Medical Research Data Management - Architectural Viewpoints - Part 2

My previous blog  Research Data Management - Architectural Viewpoints - Part 1 followed the Togaf process and developed the strategic capabilities for

1. Research data management strategy and 
2. A single repository data management strategy. 

In this blog, I will focus on each of the strategic capability and how it aligns with the strategic course of action, taken to meet clinical motivational goals and requirements.
The two strategic capabilities complement each other to provide rapid data exploration, rapid testing and continuous integration for clinical research.

Research Data Management Capability

The research data management capability realises the following primary goals for the research entity

1. Effective planning and operations to share research data
2. Enable targeted advance research and creating better opportunities for research data re-use
3. Seamless partnership and engagement with other research partners

Each of the goals is aligned with different stages of the research data lifecycle. The below set of viewpoints shows the relationship between each of the internal capabilities along the research lifecycle, functioning to deliver collaborative research with increased access to knowledge. 

Pre-research Data Management 

Peri-research Data Management

Post-research Data Management

Singular Repository Capability

The single repository capability strategy has two dimensions

1. Repository data lifecycle management 
2. Repository data management. 

Singular Repository Data Lifecycle Management

There are two different drivers that influence how the data is managed in the singular repository. 

1. Evolving the organisation into a data-driven healthcare organisation.
2. Modernizing and integrating new platforms rather than replacing old ones.

The approach to designing the singular repository is guided by the following principles

1. Augment (but don’t replace) existing data warehouse’s primary platform by adding additional data platforms and tools
2. Employing evolving data warehouse platform architectures.

As a strategy, we took two different approaches to design the singular repository based on the data lifecycle.

1. Design of a data & analytics platform bolted on a singular repository,  that would enable self-service data analytics.
2. Design of a controlled data and analytics platform bolted on a singular repository, that would enhance the current data analytics capability but also operate on a current matured data management processes.

Controlled Data Lifecycle Management Strategy

The diagram below shows the relationship between data management goal and the course of actions taken to develop the repository data lifecycle.

1. Data Acquisition

2. Data Consolidation & Standardisation

3. Data Analysis/Presentation

4. Research Data Provisioning

Repository Data Management

The single repository for clinical data management requires to meet the following prioritised primary concerns of data management.

1. Development of a platform, that handles seamlessly integrates varied and high volume data in batches and in real time.
2. To design and build a data processing platform that facilitates rapid data exploration, rapid testing and continuous integration.
3. Implement extended data management functions, leveraging metadata along the data lifecycle to enable self-service analytics and improve medical research collaboration.
4. Designing effective solutions to address the scalability problem of healthcare data privacy and security concerns.

The diagram below shows the relationship between each of the collective course of actions taken within each stage of the repository data management. 

In my next part of the research data management blog, I will focus on the research cycles, stakeholders and application architecture. 

References

https://ico.org.uk/media/for-organisations/documents/2013559/big-data-ai-ml-and-data-protection.pdf