A systematic approach for studying the signs and symptoms of fever in adult patients: the fever assessment tool (FAST)

Abstract

Background: Although body temperature is one of four key vital signs routinely monitored and treated in clinical practice, relatively little is known about the symptoms associated with febrile states. The purpose of this study was to assess the validity, reliability and feasibility of the Fever Assessment Tool (FAST) in an acute care research setting.

Methods: Qualitative: To assess content validity and finalize the FAST instrument, 12 adults from an inpatient medical-surgical unit at the National Institutes of Health (NIH) Clinical Center participated in cognitive interviews within approximately 12 h of a febrile state (tympanic temperature ≥ 38° Celsius). Quantitative: To test reliability, validity and feasibility, 56 new adult inpatients completed the 21-item FAST.

Results: The cognitive interviews clarified and validated the content of the final 21-item FAST. Fifty-six patients completed the FAST from two to 133 times during routine vital sign assessment, yielding 1,699 temperature time points. Thirty-four percent of the patients (N = 19) experienced fever at one or more time points, with a total of 125 febrile time points. Kuder-Richardson 20 (KR-20) reliability of the FAST was 0.70. Four nonspecific symptom categories, Tired or Run-Down (12), Sleepy (13), Weak or Lacking Energy (11), and Thirsty (9) were among the most frequently reported symptoms in all participants. Using Generalized Estimating Equations (GEE), the odds of reporting eight symptoms, Warm (4), Sweating (5), Thirsty (9), General Body Aches (10), Weak or Lacking Energy (11), Tired or Run Down (12) and Difficulty Breathing (17), were increased when patients had a fever (Fever Now), compared to the two other subgroups-patients who had a fever, but not at that particular time point, (Fever Not Now) and patients who never had a fever (Fever Never). Many, but not all, of the comparisons were significant in both groups.

Conclusion: Results suggest the FAST is reliable, valid and easy to administer. In addition to symptoms usually associated with fever (e.g. feeling warm), symptoms such as Difficulty Breathing (17) were identified with fever. Further study in a larger, more diverse patient population is warranted.

Trial registration: Clinical Trials Number: NCT01287143 (January 2011).

Keywords: Fever Assessment Tool (FAST); Fever signs; Fever symptoms.

Fig. 1

Fever Assessment Tool (FAST). Presents the fever assessment tool (FAST), developed from elicitation (N = 28) and cognitive interviews (N = 12) of patients who possessed fever within approximately 12 h of the interview. Sign and symptom numbers correspond to the text to assist in symptom identification

Fig. 2

Schema of Study. This figure represents the schema of the study, distinguishing between patients and time point analysis. The Fever Never subset includes all time points of patients who never experienced fever on study. The Fever Now subset include only those time points when a fever was recorded. Patients who possessed at least one fever time point but did not record a fever at another time point were categorized into the Fever Not Now subset

Fig. 3

Scatter diagram of temperatures versus time points N = 1,699. Maximum recorded temperature was 39.8 °C while minimum temperature was 34.2 °C among 125 possible fever-qualifying temperatures. Eighty percent of the total readings qualified as fever (ranged 38.0 °C to 38.6 °C)

Fig. 4

Frequencies of “yes” responses by patient subset across the 21 sign and symptom FAST assessment. “Yes” response frequencies (%) for each sign and symptom of the FAST by patient subset: Fever Now, Fever Not Now and Fever Never. The Fever Now category represents time points when a fever was recorded while the Fever Never group includes time points of patients who never experienced fever. Time points at which there was no fever (Fever Not Now) included patients who had at least one febrile time point during the study duration