Skip to main content
. Author manuscript; available in PMC: 2019 Apr 1.
Published in final edited form as: Gastroenterology. 2017 Dec 19;154(5):1290–1297.e1. doi: 10.1053/j.gastro.2017.12.006

TABLE 2.

RISK OF STROKE EVENTS BY CURRENT USE OF PPI THERAPY

TOTAL STROKE
NON-USER REGULAR USER

NHS Cases/Person-years 1,707/635,985 330/110,756
Age-adjusted 1 (Ref.) 1.14 (1.01–1.29)
Model 1 1 (Ref.) 1.07 (0.95–1.21)
Model 2 1 (Ref.) 1.01 (0.89–1.16)

HPFS Cases/Person-years 467/170,856 95/31,733
Age-adjusted 1 (Ref.) 1.03 (0.82–1.29)
Model 1 1 (Ref.) 1.00 (0.80–1.25)
Model 2 1 (Ref.) 0.89 (0.69–1.14)

TOTAL Cases/Person-years 2,174/806,841 425/142,489
Age-adjusted 1 (Ref.) 1.11 (1.00–1.24)
Model 1 1 (Ref.) 1.06 (0.95–1.17)
Model 2 1 (Ref.) 0.99 (0.88–1.11)
ISCHEMIC STROKE
NON-USER REGULAR USER

NHS Cases/Person-years 811/635,985 169/110,756
Age-adjusted 1 (Ref.) 1.27 (1.07–1.51)
Model 1 1 (Ref.) 1.18 (0.99–1.40)
Model 2 1 (Ref.) 1.08 (0.89–1.30)

HPFS Cases/Person-years 225/170,856 56/31,733
Age-adjusted 1 (Ref.) 1.21 (0.90–1.62)
Model 1 1 (Ref.) 1.18 (0.88–1.60)
Model 2 1 (Ref.) 1.08 (0.77–1.51)

TOTAL Cases/Person-years 1,036/806,841 225/142,489
Age-adjusted 1 (Ref.) 1.25 (1.08–1.46)
Model 1 1 (Ref.) 1.18 (1.02–1.37)
Model 2 1 (Ref.) 1.08 (0.91–1.27)
HEMORRHAGIC STROKE
NON-USER REGULAR USER

NHS Cases/Person-years 223/635,985 36/110,756
Age-adjusted 1 (Ref.) 0.94 (0.65–1.35)
Model 1 1 (Ref.) 0.97 (0.67–1.40)
Model 2 1 (Ref.) 0.94 (0.63–1.39)

HPFS Cases/Person-years 50/170,856 12/31,733
Age-adjusted 1 (Ref.) 1.19 (0.63–2.25)
Model 1 1 (Ref.) 1.12 (0.59–2.15)
Model 2 1 (Ref.) 0.79 (0.38–1.63)

TOTAL Cases/Person-years 273/806,841 48/142,489
Age-adjusted 1 (Ref.) 0.99 (0.72–1.36)
Model 1 1 (Ref.) 1.01 (0.73–1.38)
Model 2 1 (Ref.) 0.90 (0.64–1.27)

Model 1: adjusted for age (continuous), smoking status (never <5, 5–20, 20–40, >40 pack-years), alcohol intake (g/day, continuous), body-mass index (continuous), physical activity (metabolic-equivalent task/week, continuous), Alternative Healthy Eating Index scores (continuous), menopausal hormone use (among women, current vs. past/never), multivitamin use, regular aspirin use, regular non-aspirin NSAID use, history of hypertension, hyperlipidemia, coronary artery disease, or diabetes

Model 2: Model 1 + history of peptic ulcer disease (ever/never), history of gastroesophageal reflux disease (GERD; ever/never), history of gastrointestinal bleeding (ever/never), history of prior histamine-2 receptor antagonist (H2RA) usage (ever/never)