SESSION TITLE: Use of Technology in Training
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Tuesday, October 29, 2013 at 02:45 PM - 04:15 PM
PURPOSE: To explore variation in use of diagnostic testing in Intensive Care Units (ICUs), with emphasis on differences between teaching and non-teaching ICUs.
METHODS: We assessed all adult patients admitted 2006-2010 to five teaching and four non-teaching ICUs, excluding subgroups solely cared for in teaching ICUs (e.g. cardiac surgery patients), inter-ICU transfers, patients with prior ICU admission within 90 days, ICU lengths-of-stay (ICULOS) <12 hours, or missing death dates. Our primary outcome variable (TotalTesting), was the median number of nine laboratory tests, three radiologic tests and electrocardiograms performed in each ICU. Multivariable regression was used to identify factors associated with the median value of TotalTesting, including ICULOS, demographics, admission details, type and severity of acute illness, and specific medical interventions. We estimated the predictive power of variables using the decline in pseudo-R2 (a goodness-of-fit measure for median regression) when omitting those variables from the model. Values are presented as median (IQR).
RESULTS: 10262 patients, with median ICULOS of 2.6 days, were analyzed. TotalTesting was 27 (18-49) in teaching ICUs, and 20 (13-36) in nonteaching units. With adjustment, TotalTesting was 7.1 higher (95% C.I. 6.6-7.7) in teaching ICUs. The most influential variable was ICULOS, accounting for almost half of the variation. ICU teaching status was the second most important factor, greater than the degree of physiologic derangement, and details of medical management. TotalTesting had a small but significant association with age, and increased slightly over the four-year study period.
CONCLUSIONS: Patients in teaching ICUs had slightly but significantly more diagnostic tests done than those in non-teaching ICUs.
CLINICAL IMPLICATIONS: Prior studies have shown that excessive testing increases costs, does not improve outcomes and may cause harm. Interventions to reduce testing should be directed to all caregivers with responsibility for ordering diagnostic tests, in both teaching and nonteaching institutions. Particular attention needs to be paid to potential disparities in health care provided to the elderly, as well as the trend of increasing testing over time.
DISCLOSURE: The following authors have nothing to disclose: Jessica Spence, Dean Bell, Allan Garland
No Product/Research Disclosure Information