Table 2.

Active physical therapy intervention in the medical intensive care unit, before and after a quality improvement project

	Pre-QI (n = 120)	Post-QI (n = 123)	P Value*
Ever receiving active PT intervention, n (%)†	19 (16)	84 (68)	<0.001
Days to first active PT intervention, median (IQR)	11 (6, 29)	4 (2, 6)	<0.001
No. of PT sessions in index MICU admission, median (IQR)	0 (0, 0)	3 (0, 5)	<0.001
% ICU days with PT session, median (IQR)	0 (0, 0)	23 (0, 43)	<0.001
% ICU days with PT session after initiation, median (IQR)	18 (4, 47)	50 (33, 67)	0.003
Physiological abnormality or potential safety events, n (% of PT treatment sessions)‡	—	0 (0)	N/A
Highest daily activity during PT session during ICU stay, n (%)			<0.001
Lying or sitting in bed	113 (94)	60 (49)
Sitting at edge of bed	3 (3)	13 (11)
Standing or transfer to chair	3 (3)	28 (23)
Walking	1 (1)	22 (18)

Definition of abbreviations: ICU = intensive care unit; IQR = interquartile range; N/A = not applicable; PT = physical therapy; QI = quality improvement.

^*Calculated using Pearson chi-square, Fisher exact, and Wilcoxon rank-sum tests, as appropriate.

^†The number (%) of patients receiving active PT, by year, during the post-QI period is as follows, 2009: 18 (67); 2010: 22 (61); 2011: 31 (78); and 2012: 13 (65).

^‡Safety events were not collected for the pre-QI period. Safety events were prospectively collected during the post-QI period and defined as any of the following: (1) removal, dislodgment, disruption, or dysfunction of airway, feeding tube, chest tube, vascular access, cardiac device, or wound dressing; (2) cardiovascular/hemodynamic instability that includes hyper- and hypotension, desaturation, cardiac arrest, arrhythmia, and death; or (3) falls (37).