Technological advances in clinical thermometers have resulted in a variety of minimally invasive devices that give rapid results but may not have the accuracy necessary for use in acutely ill adults. Inaccurate temperatures can result in missed opportunities for the early identification and treatment of infection and sepsis. Following the methodology outlined by Whittemore and Knafl, the purpose of this project was to conduct an integrative review of the research on the accuracy of clinical thermometers used for acutely ill adults. The evidence was categorized using the Hierarchy of Evidence for Interventional Studies, and the quality of the studies was appraised using the indicators described by Hooper and Andrews. Forty-seven studies met the inclusion criteria; the findings on device accuracy were contradictory. Device accuracy was found in 10 (n = 27) studies on the tympanic (TM), 2 (n = 8) on the chemical dot (CH), 7 (n = 19) on the temporal artery (TAT), and 3 (n = 13) on the axillary (AX) thermometers. Two of 2 studies found the no-touch (NT) device clinically inaccurate. Diagnostic accuracy was found in 3 (n = 8) and 0 (n = 5) studies on the TM and TAT, respectively. Only 22 studies had an acceptable quality grade of A or B, limiting the validity of the evidence. The evidence did not support the use of the NT and TAT thermometers or the AX route for acutely ill adults. The CH device should be use with caution, and abnormal temperatures should be validated with a more reliable device. For thermometers in use, appropriate training and technique are essential for the most accurate results. Closing the knowledge-to-practice gap on clinical thermometers can change the culture of nursing practice, improve early sepsis identification, and increase the quality of patient care.