Twenty four studies were included for the review (n=6,539, data unclear for one study) comprising: five prospective before and after studies (n=2,558, data unclear for one study); one prospective cohort study (n=472); one retrospective cohort study (n=319); one case control study (n=416); 12 retrospective surveys (n=2,767); two case series (n=5); and two case reports (n=2), although one of the case reports was also reported in a case series study.
Before and after studies of bed rail reduction
Two studies scored eight out of 10 on the quality assessment, two scored seven out of 10, and one scored five out of 10. Results were mixed on the association between bed rail reduction and falls. One study reported a significant increase in falls following bed rail reduction (n=93, 0.94 falls versus 2.07 falls per 1,000 bed days post bed rail reduction, p=0.008). One study reported a significant increase in the number of patients who fell more than twice (RR 4, no CI reported). One study found a significant reduction in the number of falls following bed rail reduction (3.78 falls per 1,000 bed days versus 2.04 post bed rail reduction, p=0.02), although the number of falls was still significantly higher than in a control group who continued with bilateral bed rail use (p=0.01). Two studies found no significant difference in fall rate following bed rail reduction. Patients with a history of stroke (one study n= 246; RR 2, 95% CI unclear) and patients with visual impairment (one study n=93, Zelen exact p=0.01) were significantly more likely to fall. Bed rail reduction was not associated with any statistically significant changes in overall injuries, fractures or subdural haematoma.
Case Control and cohort studies
Two studies scored five out of 10 and one study scored four out of 10. A prospective cohort and a retrospective cohort study found no significant differences in injury rates between bed rail use and no bed rails once dependency and cognitive impairment were adjusted for. A case control study found significantly increased risk of falling when one or more bed rails were raised (n= 416, adjusted odds ratio 0.006 95% CI: 0.001, 0.024).
Retrospective Surveys, Case Series and Case Reports
Four out of five studies found no significant differences in fall rates between beds with bed rails raised and those with no bed rails raised. One multi-hospital study found that patients who fell from bed with no bed rails raised were significantly more likely to sustain injuries compared to patients who fell from bed where bed rails were raised (p<0.05). Four studies of reports to the US Food and Drug Administration found that half rails were significantly more likely to result in death, full rails were significantly more associated with non-fatal injury and split rails associated with near misses (one study, n=794; p<0.001). Twelve studies described injuries from bed rails or injuries from falls after bed rail use. One retrospective survey of legal claims after falls from bed found that bed rails were raised in 2.6 per cent of cases.