A meta-analysis of work sampling studies pre-and post-implementation of bedside computer charting in intensive care units
The purpose of this meta-analysis was to integrate the findings of unpublished work sampling studies to determine (1) the effect of bedside computers on the distribution of time nurses spend in direct patient care, indirect patient care, charting, and communication, and (2) the relationship between the effects of bedside computers and methodological characteristics of the primary studies. The sample included three work sampling studies that measured the impact of bedside computers on the distribution of time that nurses spend in various activities in the intensive care units (ICUs). Analyses began by coding the observations recorded in the primary studies and combining them into the four activity subsets of interest to this meta-analysis. Percent change in time that nurses spent in each activity after computer installation was converted to an effect size through probit transformation techniques. Obtained effects sizes were weighted for sample size and then combined to create a weighted mean effect size for each activity subset across studies. Findings revealed that bedside computers had no real effect on the time that nurses spent in direct patient care, indirect patient care,
charting, and communication activities when effect sizes were compared to Cohen's guidelines for effect size interpretation. There was an average increase of 2.1% in time spent in direct patient care activities and an average decrease of 2.7% in time spent in charting . The methodological characteristics considered in this study were (a) prior use of computers, (b) computer applications, (c) length of time between pre- and post-study, and (d) quality of the primary studies. After coding the primary studies, only length of time between the pre - and post-study was found to vary across studies. A Pearson Product-Moment correlational analysis was conducted to determine the relationship between length of time between the pre- and post-studies and the weighted effect sizes for direct patient care and charting activities. There was found to be no significant relationship between length of time between the pre- and post-study and the weighted effect sizes for direct patient care or charting.