Quality Improvement/Quality Assurance or Research – or Both. A decision logic.

Determining if an activity is Research or Quality Improvement/Quality Assurance (QI/QA) can be challenging. Federal regulations require human subject research to be reviewed and approved by the IRB, while strictly QI activities do not require IRB oversight. However, some QI activities may also be research and therefore need IRB approval.  Traditionally QI/QA is designed to improve patient care, compare a program/process/system to an established set of standards such as standard of care, or improve the facility practice or systems to bring about improvements in health care delivery.  Research involves a “systematic investigation including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge.”(dhhs definition) thus potentially applying beyond the individual facility and allowing for projects not expected to directly benefit all of the current patients. 

Regardless of whether IRB is needed or not, all Protected Health Information (PHI) must be transmitted, stored, analyzed and otherwise exist only on HIPAA-compliant systems within the facility electronic or paper medical records system. Where limited data sets are accessed for research purposes, PHI must be managed on GS controlled electronic systems under an approved data security plan.

Research vs. QI/QA Comparison

Please review the following guidance and determine if your activity likely needs IRB approval PRIOR to beginning the activity. Source

	RESEARCH (R)	QUALITY (Q) IMPROVEMENT	R or Q*
INTENT	Intent of project is to develop or contribute to generalizable knowledge (e.g., testing hypotheses)	Intent of project is to improve a practice or process within a particular institution or ensure it confirms with expected norms
MOTIVATION FOR PROJECT	To answer a new question or test a hypothesis	To improve patient care or service performance.
DESIGN	Designed to develop or contribute to generalizable knowledge; may involve randomization of individuals to different treatments, regimens, or processes.  May or may not benefit current subjects.	Designed to promptly benefit a current process, program, delivery or system.
MANDATE	Activities not mandated by institution or program	Activities not mandated by institution or program
EFFECT ON PROGRAM/PRACTICE EVALUATED	Findings of the study are not expected to directly affect institutional or programmatic practice – or may have additional broader application	Findings of the study are expected to directly affect institutional practice and identify corrective action(s) needed
POPULATION	Usually involves a subset of individuals – universal participation of an entire clinic, program, or department is not expected; generally, statistical justification for sample size used to ensure endpoints can be met	Information on all or most receiving a particular treatment or undergoing a particular practice or process expected to be included; exclusion of information from some individuals significantly affects conclusions
BENEFITS	Participants may or may not benefit directly – benefit, if any, to individuals incidental or delayed	Participants expected to benefit directly from the activities
DISSEMINATION OF RESULTS	Intent to publish or present generally presumed at the outset of project as part of professional expectations, obligations; dissemination of information usually occurs in research/scientific publications, presentations or forums; results expected to develop or contribute to generalizable knowledge by filling a gap in scientific knowledge or supporting, refining, or refuting results from other research studies or being generalizable to broader populations	Intent to publish or present generally not be presumed at the outset of the project; dissemination of information often does not occur beyond the institution evaluated**
COMPARISON	Study, development or implementation treatments, practices, techniques or processes that are not already well defined.  (See examples below)	Comparison of standard treatments, practices, techniques, processes – Applying what is already known into the local practice to quickly improve patient care/system within a specific practice/setting
DEVIATION FROM STANDARD PRACTICE	May involve significant deviation from standard practice	Will not be a significant deviation from standard practice
RECORDS REVIEW ONLY	Records will be reviewed outside the home organization.	Records will be reviewed solely in the home organization as an internal audit.
EXTERNAL FUNDING	The project is funded by an external research grant and requires IRB	The project is not externally funded or may be funded by a non-federal source (e.g., foundation) as assessment of service without a generalizable component

 

**Dissemination of information may occur in quality improvement publications or presentations with a statement similar to “This project was undertaken as a Quality Improvement Initiative at X hospital or clinic and as such was not formally supervised by an Institutional Review Board per their policies and is not generalizable.” 

Adapted – Working Document: University of Wisconsin-Madison Health Sciences IRBs, version date 11/2/08 Authors: Nichelle Cobb & D. Paul Moberg

QI involves implementing previously proven/tested, planned and systematic activities done to improve or satisfy quality requirements.

Examples of QI activities that are likely NOT research include:

• • Implementing a practice to improve the quality of patient care

• • Collecting patient or provider data regarding the implementation of the practice for clinical, practical, or administrative purposes

• • Measuring and reporting provider performance data for clinical, practical, or administrative uses

• • A group of affiliated hospitals implements an application to reduce prescription amount errors and collects patient prescription information from medical charts to assess whether the application helped reduce error rates as expected.

Please see HHS guidelines and FAQs for more information.

Note: A quality improvement activity may also constitute non-exempt human subject research if it meets the definition of research.

Examples of Activities that are likely QI and Research

• • A project involves introducing an untested clinical intervention for purposes which include not only improving the quality of care but also collecting information about patient outcomes for the purpose of establishing scientific evidence to determine how well the intervention achieves its intended results.

• • Collaborative (multi-site) – All the sites are trying to improve some aspects of clinical care (ex. implementing an application to help improve making clinical decisions). The whole department decides that this app will improve care, and implement the app. They collect data as the app is implemented, and in addition, analyze this data for generalizable knowledge.
  • An implementation of a new intervention to improve patient care and analysis of the data to establish scientific evidence of the interventions effectiveness.
  • A project to develop and test a new quality assessment or measurement tool.
  • Development of or implementation of a new treatment or a novel approach that has not been extensively studied.

• • A teacher implements a practice to have all students reflect on their learning by keeping a journal, with the intention of improving teaching practice.  However, the teacher also wants to prove that this method works, so they analyze student journals with grades to generalize the success of this method.

Examples of Activities that Begin as QI and Become Research

Please note that if you begin QI activities with the intent to eventually use the activity or data for research, it is best to submit to the IRB prior to beginning the activity.  However, if after a QI project is completed, and you want to study it further and make it generalizable (research), then IRB submission is required (typically using secondary data).

For example:

• • A QI project is implemented, and upon completion, the investigator realizes they want to do research about the project, and interview clinicians. The data they will collect from the interviews will be used for research, therefore, they would need to submit it to the IRB before beginning interviews.

• • A team uses biologic samples to compare two different types of tests to determine which one is better and therefore which one should be used at VCU [intent to improve care at VCU]. After they complete the comparison, they realize they want to share the success of these tests because they believe it will help other institutions [intent to contribute to generalizable knowledge]. They then submit to IRB and request to use the data collected for the QI project as secondary data for research.

• • A surgeon believes that a certain technique will improve their own practice, so they implement it and record results as part of clinical practice. They then decide that this practice would help others, so they go back to their data to systematically analyze and generalize outcomes and results. They would need to submit to the IRB prior to the review of gathered data.

• • A school decides to begin an after-school program to help with academic success. The school gathered academic data which proved that the program was successful.  After a few years of the program being a success, someone decides that they want to share that program with others.  They can submit to the IRB to be able to analyze the previously collected data.

 

Last updated: 8/1/2024