North Carolina Modeling Infectious Diseases (NC MInD) Program: Our Innovative Modeling Plan to Combat Healthcare-associated Infections and Antibiotic Resistance in North Carolina

Too often in the United States, patients are admitted into a health care facility for routine treatment, expecting routine outcomes. They may be cautioned by their healthcare team about the potential for a negative outcome resulting from an infection, but they do not anticipate that it could happen to them.

Healthcare-associated infections (HAIs)—infections patients can get while receiving medical treatment in a hospital or other healthcare facility—are all too common in today’s healthcare settings. HAIs affect many people and can extend hospital stays and make patients sicker or even lead to premature death. The Centers for Disease Control and Prevention (CDC) estimates that on any given day, about 3 percent of all patients at hospitals have an HAI—a staggering figure given the number of admissions each day. Every year approximately 722,000 HAIs are diagnosed in the United States, and about 72,000 hospital patients with HAIs die during their hospitalizations. HAIs cost the U.S. health system up to $45 billion per year.

Even more alarming, HAIs are increasingly caused by antibiotic-resistant (AR) bacteria, including those bacteria that are resistant to last-line antibiotics. These organisms are sometimes referred to as “super-bugs” in the news. According to the U.S. Antibiotic Resistance Threats Report, two of the most urgent threats are Clostridium difficile (C. difficile) and Carbapenem-resistant Enterobacteriaceae (CRE).

Although health care providers can prevent most people from getting an HAI, transmission of these infections is complicated, making HAIs difficult to control. The risk of getting an HAI involves multiple factors, including the health of the patient and the type of care and treatment they need; the type of health care facility the patient is in, like a hospital or nursing home; and the health care providers in these facilities. The movement of patients between health care facilities and their homes can also affect their and others’ risk of getting an HAI. HAI prevention could be enhanced with an improved understanding of the interconnectedness of health care facilities in a geographical area or region, more accurate estimates of HAI risk and knowledge of risk factors for getting an HAI, and early identification of high-risk areas within health care networks.

We have a plan to combat HAIs in North Carolina

The CDC has identified eliminating HAIs, including those caused by AR bacteria, as a public health priority and a “Winnable Battle.” HAI prevention is also part of the U.S. Government’s Antimicrobial Resistance Challenge. To support the fight against HAIs and AR bacteria, the CDC established the Modeling Infectious Diseases in Healthcare Network (MInD-Healthcare). In August 2017, RTI was one of five sites selected to explore HAI transmission across the health care continuum and assess HAI prevention strategies using computational models. Through this award, the North Carolina Modeling Infectious Diseases Program (NC MInD) was created.

To understand how HAIs spread, we joined forces with two major authorities on health in North Carolina: UNC Health Care and the N.C. Division of Public Health. UNC Health Care (UNC) is the state health care system of North Carolina, which includes UNC Hospitals and 11 affiliate hospitals and hospital systems across the state. It sees more than 1.4 million patients annually. Working together, we developed an innovative model using an agent-based simulation modeling (ABM) approach.

An ABM is a computational model that represents people, called agents, and that allows us to track individual characteristics (age, gender), locations (home, hospital), and infection status (infected, not infected), without using the identifying information of real-life patients.

To develop an ABM of the NC regional health care network of UNC Health Care, including hospitals, nursing homes, and the community, NC MInD used real health care and public health data and the RTI U.S. Synthetic Household Population database. The RTI U.S. Synthetic Household Population database was previously developed by our colleagues in geospatial science for the National Institutes of Health (NIH).

Patients move through the health care system in different paths. For example, a patient may be transferred from one hospital to another and then go back home, or from a nursing home to a hospital and back. Our ABM allows us to describe the movement of people among health care facilities and explore how C. difficile and CRE are transmitted. It helps us identify areas of high risk for infection—and may lead to insights into how to prevent future infections.

During the project, NC MInD will evaluate HAI-prevention strategies and develop HAI tools for public health officials to identify potential high-risk areas within health care networks. These data will provide public health decision-makers with information that could reduce and prevent HAIs, thus lessening the serious personal and financial toll on patients.

For additional information about NC MInD, please contact Sarah Rhea.

NC MInD is a collaborative effort. We would like to acknowledge the contributions of our RTI colleagues, Rainer Hilscher, Breda Munoz, Kasey Jones, and James Rineer.