Quality Assurance Toolkit
The Michelson Prize and Grants Research Quality Assurance Toolkit (QA Toolkit) is provided here to assist applicants in generating research data that are accurate, reproducible, and auditable. It is designed to facilitate best practices in the management of research personnel and equipment, validation of methods, standard operating procedures (SOPs), and maintenance of data and documents.
All documents in the toolkit may be downloaded, altered, and used free of charge by visitors to this website. The Quality research practices self assessment checklist serves as a table of contents of documents available for download. The toolkit includes sample forms and SOPs, as well as tips on creating forms and SOPs.
We recommend, but do not require, new Michelson Grant applicants to utilize these tools. Toolkit resources provide investigators with mechanisms for demonstrating process transparency, accountability, and a straightforward method for reconstructing data if required. Investigators submitting for second and renewal grants, or who are completing Data Package-1 for a Michelson Prize application, must—at a minimum—meet the standards marked as required on the Self Assessment Checklist. Found Animals will fund clinical trials to test products from one or more Prize applications, and therefore seeks applications that contain accurate, reproducible, and auditable research findings.
The QA Toolkit includes the following five sections of usable forms. Click the links provided below to download each form individually, or click the link at the bottom of the page to download the entire toolkit.
Quality practices for personnel management include maintenance of employee credentials and employee training records. The role of the employee in the research project should be easily ascertainable both for auditors and when troubleshooting errors or non-conforming results.
Quality practices for equipment management include maintenance of inventory logs of critical equipment, equipment calibration information, and the tracking of preventive and non-routine equipment maintenance. All users of a critical, designated piece of equipment should operate that equipment using a Standard Operating Procedure so as to ensure uniform results. Personnel who are responsible for equipment maintenance should be designated and provided with logs for recording their work.
Research investigators should be able to answer the following questions about the methods used to generate their data:
- How do you know that this method works?
- How do you know when this method is not working?
- Is there an effect of sample handling or environmental conditions (like laboratory temperature) on measured outcome?
- Does anything interfere with the accurate measurement or production of the result? How do you control for that interference?
- Are your results repeatable (multiple measures in the same lab) and reproducible in a different lab?
- If you make minor changes in the procedure, or if different people perform the procedure, are the results affected?
Typical validation characteristics for analytical assays include sensitivity (what is the smallest amount you can detect?), specificity (is it possible that you are measuring related compounds?), repeatability (do you get the same result in the same or subsequent runs of the assay?), accuracy (are you measuring the exact amount that is really there), precision (coefficients of variation for samples measured multiple times within and between assays), and reproducibility (do you get the same result in a different lab?).
Steps for performing a method validation are listed in the methods validations records download. The steps can be adjusted on the basis of what is already known about the method, what is required by regulations, or what is possible in light of research limitations.
Accuracy and repeatability of research findings rely on analytical methods and operation of critical equipment that are performed the same way every time. Reliability is enhanced if standard operating procedures are developed, and if users are trained in the procedures documented and follow the procedures uniformly every time an analytical method is performed or critical piece of equipment is used.
Good records accurately document what you did, when you did it, how you did it, and what materials and instruments you used. They document the results you obtained, how you analyzed the results, how you discovered errors or nonconforming results, and what you did about them. They include entries from both successful and unsuccessful activities.
Good records are tools that support accurate repetition of the project by yourself and others.
They are legible, have sequentially numbered pages, are dated, are signed or initialed, are recorded in ink or “non-erasable” form, and are auditable. Errors are not erased from good records, but rather lined through, corrected, and the correction is initialed and dated. Good records are well organized, accessible, backed up, and appropriately archived.
Complete records provide details related to identifying and handling outliers and rejected data within data sets. When corrections to research data result in significant changes in research plans or directions, details are provided that clarify the impact of the changes on the project. Documents managed in good quality research practices include the forms we have provided in the 5 sections.
This section includes templates for:
- Chemical/reagent acquisition, storage, expiration, and disposition records
- Critical facility records (i.e. temperature monitoring, water monitoring, pest control)
- Animal acquisition, care, monitoring, sampling, and disposition records
- Sample acquisition, tracking, archiving, and disposition records
- Manually and computer-generated data recording records
- Laboratory notebook [bound or electronic notebook with numbered pages, to keep all notes (primary, random, scrap notes)]
- Summary spreadsheet data records
- Nonconforming work procedures and records
The Michelson Prize and Grants Quality Assurance Toolkit was commissioned by Found Animals from the Quality Central Program directed by Dr. Rebecca Davies (firstname.lastname@example.org; 612-626-2118) at the University of Minnesota College of Veterinary Medicine. The toolkit does not provide a complete quality management system, and its use will not guarantee that quality requirements associated with specific scientific standards or the Michelson Prize and Grants program are met. Its tools are examples and should not be considered preferred to other institutional forms or to procedures in use by investigators that manage QA data.