Urine composition in terms of calcium oxalate (CaOx) supersaturation was studied in 802 patients with calcium stone disease before any intervention and during follow-up. Supersaturation was expressed as the AP(CaOx) index, a simplified estimate of the ion activity product of CaOx, and a similar index calculated for a 24-hour urine volume of 1.5 liters the AP(CaOx) index(s). The AP(CaOx) index was significantly reduced in men with and without medical treatment who remained stone-free during follow-up (p less than 0.001), but not in men who continued to form stones. For the AP(CaOx) index(s), a significant reduction was observed only in patients on medical treatment without new stone formation (p less than 0.01). In women, significantly lower AP(CaOx) index values were recorded in recurrent as well as non-recurrent stone formers on medical treatment, whereas in the group without medical treatment and without recurrences the difference did not reach a statistically significant level. This was similar to the effect on the AP(CaOx) index(s) in non-recurrent women with medical treatment. The small number of women with recurrences might have influenced the result. Significantly reduced levels of the AP(CaOx) index were recorded for patients given thiazide, thiazide + magnesium, magnesium, and alkaline citrate. The AP(CaOx) index(s) was reduced in patients given thiazide + magnesium, magnesium, and alkaline citrate. Comparison between the effects on urine composition and clinical response showed that the reduced CaOx supersaturation observed with thiazide, thiazide + magnesium, and alkaline citrate, corresponded to a low rate of stone formation during follow-up. The inefficiency of allopurinol and orthophosphate in affecting urine supersaturation was reflected in a higher recurrence rate.(ABSTRACT TRUNCATED AT 250 WORDS)