Real World Data Analytics
we value real world evidence
In today’s healthcare industry, generating and disseminating the right real-world evidence is crucial for making informed decisions and identifying new opportunities as it provides insights beyond the controlled environment of clinical trials. By leveraging real-world data, healthcare professionals can better understand patient needs, tailor treatments to specific populations, and ultimately, improve the quality of care.
At BSP, we recognize the value and importance of real-world evidence, and are committed to delivering innovative solutions that help our clients meet their needs through advanced data analytics.
BSP REal world expertise
Our wide-ranging analytics ecosystem supports clients end-to-end, from choosing an optimal dataset through evidence generation and medical communications to support your business or clinical goals.
BSP possesses extensive healthcare domain expertise that covers various therapeutic areas, industry segments, and the complete product lifecycle.
Data agnostic approach and advanced analytics solutions
BSP features expansive experience working with a wide range of real-world data (RWD), including Electronic Health/Medical Records (EHR/EMR), Claims data, instrument data, Longitudinal Prescription (LRx), patient-reported outcome (PRO), and matched datasets.
How we harness RWD:
- R&D and trial optimization – Tap into evidence networks in oncology, neurology, cardiology, and other therapeutic areas to enrich your studies
- Epidemiology assessment – Customize solutions integrating primary, secondary, and social data to identify disease patterns, treatment outcomes, and patient characteristics
- Comparative effectiveness and marketing – Assess treatment safety and efficacy, and generate insights to drive competitive advantage and commercial effectiveness for novel products and technology
- HEOR and HTA acceleration – Create market-specific analytics roadmap and organizational alignment through health economic modeling to support regulatory and clinical decision-making
- Artificial intelligence and machine learning – Use our computer-science backed platform to quickly generate insights and predictions from big data
We partner with you to simplify and expedite the process across the continuum from development to after commercialization:
Our Case Studies
1. Leveraging EHR data to demonstrate the clinical importance and value of continuous patient monitoring with advanced algorithms.
When a medical device company aimed to generate evidence to demonstrate the clinical and economic value of its advanced continuous patient monitoring, we partnered with it to identify a cohort of patients exposed to hypotension and compare their outcomes with those of patients with stable BP. Defining hypotension exposure in acute care was complex, and BSP leveraged finer details in EHR data to generate a relevant patient cohort, whose outcomes could be analyzed. BSP used various thresholds and durations of continuous BP readings to develop algorithms for mean arterial pressure (MAP), to assess the associations between hypotension and serious clinical complications. A series of retrospective analyses were conducted including propensity-matched population were used and linear regression models. The objective was also to investigate the clinical outcomes associated with patient exposure to shock to convey the value of advanced continuous patient monitoring algorithms. Advanced analytics were used for this purpose including Bayesian Information Criterion (BIC) statistics, multivariable GLMs, and logistic regression. An economic model was developed, and several peer-reviewed articles were published in top-tiered journals.
2. Using real-world data fields to describe patient journeys and demonstrate the value of compliant CPAP machine use in terms of clinical, economic, and quality-of-life outcomes.
Standards for obstructive sleep apnea (OSA) diagnosis and continuous positive airway pressure (CPAP) utilization compliance are ill-defined, posing challenges to demonstrating meaningful outcomes for a medical device company. A “fresh” approach was needed to utilize EHR/claims data fields (e.g., timeline, analysis availability/ reproducibility, potential cross-platform linkage) in order to arrive at a compliance definition and show how this matters in terms of clinical, economic, and QoL outcomes. BSP portrayed a comprehensive sleep-breathing disorder picture using EHR data to better understand the patient’s journey and time to therapy. Patients were identified based on specific inclusion criteria comprising medical codes and physiologic parameters, and specific endpoints of interest were analyzed. Machine learning algorithms also determined the most important predictors of outcomes, which can help clinicians tailor patient management accordingly. Predictor categories included patient demographics, payor types, patient comorbidities, sleep test type, and other variables.