Management of feverish children at home.

OBJECTIVES--To compare the acceptability and effects on temperature of advice to unwrap children and give paracetamol or warm sponging treatments in the management of feverish illness at home. DESIGN--A randomised, open, parallel group study using factorial design comparison of unwrapping, warm sponging plus unwrapping, paracetamol plus unwrapping, and paracetamol and warm sponging plus unwrapping. SETTING--Homes of willing families with a feverish child recruited after consulting one of 21 participating general practitioners in Southampton. SUBJECTS--52 children aged from 3 months to 5 years with axillary temperatures before treatment of > or = 37.8 degrees C and < 40 degrees C. MAIN OUTCOME MEASURES--Response to advice assessed over four hours; temperature assessed by continuous data logging from an axillary thermistor; acceptability of treatment to child and parent scored on Likert scales immediately after treatment and on return to health. RESULTS--Response to treatment advice varied; unwrapping alone had little effect on temperature. Paracetamol increased the time below 37.2 degrees C in four hours by 109 (95% confidence interval 74 to 145) minutes compared with unwrapping; warm sponging caused the fastest reduction in temperature. Parents discriminated between treatments, preferring paracetamol. CONCLUSION--Advice to give paracetamol is more effective than sponging or unwrapping in controlling temperature in children at home and is more acceptable to parents. Warm sponging has an additive effect and reduces fever more quickly than paracetamol.     

Memory keeps you at home: a mechanistic model for home range emergence

Despite its central place in animal ecology no general mechanistic movement model with an emergent home-range pattern has yet been proposed. Random walk models, which are commonly used to model animal movement, show diffusion instead of a bounded home range and therefore require special modifications. Current approaches for mechanistic modeling of home ranges apply only to a limited set of taxa, namely territorial animals and/or central place foragers. In this paper we present a more general mechanistic movement model based on a biased correlated random walk, which shows the potential for home-range behavior. The model is based on an animal tracking a dynamic resource landscape, using a biologically plausible two-part memory system, i.e. a reference- and a working-memory. Our results show that by adding these memory processes the random walker produces home-range behavior as it gains experience, which also leads to more efficient resource use. Interestingly, home-range patterns, which we assessed based on home-range overlap and increase in area covered with time, require the combined action of both memory components to emerge. Our model has the potential to predict home-range size and can be used for comparative analysis of the mechanisms shaping home-range patterns.