Development of a Mortality Prediction Model for Patients Receiving Maintenance Hemodialysis
Development of a Mortality Prediction Model for Patients Receiving Maintenance Hemodialysis
This prospective, single-center observational cohort study evaluates cardiovascular mortality risk in adults receiving maintenance hemodialysis. During the first year after enrollment, four scheduled assessments are used to characterize demographic factors, dialysis treatment, medication use, routine laboratory measures, and the proportion of serum phosphate measurements within the protocol-defined target range. Participants are subsequently followed for cardiovascular mortality. The study evaluates factors associated with serum phosphate target achievement and develops a cardiovascular mortality prediction model in the overall cohort, with a prespecified subgroup analysis among participants who undergo parathyroidectomy. No treatment or procedure is assigned by the study investigators.
Inclusion Criteria:
Exclusion Criteria:
This is a prospective observational cohort study conducted at the Hemodialysis Center of Huashan Hospital, Fudan University. Adults aged 18 to 85 years who have received maintenance hemodialysis for at least 3 months and meet the eligibility criteria are enrolled after providing written informed consent.
During the first year after enrollment, four scheduled assessments are performed. Data collected include demographic and clinical characteristics, dialysis treatment parameters, medication use, complete blood count, routine biochemical measurements, immunologic measurements, and other clinically relevant indicators. The principal exposure is the one-year serum phosphate target achievement rate, defined as the number of evaluable serum phosphate measurements within the protocol-defined target range divided by the total number of evaluable scheduled measurements during the one-year assessment period.
Following the one-year exposure-assessment period, participants are followed for cardiovascular mortality for up to 5 years. The primary objective is to evaluate the association between the one-year serum phosphate target achievement rate and cardiovascular mortality and to develop a cardiovascular mortality prediction model. Secondary objectives include identifying factors associated with serum phosphate target achievement and developing a mortality prediction model in the prespecified subgroup of participants who undergo parathyroidectomy as part of routine clinical care.
Descriptive analyses and appropriate parametric or nonparametric tests are used to summarize and compare participant characteristics. Multivariable regression is used to evaluate factors associated with serum phosphate target achievement. Cox proportional hazards regression and prespecified statistical or machine-learning methods are used to evaluate cardiovascular mortality risk and develop prognostic models. All statistical tests are two-sided with an alpha level of 0.05. No study-specific treatment is assigned, and clinical management is determined independently by the treating physicians.