While in the hospital, patients frequently have considerable volumes of blood removed for diagnostic testing which may contribute to the development of anemia associated with hospitalization. Our primary goal was to investigate the incidence of anemia acquired during hospitalization. As a secondary goal we evaluated the impact of using a smaller volume phlebotomy tube for the basic metabolic panel (BMP) on the incidence of hospital acquired anemia.


A retrospective study was conducted using patients admitted to an academic community teaching hospital in Providence, RI. Inclusion criteria included adult patients aged >18 years who were admitted to internal medicine inpatient service. Those patients who had acute medical conditions that may cause or contribute to the decrease in hemoglobin, such as gastrointestinal bleeding, transfusion dependent anemia, a previous history of anemia, hemolysis, hemorrhagic stroke, retroperitoneal bleed, chronic kidney disease/dialysis, or any hematologic malignancies or were on therapy that may affect hemoglobin levels such as iron or erythropoietin or chemotherapy were excluded. Patients who had central or peripherally central line, who received blood transfusion during hospitalization, who were hospitalized for less than 2 days, or were triaged to surgical, Intensive Care Unit, Cardiac Care Unit, maximum-care service, were also excluded.

The volume of phelobotomy tube used for BMP was reduced by approximately 4mL. Eleven percent of blood draws in this hospital included BMP. To calculate the incidence of hospital acquired anemia, hemoglobin levels on admission were compared with those at discharge. The effect of the smaller volume BMP tubes on hospital acquired anemia was evaluated by comparing the incidence a of hospital acquired anemia before and after the change to a smaller volume was implemented.


A total of 4206 hospitalizations were reviewed and 623 hospitalizations were included in our analysis. Between admission and discharge, 60% of patients dropped their hemoglobin by 1.0 g/dL or more; 56% of the men and 63% of the women. Upon discharge, 43% of the female patients and 46% of male patients who were not anemic at the time of admission had hemoglobin that had dropped into the anemic range. On average, men and women dropped their Hb by 1.3 g/dL. We also found that 61% of patients aged >65 years became anemic during hospitalization compared to 41% of patients aged ≤65 years (p<0.04). Patients with a high BMI had a smaller drop in hemoglobin on average than patients with a low BMI (p=0.01). When stratified by age, the implementation of a decrease in test tube size did not effect the decrease in hemoglobin.


Hospital acquired anemia has a high incidence and is more common in patients aged older than 65 years and has a lower frequency of occurrence in those with a higher BMI. Switching to a smaller volume BMP collecting tube did not significantly reduce the drop of hemoglobin between admission and discharge.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.