Life sciences companies increasingly are seeking new models to connect expanding sets of disparate, non-identified patient-level data streams to better develop and commercialize next-generation medicines. The need for practical, evidence-based approaches is especially acute at a time when scientific advances, the growing need for new medicines, and constrained budgets for healthcare and other social services are colliding.
According to a new QuintilesIMS Institute report, connected insights derived from big data enhance scientific and therapeutic expertise. These insights also drive informed decisions along the entire continuum of care—from pipeline to portfolio to population health. As health systems accelerate the transition to value-based use and payment approaches, mastering the collection and interpretation of data is essential to advancing medicine and medical technologies.
“Drug costs and spending are being scrutinized more intensely worldwide. At the same time, the challenges associated with developing new medicines remain extraordinarily high,” said Murray Aitken, senior vice president and executive director of the QuintilesIMS Institute. “This confluence of pressures will lead to a crisis in biomedical innovation unless the industry applies a more strategic approach—one that links actionable insights earlier in clinical development and commercialization.”
The report—Connecting Insights: Bringing Better Outcomes from Pipeline to Patient Using Data and Analytics—highlights nine evidence-based approaches to transform drug development and commercializaton.
1. Improving study design by better linking indications and target endpoints.
In every randomized controlled trial (RCT), therapeutic agents must be tested against defined potential indications. Often, a clinical study will succeed or fail based on the primary outcome endpoints that are chosen. Real-world insights, shared data and large datasets that supplement RCTs can help identify precise and useful primary outcome endpoints. They also can enable more rapid validation and acceptance of surrogate endpoints.
2. Improving patient targeting through better inclusion/exclusion criteria and biomarker-based segmentation approaches.
Biomarkers are a measurable indicator of a biological state or condition. Historically, biomarker identification has been conducted during pre-clinical research on a small scale. The advent of gene expression profiling, genetic analysis and genomic profiling now enable target and biomarker discovery to be applied beyond pre-clinical research to clinical studies and practice.
3. Optimizing site identification and improving site efficiency.
Maximizing the value of a clinical asset largely depends on smart program and trial designs. Where clinical trials are located is critically important—requiring consideration of many site-selection measures. Applying new data types and innovative analytics can increase patient-per-site enrollment while minimizing protocol amendments.
4. Executing clinical studies more efficiently via risk-based monitoring.
Clinical trial site monitoring accounts for an estimated 21 percent of the total cost of clinical trials, making it the single highest cost factor for drug development. Risk-based monitoring (RBM) is an effective way for identifying site-related risks before they create issues and unnecessary costs. These approaches require large amounts of longitudinal data covering sites, study types and indications. Using big data modelling tools, risk patterns can be identified and mitigation strategies developed—particularly for known failure points that follow study start-up.
5. Improving standards of care and outcomes analysis to determine optimal care paths.
Applying rigorous analysis to real-world treatment patterns, resource utilization and patient outcomes can yield actionable insights for healthcare stakeholders. Electronic medical records combined with registries, claims data and other sources of evidence can complement clinical research and drive innovative patient care solutions.
6. Lifting levels of patient adherence in medication use.
The economic consequences of patient nonadherence are estimated to account for 4.6 percent of global total health expenditure, representing the majority of the world’s total avoidable costs linked to suboptimal medicine use. The movement toward outcomes-based contracting, the shift of risk from payer to provider and the greater use of value-based insurance design are all contributing to a greater focus by stakeholders on this issue. Advances in technology and data analytics are enabling new approaches that: recognize the diversity of underlying drivers of patient nonadherence; apply predictive analytics to stratify patients based on risk and consequences of nonadherence; using a broader range of interventions including low-cost smartphone-based apps; and integrate adherence measures into patient records.
7. Reaching physicians and patients with appropriate medicines.
Linking real-world data with analytics is generating new insights to better understand patient habits, health journeys and engagement with healthcare professionals, as well as to measure patient outcomes. This also helps identify physicians likely to have patients who can benefit from new medicines so that manufacturers can efficiently provide relevant educational information to those prescribers.
8. Evaluating medical value and real-world evidence in decision-making.
A proliferation of frameworks and assessment approaches now exist in response to increased calls for new medicines to demonstrate “evidence of value” to patients and health systems. These new approaches increasingly draw on information and analysis that go beyond traditional RCT data and include electronic medical records, claims data, mortality data, consumer data, registries, data collected in observational studies and chart reviews.
9. Applying post-market data to drive portfolio planning and predict market dynamics.
The biopharmaceutical industry increasingly is focused on niche segments such as rare diseases and patient sub-populations. The relative scarcity of precise data within these sectors presents challenges when making commercial decisions during asset development and deployment. Companies are addressing these challenges by leveraging real-world insights to measure patient populations in complex settings and using advanced analytics to bridge gaps in data availability.
The report cites four critical hurdles to realizing the full power of connected insights: maintaining data privacy; incorporating dynamic approaches to the application of data; ensuring transparency when multiple data sources and analytics are applied; and collaborating with diverse constituents who may have conflicting interests.
Said Aitken: “Properly harnessed, data analytics can drive remarkable progress in allocating healthcare resources, developing and commercializing new medicines, as well as improving population outcomes.”
The full version of the report is available here for download.