have strong predictive and prescriptive analytical capabilities as an organization, we can leverage diagnostic analytics to form and test hypotheses, applying our best strategic judgement to determine the course of action most likely to be fruitful. For example, we may look at reasons for therapy discontinuation to see if there are unanticipated barriers patients face that we can enhance support for in our PSP offerings. We can examine copay usage by patient demographics to test our hypothesis that different patient age groups have different levels of comfort and familiarity with accessing and using our copay cards. There is much that can be learned and improved upon as we accumulate data, move beyond simple routine reporting and wait for the day a prescriptive analytical model becomes available to us.
The Data Challenge: What Needs to Be Collected?
Once the right metrics are identified, organizations must ensure that the necessary data points are being captured. Always include your analytical partner – whoever in or outside of your organization that you will rely on to store, process and analyze the data – in conversations where you’re determining what data points you need and when and how they will be collected.
A well-rounded data strategy should include:
Patient-Level Data (e.g. enrollment dates, insurance status, eligibility for financial assistance)
Engagement Data (e.g. timing of all types of support-related interactions, such as benefits investigation, benefits explanation, adherence reminders)
Operational Data (e.g. vendor performance metrics, call center wait times, case processing times)
Outcome Data (e.g. adherence patterns, therapy initiation times, reasons for patient drop-off)
Regulatory considerations—such as compliance with HIPAA and GDPR—must also be factored into data collection strategies. Organizations must balance the need for granular insights with patient privacy concerns, ensuring that data is anonymized where required and only collected with explicit consent. To that end, including your relevant Legal teams from the onset of data-related conversations is critical to successful analytics; ensure that your Legal teams understand what analytics you desire to have done and the impact of having those analytical results.
Starting on the Path to a Data-Driven PSP Strategy
A well-executed data collection strategy is the cornerstone of a high-performing PSP. By defining clear metrics, ensuring compliance, aligning all necessary stakeholders, organizations can move beyond mere reporting and leverage data-driven insights to enhance patient outcomes.
The next paper in this series will focus on data storage and maintenance, exploring how PSP data can be managed efficiently while balancing security, cost, and accessibility.