Table 2.
Association between montelukast use and risk of different out-comes (significantly protective hazard ratios in bold)a
| Study group | Cases/Total (%) | Hazard ratios (95% CI) | |||
|---|---|---|---|---|---|
| Montelukast pos | Montelukast neg | Ageb | Unadjusted | Adjusted | |
| Dementia medicinec | 489/23,521 (2.1) | 4967/178,884 (2.9) | >60 | 0.94 (0.85–1.03) | 0.89 (0.81–0.98) |
| Nursing homed | 507/23,634 (2.2) | 5459/179,768 (3.1) | 60–75 | 0.65 (0.57–0.74) | 0.67 (0.59–0.77) |
| >75 | 1.00 (0.88–1.13) | 0.99 (0.87–1.13) | |||
| Deathd | 5512/23,634 (30.4) | 55,857/179,768 (45.1) | 60–75 | 0.86 (0.82–0.89) | 0.64 (0.61–0.67) |
| >75 | 0.95 (0.92–0.98) | 0.81 (0.78–0.84) | |||
| Parkinson’s medicinee | 691/23,371 (3.0) | 4550/178,134 (2.6) | >60 | 1.21 (1.12–1.32) | 1.06 (0.98–1.15) |
| Diabetes medicinee | 1523/22,270 (7.3) | 11,247/170,157 (7.1) | >60 | 1.01 (0.96–1.07) | 0.85 (0.80–0.90) |
aThe hazard ratios are based on comparing subjects using montelukast with those using only inhaling corticosteroids. The adjusted ratios are adjusted for sex, receiving drugs for heart conditions, and having a high consumption of drugs during the follow up period. The discrepancies between the sum of users in each analysis, and the grand total of 203,473 in Table 1, are due to truncations. The total number of person-years follow-up was between 1,346,308 and 1,480,476 (average 7.0–7.3 years per subject)
bAs of 2014
cThe adjusted analysis was stratified on having a high consumption of drugs dispensed per day, because this covariate did not meet the proportional hazards assumption
dAs the effect of montelukast did not meet the proportional hazards assumption, the model was fitted by splitting into two age-periods, 60–75 years and >75 years
eThe adjusted analysis was stratified on having a high consumption of drugs dispensed per day, and prescription of cardiovascular medicine, because these covariates did not meet the proportional hazards assumption