Skip to main content

Table 4 Effect size (withdrawn compared with active treatment) 95% confidence intervals and denominators for study outcomes

From: Withdrawal of inhaled corticosteroids in individuals with COPD - a systematic review and comment on trial methodology

 

Outcome

Study

Whether exacerbation or not during study period (odds ratio)

Time to first exacerbation (mean (M) or median (m))

Frequency of exacerbation requiring course of steroids or antibiotics(rate ratio (rr) or hazard ratio (hr))

Lung function(mean difference in change from baseline)

Total SGRQ score (mean difference in change from baseline)

Walking distance in 6 minutes (mean difference in change from baseline)

Borg exercise tolerance test (units)

COPE

1.5

(0.9 to 2.5)

(n = 242)a

34.6

(M)

(15.4 to 53.8)

(not clear)

1.5 (hr)

(1.1 to 2.1)

(not clear)

-38 ml

(-79.5 to 1.6) (n = 242)

2.5

(0.4 to 4.6)

(not clear)

9.4 m

(-4.5 to 23.2)

(n = 172)

-0.3

(-0.7 to 0.3)

(n = 173)

COSMIC

0.9

(0.6 to 1.4)

(n = 373)b

Not measured

1.2

(rr)

(0.9 to 1.5)

(n = 373)

-4.1%

(-6.6 to -1.6)

(not clear)

0.9

(-.1.3 to 3.1)

(not clear)

Not measured

Not measured

WISP

1.0

(0.6 to 1.8)

(n = 260)b

19

(m)

Not givenc

1.3

(rr)

(1.0 to 1.6)

(n = 260)

-23 ml

(-101.5 to 55.5)

(not clear)

-0.45

Unable to estimate

Not measured

Not measured

O'Brien

Not measured

Not measured

Not measured

-11.3%

(-23.4 to 0.8)

Not measured

-32 ft

(-771 to 707)

Unable to estimated

  1. a all of these exacerbations were moderate - in this trial investigators did not measure exacerbations not requiring treatment
  2. b these exacerbations were mild, moderate or severe
  3. c median = 44 days in withdrawn group (CI 29-59) v. 63 (CI 53-74) days in steroid group, P = 0.05
  4. d incorrect confidence interval given in publication.