Measurement and evaluation of body temperature Implications for clinical practice

Abstract: The general aim was to explore factors influencing the normal variation and measurement of body temperature. Additional aims were to study morbidity, mortality and the clinical presentation of pneumonia and predictors for survival in elderly nursing-home residents. Two hundred and thirty seven non-febrile nursing home residents (aged 66-99 years) and 87 healthy adults (aged 19-59 years) were included. In elderly individuals, the morning ear and rectal body temperature was measured at baseline and pneumonia and survival was observed at one- two and three-year. In healthy adults the rectal, ear, oral and axillary temperature were measured simultaneously on one morning and repeated measurements were performed in three subjects.Overall, the range of normal body temperature was wider then traditionally stated. In elderly nursinghome residents, functional and cognitive impairment and BMI < 20 were related to a lower body temperature and medication with analgesics to a higher. Compared to adults < 60 years elderly persons had a higher average ear and a lower rectal temperature. Men and postmenopausal women < 60 years had lower body temperature than premenopausal women. The repeated measurements showed a wide individual variability irrespective of the site of measurement, and that replicated measurements do not improve accuracy. When comparing the rectal temperature with oral, ear and axillary readings the average difference was > 0.5°C with a wide individual variation.The yearly incidence of nursing-home acquired pneumonia varied between 6.9% and 13.7%. Functional impairment, chronic obstructive pulmonary disease (COPD) and male sex were related to a higher risk of acquiring pneumonia and presenting non-specific symptoms were common. Age and functional impairment predicted mortality, irrespective of gender, while cerebral vascular insult, a lower body mass index and malnutrition in women and heart disease, COPD, medication with sedatives and mortality rate index in men were gender specific predictors. Surviving women had a higher baseline body temperature than non-surviving, while no such difference was found in men.When assessing body temperature, it is important to consider the site of measurement, technical design, operator technique, age and gender and, in elderly nursing-home residents, physical and cognitive impairment, body constitution and medication with analgesics. The best approach is to use an unadjusted mode, without adjusting to another site. To prevent a delayed diagnosis of pneumonia, one should be aware of a low baseline body temperature and lack of specific clinical symptoms in elderly nursing-home residents. Preserving and/or improving functional, cognitive, nutritional status and preventing agitation and confusion would improve survival in nursing-home residents.