Teamwork has become a key principle of operations in virtually every industry. Even in the manufacturing sector, where workers once stood alone at their stations on the assembly line, it’s common to employ team approaches to production problem-solving. At the same time, the role of specialized knowledge — human capital — in conjunction with mentoring and sharing local knowledge is increasingly being recognized for its critical role in spurring productivity: when more experienced specialists share knowledge with less experienced colleagues, researchers have found, productivity rises.
In healthcare, productivity translates not only to better financial outcomes but more importantly to higher quality patient care. Professor Ann Bartel, along with independent researcher Nancy Beaulieu, Ciaran Phibbs of Stanford and the Veterans Administration, and Patricia Stone of Columbia University wanted to learn more about how human capital affects productivity in healthcare by determining the qualities of the most productive nursing teams.
Why nurses rather than doctors? While doctors diagnose and plan care, nurses implement these plans, so their work is critical to day-to-day quality of care. And nurses’ work is especially team-centered: nursing care requires multiple shifts per day, and nurses must regularly share information about patients’ conditions and treatment with those on other shifts. “It really is a group production process where individual knowledge workers develop communication solutions with co-workers and apply their expertise to solving problems,” Bartel says.
The researchers used longitudinal monthly data from nursing units in Veterans Administration (VA) hospitals in all 50 US states from 2003 to 2006 to study the relationship between patient outcomes and nursing staff characteristics. Unlike other hospital systems, the VA creates a separate electronic discharge record for each unit stay for each patient, enabling the researchers to link the patients to the nursing units in which they were treated. Another VA data set, Personnel and Accounting Integrated Data, allowed the researchers to identify when new nurses joined a unit, when experienced nurses left, whether nurses were regular staff or contract nurses hired to cover absences of regular staff, as well as each nurse’s education and their experience working in the hospital and specific units.
Hospitals typically have an estimate of how long a patient with a given diagnosis is expected to stay in a unit and in the hospital. By looking at the deviation of the actual length of stay (LOS) from the expected LOS, the researchers were able to get a sense of whether patients were getting good nursing care. “High LOS is a bad outcome,” Bartel explains. “Controlling for the patient’s initial diagnosis, it indicates that the patient developed complications such as infections, blood clots, or pressure ulcers, suggesting a lower quality of care.” The detailed nature of the VA data meant the researchers could even link LOS to the characteristics of the nurses on the unit in which a patient was treated, allowing the researchers to model human capital in a way that other studies have not to understand how communication, knowledge sharing, and coordination work within nursing teams.
They concluded that skill level and experience in a specific unit matter a lot, producing better patient outcomes in the form of lower LOS. Specifically, patients cared for a by a team of registered nurses that has more experience on that particular unit were discharged, on average, sooner than those cared for by RNs with less experience on the unit.The researchers estimate that if a hospital with 25 nursing units could increase the unit tenure of its RNs by about 4 years on average, each hospital could save roughly $500,000 a year while producing better patient outcomes.
The researchers also found significant team disruption effects when hospitals brought in contract nurses: LOS increased when contract nurses substituted for staff nurses. Bartel says this demonstrates the importance of a nurse’s experience on a nursing unit. “Even very skilled contract nurses don’t have the specific knowledge about the unit,” she explains. “They receive little orientation or training and are usually brought into the unit on very short notice, and they are likely to be unfamiliar with procedures, practices, and equipment in the unit as well as with their colleagues.” While the presence of contract nurses increases staffing intensity, these additional resources are not productive in improving patient outcomes.
To some degree, this isn’t surprising: studies in other industries have looked at the impact of contract workers and found reduced productivity, increased incidence of work accidents, and, when it comes to substitute teachers, negative impacts on student test scores. One reason that this kind of hyperlocal expertise is so important in the case of nursing teams, Bartel says, is that hospitals use their own systems, policies, and procedures, and with nursing units, unit managers are often free to establish their own work processes. That leaves staff nurses with a great deal of unit-specific human capital and contract nurses at a disadvantage.
Finally, the researchers found that other disruptions to the team, such as an experienced nurse leaving or a newly-hired nurse acclimating to the unit and the hospital, resulted in significant decreases in the unit’s productivity. Because the productivity of experienced nurses spills over to less experienced nurses, team performance is enhanced as the less experienced nurses learn from their more experienced mentors. A less inexperienced nurse without unit-specific knowledge can’t offer that to her team. “These kinds of exchanges just can’t occur as often when experienced nurses are out,” Bartel says. “Many of these seemingly informal relationship dynamics are lost to the team when experienced nurses are away or leave the unit.”
Ann Bartel is the Merrill Lynch Professor of Workforce Transformation and Chair of the economics subdivision in the Finance and Economics Division at Columbia Business School.
Professor Bartel is the Merrill Lynch Professor of Workforce Transformation at Columbia Business School and the Director of Columbia Business School's Workforce Transformation Initiative. She is an expert in the fields of labor economics and human resource management and has published numerous articles on employee training, human capital investments, job mobility, and the impact of technological change on productivity, worker skills, and outsourcing decisions. Bartel...
Read the Research
Ann Bartel, Nancy Beaulieu, Ciaran S. Phibbs, Patricia W. Stone
"Human Capital and Productivity in a Team Environment: Evidence from the Healthcare Sector"