Protopathic Bias May Explain Association Between PPI and Mortality

A team of investigators studied the effect of proton pump inhibitors on mortality by reviewing data from the Centers for Disease Control and Prevention’s Virtual Resource Data Center.

The association between proton pump inhibitors (PPIs) and increased mortality may be explained, in part, by protopathic bias. These findings, from an analysis of an incident drug user cohort, were published in Clinical Gastroenterology and Hepatology.

Investigators analyzed a random sample (n=840,565) of all prescription and encounter claims (N=1,930,728) from Medicare Prescription Drug (Part D) enrollees who were aged 65 years at time of enrollment between 2007 and 2017. They assessed medications and clinical characteristics from included patients up to a median follow-up of 3.8 years.

During the study period, PPI use increased from 14.6% in 2007 to 25.6% in 2017, and use of histamine-2 receptor agonists (H2RAs) increased from 2.8% to 6.4%. The PPI most frequently prescribed was omeprazole (56.5%), and the H2RA most frequently prescribed was ranitidine (66.9%).

Mortality among individuals was as follows:  14.53/1000 person-years (PY) for those who took only PPIs; 13.73/1000 PY among persons who took only H2RAs; 13.41/1000 PY for persons who took both types of medication; and 7.93/1000 PY among persons who did not use either type of medication.

Among patients who had no lag time in their Medicare coverage, any use of PPIs (hazard ratio [HR] 1.1; 95% CI, 1.08-1.12; P <.001) or short-term (£6 months) use (HR 1.26; 95% CI, 1.23-1.28; P <.001) was associated with an increased risk for mortality, and long-term (>6 months) use was associated with a decreased risk for mortality (HR 0.96; 95% CI, 0.94-0.98; P £.0.05) compared with no PPI use.

Among patients with a Medicare lag time of 90 days to correct for protopathic bias, persons with short-term PPI use were at increased risk (HR 1.13; 95% CI, 1.1-1.15; P <.001) and persons with long-term use were at decreased risk (HR 0.9; 95% CI, 0.88-0.92; P <.001) for mortality compared with patients who had never used a PPI.

Patients who had ever used a PPI were at increased risk for mortality compared with patients who had ever used an H2RA among both patients with (HR 1.1; 95% CI, 1.06-1.14; P <.001) and without (HR 1.16; 95% CI, 1.12-1.2; P <.001) a Medicare lag.

This study may have been biased by the fact that 47% of individuals covered by Medicare seek care outside the network and those data were not included in this analysis.

These data, after correcting for possible protopathic bias, did not suggest that PPI use alone was associated with increased mortality risk.


Baik SH, Fung KW, McDonald CJ. The mortality risk of proton pump inhibitors in 1.9 million US seniors: an extended Cox survival analysis. Clin Gastroenterol Hepatol. Published online January 13, 2021. doi:10.1016/j.cgh.2021.01.014