The single most common worry about keto is not whether it works but whether it is safe to keep doing. It is a fair question, and it deserves a better answer than either the cheerleading you get on keto forums or the doom you get in the occasional newspaper headline. The honest picture sits in between, and it depends a great deal on how you do it. Here is what the evidence supports, what it does not, and how to stay on the right side of the line.
The trial picture is reassuring, with one weak spot
When researchers run controlled trials of ketogenic and low-carbohydrate diets, lasting from months up to around two years, the safety findings are generally reassuring. People tend to improve on the markers that matter most for everyday health: weight, blood sugar, triglycerides and often blood pressure through weight loss. Serious harms are not a common feature of these trials. The genuine weak spot is not safety but adherence: studies lasting a year or more consistently show that many people drift off the diet, which tells you more about sustainability than danger. So for the medium term, the controlled evidence says keto is safe for most healthy people. What we lack is large, high-quality trials running for many years, so beyond the two-year mark we are reasoning from shorter studies, observational data and clinical experience rather than hard long-term trial proof. That uncertainty is worth being honest about.
The scary headline, and what it actually shows
The most-cited worry comes from a large observational study, Seidelmann and colleagues in The Lancet Public Health in 2018. Pooling data on hundreds of thousands of people, it found a U-shaped relationship between carbohydrate intake and death: both very high-carbohydrate and very low-carbohydrate eating were associated with higher mortality than moderate intake, with the lowest risk around 50 to 55 per cent of calories from carbohydrate. Taken at face value, that sounds like a strike against low-carb.
But two things matter enormously here. First, this is observational data, which can show associations but cannot prove cause; people who eat very low-carb in these population studies differ in many ways from those who do not. Second, and more revealing, the study found that the increased risk appeared when carbohydrate was replaced by animal-derived fat and protein, while replacing it with plant-derived fat and protein was associated with no harm, and if anything benefit. In other words, the data point less to “low-carb is dangerous” and more to “low-carb built on lots of meat and few plants is the risky version”. A low-carb diet rich in plants, olive oil, nuts and fish looks very different in this data from one built on processed meat. That distinction runs through almost all the long-term safety question.
The genuine concerns to manage
Setting the headlines aside, there are real issues that deserve attention rather than dismissal.
The clearest is cholesterol. In a meaningful subset of people, particularly lean, healthy individuals, a ketogenic diet raises LDL cholesterol and ApoB, sometimes substantially. This is important enough to have its own detailed treatment, and the honest picture is covered in the companion piece on keto and cholesterol. It is the long-term concern I would watch most closely, because it is common and because it is measurable.
Next is the kidneys. A ketogenic diet increases the acid load the body has to handle and is associated with a higher risk of kidney stones, a risk best documented in the strict therapeutic version used for epilepsy. For people who already have chronic kidney disease, the cautious advice is generally to avoid keto as a first choice, since gentler dietary patterns carry less theoretical risk. Healthy kidneys handle it far better, but it is a reason to stay hydrated and to be careful if your kidney function is already compromised.
Then there is nutritional adequacy. Done as a meat-and-cheese diet, keto can be low in fibre and some micronutrients, which is both a comfort problem and a long-term health one. Done as a plant-rich low-carb diet, heavy on non-starchy vegetables, it largely solves this, which is the same lesson as the mortality data. Bone health is sometimes raised as a worry, though it is more theoretical than demonstrated, and reassuringly a 2024 ten-year case report found no adverse effect on bone density, kidney or thyroid function.
The reassuring counterweight
It is worth remembering that the ketogenic diet is not some untested fad in safety terms. Its strict medical form has been used to treat epilepsy for a century, under supervision, including in children over long periods, which has generated a great deal of real-world safety experience. Recent long-duration case reports have found people maintaining ketogenic eating for many years without the feared damage to bone, kidney or thyroid. None of this is the same as a decades-long randomised trial, but it does push against the idea that long-term keto is inherently harmful.
How to do long-term keto safely
The practical upshot is encouraging, because the same choices that make keto safer also make it more pleasant and sustainable. Build it around plants and whole foods rather than processed meat and endless cheese, leaning on non-starchy vegetables, olive oil, nuts, fish and eggs, which is exactly the pattern the mortality data treats kindly. Get your bloods checked before you start and a few months in, with particular attention to your cholesterol and ApoB, and act on the results rather than assuming they must be fine. Stay well hydrated and be cautious if you have existing kidney problems. Make sure you are getting enough fibre and nutrients from those vegetables. And ask yourself honestly whether you need strict, deep ketosis indefinitely, or whether a sustainable lower-carb pattern that you can actually keep up serves you better, since the diet only helps while you are doing it.
The bottom line
For most healthy people, the evidence says a ketogenic diet is safe over the short and medium term, with the main practical limit being how long people stick to it. The frightening mortality headline is observational and, read closely, points the finger at meat-heavy low-carb rather than low-carb itself. The real long-term concerns worth managing are cholesterol in susceptible people, kidney stones and acid load, and nutritional adequacy, all of which are addressed by doing keto in a plant-rich, whole-food way and monitoring your bloods. Long-term keto is not the danger it is sometimes painted, nor the consequence-free miracle it is sometimes sold as; done thoughtfully and checked periodically, it is a reasonable long-term choice for many.
This is general information about the ketogenic diet, not medical advice. Long-term dietary choices, especially with any existing condition, should be made with a doctor, and your bloods should be monitored. If you are pregnant, have kidney disease, or take medication, seek professional advice before starting or continuing keto.
Source: Seidelmann SB, et al. Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis. The Lancet Public Health. 2018;3(9):e419-e428. Read it here.