There is a growing consensus that fasting-induced ketosis has beneficial effects on human physiology. Despite these compelling benefits, fasting-induced ketosis raises concerns in some clinicians because it is often inappropriately compared with the pathologic uncontrolled ketone production in diabetic ketoacidosis. The determinants of the inter-individual differences in the intensity of ketosis during long-term fasting is unknown. We monitored daily variations in fasting ketonemia, as well as ketonuria, which is less invasive, in a large cohort of 1610 subjects, fasting between 4 and 21 days with the Buchinger Wilhelmi program, minimally supplemented with ~75-250 kcal (daily fruit juice, vegetable soup, and honey). Ketonuria was detected in more than 95% of fasting subjects from day 4 onwards. Subjects consuming only soups, without fruit juice or honey, exhibited reduced caloric intake (72 kcal instead of 236 kcal) and carbohydrate intake (15.6 g instead of 56.5 g), leading to more intense ketonuria. Participants with high ketonuria were, in the majority, males, young, had a higher body weight, and had lower HDL-C and urea values. They had a larger decrease in blood glucose, glycated haemoglobin levels, body weight, and waist circumference. Furthermore, in the high-ketonuria group, a larger increase in blood uric acid concentration was observed. Our study showed that long-term fasting triggered ketosis, never reaching pathological levels, and that ketosis is influenced by age, gender, health, and the level of physical activity. Furthermore, it is modulated but not suppressed by minimal carbohydrate intake. Our study paves the way for better understanding how supplementation can modulate the therapeutic effects and tolerability of long-term fasting.