Research findings call on hospital leaders to take a deeper look at how staff—primarily nurses—interact with patients, to determine a way for patients at highest risk for infection to come into contact with fewer workers.
These days, the term 'social networking' is practically synonymous with 'social media,' calling to mind buzzing Twitter alerts and updating Facebook statuses. But by analyzing a hospital's social network, in the traditional sense of the phrase, researchers have discovered a new model that may help to minimize infections.
The study, developed by two researchers from the University of Maryland and one from American University, created computer models that simulated interactions between patients and healthcare workers to determine if the interactions were a source for spreading multi-drug resistant organisms.
To do this, the researchers manipulated and tracked the dynamics of the social network in a mid-Atlantic hospital's intensive care unit. They focused primarily on interactions between patients and healthcare workers, as well as on multiple competing factors—staff coverage for meetings, break, and sick leave—that can affect the transmission of infection.
Such transmissions strike one in 20 inpatients, causing tens of thousands of deaths and training billions of dollars from the national healthcare system annually, according to the Department of Health and Human Services.
"Most people think of transmission as being primarily a contact-driven problem, but they are not likely to be familiar with social network analysis and how it could be leveraged to provide insight to this very significant problem," says Sean Barnes, Ph.D., Assistant Professor at University of Maryland's Robert H. Smith School of Business. "By thinking about the structure of interactions between people in a hospital, you can start to devise alternative ways of preventing transmission."
And they did. Barnes and his fellow researchers found that there is a strong correlation of a "sparse, social network structure" with low infection transmission rates.
3 Recommendations
These findings call on hospital leaders to take a deeper look at how their staff – primarily, nurses – are interacting with patients, and to determine a way for the highest risk patients to come into contact with fewer workers. Barnes has three staffing practices that hospitals might consider.
1. Hire more nurses.
This strategy is simplest in terms of complexity, if not in terms of feasibility, Barnes says. "By limiting the number of patients each nurse comes into contact with, transmission can be limited significantly."
2. Create a strategy for patient sharing.
This practice tasks hospital leadership with deciding on a strategy for healthcare workers to cover one another's patients in a way that limits the connectivity of the contact network.
Barnes and his colleagues recommend two approaches for doing so: paired sharing (i.e. the buddy system) and revolving sharing (i.e. a circular sharing system).
3. Limit patient contact to only when medically necessary.
"Additional contact with patients only creates more opportunities for transmission," Barnes says.
Of course, it's up to hospitals to ensure that changing the way their workers interact with patients doesn't affect quality of care—something Barnes cites as the most significant issue related to minimizing interactions between patients and healthcare workers.
"We don't mean to promote the idea that having no interaction is ideal," he says. "Of course, each patient should receive an appropriate level of care to improve his or her health outcomes. The good news is that our approach supports the concept of care continuity, because seeing the same healthcare workers is the best strategy for minimizing transmission."
Barnes' team's approach is also in step with the recent study in Health Affairs that found bumping nurse staffing levels by three nurse hours per patient day provides a demonstrable and marked reduction in hospital readmissions.
"This approach actually supports a sparse social network structure, as long as the increased time is dedicated by the same nurse to the same assignment of patients," Barnes says. "If, for example, a new nurse was coming in and interacting with multiple patients across the unit, this could create new connections between patients and put previously protected patients at risk for acquisition."
While Barnes says he is not personally aware of hospitals that are using a social network analysis approach to limit contact, there are many hospitals using gloves and gowns to minimize direct contact with patients.
See Also: Cheaper Way to Stop MRSA Adds No Patient Risk
"There are also some studies looking at nurse-to-patient staffing ratios, but my impression is that they are inconclusive at this point," he says. "The challenge with these studies is data collection and integrity. Our advantage using simulation mitigates this challenge because we can experiment directly with these experimental factors and observe the effects."
Moving forward, it would be a helpful exercise for hospitals—particularly intensive care units—to start mapping the social networks within individual units, and potentially across multiple units that have a high degree of interaction, Barnes says.
"With this map, hospital administrators and infection control practitioners could analyze whether or not there are simple changes they can make to reduce the density of connections," he says.
Marianne Aiello is a contributing writer at HealthLeaders Media.