Quality Standards for regulated and

non-regulated biomedical research

laboratories.

Rebecca L. Davies, PhD

Associate Professor, Department of Veterinary Population Medicine

Director, Quality Central Program
College of Veterinary Medicine
University of Minnesota

National Academies of Sciences-Engineering-Medicine

Committee on Return of Individual-Specific Results
Generated in Research Laboratories
December 7, 2017
Personal Perspective: Scientist Driven Research Accountability
Comparative
Animal Voluntary Adoption
Endocrinology of Quality Assurance
and Diagnostic [QA] into basic
Medicine research and training
[QC] programs

Veterinary
Diagnostic Lab
AAVLD
Accreditation
[ISO 9001/17025]

My interests:

• voluntary adoption of research QA (RQA) best practices/standards into non-

regulated research environments
• Support infrastructure for RQA in non-regulated research environments
• integration of RQA into scientist training programs

Safeguarding sound science with good

research practices
 Quality Assurance
Management Systems are
designed to:

Improve and maintain the precision and accuracy

of a product and establish routine performance

22
The products we
produce are
research
data

4
 What Can Universities Do?

Quality Assurance support is rarely found in

academic basic research settings 5
 Research Quality Systems

Quality
Systems DEVELOPMENTAL PIPELINE
Basic Disease Drug Preclinical Clinical
Trials I, Manufacturing
Research Discovery Discovery Development II, III

Not Regulated GLP GCP GMP

21 CFR Part 11
Voluntary ‘Good Research Practices’?

Study Based Process Based- Control consistency required

Mentor Based
Flexibility needed for innovation and creative exploration

Slide adapted from one created by Melissa Eitzen, UTMB

Reproducibility 2020: Progress and priorities: http://www.gbsi.org/wp-
content/uploads/2017/11/F1000-Case-for-Standards.pdf

Data and Metadata

 How sound scientific principles and good quality
practices contribute to the credibility of results

(World Health Organization: Quality Practices in Biomedical Research

Handbook, 2006)

Sound Scientific Good Quality Credibility of

Principles Practices Results

Study 1 No No No
Study 2 No Yes No
Study 3 Yes No No
Study 4 Yes Yes Yes
8
 Strategies are need to fill the gaps

Research Quality Assurance is a reasonable strategy for

safeguarding research data quality and integrity throughout
the research life cycle.

20
Voluntary RQA Resources Exist

10
Questions from the Committee

Quality Standards in Research

Laboratories Using Human Biospecimens
Cost, infrastructure, and resource needs of implementing
a RQMS?
Cost Infrastructure Resources
Quality Policy
Gap
RQMS [Address the Management
Assessment
Total Testing Process]
Quality Manual
Equipment + QMS Software
Quality
Monitoring, Service Monitoring Program Documents
and Calibration
Management Review

Salaries for QA roles Time

and responsibilities Facilities and environment
that meet the RQMS and
Safety requirements Trained and
Quality Control Competent
and Proficiency Personnel
Testing
Training Program and Recurrent and
Continuing Education secure QMS
Reagent Funding
traceability
Secure Information, data
management and archive Equipment
Method and
Process Validation
and Verification
Biorepository
Accreditation or
Certification Costs
Specific minimum performance characteristics to be met
by a research laboratory?

WHO/CLSI/CDC
12 ‘Quality System Essentials’

Organization, Personnel, Equipment, Purchasing and Inventory,

Process Control, Information Management, Documents and
Records, Occurrence Management, Assessment, Process
Improvement, Customer Service, Facilities and Safety

http://www.who.int/ihr/training/laboratory_quality/introducti
on/en/

Quality Management
System Approach
What are the control processes that ensure research
results are accurate?
Demonstrate that data are accurate, reliable and fit for their
intended use. Quality Assurance

Biospecimens: Total testing process control (pre-analytical, analytical and post-

analytical processes).

Understand measurement uncertainty: analytical and biological variability

QA and QC: Assure/Monitor the quality of the testing and reporting process:

1. Define intended use

2. Validate that methods/procedures fulfill requirements for intended use
3. Design QC to confirm that the requirements have been met

Quality Assurance and

Quality Control
Can voluntary quality standard adequately protect
research participants and would external verification
process be required?
Must define or assess:

What adequate means

The risks
The value of disclosure/disclaimer/forewarning of quality limits that may exist
in a research environment.

A Voluntary Quality Standard has the advantage of establishing what quality

means through its requirements. The important deliverable is to ‘say what you
do, and do what you say’. This may provide flexibility.

External verification reduces risk.

Science-centered and risk-

based approaches?
How typical is it for an institution to have detailed
knowledge or infrastructure appropriate for the
oversight of routine and QA practices in the
research laboratory?
Typical: QA units within regulated research [GLP, GCP,
GMP]

RARE: Non-regulated research

Notes: Overlap should not be assumed

Regulated research does not ensure the presence of a full
QMS
Benefits and detriments of requiring
CLIA certification for research
laboratories?

Can Research Labs Implement RQMS?

Opposing Forces
Credible evidence of research
rigor, reliability, reconstruction
Continuous improvement
Competitive edge (investment
in research accountability)
Changes in research culture and
LACK OF mentoring

Resources/Support
Infrastructure
Need For
Buy-In
Knowledge
Training
Uniform Guidelines
Monitoring
Would clinical labs using a research
designation circumvent the
requirements (and associated costs) of
CLIA when performing tests?

Yes I think so, unless the infrastructure, funds and other critical resources

are available to meet the requirements.

Labs must have what they need to succeed. An unfunded mandate will not
meet that requirement.

My Opinion
 Data Data
Soundness Management
Big Data
Study Design
Shared Data
Bias, Report
Curation

Data
Scientists need a Quality,
research infrastructure Integrity
that provides science- Reconstruction
centered and risk-based
ways to maintain and
demonstrate the quality
of their work.

Reproducible and accurate research data

Helpful resources
1. WHO: Laboratory Quality Management System (LQMS)
Annex 1: Comparison of CLSI quality management system model to ISO 9001 and
ISO 15189.
http://www.who.int/ihr/training/laboratory_quality/introduction/en/
2. Westgard, JO and Westgard, SA. Basic Quality Management Systems. 2014.

Discussion of all ISO 15189, CLIA General Laboratory System Regulations, WHO/CDC/CLSI guidelines
Practical guide to the CLSI guidelines, ISO standards, CLIA Final Rules as well as Accreditation
Requirement. ISBN13: 978-1-886958-28-9, ISBN 1-886958-28-9

3. TDR Handbook: Quality Practices in Basic Biomedical Research, QPBR. WHO, 2006.
http://www.who.int/tdr/publications/quality_practice/en/