The impact of correlation on estimation of effect size pre- post-treatment

How does the correlation between values impact effect size?

Looking in the article "Pre-post effect sizes should be avoided in meta-analyses" the authors state that "One important reason why pre-post SMDs should be avoided is that the scores on baseline and post-test are not independent of each other." And that "The value for the correlation should be used in the calculation of the SMD, while this value is typically not known."

How DO you "use" it, the correlation value? I have done my calculations based either the t-value divided with the square root of the number of participants (or many diff div by the SD).

Is this sufficient? Does the t-value take into account the value of the correlation? Or is the proper thing to, somehow, take the value of correlation into account?

I have two groups I compare, and in addition to comparing them (or the treatments they have received) I am interested in the within-group effect size, and concerned if I should take the correlation values (0.18 and 0.42 for each group) into account?