The world population is aging, and the prevalence of chronic kidney disease (CKD) is increasing. Whether this increase is also due to the methods currently being used to assess kidney function in the elderly is still a matter of discussion. We aimed to describe the actual referral pattern of CKD patients in a large nephrology unit and test whether the use of different formulae to estimate kidney function could affect the staging and the need for specialist care in the older subset of our population. In 2019, 1992 patients were referred to our center. Almost 28% of the patients were aged ≥80 and about 6% were ≥90 years old. Among the causes of kidney disease, glomerulonephritis displayed a higher prevalence in younger patients whereas hypertensive or diabetic kidney disease were more prevalent in older patients. The prevalence of referred patients in advanced CKD stages increased with age; estimated glomerular filtration rate (eGFR) decreased with age regardless of which equation was used (chronic kidney disease epidemiology collaboration (CKD-EPI), Lund–Malmö Revised (LMR), modification of diet in renal disease (MDRD), Full Age Spectrum (FAS), or Berlin Initiative Study 1 (BIS)). With CKD-EPI as a reference, MDRD and FAS underestimated the CKD stage while LMR overestimated it. The BIS showed the highest heterogeneity. Considering an eGFR threshold limit of 45 mL/min for defining “significant” CKD in patients over 65 years of age, the variability in CKD staging was 10% no matter which equation was used. Our study quantified the weight of “old” and “old-old” patients on follow-up in a large nephrology outpatient unit and suggested that with the current referral pattern, the type of formula used does not affect the need for CKD care within the context of a relatively late referral, particularly in elderly patient