| Article Access Statistics|
| Viewed||1044 |
| Printed||19 |
| Emailed||0 |
| PDF Downloaded||33 |
| Comments ||[Add] |
Click on image for details.
|Year : 2015 | Volume
| Issue : 6 | Page : 1005-1006
Age and sex matching in case control studies
Sunil Kumar Raina
Department of Community Medicine, Dr. Rajendra Prasad Government Medical College, Tanda, Himachal Pradesh, India
|Date of Web Publication||20-Nov-2015|
Sunil Kumar Raina
Department of Community Medicine, Dr. Rajendra Prasad Government Medical College, Tanda, Himachal Pradesh
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Raina SK. Age and sex matching in case control studies. Neurol India 2015;63:1005-6
I went through with interest the article entitled "A study on cognitive decline with respect to metabolic syndrome and inflammation in elderly Indians" published in Neurology India 2015;63:537-41. It is being estimated that around 10-37% of the elderly population with dementia in developing countries are having potentially vulnerable living circumstances requiring long-term and specialized care. Keeping in view the public health importance of this finding, the study on cognitive decline with respect to metabolic syndrome is a valuable addition to literature on dementia. The authors deserve credit for this. However, I have a small concern with the description used by the authors in the methodology part of this study. The authors have divided the study participants into two groups. The authors then state that the mean age in group 1 was 69.52 ± 6.69 years (maximum = 86 years, minimum = 61 years). The mean age in group 2 was 69.04 ± 6.76 years (maximum = 82 years, minimum = 60 years). Thus, both the groups had a similar age profile, and they were age matched. In the group 1, 40% (n = 20) were female and 60% (n = 30) were male subjects. In group 2, 48% (n = 24) were female and 52% (n = 26) were male subjects. Thus, the two groups were sex-matched. The statement on age and sex matching does not seem to be accurate as the figures for both the variables (age and sex) differ in the two groups. Matching is an important way of accounting for confounding in case-control studies. In matched studies, the most efficient allocation of a fixed number of controls is usually the one that sets the ratio of controls-to-cases to be approximately equal.  Therefore, for 20 females and 30 males in group 1, we should have 20 females and 30 males in group 2 also. If this is done, then we are matching for sex. Similarly in the age matching between two groups, authors need to reflect as to how a maximum age of 86 years (in group 1) was matched with the maximum age of 82 years (in group 2). This also assumes significance in view of the fact that age in itself is an important predictor of cognitive decline.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Ghosh A, Biswas AK, Banerjee A. A study on cognitive decline with respect to metabolic syndrome and inflammation in elderly Indians. Neurol India 2015; 63:537-41.
Raina SK, Raina S, Chander V, Grover A, Singh S, Bhardwaj A. Is dementia differentially distributed- A study on the prevalence of dementia in Migrant, Urban, Rural and Tribal elderly population of Himalayan region in northern India. North Am J Med Sci. 2014;6:172-7.
Miettine OS. Theoretical Epidemiology: Principles of Occurrence Research in Medicine. New York: John Wiley and Sons Inc; 1985. p. 66.