Impact of Untreated Anemia on Hospitalization Rates in Patients with Diabetes and Chronic Kidney Disease.

The purpose of this study was to investigate whether untreated anemia is associated with an increased risk of hospitalization in patients with diabetes and chronic kidney disease (CKD). An analysis of claims and laboratory data from 01/2000 through 12/2005 was conducted. Inclusion criteria were ≥2 hemoglobin (Hb) values, ≥2 claims (within a 90-day period) with a diagnosis of diabetes mellitus preceding ≥1 claim for CKD and ≥2 glomerular filtration rate values of <60 mL/min/1.73 m². Patients were excluded if they had cancer or lupus, had received organ transplantation or chemotherapy, or were treated for anemia. A retrospective open-cohort design was used to classify patients' observation periods into anemia (Hb < 11 g/dL) and non-anemia, per the K/DOQI guidelines. The incidence rates of hospitalization (number of events / person-years of observation) were compared by anemia status using two definitions for hospitalization: (1) all hospitalizations (i.e., number of visits), and (2) the time to the first hospitalization event. 708 patients met the inclusion criteria. The incidence rates of all hospitalizations were 139.4 and 55.2 per 100 person-years for periods of untreated anemia and non-anemia, respectively (rate difference: 84.2, rate ratio: 2.53, 95% CI: 2.2-2.9, p<.001). Based on the time to first event analysis, untreated anemia increased the risk of hospitalization by 2.11 times (95% CI: 1.6-2.7, p<.001), compared to non-anemia. The majority of hospitalizations were related to cardiovascular (CV) disease, with untreated anemia increasing the risk of all CV hospitalizations and first CV hospitalization by 2.45 times (95% CI: 2.0-3.0, p<.001) and 2.21 times (95% CI: 1.6-3.0, p<.001), respectively, compared to non-anemia. Similar results were observed after controlling for covariates. Untreated anemia was significantly associated with an increased risk of hospitalization and CV hospitalization in patients with diabetes and CKD using multiple analytic approaches. [ABSTRACT FROM AUTHOR]