Does Fever at the Time of Discharge Have Any Impact on the Incidence of Readmission?

Abstract

Most physicians believe that patients who have fever within 24 hours of the planned date of discharge should be kept in the hospital until the fever resolves. A search of the literature revealed very few articles addressing this topic. The object of this study was to review a number of patient discharges from the surgical service and to document the presence or absence of fever within 24 hours of the time of discharge. The primary end point of the study was to determine the rate of readmission for both patients discharged with fever and those discharged without fever. Secondary end points were to determine whether the readmission was related to the original discharge diagnosis or the presence of fever at the time of discharge. The records of all adult patients with a hospital length of stay of > or = 5 days discharged from the surgical and gynecology services from April through July of 2007 were reviewed. Deaths were excluded. The following data elements were recorded: primary discharge diagnosis; age; highest recorded temperature within 24 hours of discharge; date time and cause of readmission within 30 days; and outcome. Fever was defined as a temperature of > or = 100 degrees F. Data were entered into an Excel (Microsoft, Redmond, WA) spreadsheet, and statistical analysis was performed using chi2 and Fisher's exact tests using Primer of Biostatistics (McGraw-Hill, New York, NY). The records of 300 consecutive patients were reviewed. Follow-up was available for 86.7 per cent of the patients, 84.4 per cent of the febrile patients, and 87.1 per cent of the nonfebrile patients. A fever of > or = 100 degrees within 24 hours of discharge was noted in 45 (15.0%) patients. The mean fever was 100.5 degrees, with a range of 100 degrees to 102.1 degrees. There were 38 readmissions. Of the 45 patients with fever, seven (15.6%) were readmitted. Of those seven, four readmissions were related to the previous admitting diagnosis. Of the patients who were discharged without fever, 31 (12.2%) were readmitted with 24 of those readmissions for diagnoses related to the first admission. The rate of readmission for fever and nonfever patients was not statistically significantly different (P = 0.697). Similarly, the rate of related versus nonrelated diagnoses in both the fever and nonfever groups was not statistically significantly different (P = 0.351). The presence or absence of fever within 24 hours of patient discharge seems to have no impact on the rate of readmission within 30 days.