Reliability of in-bed weighing procedures for critically ill infants.

The purpose of this study was to describe the intra- and interexaminer reliability of weight measurements obtained from critically ill infants on an in-bed electronic scale. Weight measurements were obtained using the in-bed scale (Smart Model 35, Olympic Medical, Seattle, Washington) for 32 infants; 16 were in an incubator, and 16 were under a radiant warmer. Two nurses each obtained two weight measurements for each infant for three consecutive days, for a total of 96 data collection sessions. The nurses were blinded to their own and to the other nurse's weight measurements. The average mean absolute difference for individual nurses' weight measurements (interexaminer reliability) was 12.58 gm for weights obtained in the incubator and 19.19 gm for weights obtained under the radiant warmer. The average mean absolute difference for pairs of nurses' weight measurements (interexaminer reliability) was 14.29 gm for weights obtained in the incubator and 24.42 gm for weights obtained under the radiant warmer. The average mean absolute differences for weights obtained in the two bed types differed significantly for both intra- (Z = -2.46, p = .0141) and interexaminer (Z = -3.11, p = .0019) reliability. The number of pieces of equipment that had to be held during the weight measurement was weakly correlated with both the intra- (rs = .1878, p = .0091) and interexaminer (rs = .1600, p = .0266) mean absolute differences. These findings suggest that weight measurements of critically ill infants obtained using the Smart Model 35 in-bed electronic scale are sufficiently reliable for calculation of medication, parenteral fluid, blood replacement, and nutritional requirements.(ABSTRACT TRUNCATED AT 250 WORDS)