Objectives This study aims to identify the factors affecting the survival of hemodialysis patients using a two-stage survival/multivariate longitudinal modelling. Methods & Materials The is a retrospective cohort study on 395 patients aged > 60 years who underwent hemodialysis in Hasheminejad Hospital in Tehran, Iran. Data were collected over a 15-year period from 2004 to 2019. Phosphate, calcium, parathormone and creatinine variables were considered as longitudinal variables and mortality as a survival response. The two-stage survival and multivariate longitudinal modeling for long-term hemodialysis patients were fitted and the results were compared according to theirs likelihood function. Data analysis was performed in RStudio v. 3.4.3 and SAS v. 4.9 applications. The significance level was set at 0.05. Results The mean age of patients was 70.41±6.11 years. Of 395 patients, 249(63%) were male. Based on the log-likelihood function, the communication structure of the cumulative effect of longitudinal biomarkers and survival data was selected. Assuming that other variables were constant, the significance of values for mobility impairment, calcium level and parathormone level indicated that these variables were good predictors of the occurrence of death in hemodialysis patients over time. By each unit increase in each variables of mobility impairment, calcium level, and parathormone level, the risk of death increased by 1.391, 0.927, and 0.967, respectively. Conclusion By using a two-stage modelling, the variables of mobility impairment, calcium level and parathormone are found as factors affecting the risk of death in hemodialysis patients. Therefore, therapists should focus on changing health behaviors of these patients in consuming calcium and parathormone.
Read full abstract