more explanation for above problem

Any help would be highly appreciated

Thanks in anticipation

I'm not sure I understand what you are trying to do. However, the reference category changes between binomial and multinomial logistic in your examples and .034 is essentially 1/29.2 - so the analyses look pretty much equivalent. Adding covariates changes the estimates. This is not surprising - this is why one adds covariates (presumably they are correlated with the outcome and/or other predictors).

Thanks a lot Dr Baguley. I got it. 

Yes, as you said, the covariates are related to outcome that's why the estimates are changing on increasing the number of covarites.

Now I am not using multinomial as my outcome is dichotomous categorical variable. 

Thanks a lot again