[MUSIC] In this video, we're going to talk about the Michigan Keystone Project as a case study of the difference between quality improvement and research. In a study that was approved as exempt by the Institutional Review Board at Johns Hopkins that ended up going through a review by the Office of Human Research Protection, which is an arm of Health and Human Services in the United States that oversees research. And how that affected the project, and ultimately what the outcome was. So to give you some background, the Michigan Keystone ICU study was a study that was funded by the US Agency for Healthcare Research and Quality. Dr. Peter Pronovost of Johns Hopkins University School of Medicine was the principal investigator. And it was funded under the name Statewide Efforts to Improve Patient Care in Intensive Care Units, which is kind of a mouthful. But we did it in partnership with the Michigan Health and Hospital Association, and we called it the Michigan Keystone ICU Project. The study hypothesized that we could improve patient safety, improve safety culture, and reduce ICU mortality, blood stream infections, aspiration pneumonia, and ICU length of stay. The interventions implemented were the Comprehensive Unit-based Safety Program, or CUSP, which is designed to improve patient safety culture. Interventions to improve teamwork, and communication, since we know both are associated with patient safety and the occurrence of medical errors. Interventions to reduce or eliminate catheter-line associated bloodstream infections. And interventions to improve care for patients on mechanical ventilation. So the study began in 2003, in April of 2003, ARC had a request for applications. In September, the project was funded. The IRB was submitted. And in October, the IRB approved the project as exempt research. And in October of 2003, the project was launched. Following implementation of the study and analysis of the results, the project was published in the New England Journal of Medicine in December of 2006, in an article entitled, An Intervention to Decrease Catheter-Related Bloodstream Infections in the ICU. Following publication of that article, the OHRP received an anonymous complaint. The complaint stated that the study constituted the research, that the rules of research were bypassed, that informed consent was not appropriately obtained from subjects, and that an applicable regulation under protection of human subjects regulations had not been followed. As a result of this, In February, the Office of Human Research Protections sent a letter to Johns Hopkins University indicating that a complaint had been received, outlining the issues of the complaint and opening up a process of review for the Michigan Keystone ICU study. In July of that year, Johns Hopkins University suspended the study. At that point, all of the interventions and all of the hospitals participating had been enrolled. And at that point, only follow-up data was being collected. So no new hospitals joined the project and no new interventions were put in place. And the follow-up data was not collected and sent to Johns Hopkins. Over the period of time, there was a correspondence, there was investigation, there were multiple publications in the medical literature around this issue and what constituted quality improvement. Was the Michigan Keystone ICU project, in fact, quality improvement? And discussion of uncertainty and ambiguity in the guidelines that, at that time, existed outlining what was quality improvement, what was research, how IRBs should review these types of project. And many of them supported the project as it had been reviewed, but they also called for clarification of the regulations. After this, and after the investigation, in November the Office of Human Research Protection made a determination and the study was reinstated. And at that time, given that, as I said, the study was collecting follow-up data, those data could continue to be collected because we could go back through the records and obtain data that we might not otherwise have been able to, given that the study was reinstated. This issue raises some questions about the difference between quality improvement, exempt research, and expedited research. So in the United States, the federal guidelines consider exempt research to be, research involving the collection or study of existing data, documents, records, pathological specimens, or diagnostic specimens, if these are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified. So exempt research is subject to a less intensive review than is expedited or other research. But it still is reviewed by a committee to ensure that it meets the guidelines for exempt research and there is follow-up of the project over time. In expedited research, Projects are considered to be research. They go through a bit of a more thorough investigation, and they usually include annual reports to the Institutional Review Board and review and approval on an annual basis. This is from an article that was published in New England Journal of Medicine during this time that discussed these issues. And just this quick quote, nevertheless, the research could have been reviewed in an expedited fashion by the IRB chair alone, since it posed no more than minimal risks, and fit within two categories for expedited review specified by the OHRP. Collection of data through noninvasive procedures, not including anesthesia or sedation, routinely employed in clinical practice. And research including materials, data, documents, records or specimen that have been collected or will be collected solely for nonresearch purposes, such as medication treatment or diagnosis. So this category seemed to fit the Michigan Keystone ICU study, in that it was a minimal risk study, the data were collected through noninvasive procedures, and in fact no changes in care were made. Some evidence-based interventions were implemented, but they were proven to improve care. And by that definition, at one level, qualify as quality improvement. So this is the issue that OHRP ultimately had to decide was, first of all, whether this project was quality improvement or research. And then second of all, had the IRB at Johns Hopkins acted correctly, and had the review and approval been within the guidelines, contrary to the anonymous complaint that they received. So on July 30th of 2008, Dr. Pronovost previously had sent them a letter, asking them to clarify the letter that they made the determination that the study could go on. And this is, in part, what he received back from OHRP. We believe that the actual implementation of the five part catheter-related bloodstream infection and reduction program in the participating hospitals is a quality improvement activity that does not meet the regulatory definition of research. Second, we believe that the activity involving the analysis of the aggregate data does not fall under the regulations, and therefore does not meet, or rather, does not need to meet regulatory requirements, including the requirements for IRB review and approval. The planned activity does meet the regulatory definition of research, and that refers to the Keystone ICU project in general, because it is a systematic investigation that is designed to improve the scientific understanding of how to implement this quality improvement initiative on a wide scale. So in other words, the OHRP is deciding that for the participating hospitals and for the subsequent data analysis, those do not meet the regulatory definition of research. But the ongoing project, in fact, did meet the regulatory definition of research. However, obtaining and analyzing the aggregated data about the rate of infections at the participating hospitals does not meet the regulatory definition of human subjects. The two surveys represent research involving human subjects under the regulatory definition. But since the survey information is being collected anonymously, the research activity, including the comparison with the aggregate data about the rates of infections, is exempt from the regulatory requirements under the federal guideline. So, in essence, what they're saying is the Michigan Keystone ICU project, in general, at the Johns Hopkins level was research. But it did not meet the regulatory definition of human research in the data analysis phase. It was human subjects research, we did a number of surveys in the hospitals, that that was research, but since those data were collected anonymously, they were exempt from the federal regulatory requirements. And in fact, the study had been approved as exempt. So what is research, and what is quality improvement? Under the federal regulations in the United States, research is a systematic investigation, including research development, testing, and evaluation, designed to develop or contribute to generalizable knowledge. Quality improvement is the implementation of projects that seek to improve the process or delivery of care with established/accepted methods. Research seeks to answer questions and create generalizable knowledge. Quality improvement seeks to implement established or accepted methods to improve the process or delivery of care. Different institutions, different countries look at these issues in different ways. So if in doubt, I suggest that you submit any project that you're considering doing to your IRB. And allow them to decide whether you're engaged in human subjects research or quality improvement. And if you are involved in human subjects research, what guideline fits that research and what review is required based on that determination. I guess I kind of take a better safe than sorry approach. Thinking that, if the IRB tells me that what I'm doing is not human subjects research, but is in fact quality improvement, as I think, then I've done nothing more than had to create forms and do a submission according to the guidelines. But if in fact, what I'm doing is not quality improvement, and qualifies for research, and I don't submit it, I am violating the rules, regulations, and laws surrounding human subjects review of research projects. So the IRB is a better arbiter of that than am I or probably you, for that matter.
