Fall-prevention strategies and patient characteristics that impact fall rates after total knee arthroplasty.

Background: Fall prevention has emerged as a national quality metric, a focus for The Joint Commission, because falls after orthopedic surgery can result in serious injury. In this study, we examined patient characteristics and effects of fall-prevention strategies on the incidence of postoperative falls in patients undergoing total knee arthroplasty.

Methods: We reviewed electronic records of all patients who fell after total knee arthroplasty between 2003 and 2012 (10 years). Patient demographics, including age, sex, and body mass index, were analyzed. The impact of various fall-prevention efforts, including provider and patient education, Hendrich II Fall Risk Model, fall-alert signs, and the use of patient lifts on the incidence of falls, also was studied.

Results: Between January 2, 2003, and December 31, 2012 (10 years), 15,189 total knee arthroplasties were performed at Methodist Hospital, Mayo Clinic Rochester, MN. The overall fall rate was 15.3 per 1000 patients (95% confidence interval [CI]: 13.4-17.4). The rate varied significantly (P < 0.001) during the 10-year period with an initial increase followed by a gradual decrease after the initiation of the fall-prevention strategies. From multivariable analysis adjusting for the temporal trends over time, the odds of falling were found to increase with older age (odds ratio = 1.7 and 2.0 for those 70-79 and ≥80 compared with those 60-69 years of age; P < 0.001) and were lower for patients undergoing revision compared with primary total knee arthroplasties (odds ratio = 0.6, P = 0.006). There was no statistically significant difference in fall rates by sex or body mass index. Most patient falls (72%; 95% CI: 66%-78%) occurred within their own rooms. Elimination-related falls (those that occurred while in the bathroom, while going to and from the bathroom, or while using a bedside commode) comprised a majority (59%; 95% CI: 53%-65%) of the falls. Most patients who fell were not considered high risk according to the Hendrich II Fall Risk Model. Twenty-three percent of falls were associated with morbidity, including 7 return visits to the operating room and 2 new fractures.

Conclusions: Our data demonstrate a reduction in fall incidence coinciding with the implementation of a multi-intervention fall-prevention strategy. Despite prevention efforts, patients of advanced age, elimination-related activities, and patients in the intermediate phase (late postoperative day 1 through day 3) of recovery continue to have a high risk for falling. Therefore, fall-prevention strategies should continue to provide education to all patients (especially elderly patients) and reinforce practices that will monitor patients within their hospital rooms.