Skip to main content
. 2019 Sep 18;366:l5101. doi: 10.1136/bmj.l5101

Table 4.

Safety outcomes by aerobic physical fitness training or relaxation session (control group)

Events Aerobic physical fitness training (n=105) Relaxation sessions (n=95) Total cohort (n=200) Incidence rate ratio (95% CI)*
Serious adverse events† (from baseline to three months after stroke)
Median (interquartile range) follow-up (days) 68 (56-78) 69 (54-77) 66 (56-78)
Total No 22 9 31
Incidence rate/100 person months (95% CI) 13.19 (9.22 to 18.86) 7.28 (4.39 to 12.08) 10.38 (7.75 to 13.90) 1.81 (0.97 to 3.36)
Cardiovascular event 0 0 0 -
No with recurrent stroke 8 3 11 -
Incidence rate/100 person months (95% CI) 3.52 (1.76 to 7.03) 1.46 (0.47 to 4.52) 2.54 (1.41 to 4.58) 2.41 (0.64 to 9.10)
No of hospital admissions 14 5 19 -
Incidence rate/100 person months (95% CI) 6.15 (3.64 to 10.39) 2.43 (1.01 to 5.83) 4.39 (2.80 to 6.87) 2.53 (0.91 to 7.04)
No of deaths 0 1 1 -
Incidence rate/100 person months (95% CI) - 0.49 (0.07 to 3.45) 0.23 (0.03 to 1.64) 0.30 (0.01 to 7.42)‡
Self reported adverse events§ (during intervention period)
Median (interquartile range) follow-up time (days) 33 (29-36) 22 (28-36) 33 (29-36)
No of falls 36 14 50 -
Incidence rate/100 person months (95% CI) 32.40 (23.37 to 44.92) 13.84 (8.19 to 23.36) 11.8 (8.9 to 15.5) 2.34 (1.26 to 4.34)
No of fractures 0 0 0 -
No with pain 68 44 112 -
Incidence rate/100 person months (95% CI) 61.20 (48.25 to 77.62) 43.49 (32.36 to 58.44) 26.1 (21.7 to 31.4) 1.41 (0.96 to 2.06)
No with fatigue 29 21 50 -
Incidence rate/100 person months (95% CI) 26.10 (18.14 to 37.56) 20.76 (13.53 to 31.83) 23.55 (17.85 to 31.07) 1.26 (0.72 to 2.20)
No with dizziness 5 14 19 -
Incidence rate/100 person months (95% CI) 4.50 (1.87 to 10.81) 13.84 (8.19 to 23.36) 8.95 (5.71 to 14.03) 0.33 (0.12 to 0.90)
Other 8 4 12 -
Incidence rate/100 person months (95% CI) 7.20 (3.60 to 14.40) 3.95 (1.48 to 10.53) 5.65 (3.21 to 9.95) 1.82 (0.55 to 6.05)
*

Calculated using Poisson regression.

Serious adverse event can belong to more than one category (eg, recurrent stroke and hospital admission). Thus total number of participants with a serious adverse event does not equal sum of individual categories of serious adverse events.

Calculated using penalised maximum likelihood logistic regression (firthlogit command in stata).

§

Adverse events can occur multiple times in a participant. Time interval ranges from baseline to follow-up at three months after stroke or until last observation, if participant dropped out.