Human Subject Research or Quality Improvement Project?

Most physicians are learning to adapt to quality-based reimbursement. It might be helpful to remember that the initiative to define and improve quality is a decade and a half old and is ongoing. Seminal publications from the Institute of Medicine in 2000 described the problems and potential solutions, ^1-2 and more recent publications document continued efforts at quality improvement (QI). ³ Quality measures have been implemented not only by the American College of Surgeons through the American College of Surgeons National Quality Improvement Program (ACS NSQIP), ⁴ but also by the Centers for Medicare and Medicaid Services through its Surgical Care Improvement Project. ⁵ Otolaryngologists need to be involved in quality improvement projects (QIPs) in order to help determine appropriate quality definitions and measures for our specialty. However, the differences between QIPs and human subjects research (HSR) requiring institutional review board (IRB) approval are not always clear. This subject was reviewed by Raval et al in 2014, and areas of confusion and controversy have not been clarified much in the intervening years. ⁶

Research is “a systematic investigation, including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge,” as defined by the Federal Code of Regulations (32 CFR 219.102 [D]). ⁷ Research is designed to clarify and expand understanding of a specific disease, problem, or treatment, and to add to the profession’s fund of knowledge. Quality improvement projects are not designed to expand generalizable knowledge. Rather, they are used to determine whether an accepted behavior or practice is being utilized locally and if not to seek to change the behavior of health-care providers to bring their actions into compliance with accepted norms.

Raval et al summarized the effects of QIP, as including interventions (including physical procedures such as venipuncture or biopsies) through which data are gathered; manipulation of a subject, or environment for QI research purposes; communication or interpersonal contact between the researcher and subject; and acquisition of personal health information which is not made public, but which is available to the investigator. In general, QIPs do not include control groups. The QIP generally does not involve assembling a data set that has been built and then had the subjects deidentified. However, most QIPs are, or come close to being, HSR.

Human subjects research is well known to most of our readers. It is performed for the purpose of enhancing knowledge in the field generally and acquiring information that will expand our understanding of diseases and/or alter evaluation and treatment. It requires IRB approval, and subjects must be deidentified. An IRB may determine that formal IRB application and approval, and informed consent can be waived if the research does not adversely affect the rights and welfare of the subject, poses no more than minimal risk, could not be performed if consent were required, and meets other specific criteria. For research, the Health Insurance Portability and Accountability Act requires deidentification of personal health information for subjects to be used in research. There are 2 methods of deidentifying data. Specific information can be deleted from data sets (the safe harbor method). Alternatively, personal identifiers can be removed from each subject, to the point at which the data cannot be traced back to individual subjects. There are national data sets available that have been deidentified. For example, deidentified medicare billing data are available. So is information from the Agency for Health Care Research and Quality’s Healthcare Cost and Utilization Project, the ACS NSQIP participant use file. The SEER database also is deidentified. So, a study using these databases typically does not require formal IRB review and approval. However, in most cases, it is appropriate to submit the proposed re-cert to an IRB in order to have a formal determination and letter stating that the project has been evaluated and found to be exempt.

Although it is important for us to perform QIPs as well as human research, it is essential that we take all necessary precautions to avoid violating the rights of patients, the laws governing human subject research, and the ethics of such investigations (summarized briefly by Raval et al ⁶ ). Nevertheless, otolaryngologists must be involved. If we do not perform QIPs and the appropriate research within our specialty and define what should be measured and what constitutes “quality,” someone else will do it for us, and that is not likely to result in valid, reliable, or practical measures. Interested otolaryngologists should proceed enthusiastically, but knowledgably. Consultation with a senior representative of the IRB, and with a statistician, prior to beginning research is invaluable.