The cornerstone of management is the determination of the infant's risk of having serious illness. If the infant appears ill, then the risk is obviously high. However, since well-appearing febrile infants also have a significant risk of serious bacterial illness, clinical impression alone cannot decide management. Similarly, individual predictors of height of fever and peripheral blood WBC count are unreliable. Thus, investigators sought to combine clinical impression with history, physical examination, and a variety of laboratory tests to develop a set of criteria that could separate those febrile infants at risk for serious bacterial infection from those who may be safely managed as outpatients. The most common strategies for managing febrile infants are shown in Table 51–6.2,4–6,30 There are notable differences among the studies including the age of infants studied, peripheral WBC cut-off, and use of empiric antibiotics. Most strategies do not include infants younger than 1 month, because of their higher risk of serious bacterial illness and their limited clinical clues on observation. Thus, for infants younger than 1 month, a complete blood count, urinalysis, CSF cell count, protein and glucose as well as cultures of the blood, urine and CSF are part of a standard evaluation; these infants should be admitted to an inpatient service for observation and given empiric parenteral antibiotics pending negative culture results. For infants older than 1 month who appear well, the clinician may defer the lumbar puncture if the other laboratory parameters fall in the low-risk range and follow-up can be assured. Recently, the Pediatric Research in Office Settings network has challenged this dogma.7 In their study of the management of febrile infants younger than 3 months by practitioners in the office setting, 64% were managed as outpatients and 24% had no laboratory tests performed.7 Compared to accepted guidelines for management of these infants, the Pediatric Research in Office Settings clinicians detected as many cases of bacteremia and bacterial meningitis while performing fewer tests and hospitalizing fewer infants. However, this management approach required close and reliable follow-up often only obtainable in the office setting. Furthermore, the results may not be generalizable to other populations of febrile infants. Nevertheless, the variety of existing management schemes highlights the importance of individual decision modifiers such as practice setting, experience of practitioner, ease and reliability of follow-up and patient demographics.