The dichotomy of big data: Reconciling individualized intervention with population health goals
By: John Doyle, DrPH, MPH | July 27, 2016
This is the sixth in a series on trends impacting the biopharma industry.
"The interest of manufacturers in communicating RWE (real world evidence) now appears to be converging with the interest of many payers in using RWE to make coverage and reimbursement decisions," notes a September 2015 brief from the Network for Excellence in Health Innovation. The brief argues that RWE “is not just ‘big data,’ it’s the integration of multiple sources of data.”
Within the patient-centric model of care, value will be determined based on the ‘triple aim’ of improving the experience of care and health of populations, and reducing per capita costs of care. The future is not about a siloed dialogue around one drug, one physician, and one patient. The dialogue will be driven by clinically relevant information aimed at reconciling individualized interventions with system-wide efficiency and cost savings.
As a result of this shift, biopharma companies are migrating from a focus on achieving product approval based on safety and efficacy to one based on benefit-risk and economic profile. This is driving changes in the role of pharmacoepidemiology – the study of the utilization and effects of drugs in populations – throughout the drug development and commercialization cycle. This discipline applies methods of clinical epidemiology to understanding beneficial and adverse drug effects, impacts of genetic variation on drug effect, duration-response relationships, and the effects of medication non-adherence.
The pharmacoepidemiology function is a key element in evaluating RWE, transitioning from a reactive, technical specialization to a forward-looking strategic role focused on demonstrating value to external stakeholders. A focus on value requires better internal integration of clinical and commercial functions. Companies that achieve this and respond to the ‘evidence-based marketplace’ – solving provider, payer and patient needs with medicines that deliver broad population outcomes – will thrive.
As discussed at PharmExec’s Roundtable, ‘Epidemiology Arising’ (August 11, 2015), a major transition is underway in healthcare financing and delivery. To define value, a clear perspective is needed on what proof points stakeholders are seeking. They are also more patient-focused, with all key players endorsing patient centricity as a key goal. The challenge now is that stakeholders differ in their progress towards adopting the patient-driven value metric. There remain gaps in adoption, and geographic variations in readiness. To address these, a stronger commitment to integration of service and information is required. A common vision of population health linked to better value and outcomes is also needed. Pharmacoepidemiology can contribute to a strategy to make this happen.
We are entering a new era of healthcare catalyzed by real-world research. Demonstration of value will increasingly determine market access for a new product. Efficacy will give way to effectiveness, which includes cost as a key variable. Pharmacoepidemiology will play a vital role in integrating data-driven functions in these areas, including health economics and outcomes research and medical affairs.
What does this mean for biopharma?
Biopharma firms face continuing scrutiny over drug costs, and they have little choice but to demonstrate the benefits, risks, and outcomes of their products. Harnessing the power of ‘big data’, biopharma companies should upgrade and accelerate the way they collect, analyze, and apply real-world data to attain a premium position in patient centricity. Other healthcare system players are advancing rapidly on this front, with integrated hospital systems such as Kaiser and Geisinger applying ‘big data’ to develop guidelines and manage formularies. In parallel, managed care organizations, major retail pharmacies and drug distributors are increasingly leveraging claims data to understand better what motivates the patient.
Analysis of retrospective data can help elucidate a drug’s impact on value and outcomes as well as the effect of payers’ utilization controls such as tiered co-pays on these outcomes. As technological advances facilitate real-time data collection and advanced analytics, these retrospective evaluations are increasingly potentiating prospective observational studies. Given that these studies are generally not randomized, special attention must be paid to potential bias and confounding when conducting real-world research.
Pharamacoepidemiologic methods to address potential sources of bias and confounding should be deployed in the study design and analytic phases of research, and should set the standard for observational research, regardless of stakeholder or setting. Approaches such as the Good ReseArch for Comparative Effectiveness (GRACE) checklist, which aims to recognize non-interventional studies that are good enough for decision support, are helpful here.
Looking ahead, demonstration of real-world value and outcomes will increasingly determine initial and sustained market access for a new product, as efficacy and safety will give way to effectiveness and benefit-risk profiling. It will be incumbent on pharmaceutical companies to link the various data-driven, go-to-market functions, including health economics and outcomes research (HEOR), market access, commercial operations and medical affairs in a way that mirrors the integrated market they seek to serve. As this market continues to drive toward evidence-based decision-making, pharmacoepidemiology will play a crucial role in designing and vetting value and outcomes demonstration projects. The validity and reliability of this new order, real-world data hinges on sound research methodology and biopharma is uniquely positioned to provide this scientific oversight and patient insight.