Objective: To observe the variability of hemoglobin (HB) level in patients with renal anemia, and to analyze its relationship with effect of repeated blood transfusion therapeutic in patients., Methods: A retrospective cohort study and propensity score matching method were used, 60 patients with renal anemia who had effective treatment with repeated blood transfusion in Changzhou No.2 People's Hospital from May 2018 to May 2021 were retrospectively analyzed and set as the effective group; 153 patients with renal anemia who had ineffective treatment with repeated blood transfusion in the hospital in the same period were collected and set as the ineffective group, the propensity score matching method was used, the patients who were effective and ineffective in repeated blood transfusion were matched 1∶1 for analysis; the medical records and laboratory indexes of the two groups were checked; the Hb level of patients within 6 months (1/month) were recorded, the residual standard deviation (Res-SD) of Hb of patients was calculated according to the Hb level and evaluated the variability of Hb level; the relationship between HB variability level and therapeutic effect of repeated blood transfusion in patients with renal anemia was analyzed., Results: After propensity score matching, there was no statistical significant difference between the two groups in terms of baseline data such as age, sex, dialysis age and BMI ( P >0.05). The levels of serum albumin and transferrin of patients in the ineffective group were significantly lower than those of patients in the effective group ( P <0.05); at 1 and 2 months of the observation period, there was no statistical significant difference in Hb levels of patients in both groups ( P >0.05); the Hb level of patients in the ineffective group was significantly lower than that of patients in the effective group at 3, 5 and 6 months, and significantly higher than that of patients in the effective group at 4 months ( P <0.05); the Res-SD of male patients and female patients in the ineffective group were respectively significantly higher than that of male patients and female patients in the effective group ( P <0.05). Logistic regression analysis results showed that high variability of Hb level (Res-SD) was a risk factor for the ineffective treatment of repeated blood transfusion in patients with renal anemia ( OR >1, P <0.05); the decision curve results showed that, when the high-risk threshold was 0.0-1.0, Res-SD predicted the net benefit rates of male and female patients with renal anemia were greater than 0, which was clinically significant, the smaller the high-risk threshold in the above range, the greater the net benefit rate., Conclusion: The therapeutic effect of repeated blood transfusion in patients with renal anemia may be related to the variability of Hb level.