Background: Serum phosphorus (SP) level is closely associated with overall mortality and cardiovascular events, while the role of SP controlled duration is not fully recognized. Here, we conducted a retrospective cohort study in our department to identify the relationship of SP controlled duration with clinical outcomes in patients undergoing peritoneal dialysis (PD). Methods: PD patients in our center from January 1, 2009, to June 30, 2019, were followed up at 2-month (the first year) or 5-month (the next follow-up period) intervals, and until death, until PD withdrawal, or until June 30, 2019. Data at each follow-up point were collected from their medical records. SP levels, changed degree of SP over baseline, and SP controlled duration were analyzed with overall mortality, PD withdrawal (including death, transferred to hemodialysis, and received renal transplantation), and combined endpoint (including death, acute heart failure, cardiovascular event, and stroke). Results: A total of 530 patients entered the analysis. Of them, 456 (86.0%) had hyperphosphatemia before dialysis, and the SP levels decreased soon after dialysis. The degree of SP change over baseline was the maximum at the 3rd month after dialysis (−31.0%), and lower degree was associated with higher overall mortality (hazard ratio [HR], 1.012; 95% CI, 1.004-1.020; p = 0.003). The median SP controlled duration was 13 (5-28) months, and longer duration was significantly associated with lower overall mortality (HR, 0.968; 95% CI, 0.956-0.981; p < 0.001). After categorization, duration more than 12 months greatly improved overall mortality with a HR of 0.197 (0.082-0.458; p < 0.001 vs. SP never controlled group) and 0.329 (0.150-0.724; p = 0.006 vs. duration <12 months group). Longer SP controlled duration also improved PD withdrawal and combined endpoint. Conclusions: In summary, both degree and duration of SP control were tightly associated with overall mortality. We should control SP levels as early, as possible, and as long as we could.