Review of the Method and Results in Curley Hibberd Fineman Wypij Shih Et Al 2005

Curley, Hibberd, Fineman, Wypij, Shih, et al. (2005) tested their hypothesis that at the end of 28 days children and infants and who had acute lung injuries that were treated using prone positioning would have significantly more ventilator-free days than children and infants with acute lung injuries who were with supine positioning. The primary outcome variable for this hypothesis was number of ventilator free days for the groups at the end of 28 days treatment. The hypothesis was tested with an independent t-test because there were two independent groups (the independent variable here is those treated prone vs. those treated supine [both groups being randomly assigned]) that is being measured on a ratio-level dependent variable (number of ventilator free days).

The results indicated that mean number of days on the ventilator was 15.8 (standard deviation = 8.5) for supine and 15.6 (standard deviation = 8.6) for prone. The primary hypothesis was not supported (the null hypothesis that the two groups were equivalent on the dependent measure was not rejected; t-test P = 0.91 (95% confidence level -3.6 to 3.2 days).

The researchers also tested a number of secondary outcomes. For instance they controlled for the age of participants, their Pediatric Risk of Mortality III score, direct versus indirect cause of the lung injuries, and the mode of mechanical ventilation at the enrollment of the participant using multiple regression and the difference on the outcome variable was only 0.3 days so again the null hypothesis could not be rejected (Wald test P=0.87; 95% confidence interval, ?3.0 to 3.5 days). The researchers wondered if the age of the participant was a confounding variable but controlling for this resulted in no significant difference on the outcome variable (difference was 0.4 days when age was controlled for; 95% confidence interval -2,9 to 3.7 days).