Heartburn drugs linked to premature death

"Millions of people taking common heartburn and indigestion medications could be at an increased risk of death," The Guardian reports after a US study found people taking proton pump inhibitors (PPIs) had a slightly higher risk of death than the control group.

PPIs reduce the amount of acid in the stomach. As well as being used to treat heartburn, they're often given to people as a protective measure if they're thought to be at risk of a stomach ulcer – for example, people who take daily low-dose aspirin, which is known to irritate the lining of the stomach.

This headline is based on research in 350,000 predominantly male US veterans who were prescribed PPIs or H2 blocker drugs to either treat heartburn or protect the stomach. PPIs and H2 blockers both work by reducing stomach acid.

The researchers found people who took PPIs had a greater risk of death from any cause compared with those who took H2 blockers or nothing at all.

But there was no proof that the increased risk of death was directly caused by the PPI drugs. The researchers tried to adjust for underlying health factors, such as cardiovascular disease, which is often treated with daily aspirin, but it's possible the effects of these or other factors could still have influenced the results.

If you've been prescribed PPIs, you shouldn't stop taking them without first consulting your GP. The risk of not taking them (such as a stomach bleed) may be greater than any risk associated with taking them.

Where did the story come from?

The study was carried out by researchers from VA Saint Louis Health Care System, Washington University School of Medicine, and Saint Louis University in the US.

No information on funding was provided, but the data the researchers analysed came from the US Department of Veterans Affairs.

The study was published in the peer-reviewed journal BMJ Open and is open access, so it's free to read on the BMJ website.

The UK media's coverage of the story was generally accurate, but the headlines failed to reflect the inherent limitations of the study – including the fact that the conditions people were taking PPIs for in the first place may also have been one of the main causes of death.

What kind of research was this?

This large cohort study of US veterans aimed to look at whether PPIs or H2 blockers were associated with risk of death.

H2 blockers are drugs like ranitidine (Zantac) that reduce stomach acid, and are commonly used to treat acid reflux or heartburn.

PPIs such as omeprazole work in a slightly different way, but are also used to protect the stomach, often in people who have ulcers or those at risk because they take anti-inflammatories or aspirin long term.

Both types of drugs are available on prescription, and some can be purchased over the counter in pharmacies.

As this was a cohort study, it can't prove that taking one drug directly causes death – it can only show there's an association. It might be the case that other health, sociodemographic or lifestyle factors, such as high body mass index (BMI), contributed to the higher risk of death.

A randomised controlled trial (RCT) would give more reliable evidence on the direct effect of either taking the different drugs or doing nothing (control group) while controlling for other factors.

But RCTs can be expensive and time consuming to carry out. Cohort studies can be useful to assess potential adverse effects, as they're able to follow an extensive number of people (in this case 349,312) over a long period of time.

What did the research involve?

Researchers used the US Department of Veterans Affairs national databases to identify 349,312 people (average age 61, 94% male) who'd been prescribed acid suppression therapy (PPIs or H2 blockers) between 2006 and 2008. They looked at their likelihood of death by any cause over 5.71 years on average.

Information on deaths is routinely gathered by the Veterans Benefit Administration for all US veterans.

The 275,977 participants whose first acid reflux drug was a PPI were placed in the PPI group, while the 73,335 participants who received H2 blockers first were the reference group.

In the H2 blocker group, 33,136 participants were later prescribed a PPI and were placed in the PPI group from the point they started taking PPI drugs.

The main outcome of interest was drug use in relation to death. The researchers also looked at how long the drugs were prescribed for.

They adjusted their data to take into account a number of things that could have influenced the results, including:

age
race
gender
kidney function
number of hospitalisations

They also took into account a range of chronic illnesses, including:

diabetes
hypertension
cardiovascular disease
peripheral artery disease
stroke
chronic lung disease
hepatitis C
HIV
dementia
cancer
a range of gastrointestinal illnesses

What were the basic results?

Overall, 23.3% of the entire cohort died over the 5.71-year follow-up. The rate was 12.3% in those using H2 blockers at the start of the study, 24.4% in those using PPIs at the start of the study, and 23.4% in those who'd ever used PPIs.

The researchers found:

PPI use was associated with increased risk of death compared with H2 blocker use (hazard ratio [HR] 1.25, 95% confidence interval [CI] 1.23 to 1.28)
PPI use versus no known exposure to acid suppression therapy (PPIs or H2 blockers) was also linked with a similar increased risk of death (HR 1.23, 95% CI 1.22 to 1.24)

Risks were similar when only looking at participants with no known gastrointestinal problems:

PPI versus H2 blocker use (HR 1.24, 95% CI 1.21 to 1.27)
PPI versus no known acid suppression therapy (HR 1.22, 95% CI 1.21 to 1.23)

Compared with participants taking PPIs for 30 days or less, risk of death gradually increased with the length of time they were taking them:

31-90 days (HR 1.05, 95% CI 1.02 to 1.08)
91-180 days (HR 1.17, 95% CI 1.13 to 1.20)
181-360 days (HR 1.31, 95% CI 1.29 to 1.34)
361-720 days (HR 1.51, 95% CI 1.47 to 1.56)

How did the researchers interpret the results?

The researchers concluded that, "The results suggest excess risk of death among PPI users; risk is also increased among those without gastrointestinal conditions and with prolonged duration of use. Limiting PPI use and duration to instances where it is medically indicated may be warranted."

Conclusion

This larger set of observational data finds that PPI drugs are associated with an increase in the risk of early death compared with either H2 blockers or no acid suppression drugs. This was the case for participants both with and without gastrointestinal problems.

It also appears as though the longer the PPIs drugs are taken, the greater the risk of death.

Considering that these drugs are widely used in the UK, these findings may cause concern. But the research has a number of important limitations:

The study was conducted in a population of mostly white, older US male veterans, which might limit the ability to generalise the results to the whole UK population.
Deaths can't be linked directly to the use of PPIs. The researchers have tried to adjust for many health and other characteristics that could be linked with both PPI use and higher risk of death, such as cardiovascular diseases, but we still can't be certain the influence of the disease has been fully taken into account.
Many of the deaths occurred in the first year, so could well be linked to underlying causes. There was also no information on cause of death.
The follow-up period only lasted around five years. Longer term death outcomes weren't examined – it may be that PPIs are associated with better outcomes for participants in the long term, but we can't say for sure either way.
The length of follow-up in the PPI group was more than two years longer than in the H2 blocker group, so it's unsurprising there was a greater risk of death given the extra two years of data collection.
The drugs were all prescribed in outpatient settings. Some brands of these drugs are available over the counter in the UK. There might be a difference between the groups of people who have their drugs prescribed and those who buy them over the counter, both in terms of risk and in the dose of the drugs.
This study can't attribute risk to any individual PPI drug. If there is a direct mortality risk from PPIs, it may differ according to which drug it is – but this study isn't able to tell us this.

Overall, this large study of good-quality data raises a clear link that needs further examination.

But people who have been prescribed PPIs shouldn't stop taking them – the risk of not doing so may be much greater than any risk the drugs pose. For example, a bleeding stomach ulcer can be very serious and potentially life threatening.

If you're concerned about your medication, you should discuss your treatment options with your GP or the doctor in charge of your care.