With Sureeta Mathur and Rachana Khatkhate, Directors at NNIT Migration Powerhouse
While data migration is rarely simple, it is further complicated by the highly regulated nature of the life sciences industry. In this article, experts from NNIT’s Migration Powerhouse provide valuable insights that may save you both time and money on your next data migration project.
With the increasing digitalization of the pharmaceutical industry, most life sciences organizations face the need to migrate large quantities of data sooner or later.
Regulatory demands are often the main driver for data migration, as older in-house systems become unable to keep up with increasing requirements without the support of a large internal team. When upgrading to a more modern system, many companies take the opportunity to switch from on-premise solutions to cloud-based services.
Data migration projects may also be triggered by mergers and acquisitions or a desire to consolidate multiple systems into a single platform. Regardless, any major data migration project is bound to have a significant impact on the organization, especially in the highly regulated life sciences industry. And without careful management and planning, troublesome surprises have a nasty tendency to increase both the timeline and budget.
Sureeta Mathur and Rachana Khatkhate are both directors in NNIT Migration Powerhouse (formerly Valiance). Between them, they have decades worth of experience assisting numerous life sciences clients with migration projects. Below, they provide insight into five factors that often cause surprises in life sciences organizations when migrating their data.
Common surprises during life sciences data migration
1. The amount of time and resources you need for data harmonization and cleanup
Before your data can be migrated, it needs to be evaluated and prepared. It is almost never possible or desirable to simply copy and paste everything from the source system(s) to the target system, as this would involve moving large quantities of corrupted, incomplete or irrelevant data.
Source systems typically have inconsistencies in naming conventions or missing information, which require significant effort to analyze, harmonize and enrich. This is especially true if your organization has used several systems without proper data governance and a single source of truth. There are gap analysis tools available to assist in identifying problematic data, but you still need specialist knowledge about life sciences business processes and regulatory requirements to do a comprehensive evaluation.
Don’t go overboard and turn your data migration project into a data quality project, as this will likely place too much strain on your organization and your subject matter experts (SMEs). You don't need to fix everything, so it is important to be able to prioritize which projects or products should be a focus area from day one. Some data is just moved for compliance purposes and have low impact on business.
2. The impact of changed regulatory requirements and guidelines
As mentioned, shifting regulatory requirements are one of the main drivers for data migration. As such, regulatory requirements typically have a significant impact on the migration project, both in terms of which data needs to be moved and how the target system is configured.
It is not uncommon for organizations to overlook their reporting requirements and audit reports during migrations, and in many cases, there is a lot of data missing from standard procedures and other critical documentation. On the other hand, much of the existing data may be historic and have been created due to obsolete requirements that are no longer relevant. So, it is important to be able to ensure compliance without cluttering your new system with irrelevant data.
Brexit is a good example of the impact of changed regulatory requirements. When the UK withdrew from the EU, regulatory responsibilities for clinical trials were transferred to the Medicines & Healthcare Products Regulatory Agency (MHRA). This increased the administrative burden of life sciences companies as clinical trial authorizations and market authorization applications need to be obtained under a new and different legal framework.
3. Data may be managed and reported differently in the target system
Each system has different requirements for license management and change management. Often, these differences in data granularity between the source and target systems means that the data needs to be managed and reported in different ways. For example, there can be product or region-specific challenges when moving data for flu vaccines.
Frequently, missing functionality in the source system has caused life sciences organizations to resort to creative workarounds to keep their system working for products that don’t fit standardized templates, such as many oncology products. These workarounds need to be accounted for when migrating to the target system.
Moving from a non-GXP system to a highly regulated system can also be a source of complication. In these cases, users often have trouble adjusting because the target system is more rigid and automated than what they are used to. This means that you may need to devote extra time for training and familiarization before users are fully proficient.
4. The importance of multiple dry runs before the actual migration
Before migration, NNIT’s migration methodology recommends performing at least one partial dry run with selected data and products and a full dry run. The partial dry run helps ensure that all rules are correctly configured, while the full dry run serves as an opportunity to test the target system, identify all data issues and acts as a hard deadline for data cleanup and harmonization. After the dry runs, users perform tasks such as looking up documents and generating reports to check if they have the right data, user access etc. This is a good way to involve relevant stakeholders, as it usually reveals multiple issues that need to be addressed.
While bringing the target system online, it is also recommended to analyze how the source system lines up with the default target system configuration. This approach, which is referred to as Phase Zero, can lead to a lot of “aha” moments, for example when migrating documents with hidden properties from a folder system into a target system that uses document classifications.
Once the dry run is complete, all of the requirements are locked and approved so the full migration can take place.
5. Validation of the migrated data is not viable without automation
Once a migration is complete, everything needs to be checked and verified before the target system can be released for production. This involves examining vast quantities of data and content, settings and integrations. In one example, the migration involved approximately 15 million data points and 1500 different rules. Without automated tools, using a sampling method, manually comparing thousands of records – data and documents – is a virtually impossible task.
To ensure that the migration and all data transformations are completed according to specifications, NNIT uses TRUcompare™. This unique automated data and content testing software is able to provide 100% pre- and post-migration testing. Equally important, the software can generate comprehensive reports that serve as documentation of a successful migration for compliance purposes.
However, even with a fully automated test of all the migrated data and content, it is still important to perform a user acceptance test before the migration is finalized. Because while the software can compare and document that all data was transferred to the correct fields, and content is correct, live users need to test and verify that everything is working.
Need migration? The NNIT Migration Powerhouse is here to help
NNIT’s Migration Powerhouse is the result of Valiance Partners’ merger with NNIT. We bring unmatched expertise, experience and knowledge of life sciences business processes to every client engagement.
Our unique migration methodology and proprietary software can ensure expedient and efficient migrations with 100% verification of migrated data. With the assistance of NNIT’s Migration Powerhouse, your organization can successfully complete migration projects in a timely, accurate, compliant, and cost-effective manner.