Background: Studies have shown that overutilization of labs is common and can cause adverse health outcomes.  No study to date has compared whether variation in lab ordering tendencies exists between different services within the same hospital.      

Methods: This observational study combined a survey of internal medicine residents and hospitalists and a retrospective analysis of lab ordering data.

The survey asked residents and hospitalists about their own practices with regard to routine lab ordering, their preferences for their colleagues’ ordering, and how their use compared to that of their colleagues. Retrospective data was obtained from our data warehouse for all patients admitted to either the teaching service or non-teaching hospitalist service during 2014.  The average number of CBCs per patient per day and chemistry panels per patient per day was calculated for both the teaching service and hospitalist service, and multivariate regression was performed to control for patient characteristics.       

Results: Overall, 44 of 120 (37%) residents and 41 of 53 (77%) hospitalists responded to the survey.  Forty-four (100%) residents reported ordering a daily CBC and chemistry panel rather than one time labs at patient admission compared to 22 (54%) of hospitalists (p<0.001).  Most residents (77%) reported considering the need for subsequent day labs for fewer than half of their patients. On the other hand, most hospitalists (78%) considered the need for subsequent day labs for more than half their patients (p<0.001). 

Overall, 58.8% of teaching service patients and 29.1% of hospitalist patients had a daily CBC order placed on the day of admission (p<0.001).  For basic chemistry panels, 65.6% of teaching service patients and 30.2% of hospitalist patients had a daily basic chemistry panel ordered on the day of admission (p < 0.001).

Teaching service patients had a mean 1.72 ± 0.91 CBCs per day compared to a mean of 1.43 ± 0.79 CBCs per day for patients on the hospitalist service (p<0.001).  After adjustment with multivariate regression, teaching service patients experienced 1.75 CBCs per day and hospitalist service patients experienced 1.45 CBCs per day (p<0.001).  Teaching service patients had a mean 1.96 ± 0.92 chemistry panels per day compared to a mean 1.78 ± 0.94 chemistry panels per day on the hospitalist service (p<0.001).  After multivariate regression, teaching service patients experienced 2.02 chemistry panels per day and hospitalist service patients experienced 1.85 chemistry panels per day (p<0.001).         

Conclusions: Residents’ self-reported and actual use of CBCs and chemistry panels is significantly higher than hospitalists in the same hospital.  Residents may use lab tests more often than hospitalists due to their higher levels of clinical uncertainty or ingrained cultural habits.