Looks like the most logical outcome for me, but I don't know how to test it - or even if there is enough data in the world to test it now. Anyway, the usual correlation is not causality applies here.
In this case the distinction between correlation and causation is actually very important. The alternative causal chain, that happy people are more likely to become religious, is highly plausible. Further, there are myriad plausible confounding variables. In many cases correlation can be taken as evidence of causation, but in this particular case it really should not be.
By the time people are adults, if they haven't grown up with religion, I suspect the response to such a commandment would be something along the lines of 'That instruction set isn't implemented on this architecture - and I'm not sure I can, or would like, to become the sort of architecture where it can be.'
So, that might be one line of thought as to why it was left out.
""[9] and a review of 498 studies published in peer-reviewed journals concluded that a large majority of them showed a positive correlation between religious commitment and higher levels of perceived well-being and self-esteem and lower levels of hypertension, depression, and clinical delinquency.[12] A meta-analysis of 34 recent studies published between 1990 and 2001 found that religiosity has a salutary relationship with psychological adjustment, being related to less psychological distress, more life satisfaction, and better self-actualization.[13] Finally, a recent systematic review of 850 research papers on the topic concluded that "the majority of well-conducted studies found that higher levels of religious involvement are positively associated with indicators of psychological well-being (life satisfaction, happiness, positive affect, and higher morale) and with less depression, suicidal thoughts and behavior, drug/alcohol use/abuse."[14]"
[1] http://en.wikipedia.org/wiki/Religion_and_happiness