1. To assess the risk of end-stage renal disease (ESRD) associated with the regular use of three classes of non-narcotic analgesics, we performed a case-control study of 340 patients with ESRD on a haemodialysis maintenance program and 673 hospital controls. 2. The overall odds ratio estimate for non-narcotic analgesics taken at least every other day for 30 days or longer before the first symptom of renal disease was 2.89 (95% CI, 1.78 to 4.68). 3. The risk increased in relation to the use duration. 4. The previous regular consumption of combinations containing phenacetin was strongly associated with ESRD (odds ratio, 19.05; 95% CI, 2.31 to 157.4). The odds ratio for previous regular consumption of salicylates was 2.54 (95% CI, 1.24 to 5.20) and for pyrazolones 2.16 (95% CI, 0.87 to 5.32). 5. An analysis for possible confounding by a history of repeated headaches, arthritis, kidney stones, hypertension, and diabetes did not alter the results. 6. The odds ratio estimates for different pathological subgroups of ESRD patients in relation to previous use of any non-narcotic analgesic were glomerulonephritis. 10.57 (95% CI, 1.25 to 89.0), interstitial nephritis, 3.33 (95% CI, 1.21 to 9.17), cystic kidney disease, 0.71 (95% CI, 0.25 to 1.97), and unknown, 5.15 (95% CI, 2.29-11.57). 7. The results of this study suggest that the regular consumption of analgesics should be routinely considered as a risk factor for any non-congenital cause of chronic renal failure. They also suggest that the risk of ESRD associated with the regular consumption of phenacetin is much higher than the risk associated with other non-narcotic analgesics.