Table 3

Subgroup analysis according to use of oral corticosteroid

Tuberculosis (N=366), n (%)Control (N=977), n (%)UnadjustedAdjusted†
OR (95% CI)p ValueOR (95% CI)p Value
OCS users*
ICS
 Non-ICS users‡139 (38.0)515 (52.7)1 (ref)1 (ref)
 ICS users§227 (62.0)462 (47.3)1.44 (0.97 to 2.14)0.0751.35 (0.86 to 2.12)0.192
ICS cumulative dose (μg)
 Median (Q1, Q3)45 000 (15 875 to 92 000)23 500 (15 000 to 53 000)
p Value for trend¶0.0010.005
 0 dose105 (28.7)410 (42.0)1 (ref)1 (ref)
 1–15 00059 (16.1)224 (22.9)0.54 (0.28 to 1.06)0.0740.47 (0.22 to 1.00)0.051
 15 001–45 00076 (20.8)182 (18.6)1.32 (0.73 to 2.41)0.3601.22 (0.60 to 2.46)0.584
 45 001–75 00046 (12.6)72 (7.4)2.17 (1.09 to 4.31)0.0272.40 (1.11 to 5.20)0.026
 75 001+80 (21.9)89 (9.1)2.22 (1.16 to 4.22)0.0161.95 (0.93 to 4.05)0.076
Non-OCS users**
Tuberculosis (N=3783)Control (N=19 606)    
ICS
 Non-ICS users‡3155 (83.4)16 492 (84.1)1 (ref)1 (ref)
 ICS users§618 (16.3)3114 (15.9)1.22 (1.08 to 1.37)0.0021.17 (1.03 to 1.33)0.015
ICS Cumulative dose (μg)
 Median (Q1, Q3)15 625 (6500 to 53 000)13 000 (2750 to 26  000)
p Value for trend¶<0.001<0.001
 0 dose2864 (75.7)13 935 (71.1)1 (ref)1 (ref)
 1–15 000440 (11.6)3905 (19.9)0.51 (0.46 to 0.58)<0.0010.51 (0.46 to 0.58)<0.001
 15 001–45 000227 (6.0)1117 (5.7)1.07 (0.91 to 1.27)0.4211.03 (0.87 to 1.23)0.728
 45 001–75 00098 (2.6)306 (1.6)1.79 (1.39 to 2.31)<0.0011.66 (1.27 to 2.17)<0.001
 75 001+144 (3.8)343 (1.7)2.57 (2.05 to 3.22)<0.0012.18 (1.72 to 2.77)<0.001
  • *Defined if cumulative dose of 1680 mg of hydrocortisone equivalents or more during 1 year before the index date was identified.

  • †Adjusted for LAMA use, SABA use, SAMA use, OCS use, presence of TB sequelae, immunosuppressant use, other comorbidities (malignancy, diabetes, chronic renal failure / dialysis, silicosis, malabsorption, HIV/AIDS and transplantation), Charlson Comorbidity Index and healthcare usage.

  • ‡Defined if inhaler prescriptions shorter than 30 days during 1 year before the index date was identified.

  • §Defined if inhaler prescriptions 30 days or longer during 1 year before the index date was identified.

  • ¶Conditional logit model: used each group median

  • **Defined if cumulative dose of less than 1680 mg of hydrocortisone equivalents during 1 year before the index date was identified.

  • COPD, chronic obstructive pulmonary disease; ICS, inhaled corticosteroid; LAMA, long-acting inhaled muscarinic antagonist; OCS, oral corticosteroid; SABA short-acting inhaled β2 agonists; SAMA, short-acting inhaled muscarinic antagonist.