While the study by Diem and colleagues1 adds to a growing body of evidence regarding the possible deleterious effects of selective serotonin reuptake inhibitors (SSRIs) on bone health, their conclusion about the lack of a similar effect with tricyclic antidepressants (TCAs) may be mitigated by methodological concerns.
Of note, there are concerns that confounding by indication may have selected a TCA group that possesses different characteristics from SSRI users. In particular, the majority of TCA users in the analysis may have been prescribed TCAs for indications other than depression (eg, neuropathic pain and insomnia), a postulation corroborated by the observation that TCA users were less severely depressed and more physically impaired compared with SSRI users. This is further compounded by the exclusion of 31 subjects who reported the use of both SSRIs and TCAs, a group that is likely to comprise those with significant depression. This may have an impact on both the duration and dose of exposure. For instance, subjects who were prescribed TCAs for acute pain (hence a shorter duration of exposure) are more likely to be included compared with those with chronic pain, since the latter are often associated with significant depression (and those more likely to be using dual agents). Strikingly, the number of TCA users declined during the study period from 86 to 74, while the corresponding number of SSRI users increased almost 3-fold.