Services
01–03
Three engagement types. Most projects combine two.
Each begins with the question you need answered, not with a package.
Who uses real-world evidence
The same data answers different questions depending on who is asking. We have worked on the payer, provider, and manufacturer side of each of these — and inside a manufacturer, with market access, commercial, R&D, and medical affairs on the same program. See how that worked on mSToPS.
Payers
- Formulary decisions
- Care and disease management
- Risk stratification
- Utilization and outcomes
Manufacturers
- Repositioning and new indications
- Safety and efficacy in large, longitudinal populations
- New and linked data sources
- New trial designs and precision therapy
- Access and reimbursement, from the payer and provider perspective
Providers
- Care delivery
- Adherence
- Cost
- Putting interventions into practice
Regulators
- Safety
- Efficacy
Patients
- Patient-reported outcomes
- Treatment choices
- Analysis and visuals they can use
Global
- Diverse data sources
- Electronic medical records
Service 01 — Evidence generation
The problem
Your trial answered a controlled question. Payers and HTA bodies are asking an uncontrolled one.
Our approach
We start with the decision the evidence has to support, not with a study design. Feasibility comes first — whether the question can be answered with data that exists — and only then do we write a protocol. Senior researchers stay on the work from design through interpretation.
How it runs
- 01
Feasibility
Whether the question can be answered with data that already exists, before anyone writes a protocol.
- 02
Protocol & SAP
Design, endpoints, and the analysis plan agreed and fixed in writing before any analysis is run.
- 03
Execution
Cohort built, analysis run, results checked against the plan rather than against expectations.
- 04
Interpretation
What the result supports, what it does not, and how it holds up when a payer pushes back.
Methodology
- Study design and feasibility assessment
- Protocol development
- Survival analysis
- Propensity scoring
- Multivariate modeling
- Incidence and prevalence studies
- Event rates defined from the literature
- Post-marketing adverse-event surveillance algorithms
Deliverables
- Protocol document
- Statistical analysis plan
- Analysis datasets and code
- Results report with interpretation
Service 02 — Real-world data analytics
The problem
You hold claims or EHR data and no analytic capacity to interrogate it.
Our approach
We build the cohort, run the statistics and hand back something you can act on — including the code, so the work is reproducible rather than a black box. Extraction and analysis are done by the same team, which is why the cohort definition survives contact with the data.
How it runs
- 01
Data assessment
What is actually in the source, what is missing, and which questions it can carry.
- 02
Cohort definition
Study period, population, and outcome fixed first; inclusion and exclusion applied step by step, with every drop accounted for and comorbidities controlled for.
- 03
Analysis
Models fitted, assumptions tested, sensitivity analyses run before anything is reported.
- 04
Handover
Results, interpretation, and the code behind them, so your team can re-run and extend the work.
Three retrospective designs
Each runs on the same spine — baseline data, an event or intervention, outcomes data — and is built for a different purpose.
Methodology
- SQL extraction across Hadoop, Teradata, and TOAD
- Analysis in SAS and SPSS
- Logistic regression and ANOVA
- Machine learning on hierarchical files
- Cohort development and risk stratification
- Matched control populations built in PL/SQL
- Extraction from MySQL and Oracle into SAS, SPSS, Minitab or Excel
- Paper or web-based data collection where no database exists
Deliverables
- Defined cohort with attrition accounting
- Risk stratification model
- Analysis code
- Results and interpretation
Service 03 — Access & value strategy
The problem
You have the evidence. The payer conversation still stalls.
Our approach
Evidence does not move an access decision on its own; it has to be framed for the person making it. We have sat on the payer side of that table, and we build the value case in the terms a plan actually evaluates — clinical and administrative measures that are simple enough to execute and quick enough to show a return.
How it runs
- 01
Evidence review
What you already hold, where the gaps are, and which of them actually matter to the decision.
- 02
Value framing
The case rebuilt in the terms a plan evaluates, not the terms the trial was designed around.
- 03
Contract design
Measures simple enough to execute and quick enough to show a return on both sides.
- 04
Payer-ready output
A dossier and a summary that survive being forwarded without you in the room.
Methodology
- Value proposition development
- Value-based and outcomes-based contract design
- Care management program design
- Payer–provider value propositions
Deliverables
- Value dossier
- Contract framework
- Payer-ready summary
Not sure which of these fits?
Describe the question you are trying to answer and we will tell you what it would take to answer it — including if the answer is that the data does not exist.