No abstract

The aim of this study was to evaluate the effectiveness of a practice magnetic resonance unit, in preparing children to undergo magnetic resonance procedures without general anaesthesia (GA) or sedation. The records of children who attended the practice MRI between February 2002 and April 2004 were retrospectively reviewed. Each record was assessed as to whether the child had passed or failed the practice MRI intervention. Those children who were considered to have passed and were proceeded to a clinical non-GA MRI had the report of the clinical scan reviewed. If the scan had been reported as non-diagnostic because of movement artefact it was classified as a failed scan, otherwise it was considered a pass. One hundred and thirty-four children undertook a practice MRI (age range 4.1-16.1 years, median age 7.7 years, 47% boys) and 120/134 (90%) passed the practice session. In all, 117/120 (98%) subsequently had a clinical non-GA MRI and 110/117 (94%) passed (median age 7.8 years, 47% boys). Preparation is a safe and effective method to reduce the need for sedation and GA in children undergoing a clinical MRI scan. It provides a positive medical experience for children, parents and staff, and results in cost savings for the hospital.

No abstract

as are their covariance matrices Vi. Taking the V, as fixed, we assume the distribution of the bi to be (i-N (i,Vi) where that is, the distribution of the 3 depends on the year i by way of the parameter xi. Assuming a linear trend in the parameter estimates we constrain xi such that xi-1x Estimation and maximisation now proceed according to the computational details given by Dempster et al'2 with the exception of x, an estimate of which may be achieved by maximum likelihood. Calculation of the profile All variable differences are standardised to a normal N(O,1) distribution using the parameter estimates from Scotland and from the hospital alone, together with the two standard errors of these estimates. Using figure 2 as an example, the parameter estimate (SE) for the variable associated with no antenatal care obtained from the Scottish model for death in the first week (1984-7) was 2 101 (0 386), while for hospital A these figures were 2 534 (1 048). The z score for the difference in the parameter estimates (hospital A-Scotland) is then z = (2 5 34-2 10 1)/ \V(0 3862+ 1 0482) =0 388 Because the comparison is being made across eight variables, the confidence intervals are placed at ±2-734, so that there is a probability of 0-025/8 in each tail. The point by this variable in figure 2 is therefore positioned 14% of the distance between the central vertical line (the mean for Scottish, representing zero difference) and the right hand edge of the shaded area (95% simultaneous confidence interval).