Year : 2015  |  Volume : 57  |  Issue : 1  |  Page : 110-

Accounting for confounding

Sunil Kumar Raina 
 Department of Community Medicine, Dr. RPGMC, Tanda, Kangra, Himachal Pradesh, India

Correspondence Address:
Sunil Kumar Raina
Department of Community Medicine, Dr. RPGMC, Tanda, Kangra, Himachal Pradesh

How to cite this article:
Raina SK. Accounting for confounding.Indian J Psychiatry 2015;57:110-110

How to cite this URL:
Raina SK. Accounting for confounding. Indian J Psychiatry [serial online] 2015 [cited 2019 Nov 16 ];57:110-110
Available from:

Full Text


Went through an article entitled "impact of psychiatry training on attitude of medical students toward mental illness and psychiatry" published in Indian Journal of Psychiatry. [1] The authors deserve credit for touching on the important issue of "role of training" in medical sciences. However, I have a concern. The authors stated aims are: (1) To compare the attitude of medical students and interns in a medical college toward mental illness and psychiatry and (2) to assess the impact of psychiatric training on attitude toward the mentally ill person and mental illness. The authors have done full justice to the first aim of this study in presenting comparable data about the attitude of medical students and interns. However, the second aim of this study, that is, impact assessment has been handled inadequately. The fact that the difference between attitudes of interns and undergraduate medical students (not yet fully exposed to clinical subjects) can provide an idea on impact of training, does not seem logical. The comparison groups are diverse enough to be included for comparison. The fact that interns have undergone clinical training in other disciplines (other than psychiatry) during their under graduation as well as internship period may influence their attitude to psychiatry and mental illness as well and thus act as a confounder in this study. This has not been accounted for in the study. In statistics, a confounder is an extraneous variable that correlates (directly or inversely) with both the dependent variable and the independent variable. A perceived relationship between an independent variable and a dependent variable that has been misestimated due to the failure to account for a confounding factor, gives rise to a spurious relationship.

The methodology to estimate the impact of psychiatry training should have been modified and study could have been conducted on interns only before and after they had received their training in psychiatry.


1Gulati P, Das S, Chavan BS. Impact of psychiatry training on attitude of medical students toward mental illness and psychiatry. Indian J Psychiatry 2014;56:271-7.