Quality science and quality assurance: observations of an environmental scientist

The purpose of this manuscript is to examine the relationship between quality science (QS) and quality assurance (QA). Many research scientists definitely want to do QS, but are afraid or do not want to do QA because they are intimidated by the QA process or they do not appreciate the benefits of QA. Therefore, the relationship between QS and QA is examined in this manuscript by an environmental scientist who has conducted 30 years of research in university, contract and government laboratories. To start, QS is defined in this paper as data that are published in the peer-reviewed literature. The quality of the research data is assumed by the general scientific population to be directly proportional to the status of the journal. For example, it is highly prestigious to have an article published in Science. At the U.S. EPA, the procedure for sending a manuscript to a journal for publication is the responsibility of the senior author. The senior author of an EPA-sponsored manuscript is expected to have the manuscript reviewed by the coauthors (they should also review the data), then the manuscript must be reviewed by at least two other scientists, one of whom must be from outside the authors' division. After this review and approval by management, the manuscript is sent to a peer-reviewed journal, where it is reviewed by several anonymous scientists as determined by the journal. After the comments of the reviewers are addressed, the manuscript can either be accepted or rejected for publication by the journal. For the purpose of this manuscript, the definition of QA is defined as the guarantee from a review team that the entire study was adequately and correctly conducted and recorded according to the study protocol. Many scientists view QS and QA as separate entities. From the scientist's perspective, QA procedures are not applicable to research studies, and should be used only for studies that will be submitted to either the EPA or the FDA for regulatory approval (i.e., Good Laboratory Practice [GLP] studies). However, QA can be applied to both types of studies. A QA review will examine all aspects of the study including data files (notebooks, protocols), as well as equipment, sample storage, actual experimental organisms (animals or cells) and the management of all study records. The data from a QA-reviewed study are therefore more defensible in a court of law, and more reproducible due to more through, chronological records. Generally speaking, few coauthors of a scientific manuscript analyze the raw data in the laboratory notebooks or inspect the laboratory equipment. Furthermore, coauthors generally have not been in the laboratory where the research was conducted in order to observe quality control measures. These are the areas where a QA review is extremely beneficial. In summary, data in the peer-reviewed literature do not undergo the same type of review as do data that have undergone a QA review. QA reviews assist EPA scientists in conducting and improving their research studies by identifying both excellent study practices and study deficiencies to be addressed, which thereby produces higher quality scientific data. In the opinion of this EPA Scientist and QA Manager, although QA reviews do require effort from the scientist, data from research studies are strengthened by QA review when compared to data from peer-reviewed studies that have not undergone a QA review. QA reviews should be viewed as part of the entire research process--a part that improves the overall quality of the data.