Question

Researchers designed a study to determine risk factors for late-onset candidemia among infants in the neonatal intensive care unit. The researchers performed a study from March 2010 to January 2013 in two level III-IV neonatal intensive care units. Case subjects had candidemia diagnosed more than 48 hours after hospitalization. For every one case, three children were selected for a control group with the same study site, birth weight, study year, and date of enrollment as the case. Potential risk factors that were evaluated included medical devices, medications, gastrointestinal pathology (congenital anomalies or necrotizing enterocolitis) and previous bacterial bloodstream infections.

1. What was the reason researchers decided to select three children rather than one child for every one case as a control group?

Answer 1

The control group is selected in proportion of the exposure of interest. The objective of the control group is to find out the relative size of the exposed and unexposed components of the source population.

In case-control studies comparable control group is necessary to estimate the prevalence of exposure in the population.

Here the three children are selected rather than one child for the comparison of risk factors. Because it gives more statistical significance. But do not include more than four or five controls per case because statistical power decreases by further increasing this ratio.