Care Redesign
Relentless Reinvention

Solving Puzzles and Problems: Linking Biomedical Research and Improvement Science

Article · February 5, 2017

Patient harm resulting from health care has been a prominent focus for over 15 years, but reducing such harm has been slow and difficult. One success story during this time period has been the 80% reduction in central line-associated bloodstream infections (CLABSIs) across the United States. Although this improvement is often attributed to the use of a central line insertion checklist, the reasons for the plummeting CLABSI rates are complex and are related to the link between traditional biomedical research and improvement science.

Let’s start by differentiating between the two concepts. Biomedical research generally “feeds forward,” starting with a hypothesis of whether one therapy is better than another and then testing the hypothesis by building a corpus of knowledge about the mechanism of disease and the effectiveness of the therapies. In contrast, improvement science “feeds backward,” starting with the goal, drawing upon theory, and then designing a multifaceted intervention to achieve the goal. In other words, biomedical research focuses on solving puzzles whereas improvement science focuses on solving problems. When biomedical research has sufficiently matured the science and evidence pertaining to a particular problem (e.g., CLABSIs), improvement science can design, apply, and evaluate a package of interventions to achieve the desired goal (e.g., reducing or eliminating these infections).

Investing in Biomedical Research

Investment in biomedical research made the improvement in the rate of CLABSIs possible. Starting in the 1970s, the Centers for Disease Control and Prevention (CDC) established evidence-based definitions, developed a valid and reliable measure of bloodstream infection, standardized the data collection process, and created a common measurement system (the National Healthcare Safety Network).

A common measurement and surveillance system made comparability across hospitals feasible and national benchmarking possible by organizing all stakeholders (e.g., researchers, policymakers, consumers, providers, and quality measurement groups) around a common goal and purpose.

The investment of federal and nonfederal funds in biomedical research supported rigorous study of the biology, causes, prevention, and treatment of these infections. Specifically, the National Institutes of Health and the CDC examined the pathogenesis of CLABSIs, investigated the causes of infection outbreaks, revised the clinical practice guidelines, and outlined strategies to reduce these infections.

The Role of Improvement Science

While the role of biomedical research is to solve puzzles by increasing the understanding of disease biology, the role of improvement science (also called translational research or implementation science) is to solve problems by drawing on social science theory in order to help translate evidence into practice. In fact, some of the greatest breakthroughs throughout history, from the 19th-century discoveries by Louis Pasteur to more recent developments at Bell Labs, have been accomplished by linking basic and applied research. Without this link, and a concerted effort to integrate evidence into practice, groundbreaking biomedical discoveries can take years to reach clinicians and patients.

Translating evidence-based strategies for the prevention of CLABSIs to the frontlines of care required an evaluation period. Two large-scale collaboratives in Pittsburgh and throughout Michigan were funded to test the feasibility of implementing the practice recommendations (in checklist format) within diverse intensive care units (ICUs) and to evaluate any associations with reductions in the rate of CLABSIs. Then, a cluster-randomized trial comparing an intervention group with a control group was performed to more strongly correlate the effectiveness of the evidence-based practice interventions in decreasing the rate of infections. The collaboratives demonstrated a 66% to 68% relative reduction in infection rates, and the trial demonstrated a 70% reduction in the intervention group (compared with only a 21% reduction in the control group).

Applying Improvement Science Locally

While these organized studies were successful, it was essential to engage local stakeholders in applying the CLABSI checklist in a manner that fit their local context, thereby motivating them to own the problem and the solution. In hospitals that were ultimately successful in decreasing and sustaining low infection rates, a clearly communicated organization-wide commitment to eliminate infections was backed by structured support from hospital and local ICU leaders to implement the checklist, to monitor and transparently report on performance, and to be accountable for infection rates. For example, if infection rates in an ICU were high for two consecutive quarters, a peer-to-peer review was conducted to uncover the problem and to work with unit managers, department directors, and hospital leaders to develop an improvement plan.

A coordinating team, comprising clinical quality staff and infection control practitioners from the hospital as well as ICU physician and nurse leaders, coached frontline clinicians on the technical components of improvement science (e.g., interventions to prevent infections, data collection procedures, etc.) and supported implementation of the CLABSI checklist. A network of unit-based improvement teams used the comprehensive unit-based safety program to stimulate a culture focused daily on safety and quality improvement. All teams connected as a community to learn and share successes and barriers. The individual team members were instrumental in transitioning the CLABSI interventions into daily practice, educating colleagues about infections, reporting infection rates, and intrinsically motivating coworkers to improve performance.

Underlying Motivations for Improvement

While biomedical research and improvement science were essential in decreasing the rate of CLABSIs, equally important were the underlying motivations of the managers and frontline clinicians who were directly involved in the improvement effort. An ethnographic inquiry of one group of clinicians from different ICUs in Michigan revealed an important discovery, namely, a shift in norms and a change in the narrative from “These infections are inevitable” to “These infections are preventable, and I can help to eliminate them.”

For example, a nurse in a rural Michigan hospital challenged a senior physician who had not donned a sterile gown when preparing to insert a central line, saying, “Bloodstream infections are preventable, and it’s my job to ensure that they are, so go back and put on a gown.”

In addition, when frontline staff listened to and told stories about patients who had been harmed and looked at the faces of those who had died of bloodstream infections, the problem became more personal and real. Similarly, when staff observed the infection rates dropping after the implementation of the intervention at their own hospital or at other hospitals, they understood that such infections were preventable and that they had a role in preventing them.

Aligning National Support for Improvement Science

To help ensure that these discoveries found their way into routine patient care, policymakers aligned the nation around the common goal and sparked a movement to reduce CLABSIs. The Department of Health and Human Services collaborated with patient and professional stakeholder organizations to set realistic goals and launched the National Action Plan to Prevent Health Care-Associated Infections (HAI Action Plan) in 2009. Patient advocacy groups pressured hospital leaders to reduce bloodstream infections and to transparently report their infection rates, which in turn prompted additional state and federal policy initiatives, including a national goal to reduce CLABSIs by 50% between 2009 and 2013. Congress created an incentive to prevent CLABSIs by including the HAI Action Plan in the Affordable Care Act, and the Centers for Medicare & Medicaid Services started publicly reporting CLABSI rates through their Hospital Compare website.

Improvement science encompasses more than a local improvement project; it is a comprehensive approach that draws on social science theory to produce and implement multifaceted interventions for the purpose of accelerating the translation of biomedical research into practice. It brings policymakers and clinicians together, aligning all parties around a common purpose and coordinating efforts at national, organizational, local, and sociocultural levels to achieve a goal. Linking improvement science to biomedical research makes it possible for those in health care to both think like engineers to solve problems and think like biomedical researchers to solve puzzles. While extrinsic motivators such as pay for performance and public reporting can help to engage hospital leaders, intrinsic motivators such as the faces of patients who have been harmed and the belief among clinicians that they can prevent harm will accelerate improvement. In the end, change progresses at the speed of trust, and policymakers and clinicians must create mutually respectful relationships that assume positive intent and are guided by shared leadership accountability.


This article originally appeared in NEJM Catalyst on January 11, 2017.

New Call for Submissions ­to NEJM Catalyst


A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »

More From Care Redesign
Relentless Reinvention

Adopting Innovations in Care Delivery — The Case of Shared Medical Appointments

Given the effectiveness of group interventions, why aren’t doctors routinely using them to treat physical and mental conditions?

Relentless Reinvention

“Being the Best at Getting Better” — Creating a Culture of Change

How Cincinnati Children’s Hospital Medical Center built a culture focused on broad-based change that is transformational for children and their families.

Relentless Reinvention

Rural Health Care: Thirty Miles at Sea — Providing Consistent Care in an Inconsistent Environment

How one of the smallest hospitals in Massachusetts addresses the needs of its unique population.

Relentless Reinvention

Lessons from Oregon in Embracing Complexity in End-of-Life Care

Persons with chronic progressive medical illness require more care in the ICU and more hospitalizations, and often receive late or no referrals to hospice care. These utilization patterns are strikingly different in Oregon.

Relentless Reinvention

The Move to Value-Based Care in Navy Medicine

Achieving the mission of Navy Medicine to “keep the Navy and Marine Corps family ready, healthy, and on the job” requires rethinking current health care delivery models.

Relentless Reinvention

Improving Access to Specialist Expertise via eConsult in a Safety-Net Health System

Electronic referral system supports communication between primary care and specialty providers.

Relentless Reinvention

Redesigning the Delivery of Specialty Care Within Newly Formed Hospital Networks

As the trend toward hospital mergers and consolidations continues, how can newly formed health care networks optimize their delivery of specialty care? They will need to consider a redesign of service lines that includes both centralizing and decentralizing strategies.

Relentless Reinvention

Survey Snapshot: Genomic Data Is Far from Clinical Use

NEJM Catalyst Insights Council members say that clinical and cost data will continue to be the most useful data sources.

Relentless Reinvention

Why Every Health Care Organization Needs a Data Science Strategy

Data science strategy can help providers tap into the power of their data, improve its quality, and keep it safe.

Relentless Reinvention

Learning to Drive — Early Exposure to End-of-Life Conversations in Medical Training

The importance of listening to the patient at the end of life.


A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »


Coordinated Care

87 Articles

Patient Engagement Survey: How to Hardwire…

Technology and social networks can help, but nurses and care teams remain essential, say NEJM…

Social Needs

41 Articles

Evaluating Complex Care Programs: Is It…

Policymakers see programs for complex patient populations as a way to bend the health care…

Reading List: Dave Chokshi and François…

NEJM Catalyst Thought Leaders weigh in on the most influential and inspiring texts of their…

Insights Council

Have a voice. Join other health care leaders effecting change, shaping tomorrow.

Apply Now