Does Ozempic Make You Pee More? GLP-1 Drugs, Urination and Thirst

No, on the evidence a GLP‑1 does not make you pee more. In the first hours, these drugs push extra sodium into the urine. After a few weeks of a weekly shot, the two trials that measured a full day of urine found less of it, not more, because people drank less. Frequent urination and thirst are not listed as side effects on the Ozempic, Wegovy, Mounjaro or Zepbound label. If you are running to the bathroom and always thirsty on one of them, the drug is the least likely explanation. High blood sugar, an SGLT2 pill, a urinary infection and plain extra drinking come first. The fluid problem these drugs actually carry runs the other way: too little in, not too much out.

What the labels list: nothing about urination or thirst

We searched the full text of all four current GLP‑1 labels for polyuria, pollakiuria (the medical word for frequent urination), nocturia, urination and thirst. None of the words appears, in the semaglutide labels^[1][2] or in the tirzepatide ones.^[3][4] The adverse-reaction tables are dominated by the stomach. The nearest entries are a kidney warning about dehydration, a urinary tract infection line in one pediatric trial, and dry mouth on Zepbound. For contrast, the Jardiance label, an SGLT2 inhibitor, has its own line for increased urination and its own figure for thirst.^[5]

Absence from a label is not proof of absence. It means that in trials of thousands of people, increased urination and thirst did not show up often enough on drug versus placebo to make the cut. Jardiance’s effect, at about three people in a hundred, did.^[5]

The sodium effect is real, small and mostly early

GLP‑1 does act on the kidney. The human studies place the effect in the proximal tubule, the first stretch of each filtering unit, where less sodium gets pulled back into the blood. In Basel in 2004, a three-hour intravenous GLP‑1 infusion after a salt load roughly doubled sodium excretion in healthy men, from 74 to 143 mmol over the three hours, and raised it from 59 to 96 mmol in men with obesity.^[6] Three weeks of liraglutide, the daily drug in Saxenda and Victoza, raised 24-hour and overnight sodium in the urine of people with type 2 diabetes and high blood pressure.^[8]

Sodium out is not the same as more trips to the bathroom. When Amsterdam researchers gave exenatide intravenously to people with type 2 diabetes, sodium excretion went up while urine flow and free water clearance went down.^[7] The urine became more concentrated, not more plentiful. And the effect fades. Twelve weeks of liraglutide left no lasting change in tubular function,^[9] and after three weeks of weekly dulaglutide (Trulicity), sodium excretion was unchanged or lower, depending on how it was measured.^[10]

After a few weeks: less urine, because less drinking

Two randomized crossover trials from the same Basel group measured what most readers actually want to know: how much people drink and how much they pass. Both used dulaglutide 1.5 mg once a week for three weeks against a saline placebo.

In 20 healthy volunteers, free to drink water as they liked during an eight-hour visit, 16 drank less on dulaglutide. Median intake was 1,300 mL against 1,600 mL on placebo, a difference that fell just short of significance. Their 24-hour urine was lower on the drug, 1,250 mL against 1,680 mL, and that difference was significant. Thirst ratings did not differ.^[11] In 34 people with primary polydipsia, a condition of compulsive water drinking, dulaglutide cut intake by 490 mL, about 17%, and cut 24-hour urine by 943 mL. Here, thirst ratings fell too.^[12]

The likely route is the brain’s water hormone, vasopressin. In the pooled participants of the two trials, dulaglutide lowered copeptin, a stable marker of vasopressin, by about 12%.^[13] A 2026 before-and-after study found liraglutide lowered vasopressin itself in healthy people.^[14] In principle, less vasopressin means more dilute urine, and less drinking means less of it. That the second effect wins is what the urine collections show.

Thirst: the drug is more likely to turn it down

If a GLP‑1 changes thirst, the evidence points to less of it, not more. That matters because the one fluid warning on every label is about running dry. The Ozempic label sets it out:

There have been postmarketing reports of acute kidney injury, in some cases requiring hemodialysis, in patients treated with semaglutide. The majority of the reported events occurred in patients who experienced gastrointestinal reactions leading to dehydration such as nausea, vomiting, or diarrhea.

— Ozempic prescribing information, section 5.6

Mounjaro and Zepbound carry the same warning for tirzepatide.^[3][4] In Wegovy’s weight-loss trials, kidney injury occurred in 7 adults on the drug against 4 on placebo, more often during dose increases and in people with a history of reduced kidney function.^[2] On Zepbound, low blood pressure was reported in 1.6% against 0.1% on placebo, sometimes alongside dehydration.^[4] The patient leaflets put the practical point plainly: “It is important for you to drink fluids to help reduce your chance of dehydration.”^[1]

Put that next to the Basel trials and a pattern appears, though it is our inference rather than a tested finding: a drug that lowers the urge to drink can let fluid run low before thirst sounds the alarm. On a normal day that is minor. On a day of vomiting it is the setup the kidney warning describes. The longer-run kidney picture, which is favorable, is in GLP-1 drugs and your kidneys.

A dry mouth is a separate complaint from thirst, though people often describe them together. Zepbound’s label reports dry mouth or dry throat in 1% of users against 0.1% on placebo.^[4] What a dry mouth means for teeth, and how thin the evidence is for semaglutide, is covered in GLP-1 drugs, dry mouth and dental care.

What the adverse-event reports show

The FDA Adverse Event Reporting System collects reports from patients, clinicians and manufacturers. Anyone can file one, nothing is confirmed, and a report says nothing about how often an event happens or whether the drug caused it. With that caveat, the counts are a useful sense of proportion.^[15]

Frequent urination appears in well under 1% of reports for either drug. Dehydration appears about seven to ten times as often. That ratio fits the trial data: the recurring fluid story with these drugs is loss and low intake, not excess urine. A 2025 study of the same database found acute kidney injury in 1.07% of semaglutide reports and 0.47% of tirzepatide reports, a disproportionality signal for semaglutide.^[16] Its authors note that causality cannot be confirmed. A reporting signal is a reason to look, not a measured risk. We found no published disproportionality study of urination or thirst on any GLP‑1.

If you are peeing more and thirsty, the usual suspects

High blood sugar

This one comes first. The American Diabetes Association lists polyuria, polydipsia and unexplained weight loss as the classic symptoms of hyperglycemia; with them, a single random glucose of 200 mg/dL or more is enough to diagnose diabetes.^[17] Glucose spilling into the urine pulls water with it. On a GLP‑1 the third sign, weight loss, is expected anyway, which makes the first two easier to dismiss. That is our reading, not a guideline statement.

Three groups should take the pair of symptoms seriously. People with type 2 diabetes whose sugars are not yet controlled. People on insulin who cut their dose on their own after starting the drug: a 2026 case report describes a man in his 50s who halved his basal insulin for fear of lows after starting tirzepatide, stopped keeping food and fluids down, and arrived severely dehydrated in diabetic ketoacidosis.^[19] And people with type 1 diabetes taking a GLP‑1 off-label, covered in GLP-1 drugs in type 1 diabetes.

An SGLT2 inhibitor

Jardiance, Farxiga and Invokana lower blood sugar by sending it out in the urine, and water follows. Jardiance’s label reports increased urination in 3.4% of people on 10 mg against 1.0% on placebo, and thirst in 1.7% against none.^[5] Many people take one alongside a GLP‑1. If you do, the extra urine is almost certainly the SGLT2, and the overlap that matters is fluid and ketones on sick days; see an SGLT2 and a GLP-1 together.

A urinary tract infection

Frequency with burning, urgency or cloudy urine points to infection, not the drug. Wegovy’s teen trial listed urinary tract infection in 4% of those on the drug against 2% on placebo.^[2] Across the wider evidence, a 2026 network meta-analysis of 105 randomized trials and more than 219,000 participants found no link between GLP‑1 drugs and infections, urinary tract infection included.^[18] Kidney stones, which can also bring frequency and pain, are covered in GLP-1 drugs and kidney stones.

Drinking more on purpose

Many people are told to drink more when they start a GLP‑1, and do. More water in means more urine out. Coffee counts toward fluids, as set out in coffee and caffeine on a GLP-1; alcohol is its own question, covered in alcohol and GLP-1 drugs.

Lithium

Lithium depends on kidney function and fluid balance, and its safe range is narrow. A 2026 review counted six separate published reports of lithium levels rising, or toxicity, after a GLP‑1 was started, switched or increased, and noted that the European Medicines Agency’s safety committee opened a signal on the interaction in June 2026.^[20] The authors argue the GLP‑1’s own sodium effect pushes the other way and does not last; reduced intake and fluid loss explain the acute cases better. Anyone on lithium who notices a change in thirst, urine or stomach symptoms after a GLP‑1 change has a reason to ask for a level.

When to call

• Thirst and frequent urination together, with high readings or none taken. Ask for a glucose check, sooner if you have diabetes or take insulin.
• Vomiting or diarrhea with very little urine. Dark urine, dizziness when you stand, or a day without keeping fluids down is the pattern behind the kidney warning, and more so with a diuretic or an SGLT2 inhibitor; dizziness and blood pressure pills on a GLP-1 covers that overlap.
• Burning, fever, or pain in the side or back. That is infection or a stone until shown otherwise.
• Nausea, belly pain and unusual tiredness on an SGLT2 inhibitor or insulin. Ketoacidosis can present this way. On an SGLT2, the Jardiance label warns glucose may read lower than expected, under 250 mg/dL.^[5]
• Lithium. Any new thirst, urine or stomach change after starting or raising a GLP‑1.