• Gap analysis – summary of current state vs future vision and current vs future technical architecture
• Use case assessment – definitions of users and use cases and map to analytics capabilities
• Build vs buy assessments – analysis of strategic and cost impact of various options
• Policy and compliance review – review of federal and state reporting requirements
• Organizational resource assessment and planning – assessment of technical and business resources required for proposed solutions
• Timeline and implementation planning – recommendations for implementation sequencing and cadence 

• KPIs and metrics – well-formed healthcare KPIs and metrics that provide actionable knowledge to users
• Dashboard user experience – dashboard designs that are intuitive and visually-appealing
• Reporting data architecture – optimized structures of underlying data sources appropriate to the analytics tool and user need
• BI tool recommendation –  assessment of options and client-specific recommendation
• User access and security – recommendations for ensuring secure access to data and structuring reporting data marts to ensure protection of PHI

We help develop data pipelines and curated datasets to help organizations get more value of their key data assets.  Our data integration work includes:

• Data acquisition – data acquisition and data profiling for net new data sources
• Data integration – development of data pipelines through stream or data warehousing for normalized and curated datasets; integration of vendor solutions, especially healthcare-specific tools, to accelerate commoditized data normalization of claims, EHR, administrative, and reference data
• Data maintenance – processes to maintain updated and relevant data (client data and healthcare-specific datasets and ontologies such as diagnosis code definitions often change over time)
• Data governance – supplemental data documentation and data visibility and compliance with organizational governance standards

We help develop and deploy data science capabilities such as predictive models and data labeling for healthcare applications.  We address particular challenges in deploying healthcare models such as the need to develop clinically-comprehensible “explainable” models as well as known risk in reinforcement of bias from underlying data.  How we can help:

• Model prototyping and feasibility assessment – assessment of whether an AI model can effectively solve a particular problem
• Risk stratification models – development (or deployment of vendor solutions that implement) risk stratification, prediction, revenue cycle, and NLP models
• Model maintenance – management of model performance shift and deployment
• Model integration into existing operations – integration of model outputs into existing workflows such as case management outreach