Article Text

  1. T. C. Tan,
  2. K. Hamilton,
  3. J. T. Martin
  1. Western University of Health Sciences, College of Osteopathic Medicine, Pomona, CA, and Doctor's Hospital Medical Center of Montclair, Montclair, CA


There is increasing evidence that early developmental history may predict susceptibility to later adult disease. Developmental instability (DI) caused by allelic incompatibility or fetal environmental insults could contribute to later disease. DI is normally assessed by a composite of differences in bilateral trait measurements, and this is referred to as fluctuating asymmetry (FA). We have used such FA of the hand and lower arms of patients to assess DI in patients with coronary artery disease and diabetes mellitus. Methodological factors may influence the integrity of the results and have not been widely discussed in the literature. In this study we assess the importance of three of these factors: measurement accuracy for hospitalized patients versus controls, observer reliability, and subjects' handedness. We hypothesized that FA measurements of inpatients would have a higher variance than those of controls, that FA accuracy would improve with experience, and that handedness might be an important covariant in FA studies. To test these hypotheses we calculated the variance in the triplicate measurements of six bilateral traits and the composite FA. These include looking at the differences in variation in the control patients compared to the coronary artery disease and diabetes mellitus patients, the differences in asymmetry between left-handers and right-handers, and how variance changed with experience of the investigator. Our results showed that there was greater variation in measures of patients with coronary artery disease and diabetes mellitus patients when compared to the controls. This can be problematic because the greater variation can adversely affect the estimate of the true mean. It was found that the asymmetry in right-handers is greater than the asymmetry seen in left-handers. This finding needs to be taken into account due to the fact that this can cause bias in the analysis of the data. A decreased variance was also seen with increased experience of the investigator. These findings should be considered in all future studies of developmental instability and FA in coronary artery disease or diabetes mellitus.

Statistics from

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.