Dear all!

We are investigating the longitudinal association between FTO SNP and BMI. We have repeated measure from the same individuals at age 15, 18 and 25 years. How can I adjust my model for daily energy intake and physical activity score? They are both continuous variables.

My model currently looks like this:

Code:
mixed BMI c.time i.FTO if sex==1 || kood: aeg, reml cov(unstructured)

Performing EM optimization:

Performing gradient-based optimization:

Iteration 0:   log restricted-likelihood = -2970.1662  
Iteration 1:   log restricted-likelihood = -2969.6672  
Iteration 2:   log restricted-likelihood = -2968.7511  
Iteration 3:   log restricted-likelihood = -2968.7497  
Iteration 4:   log restricted-likelihood = -2968.7497  

Computing standard errors:

Mixed-effects REML regression                   Number of obs     =      1,284
Group variable: kood                            Number of groups  =        545

                                                Obs per group:
                                                              min =          1
                                                              avg =        2.4
                                                              max =          3

                                                Wald chi2(3)      =    1276.54
Log restricted-likelihood = -2968.7497          Prob > chi2       =     0.0000

------------------------------------------------------------------------------
         BMI |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        time |   .4551195   .0127798    35.61   0.000     .4300715    .4801675
             |
         FTO |
          2  |   .7402651   .2758039     2.68   0.007     .1996994    1.280831
          3  |   .8363359    .344866     2.43   0.015     .1604111    1.512261
             |
       _cons |   13.25358   .2945753    44.99   0.000     12.67622    13.83093
------------------------------------------------------------------------------

------------------------------------------------------------------------------
  Random-effects Parameters  |   Estimate   Std. Err.     [95% Conf. Interval]
-----------------------------+------------------------------------------------
kood: Unstructured           |
                    var(aeg) |   .0293513   .0057656       .019972    .0431352
                  var(_cons) |   8.557252   2.078842      5.315561    13.77589
              cov(aeg,_cons) |  -.2978246   .1025787     -.4988752   -.0967741
-----------------------------+------------------------------------------------
               var(Residual) |    1.94703   .1534048      1.668427    2.272156
------------------------------------------------------------------------------
LR test vs. linear model: chi2(3) = 713.17                Prob > chi2 = 0.0000
Kind regards,
Urmeli