Our solution – SHAARPEC for population

Our solution

The solution to the challenges of the healthcare systems is to link all clinical, financial and resources data in one data layer and then analyze it from the perspective of the patient’s pathway rather than the organizational structure. We call this strategic, end-to-end analysis solution SHAARPEC.  

Patient pathway

Patient pathways are the time-sequences of patient meetings in different parts of the healthcare system over time. These meetings – patient encounters – are linked to different encounter attributes, such as diagnoses that the patient received, lab values, medications prescribed, cost of the encounter.  

Patient pathway analysis

SHAARPEC enables data-driven decision-making by arranging all healthcare data into patient pathways. Through patient pathways, SHAARPEC links care now to its impact on care consumption in the future. 

Patient pathway analysis enables modelling and calculating the effect of proposed improvement measures in healthcare.  

Preventive care aims at altering these patient pathways. Through screening, early detection, medication and similar measures we can alter the patient pathway. By keeping patients healthier, healthcare systems can achieve better health outcomes and save resources. 

Detailed: Five time-stamped patient encounters (blue) and encounter attributes (e.g. diagnose, lab value).

All encounters by a certain patient in health care

Data-driven management

SHAARPEC creates a foundation for  

  • System level analysis of resource utilization, outcomes and cost, helping to optimize patient health outcomes and keeping track of costs and resource utilization. 
  • Identifying opportunities for improvement at the system level through variability analysis of utilization, outcomes, cost for populations of interest as well as by quantifying the variability 
  • Evaluating and prioritizing proposed new initiatives for process change based on the likely impact on outcomes, cost, social equity etc. 

The analysis is built from the lowest possible level – patient encounters – which enables aggregation in many dimensions: by organizational unit, diagnose code, encounter type, patient cohort, etc.

The SHAARPEC cost allocation model considers all costs, including the cost of unused capacity. The platform is based on the principles of Time Driven Activity Based Costing (TDABC), adjusted to the existing data available at healthcare systems. 

Personalized precision medicine

Evidence-based guidelines  

  • Analysis at the patient and population level to determine which patients meet inclusion criteria for which evidence-based guidelines 
  • Tracking compliance with the diagnostic and treatment recommendations for each guideline to determine if patients’ care is compliant.
  • Providing feedback to physicians about which patient’s care could be improved

AI-based machine learning predictive modeling  

  • Determine which patients are at risk for specific health outcomes of interest so that they can be started on proven treatments or preventative strategies earlier, thereby improving outcomes and reducing downstream healthcare system expenditures 

Within SHAARPEC, all of the data is organized from a patient perspective enabling

In short, SHAARPEC is an effective solution to persistent challenges within healthcare systems

Technical design

SHAARPEC is designed to enable sustained value creation for healthcare systems regardless of underlying systems and financial models.


Privacy within SHAARPEC is protected by hosting the platform and data within the healthcare system and pseudo-anonymizing the clinical data. 

Shaarpec is uniquely designed to keep the data with the users within the healthcare system, compliant with national and international guidelines and regulations. Data access is restricted to the authorised users. 

The platform is designed to operate within Swedish and EU laws and regulations respecting confidentiality and privacy pertaining to patient and corporate data.  

Research and benchmarking are made feasible without breaching GDPR and HIPAA by applying several techniques, such as 

  • Secure multiparty computation 
  • Homomorphic encryption 
  • Federated learning 

Grouped together, we call these techniques Base Jumping. Base Jumping makes it possible to run queries across all systems and return the results to one centralized node, while the data itself stays within the local healthcare system.

This allows not only for benchmarking between the systems, but also faster and more accurate training of AM/machine learning algorithms. 

Get in contact with us