The confusion is caused by the phrasing of the two bullet points at the bottom of page 312. When it is says "reject OOmph" it does not mean reject the null hypothesis about Oomph. On the contrary the Z statistic for the null hypothesis ('no change with Oomph') is 'not significant' and so we cannot be reject the hypothesis, which means that there are no grounds for accepting Oomph as being an effective weight loss treatment. So this is the sense in whihc we 'reject' Oomph as being an effective treatment. Conversely because the Z statistic for Precision is 'significant', we can reject the null hypothesis for Precision and hence 'accept' that Precision is an effective treatment.
I will suggest a rewording of the bullet points on the errata page.