Jonathan S. Lee, MD*, University of California, San Francisco, San Francisco, CA;Neelaysh Vukkadala, BS, School of Medicine, University of California San Francisco, San Francisco, CA;Ebrahim Barkoudah, MD, MPH, Brigham and Women's Hospital, Harvard Medical School, Boston, MA;Dr. Shoshana J. Herzig, MD MPH, Beth Israel Deaconess Medical Center, Brookline, MA;Kathryn Levy, MD, University of Michigan Health System, Ann Arbor, MI;Greg Ruhnke, MD, University of Chicago, Chicago, IL;Ronald Weir Jr., BS, Pritzker School of Medicine, Chicago, IL;Dr. Joshua Allen-Dicker, MD MPH, Beth Israel Deaconess, Boston, MA and Andrew D Auerbach, MD, MPH, University of California San Francisco, San Francisco, CA

Meeting: Hospital Medicine 2017, May 1-4, 2017; Las Vegas, Nev.

Abstract number: 303

Categories: Research Abstracts, Value in Hospital Medicine

Background: Avoiding repeated complete blood count (CBC) testing in the face of clinical and laboratory stability is a Choosing Wisely initiative recommendation endorsed by the Society of Hospital Medicine. We investigated the prevalence and clinical utility of this practice in patients hospitalized with community-acquired pneumonia (CAP).


This was a multicenter study of 61 patients with clinical and radiographic evidence of CAP who were discharged alive from 5 academic medical centers from 2010-2016. We excluded patients who were immunosuppressed or who met criteria for healthcare-associated pneumonia. We performed structured chart abstractions collecting patient demographics, comorbidities, daily vital signs and CBC values, and documentation that a daily CBC value was noted and contributed to changes in clinical management (e.g., a new or changed order). We calculated positive likelihood ratios to characterize the utility of CBC testing for producing changes in clinical management. We performed the same analyses stratified by clinical stability based on vital signs criteria.


Thirty patients (49%) were female, mean age was 62 years, mean length of stay was 3.9 days and 38% of patients had a CURB-65 score ≥2. All patients had CBCs drawn on admission. On days following admission, 169 CBCs were obtained. Of subsequent day CBCs, 145 (86%) included at least one abnormal value, but only 36 (21%) were mentioned in daily assessment and plans, and only 12 (7%) were associated with 19 documented management changes (3 escalations and 7 de-escalations related to white blood cell counts, 6 escalations and 3 de-escalations related to hemoglobin) (Table 1). On the day of discharge, 47 patients (77%) had CBCs drawn, 39 (83%) included at least one abnormal value, and 5 (11%) were associated with management changes. The positive likelihood ratio of abnormal CBCs obtained after admission for producing a change in clinical management was 1.07 (95% CI, 0.89-1.29) representing a non-significant 7% increase in the odds of a change in management based on the CBC result. Stated alternately, in the setting where a provider had a pre-test probability of a change in management of 50%, an abnormal CBC would result in a post-test probability of 52%. We found similar positive likelihood ratio estimates amongst clinically stable (1.09; 95% CI, 1.00-1.19) and unstable (1.06; 95% CI, 0.80-1.39) patients. 


CBCs are commonly obtained in patients with CAP on days following admission but may not contribute meaningfully to changes in management after admission. Interventions seeking to reduce CBC use in CAP patients may need to examine nuances of clinician decision-making such as need for clinical reassurance in addition to lack of clinical impact in their data collection and program design.

Table 1. CBCs and Clinical Management Changes in Patients with CAP

Initial evaluation (admission day)

All days of hospitalization (excluding admission day)

Day of discharge

Patients with CBCs ordered (n, %)

61 (100%)

169 (N/A)1

47 (77%)

CBCs with any abnormal value (n, %)

53 (87%)

145 (86%)

39 (83%)

CBCs with any mention in note (n, %)

27 (44%)

36 (21%)

14 (30%)

CBCs with any associated management changes (not restricted to pneumonia) (n, %)

11 (18%)

12 (7%)

5 (11%)

1 Represents number of days a CBC was ordered

To cite this abstract:

Lee, JS; Vukkadala, N; Barkoudah, E; Herzig, SJ; Levy, K; Ruhnke, G; Weir, R Jr.; Allen-Dicker, J; Auerbach, AD . REPETITIVE CBC TESTING IN PATIENTS WITH COMMUNITY-ACQUIRED PNEUMONIA: A MULTICENTER STUDY. Abstract published at Hospital Medicine 2017, May 1-4, 2017; Las Vegas, Nev. Abstract 303. Journal of Hospital Medicine. 2017; 12 (suppl 2). Accessed April 3, 2020.

« Back to Hospital Medicine 2017, May 1-4, 2017; Las Vegas, Nev.