Improving Patient Safety in Inpatient Units—A Canadian Context
Trillium Health Centre–Mississauga’s new west wing was designed as an optimum inpatient care facility modeled with a focus on patient safety and improved patient outcomes. Based on the approved inpatient design from 2000-2001, Trillium Health Centre started to build the new wing in March 2007.
A $100 million, 177,000-square-foot addition, it marks a shift in inpatient unit design from the traditional inpatient settings seen elsewhere on Trillium’s Mississauga site and is the first inpatient wing design of its kind in Ontario, Canada.
With a predominant focus on patient safety and the nursing work environment, the new wing features decentralized nurses’ units, bringing nurses closer to patients and improving ergonomics for nurses. Along with the increase in the number of private rooms (50% maximum approved based on 2003 standards), the new wing also includes healthcare provider sinks at each room entry, technological enhancements to connect patients directly with their clinicians, and innovative solutions to make semi-private rooms function optimally (individual access to washrooms, sunlight, and doorways).
Post SARS (severe acute respiratory syndrome), Canadian healthcare providers are more aware of the importance of being able control the spread of infection, isolate patients, practice good hand hygiene, monitor the distribution of patient medications, and provide timely bedside care. Traditional inpatient hospital design, however, has sometimes made it difficult for healthcare providers to implement these practices.
The lack of handwashing is the strongest predictor of cross contamination (Larson, 1998). Research has shown that the location of sinks improves handwashing. The provision of healthcare provider sinks in each entryway should encourage greater handwashing compliance. Medications and supplies are located in proximity to serve a cluster of patient rooms, thereby decreasing the potential for interruptions during medication administration and order transcription.
Specifically related to patient safety, we were interested in determining the benefits of an increased number of private rooms and strategically placed handwash sinks on infection rates. The hypothesis was that these infections would be reduced through the use of private rooms and semi-private rooms with individual access (Chaudhury, Mahmood, and Valente, 2005); an improved ratio of handwash sinks with one at each room entry (Kaplan and McGukin, 1986); and an improved patient/toilet ratio (Korpela et al., 1995).
We also wanted to determine whether medication errors would decrease as a result of the isolated nursing alcoves and improved light levels. The hypothesis was that medication errors in the new wing would be reduced through a decentralized design (Reynolds et al., 1978) that improves the proximity of supplies to the patient and provides increased privacy with improved lighting levels.
In addition, we wanted to provide information regarding the impact the room design had on the number of patient falls. The hypothesis was that falls in the new wing would be reduced through the room design.
The new design provides more direct access to the bathroom with proximity from the headwall to the bath door, wide bath doors, and double-leaf doors in private rooms, which should reduce patient falls (Lieberman, 2004; Joseph, Fabacher, and Rubenstein, 1991, as cited in Dickinson, et al., 2004; Reiling et al., 2004). Having nursing staff closer to the patient would also reduce falls due to increased patient surveillance (Brown 2006, Stichler, 2007).
To assess the impact of the physical environment on safety, a pre-/post-intervention design using mixed methods was used to compare various clinical outcomes and safety metrics before and after the move, including nosocomial infections, patient falls, and medication errors.
Pre-move antibiotic resistant organism (ARO) data was collected retrospectively and included data from November 2005 through December 2008. The pre-move data was then compared to post-move ARO data for the same units that now occupy the new wing. The start period for data collection for the new wing was from November 2009 through March 2010.
Pre-move medication errors and patient fall data were collected retrospectively from January 2006 through December 2008. This pre-move data was then compared to post-move data for the same units that now occupy the new wing. The start period for data collection for the new wing was from November 2009 through March 2010. (Medication errors and patient falls are voluntary entries into the database and, therefore, the data cannot be assumed to be comprehensive of all the medication errors and patient falls that occurred.)
For all metrics, pre-move data collection end dates were selected to minimize confounding variables (pre-move activities). The start date of post-move data collections was selected to correct for a washout period.
For the pre-move time period, there were 4,366 discharged patients and 105 hospital-acquired infections constituting an infection rate of 2.40%. For the post-move time period, there were 2,528 discharged patients and 21 hospital-acquired infections constituting an infection rate of 0.83%. The decreased error rate was statistically significant at the 95% confidence level (p=0.000<0.05).
As mentioned, handwashing is the simplest, most effectively proven method for reducing the incidence of nosocomial infections. However, the work environment has often been blamed for impeding compliance amongst healthcare workers. Eighty percent of the nurses interviewed in a 2004 study stated that easy access to sinks and handwashing facilities would increase the likelihood of handwashing amongst their peers (Heindrich et al., 2004).
Indeed, the hand hygiene audits conducted in the new and old patient units support this notion. With the placement of sinks at the point of care for each patient bed and alcohol dispensers in the hallway, compliance was higher in the new patient units than the older patient units. Staff in both units had received the same type of training; the only difference was the placement of the handwashing facilities.
This in turn led to a significant decrease in the number of nosocomial infections.
For the pre-move time period, there were 7,100 discharged patients and 136 reported medication errors constituting an error rate of 1.92%. For the post-move time period, there were 2,528 discharged patients and 33 reported medication errors constituting an error rate of 1.31%. The decreased error rate, while small, was still statistically significant at the 95% confidence level (p=0.014<0.05).
The decentralized environment with its nursing alcoves ensures that nurses are not interrupted as they dispense medication. Secondly, having a room number for each patient bed minimized the risk of sending the wrong medication to patients. A combination of these environmental factors contributed to a reduction of the medication error rates.
For the pre-move time period, there were 2,731 discharged patients and 95 falls constituting a fall rate of 3.48%. For the post-move time period, there were 2,528 discharged patients and 84 falls constituting a fall rate of 3.32%. While the fall rate decreased from 3.48% to 3.32%, the result was not statistically significant at the 95% confidence level (0.378>0.05).
The decrease is still attributed to the environment in the new wing. The large double-door entrance at the bathroom, for example, is large enough to accommodate a patient with an assistive device or two caregivers with a patient. With adequate maneuvering space for proper body mechanics, opportunities for caregiver and pa
tient injury are reduced.
While the body of literature supporting evidence-based healthcare design coupled with technological innovations is growing, few empirical studies have been conducted that incorporate a provincial or Canadian context. The goal of this study was to investigate the impact of the new wing’s design features (decentralized nursing, patient room design, and wireless technology) on patient safety and staff outcomes, cost effectiveness, and overall patient satisfaction.
Nursing environment scales demonstrated significant positive results, and additional detail will be described in July’s report. While the focus of this brief report is patient safety, additional results related to other aspects of the study will be shared at a later time. HCD
Patti Cochrane, RN, BSCN, MHSC, is Vice President, Quality and Patient Safety and Chief Nursing Officer at Trillium Health Centre site of The Credit Valley Hospital and Trillium Health Centre; Juliet Nishimura, MS, M.ENG, is a Quality and Patient Safety Advisor at Trillium Health Centre site; Gary Spencer, MSC, MBA, is the Director Decision Support at Trillium Health Centre site; Ellen Taylor, AIA, MBA, EDAC, is Director of Pebble Projects for The Center for Health Design. Joseph Szmerekovsky, PhD, provided statistical analysis.
On December 1, 2011, Trillium Health Centre merged officially with The Credit Valley Hospital. The Credit Valley Hospital and Trillium Health Centre is one of Ontario’s largest community-based academic health networks, serving the communities of Mississauga, West Toronto, and the surrounding regions. For more information, see www.trilliumhealthcentre.org/www.cvh.on.ca.
Brown, K. (2006) Impacting Patient Outcomes Through Design: Acuity Adaptable Care/Universal Room Design. Critical Care Nursing Quarterly, 29 (4), 326-341.
Chaudhury, H., Mahmood, A., and Valente, M. (2005) Advantages and Disadvantages of Single-Versus Multiple-Occupancy Rooms in Acute Care Environments. Environment and Behavior, 37(6), 760-786.
Hendrich A. (2006). Hospital Work Environments: Implications for Nursing Practice and Patient Care Quality. In Proceedings of Healthcare Environments Research Summit 2006: Developing the Research Roadmap. February 8-9, 2006: Georgia Tech, Atlanta
Kaplan, L.M., and McGukin, M. (1986) Increasing handwashing compliance with more accessible sinks. Infection Control, 7 (8), 408-410.
Korpela J., Karpanoja P., Taipalinen, R., Sitonen, A. (1995). Subtyping Of Shigella Sonnei For Tracing Nosocomial Transmission. Journal of Hospital Infection, 30(4) 261-266.
Reynolds, D. M., Johnson, M. H., & Longe, R. L. (1978). Medication delivery time requirements in centralized and decentralized unit dose drug distribution systems. American Journal of Hospital Pharmacy, 35(8), 941-943.
Reiling, J.G.,Knutzen, B.L., Wallen, T.K., McCullough, S., Miller, R., Cherno, S. (2004).Enhancing the traditional hospital design process: a focus on patient safety. JT Comm J Qual Saf, 30(3), 115-124.
Stichler, J. F. (2007) Enhancing Safety with Facility Design FACHE JONA 37(7/8),319-323 Copyright B 2007 Wolters Kluwer Health. Lippincott Williams & Wilkins
Tucker, A.L., and Spear, S. (2006) Operational Failures and Interruptions in Hospital Nursing. Health Services Research, 41(3 Pt 1), 643–662.