Real-World Outcomes Solutions: Data-driven, Value-based

Harness your data’s full value to drive real-world outcomes. Whether your data comes from EHRs, claims, clinical programs, or incentive & reward programs, you can use the power of our analytics platform, REFS™ to extract underlying cause-effect relationships and find the mechanisms that explain observed phenomena.

Understand & Compare Effectiveness

Reveal what your data is saying about what works for whom. Use REFS™ high-resolution and interactive capabilities to identify optimal care paths or conduct rigorous head-to-head treatment comparisons using real-world data.

  • Discover hidden interactions of treatments, care processes, conditions and patient characteristics that compromise safety or cause adverse events;
  • Chart the structure of complex treatment or service pathways that improve health;
  • Recognize which treatments or services are more effective or where gaps exist.

Real World Example

Treatment Algorithms for an Approved Multiple Sclerosis Drug

In collaboration with a leading biopharmaceutical company, GNS Healthcare developed a treatment algorithm to match Multiple Sclerosis (MS) patients with therapeutic interventions. Using patient demographic and clinical data along with molecular and imaging biomarkers, REFS™ predicted disease exacerbation at 6 months with 98.6% accuracy, correctly predicted the results of a Phase II clinical study, and identified biomarker combinations that could identify patients and time points more likely to benefit from the drug.

Improve Targeting & Prediction

Identify how you can reduce adverse impacts or amplify positive results. Use REFS™ to create models that target at-risk patients and fuel intervention programs that improve health.

  • Uncover the opportunities where you can make meaningful changes;
  • Target the specific circumstances and patient characteristics with the highest potential for impact

Real World Example

Automatic Detection of Adverse Drug Effects (ADEs)

In collaboration with a leading Pharmacy Benefit Manager (PBM), GNS Healthcare predicted adverse drug-drug interaction events from demographic, medical claims and pharmacy data. Such adverse drug reactions are estimated to cost us Billions annually1,2. Our REFS™ platform automatically identified drug-drug combinations producing both adverse (negative) and synergistic (positive) effects.

Accelerate Design & Innovation

With REFS™ ‘calculus of causality’, transform your massive-scale data assets and computing architectures into a powerful scientific instrument. Use it for synthetic design to transform how you consider changes to your offerings, or as a powerful strategic lens to identify new opportunities.

  • Quickly simulate the range of possible outcomes from changing treatments or services. Accelerate exploring alternatives, avoid changes with little or no value, and identify changes holding the greatest value.
  • Discover new value streams – Use REFS™ as a strategic tool to explore your ‘opportunity landscapes’, formulate new problems, and prioritize areas for further development based on strong data-driven evidence.
Real-World Value

REFS™ helps you realize value quickly. By harnessing the power of real-world data, you can easily translate models into analytic forms that fit your organization’s operations and data flows to capture value right away.

Examples of Uses

  • Drive use of effective treatments
  • Reduce avoidable readmissions and inpatient complications
  • Improve Drug Safety
  • Reduce fraud, waste and abuse
  • Reduce gaps in care
  • Refine patient monitoring
  • Enhance compliance
  • Enhance care processes
  • Personalize care plans
  • Accelerate value-based strategies
  • Detect new drug indications
  • Prioritize data investments


1 Drug-related morbidity and mortality. A cost-of-illness model. Johnson JA, Bootman JL. Arch Intern Med. 1995. 155(18):1949–1956. 2 Systems analysis of adverse drug events. ADE Prevention Study Group. Leape LL, Bates DW, Cullen DJ, Cooper J, Demonaco HJ, Gallivan T, et al. JAMA. 1995. 274(1):35–43.