March 27, 2012

Know Your IT Systems Vendors

Posted in Cloud, Computer Validation, IT, Vendor Audits tagged , , , , , , , , , , , at 9:38 AM by Solutions2Projects, LLC

IT systems and infrastructure are critical to any organization.  This is especially true for life sciences companies selecting and implementing IT systems critical to the business functions supporting compliance functions. Regulatory bodies expect life sciences companies to demonstrate control over these elements regardless of whether they are the ones developing or maintaining the IT systems (infrastructure, software, etc.).

Companies cannot simply toss the responsibility over the fence to the vendors. Life sciences companies are still responsible for the integrity of the data and control over the systems.  They may delegate but only after verifying the vendor can meet the compliance and control requirements. 

This is where vendor audits come in to play. 

Vendor audits for software are not new.  Over the past decade I’ve seen the importance of vendor audits for software wax and wane and wax again.  In light of the increase in cloud and hosted solutions chosen by companies to decrease overall spend, the need for vendor audits is critical.

And, as biotechs become more virtual and more services are outsourced (CRO, CMO, data management, complaint handling, etc.), it is imperative that companies verify their vendors meet compliance requirements as well as their own procedural and process requirements.   The vendor’s IT systems and controls must meet the requirements as if they were hosted by your own company.  Not all vendors perceive the need to meet compliance requirements at the same level and you need to know before you enter any agreements.  Once you’ve signed the contracts, you’ve lost your leverage for process improvements and controls. 

Why conduct the audits? 

  • Gain high level of confidence that the computerized system will meet technical, commercial and regulatory requirements (GAMP 5)
  • Confirm the supplier builds quality and integrity into the software product during development
  • Leverage knowledge, experience and documentation of supplier (GAMP 5) to potentially reduce validation effort
  • Confirm processes and controls when  outsourcing IT / software functions (SaaS, PaaS, IaaS, hosted solutions, co-locations)

When should audits be performed?

  • For high risk systems / outsourced services
  • Before any contracts are signed!
  • Scheduled follow up audits based on
    • Audit results
    • External audit program
    • Risk assessment
    • Significant vendor business changes
    • When there are issues with the vendor

How are audits performed?

  • Similar to other vendor audits for CMOs or other critical suppliers
  • Plan for the audit and communicate expectations to the vendor
  • Conduct the on-site audit (for IT systems, Quality and IT representatives should participate)
  • Summarize findings with the vendor at the end of the audit
  • Document findings in an audit report and provide to the vendor for a response
  • Follow up on observations and document

The financial cost of, and risk associated with, software solutions has increased exponentially which means that it is imperative for organizations to understand what they are getting into before they sign on the dotted line.  The cost of a software or IT system blunder can be expensive in terms of resources, time and can make or break a life sciences company.  If you cannot demonstrate control, and therefore the integrity of your data, for systems supporting drug product administered to patients, a regulatory body may not grant approval for your product or could shut down manufacturing operations.  Your company owns the data and the responsibility even if it service is outsourced. 

Knowing your IT vendors gives you the knowledge to reduce the risks associated with the IT solutions in your life sciences company.  Without this knowledge, you are powerless to defend your risk assessment and risk mitigation strategy to regulatory agencies.

January 31, 2012

IT Deviations: Tool for Improving Overall Performance

Posted in IT tagged , , , , , at 7:36 AM by Solutions2Projects, LLC

I am a firm believer in documenting deviations from performance expectations for IT systems.  This includes functional and procedural errors, both system and human.  Documenting deviations provides data for better management and control over IT systems.   The process for capturing and managing IT deviations should be simple to be most effective as we know most IT folks do not like cumbersome documentation tasks especially if they think they will be penalized as part of the process (which should not be the case, ever!). 

 A basic deviation has the following elements:

  • Deviation description
  • Date(s) of occurrence
  • Categorization for tracking purposes (planned, unplanned)
  • How it was discovered and by whom
  • Investigation into source of deviation
  • Recommended resolution or actions to be taken
  • Approvals of investigation and recommended resolution
  • Documented action taken to resolve issue
  • Approval of action taken

 Deviation examples include:

  • Backup failures
  • Failure to follow documented procedures
  • Unexpected system behaviors
  • Unplanned outages

By tracking and trending the deviations, you can see if there are recurring issues and answer the following questions to improve data integrity and reliability. 

  • Is additional training required?
  • Should work instructions be developed or be more detailed? 
  • Is there a technical issue that needs to be addressed due to a repeated failure? 
  • Is there a configuration issue that needs to be addressed?
  • Should the issues be addressed with a vendor? 
  • Does the system need to be replaced?

 IT personnel and business system owners gain visibility into issues and can track patterns if deviations are documented.  It is too easy to forget when something happened (let alone if it happened at all) and how it was resolved.  We have too much to keep track of these days.  Deviations are often seen as something that is bad rather than as a useful tool to gather information to improve overall performance.  For those of us in IT that like data, deviations should be seen as an effective tool to gather data to improve overall system performance.