Childhood fever is one of the commonest reason for medical consultation in children, being responsible for 15-25% visits in primary care, and also presentations to the emergency departments (ED) [1,2], and is known to cause significant anxiety in parents [3]. Most children undergo evaluation for at least one febrile illness before their third birthday [4]. Western studies report good parental awareness about fever [5], but studies from India [6,7] have shown conflicting results. Frequently, parents do not document temperature or record it improperly, leading to undue anxiety and over-crowding of the ED [6-8]. We studied the correlation of temperatures measured at home by parents with recordings done at presentation in the ED among infants with acute illnesses.

This cross-sectional study was conducted from April, 2018 to January, 2019 at the pediatric ED of a public hospital in northern India, after taking clearance from the Institutional Ethics Committee. Febrile children aged 3 month to 2 year, with fever of at least 4 days were considered for enrollment. A febrile child was defined as one with history of fever e"38ÚC recorded at least once at home in previous 24 hours. Those suffering for fever for >7 days, children with any underlying heart disease, and children with any diagnosed immunodeficiency disorder or conditions predisposing to recurrent infections (like type 1 diabetes, vesico-ureteric reflux) were excluded. Consecutive children were enrolled on one pre-decided day every week.

After taking written informed consent, enrolled children were evaluated clinically and initial management provided. Subsequently, based on history, and clinical and laboratory information, they were treated as inpatient or outpatient. For all enrolled children, demographic details, contact information and details of education and income of parents were collected. History was taken regarding highest temperature recorded at home and any associated symptoms, treatment taken if any before presentation, relevant history of co-morbidities, immunization and feeding history. Anthropometric measurements were taken for all included children as per standard guidelines, and Z-scores were calculated using Anthrocalc application.

All the data were recorded in a structured pre-tested form. Rectal temperature was taken at presentation for all enrolled children, The various diagnoses were made and management carried out according to the departmental protocols guided by standard management guidelines [9,10]. Mean (SD) or median (IQR) were calculated for the baseline characteristics. Pearson correlation coefficient was calculated for temperature documented at home and in the ED. Comparisons were done between children with fever documented in ED, and those without fever documented in ED.

Out of overall study population of 150 children with history of fever of 4-7 days in respective age group attending ED, only 108 (68.3% boys) had documented fever at home. The median (IQR) age of the study population was 12 (3-20) months. The median Z scores for all anthropometric variables (weight, length and head circumference) were greater than –3. Majority (88%) of children belonged to the lower middle (III) and upper lower (IV) socioeconomic classes, and majority of mothers (62%) had at least secondary school education (Table I). Along with fever, the most common presenting complaint was respiratory problems.