The ketogenic diet has transitioned from a highly specialized therapeutic intervention to a widely promoted dietary strategy for weight loss, metabolic health, inflammation, and, in some circles, cancer prevention or treatment. Its popularity has outpaced the strength of the evidence supporting its long-term use in the general population.
As a registered dietitian, I would not recommend a ketogenic diet as a default approach to healthy eating or long-term weight management. While there are specific clinical circumstances in which ketogenic therapy is evidence-based, the available evidence does not establish that prolonged nutritional ketosis provides superior health outcomes for the general population. More importantly, emerging research raises legitimate concerns about the potential metabolic consequences of sustained, very-low-carbohydrate, high-fat intake.
The distinction between short-term efficacy and long-term safety is particularly important.
Ketogenic therapy has legitimate medical applications—but that does not make it a universally appropriate diet
The ketogenic diet has a well-established therapeutic role in the management of certain seizure disorders, particularly drug-resistant epilepsy. In this context, it is a structured medical therapy implemented with clinical supervision and careful attention to nutritional adequacy.
That application should not be confused with the increasingly common use of ketogenic diets for general wellness.
Very-low-carbohydrate diets can produce short-term weight loss and improvements in certain metabolic markers. These effects may be meaningful for some individuals. However, weight loss alone does not establish that a dietary pattern is optimal for long-term health.
In many cases, the initial weight reduction associated with ketogenic diets reflects reduced overall energy intake, changes in appetite, alterations in glycogen and water stores, and the difficulty of maintaining a highly restrictive dietary pattern. There is insufficient evidence demonstrating that nutritional ketosis itself is necessary to achieve or maintain long-term weight loss.
This distinction is clinically relevant. A dietary intervention should not be considered superior simply because it produces a rapid initial change in body weight.
The long-term evidence is the central concern
One of the most significant limitations surrounding ketogenic diets is the relative lack of high-quality, long-duration human research.
Much of the existing literature focuses on relatively short intervention periods. A 2020 review identified several concerns associated with ketogenic diets, including potential nutritional inadequacy, questions surrounding long-term adherence, and increases in LDL cholesterol in some individuals.
From a clinical perspective, this matters because cardiovascular and metabolic consequences develop over years, not weeks.
A diet can produce favorable changes in body weight or blood glucose over several months while still raising questions about its effects over substantially longer periods. Consequently, it is premature to characterize ketogenic eating as a universally health-promoting dietary pattern when the evidence regarding prolonged adherence remains incomplete.
There is also substantial variability in the quality of foods consumed on a ketogenic diet.
A well-formulated ketogenic diet can include fish, olive oil, avocado, nuts, seeds, and non-starchy vegetables. However, ketogenic eating can also result in high consumption of butter, processed meats, high-fat dairy products, and other sources of saturated fat.
That distinction is not trivial.
A dietary pattern that substantially increases saturated fat intake may adversely affect LDL cholesterol in susceptible individuals. Therefore, describing a diet simply as “ketogenic” provides insufficient information about its potential cardiovascular consequences.
Emerging research raises additional questions about prolonged ketogenic intake
Recent preclinical research provides further reason to avoid prematurely declaring ketogenic diets safe for indefinite use.
A 2025 study published in Science Advances examined mice maintained on a ketogenic diet for up to a year. Although the ketogenic diet prevented weight gain compared with a high-fat Western diet, the researchers observed several adverse metabolic changes, including hyperlipidemia, liver dysfunction, and impaired glucose regulation associated with impaired insulin secretion.
These findings should not be interpreted as proof that humans following ketogenic diets will develop the same conditions. The study was conducted in mice, and animal findings cannot be directly extrapolated to clinical outcomes in humans.
However, the findings are important because they challenge the assumption that preventing weight gain necessarily equates to improved metabolic health.
This is precisely the type of research that should prompt additional long-term human investigation rather than reassurance that the diet is inherently safe.
Cancer claims require particular caution
The promotion of ketogenic diets as a means of “starving” cancer is an area in which the distinction between biological plausibility and clinical evidence is especially important.
Cancer metabolism is complex, and researchers continue to investigate the relationships among glucose availability, insulin signaling, fatty acids, ketone bodies, and tumor biology. However, these mechanistic observations do not establish that eliminating carbohydrates from the diet will selectively starve cancer cells or improve cancer outcomes.
There is currently insufficient clinical evidence to recommend a ketogenic diet as a general cancer treatment.
This distinction becomes even more important for individuals actively undergoing cancer treatment. Maintaining adequate energy, protein, hydration, and micronutrient intake is often a primary nutritional objective. A highly restrictive diet may make these goals more difficult to achieve, particularly for patients experiencing nausea, diarrhea, early satiety, appetite loss, taste changes, or unintentional weight loss.
In these circumstances, imposing additional dietary restrictions may be counterproductive.
New animal research illustrates why oversimplification is dangerous
A 2026 preclinical study examining ketogenic diets and gastrointestinal tumor biology provides another example of why the subject requires careful interpretation.
Researchers found that ketogenic feeding had different effects in different portions of the gastrointestinal tract in genetically predisposed mice. The diet appeared to suppress tumor development in the colon while promoting tumor development in the small intestine. The researchers attributed the latter effect to lipid metabolism rather than ketone bodies themselves.
This research does not demonstrate that ketogenic diets cause cancer in humans.
Its significance is instead that biological responses to diet are considerably more complicated than the popular narrative that ketosis is uniformly beneficial. A dietary intervention can have different effects depending on tissue, metabolic environment, genetics, disease state, and the composition of the diet itself.
For clinicians, that complexity should reinforce the need for caution rather than encourage broad dietary prescriptions based on preliminary findings.
Nutritional adequacy is another concern
Restricting carbohydrates to ketogenic levels necessarily eliminates or substantially limits numerous foods that contribute important nutrients and dietary components.
Depending on how the diet is constructed, this can mean reduced intake of fruits, whole grains, legumes, and certain vegetables. These foods provide dietary fiber, vitamins, minerals, phytochemicals, and other bioactive compounds associated with health.
It is possible to formulate a nutritionally adequate ketogenic diet, but doing so requires intentional planning.
That is an important distinction between saying that a diet can be nutritionally adequate and saying that it is the most appropriate dietary pattern for the population at large.
From a dietetic perspective, I would be particularly hesitant to recommend a highly restrictive dietary pattern when comparable health goals can often be pursued through less restrictive approaches that allow greater dietary variety.
The issue is not carbohydrates themselves—it is dietary quality
One of the most persistent problems in the ketogenic conversation is the tendency to treat carbohydrate as a single nutritional category.
Carbohydrates from sugar-sweetened beverages, refined sweets, legumes, berries, whole grains, and vegetables do not have equivalent nutritional profiles.
Reducing excessive intake of refined carbohydrates and added sugars can be a reasonable nutritional goal without eliminating carbohydrate-containing foods as a category.
For most individuals, I would prioritize an eating pattern emphasizing vegetables and fruits, legumes, whole grains, nuts and seeds, fish and other lean protein sources, and predominantly unsaturated fats, while limiting highly processed foods, excessive added sugars, and processed meats.
This approach provides substantially greater dietary flexibility while supporting fiber intake and overall nutrient adequacy.
A more appropriate clinical standard
The question should not be whether a ketogenic diet can produce weight loss. It clearly can.
The more important question is whether we have sufficient evidence to recommend prolonged nutritional ketosis as a preferred dietary strategy for generally healthy individuals.
At present, I would argue that we do not.
The evidence supporting ketogenic diets is strongest in specific therapeutic contexts, such as certain seizure disorders. Evidence for weight loss and some metabolic outcomes is more established in the short term but does not demonstrate clear long-term superiority over other dietary approaches. Evidence regarding prolonged exposure continues to develop, while emerging preclinical research raises questions about lipid metabolism, hepatic function, glucose regulation, and tissue-specific effects.
That evidence profile does not justify positioning the ketogenic diet as a universally healthy or preventive dietary pattern.
The bottom line
As clinicians, we should be careful not to confuse rapid results with long-term health.
A ketogenic diet may be appropriate for selected patients under appropriate medical supervision. However, outside established therapeutic indications, I would not recommend routine long-term ketogenic dieting as a first-line approach to healthy eating or weight management.
The potential concerns—including elevated LDL cholesterol in some individuals, challenges with nutritional adequacy, limited long-term safety data, and emerging evidence of adverse metabolic effects in animal models—warrant a more cautious approach.
Most importantly, patients should not interpret emerging research as evidence that they need to eliminate carbohydrates to prevent disease or “starve” cancer. Nor should they assume that a reduction in body weight automatically means that every underlying metabolic process has improved.
A more defensible clinical approach is to emphasize dietary quality, adequacy, variety, sustainability, and the individual’s medical circumstances. Until stronger long-term human evidence demonstrates that sustained ketogenic eating offers benefits that outweigh its potential risks, there is insufficient justification for recommending it broadly as a superior dietary pattern.
Author Bio
