Skip to main content
. 2019 May 8;14(5):e0216750. doi: 10.1371/journal.pone.0216750

Table 3. Effects of medications on the risk for post-stroke pneumonia.

Variable adjusted HR [95% CI] P value
Time-independent variables
    Sex, male 1.19 [1.09–1.31] < .001
    Age, per 5-year 1.21 [1.18–1.24] < .001
    Hypertension 1.05 [0.94–1.16] .393
    Diabetes mellitus 1.25 [1.13–1.37] < .001
    Myocardial infarction 1.10 [0.93–1.29] .254
    Atrial fibrillation 1.19 [1.02–1.38] .025
    Chronic obstructive pulmonary disease 1.28 [1.15–1.43] < .001
    Thrombolysis (intravenous or intraarterial) 1.03 [0.69–1.54] .885
    Household income
        low Ref
        middle 0.96 [0.86–1.07] .439
        high 0.97 [0.87–1.08] .576
    Hospital type
        general hospital Ref
        hospital or clinic 1.24 [1.10–1.39] < .001
    Length of hospital stay
        ≤17 days Ref
        >17 days 1.00 [0.91–1.09] .986
    Year of index stroke
        2002–2005 Ref
        2006–2009 0.93 [0.84–1.03] .148
        2010–2013 0.93 [0.81–1.07] .337
Time-dependent variables
    Proton pump inhibitor 1.56 [1.24–1.96] < .001
    H2-receptor antagonist 1.40 [1.25–1.58] < .001
    Mucoprotective agents* 0.89 [0.78–1.01] .062

Data were obtained from multivariate time-dependent Cox proportional hazard regression model. Adjustments were made to the variables listed in this table.

*rebamipide, teprenone, irsogladine, ecabet sodium, polaprezinc, sofalcone, sucralfate, and misoprostol.

Abbreviations: HR, hazard ratio; CI, confidence interval.