Keywords

IRB, quality improvement

 

Authors

  1. Nerenz, David R.
  2. Stoltz, Patricia K.
  3. Jordan, Jack

Abstract

Improving health care quality is increasingly recognized as a national priority in the United States. As a result, more and more health care organizations can be expected to undertake quality improvement (QI) initiatives. A question being raised with increasing frequency is: "Which QI activities need review by an institutional review board (IRB)?" Structured data collection and analysis is a common characteristic of most QI activities. For some QI projects, the fundamental goal is improved understanding of phenomena presumed to be generalizable to settings other than those directly studied. These activities are research. For other projects, the fundamental goal is improvement in specific processes and systems within specific organizations. These activities are not research. This article proposes that this difference in intent and the fundamental nature of the activity is crucial in deciding which QI initiatives need IRB review. The article presents test questions and markers to distinguish research from other types of QI activities. Those that are not research do not require IRB review. However, because such activities may still put patients at risk, some other review may be necessary. The article proposes five levels of risk and makes recommendations for review by an entity other than the IRB.

 

Quality improvement (QI) is a generic term for almost any sort of activity that has the desired effect of improving some aspect of the health care process, either in a single setting or in many settings. A question that arises with increasing frequency is: Which QI projects require review and approval by an institutional review board (IRB)?1-4

 

The issue is particularly salient today. In two landmark reports on US health care, one on patient safety and medical errors (1999) and one on the "quality chasm" between the care possible today and the care actually delivered (2001), the Institute of Medicine (IOM) has called for a major commitment to improvement in quality in the next decade.5,6 Expanding QI activities will inevitably lead to questions about the need for IRB review for those activities and possible written informed consent by patients, clinicians, or support staff who are involved.

 

Quality improvement research is another, newer, generic term with an even less precise meaning. One meaning is research in which the phenomena under study are quality improvement activities per se. This research would be exemplified by studies of the effects of QI programs or continuous quality improvement (CQI) initiatives as independent variables on some dimension of quality as the dependent variable.7 Another possible meaning is the study of changes in health care structures or processes (perhaps organized outside the context of CQI) and consequent changes in measures of quality. This type of QI research would be exemplified by studies of the effects on quality of implementation of clinical guidelines or physician reminder systems.8-10 Our task in this article is to address the issue of various kinds of QI activities and requirements for IRB review. We will argue that any activity whose fundamental objective is to gain knowledge about generalizable phenomena (processes and relationships) is research and should be subject to federal regulations about IRB review, regardless of whether its topic is quality improvement, organization of health services, or basic biological phenomena. We also argue that any activity whose fundamental objective is improvement in local processes is not research and is not subject to federal regulations about IRB review, regardless of whether the activity affects human participants or whether a description of the activity may eventually be presented at meetings or published in professional journals. We recognize that this distinction is different from those proposed by others who have taken up the same question1-4 about IRB review for QI activities; we will attempt to show how our distinction is more consistent with applicable federal regulations and less likely to result in undue and unnecessary burden on IRBs and delay in bringing the benefits of QI to health care organizations and those who depend on them.