Acid reflux is linked to tinnitus in large population studies, but no study has shown that reflux causes it, and we found no trial testing whether treating reflux makes tinnitus better.
The question comes up because both conditions are common and often turn up in the same person. NIDDK reports that researchers estimate about 20 percent of people in the United States have gastroesophageal reflux disease (GERD), the persistent form of stomach contents coming back up into the oesophagus. What follows is the best available evidence on whether one affects the other, and why its most striking number deserves caution.
The largest study: a sixfold difference in diagnoses
A 2024 study in Scientific Reports used South Korea's national health insurance sample to follow 669,159 adults, with records running to the end of 2019. It compared the time after someone's first reflux diagnosis with time without one. Both reflux and tinnitus were identified from diagnosis codes in insurance claims.
After a reflux diagnosis, tinnitus was newly diagnosed at 14.9 per 1,000 person-years, against 1.74 in time without reflux. After adjusting for age, sex, income, other conditions including hearing loss and vertigo, several medicines, and the number of doctor visits and endoscopies, the hazard ratio was 6.65 (95% CI 6.50–6.81). Time after a reflux diagnosis carried more than six times the rate of new tinnitus diagnoses.
Why a number that large calls for caution
A ratio that large is the kind of result that deserves a second look before it is believed. Several features of the study suggest part of the gap has other explanations, and the authors name most of them.
- It counted diagnoses, not tinnitus. Tinnitus entered the data only when someone raised it with a doctor and it was coded. People in the reflux group had made more medical visits in the year before, 17.41 on average against 14.04. The authors adjusted for visit numbers, but list surveillance bias and differences in health-seeking behaviour among the things that could still sway the result.
- The gap narrowed with time. When tinnitus diagnoses in the first 180 days, first year or first two years after the reflux diagnosis were set aside, the adjusted hazard ratio fell from 6.65 to 5.74, 5.05 and 4.01. Even at two years the association stayed about fourfold, and the authors read these results as in line with the main finding. Still, a good share of the excess arrived soon after the reflux diagnosis, which is also consistent with both being recorded during the same stretch of medical attention.
- Anxiety and depression were largely invisible. The authors describe both as major risk factors for tinnitus, but mental-health codes are often masked in this database, so they used antidepressant and benzodiazepine prescriptions as stand-ins. The NHS lists stress and anxiety among the things that cause or worsen heartburn, and tinnitus and anxiety feed each other, so a shared driver is plausible.
- One shared medicine was left out. NIDDK lists NSAID painkillers among the medicines that can cause or worsen reflux, and frequent NSAID use has its own association with tinnitus, covered in aspirin, NSAIDs and tinnitus. The study's medication adjustments did not include NSAIDs.
- No hearing tests. Recorded hearing-loss diagnoses were adjusted for, but audiogram results were not available, and the reflux code itself had a previously validated diagnostic accuracy of 74%.
None of this means the association is false. It means the study shows reflux and tinnitus diagnoses travelling together in a health system, and cannot say how much of that is reflux acting on the ear.
Reflux medicine did not lower the risk
The same study asked the obvious follow-up. If acid reaching the ear were driving tinnitus, suppressing acid might be expected to reduce it. Among people with reflux, the time counted from a first prescription for a proton pump inhibitor (PPI), the omeprazole family of acid-reducing drugs, carried a higher rate of new tinnitus diagnoses, not a lower one: 17.22 per 1,000 person-years against 10.80, an adjusted hazard ratio of 1.35. The study's abstract describes this only as no significant decrease; the full paper reports the increase.
That is weak evidence that PPIs cause tinnitus, because the people prescribed them were a different group. They were older (mean age 49.2 against 43.5), carried more other illness (22.2% had a comorbidity score of 3 or more, against 8.5%) and had nearly twice as many doctor visits (27.92 against 15.13). The analysis adjusted for these, but gaps this large can leave differences that adjustment does not fully capture, raising a diagnosis rate without the drug doing anything. With a two-year lag the hazard ratio fell to 1.12. The authors float possible mechanisms, including low magnesium with long-term PPI use, as proposals rather than findings.
Other studies do not agree with each other:
- A Korean study of people with type 2 diabetes linked PPI use to a combined outcome of sensorineural hearing loss or tinnitus, with a hazard ratio of 1.50 across 17,233 matched pairs, and the association was stronger in people without reflux.
- A US study of 54,883 women in the Nurses' Health Study II found that daily reflux symptoms were linked to a higher risk of self-reported hearing loss (relative risk 1.33), but once symptoms were accounted for, neither PPIs nor H2 blockers were.
All three are observational. The NHS advises against stopping any prescribed medicine without speaking to a doctor first, and nothing here is a reason to.
The genetic study: a different design, its own caveats
Mendelian randomisation uses genetic variants linked to a condition as a natural experiment. Because genes are fixed at conception, they are not shaped by lifestyle or by the disease itself, which sidesteps some of the confounding above, provided its assumptions hold. A 2025 study applied it to reflux and ear disorders using European-ancestry genetic datasets.
Genetically predicted reflux was associated with constant tinnitus (odds ratio 1.019), tinnitus most of the time (1.007) and occasional tinnitus (1.014), and with a lower chance of never having tinnitus (0.939). It found no causal association with otitis media.
Three things limit what those numbers can carry:
- The numbers do not measure the effect of having reflux. Reflux is a yes-or-no exposure here. Methodologists Stephen Burgess and Jeremy Labrecque explain that Mendelian randomisation with such an exposure is best read as a study of an underlying, possibly unmeasured, tendency towards it, and that turning the result into an effect size needs further assumptions. So these odds ratios describe a higher genetic liability to reflux, not reflux against no reflux, and cannot be set beside the Korean hazard ratio.
- They lean on one method. For constant tinnitus and for tinnitus most of the time, only the main analysis reached significance; the four supporting methods did not. The paper's own heterogeneity table flags significant heterogeneity for three of its four tinnitus outcomes, although its text reports none.
- The reflux genes carry other signals. The genetic markers came from a 2022 analysis in Gut that deliberately combined reflux with genetically related traits, among them education, depression and body mass index, and that identified a subset of reflux loci driven by depression. Depression is tied to tinnitus in its own right, as tinnitus and depression covers, so markers that partly track mood can make reflux look like a cause.
The authors conclude that, under Mendelian randomisation assumptions, reflux may increase tinnitus risk. "May" is the right word.
How reflux could reach the ear
The proposed route is anatomical. The Eustachian tube links the middle ear to the back of the nose and throat, and reflux that rises high enough can reach its opening. Reflux that reaches the throat is usually called laryngopharyngeal reflux.
A 2026 systematic review in the Journal of Voice gathered 20 clinical studies covering 2,100 patients:
- Among people with Eustachian tube dysfunction, throat reflux was found in 20.3% to 32.6% using symptom-based approaches, and in up to 58.0% with objective testing.
- Among people with fluid behind the eardrum (otitis media with effusion), the figures ran from 21.0% to 73.3% on symptoms, and up to 77.4% with objective testing.
- Pepsin and trypsin, both digestive enzymes, were detected in up to 39.6% of middle-ear fluid samples, and animal experiments supported refluxed material inflaming the tube.
The review calls the association probable, but rates the studies low to moderate in quality, notes that reflux was mostly diagnosed from symptoms, and says the evidence does not allow causal conclusions. Tinnitus was not its outcome. It looked at tube dysfunction and middle-ear fluid, which can themselves bring ringing, as otitis media and tinnitus explains. The Korean authors also suggest refluxed acid could affect the round window, a membrane between the middle and inner ear, while describing that mechanism as unclear.
Treating reflux to help the ear has been studied for middle-ear fluid; we found no such study for tinnitus. A 2024 meta-analysis of 16 studies and 21,744 patients found otitis media with effusion more common in people with reflux (risk ratio 1.58), but when reflux treatment was compared with a control group, the improvement did not reach statistical significance (odds ratio 1.65, 95% CI 0.95–2.85).
Throat reflux after sudden hearing loss
One more finding appears in the article on sudden sensorineural hearing loss. A 2026 study followed 195 patients with new tinnitus after sudden hearing loss and found that a higher throat-reflux symptom score, a Reflux Symptom Index of 13 or more, was associated with less favourable tinnitus outcomes, alongside higher anxiety and a history of vestibular disorders. It is one observational finding in one group of patients, and we found no study testing whether treating the reflux changes it.
If you have both
- Treat reflux for its own sake. The NHS self-help advice includes smaller, more frequent meals, not eating within 3 or 4 hours of bed, raising the head of the bed by around 10 to 20cm, and not smoking. It is aimed at heartburn, and we found no evidence that it changes tinnitus.
- Cutting coffee is a heartburn decision. The NHS lists coffee, alcohol, chocolate and fatty or spicy food among heartburn triggers. The tinnitus research does not find caffeine making tinnitus worse, as diet, caffeine and alcohol sets out, so the reason to cut coffee would be the heartburn.
- Do not stop a prescribed PPI because of the tinnitus studies. They do not show that PPIs cause tinnitus. If it worries you, raise it with whoever prescribed it.
- Describe ear and throat symptoms together. Ear fullness, popping or muffled hearing alongside hoarseness or a persistent cough, which NIDDK lists among reflux complications outside the oesophagus, is worth mentioning as one picture, since the Eustachian tube is the best-supported link between the two. How tinnitus is assessed explains what a work-up usually involves.
- Treat the tinnitus directly as well. Tinnitus with an uncertain cause is managed like any other, and the treatment comparison sets out which approaches have evidence.
When to see a clinician
NIDDK advises seeing a doctor if you think you have GERD or if symptoms do not improve with over-the-counter medicines or lifestyle changes. It also advises seeing one for chest pain, loss of appetite, persistent vomiting, trouble or pain swallowing, signs of bleeding such as vomit that looks like coffee grounds or black, tarry stools, and unexplained weight loss.
On the ear side, NIDCD calls sudden hearing loss a medical emergency and notes that people often delay because they put it down to allergies, a sinus infection or earwax. Putting it down to reflux would be the same mistake. The when to seek care checker and is tinnitus a sign of something serious? cover the other tinnitus patterns that need a prompt look.