The adverse skin conditions that are well-recognized to result from or be associated with systemic corticosteroid use are listed in Table 2. Among these, rash, easy bruising, and slow healing differ between the groups, suggesting some systemic absorption. The reports of rash, however, are significantly higher in the placebo group. Also in Table 2 are the reports of mild adverse skin conditions from the subsample of participants who were considered good compliers with their prescribed inhaler (at least 6 puffs per day of a recommended 12 puffs per day). In this subsample, only reports of easy bruising and slow healing differed significantly between the treatment groups. In the subsequent analysis, only easy bruising and slow healing were considered.
Data on participants who described any bruising are shown in Figure 1 by treatment group and compliance with the study drug, as measured by canister weights, at the semi-annual visits between months 24 and 42. Because relatively few questionnaires were administered at both the 18-month and 48-month visits, data from these visits are not shown.
The percentages of participants in the ICS and placebo groups who were classified as good compli-ers were 54.7% and 59.6%, respectively. The percentages of participants in the placebo group reporting any bruising was uniformly around 10% at all visits shown, and no significant differences were noted between the poor compliers and good compli-ers. In contrast, the percentage of good compliers in the ICS group reporting one or more bruises was close to 20% at all visits, while the percentage of poor compliers assigned to receive ICSs who reported bruising was generally about half that of the good compliers who were assigned to receive the active study drug and was similar to the percentage of placebo-assigned participants with reported bruising, irrespective of compliance. The differences between reports of bruising by the ICS group who complied with their medication and all other subgroups from 24 to 42 months was highly significant (p < 0.002).
Figure 1. Percentage of participants who reported one or more bruises by drug treatment assignment and compliance. Values above the bars represent the number of participants who reported bruising. Differences between ICS, 6+ puffs per day, and all other subgroups were highly significant (p < 0.002). PLA = placebo.
Table 2—Reports of Adverse Skin Conditions per 100 Person-Years by Randomized Group Assignment
|Skin Condition||TAA||Placebo||p Value t|
|Easy bruising||7.94 (22.6)||3.55 (11.4)||0.001|
|Rash||4.48 (13.9)||6.58 (15.7)||0.01|
|Slow healing||1.59(7.1)||0.94 (8.3)||0.02|
|Dry skin||30.1 (47.7)||30.6 (49.4)||0.99|
|Acne||0.85 (7.4)||0.58 (4.0)||0.99|
|Other||8.53 (20.1)||10.6 (23.9)||0.35|
|Good compliers (a 6 puffs/d)§|
|Easy bruising||11.2 (27.0)||3.68 (11.5)||< 0.0001|
|Slow healing||2.38 (8.6)||0.50 (3.6)||< 0.001|
|Dry skin||33.9 (49.9)||29.5 (49.1)||0.24|
|Acne||1.08 (9.3)||0.70 (4.4)||0.72|
|Other||9.78 (22.2)||9.41 (21.4)||0.72|
|Easy bruising||3.72 (14.8)||3.34(11.3)||0.85|
|Slow healing||0.70 (4.5)||1.60(12.3)||0.61|
|Dry skin||25.4 (44.4)||32.3 (49.8)||0.07|
|Acne||0.57 (4.0)||0.39 (3.4)||0.59|
|Other||7.03 (17.1)||12.3 (27.2)||0.35|