Elderly patients, use of antidepressants, and hip fracture

Abstract

Barbara Liu and colleagues (May 2, p 1303)1Liu B Anderson G Mittman N To T Axcell T Shear N Use of selective serotonin-reuptake inhibitors or tricyclic antidepressants and risk of hip fractures in elderly people.Lancet. 1998; 351: 1303-1307Summary Full Text Full Text PDF PubMed Scopus (332) Google Scholar show that the use of antidepressants increases the risk of hip fracture, and that selective serotonin-reuptake inhibitors (SSRIs) do not offer any advantage over tricyclic agents (TCAs). Some aspects of the findings need to be clarified. For example, the absence of significant differences across doses of SSRIs is somewhat surprising. This circumstance might be explained by the method of calculating the dose. Specifically, in patients without an additional prescription in the 90 days preceding the one closest to the index date, the dose was that estimated in the rest of the population. Although this method is plausible, it rests on the assumption that a few patients were attributed a dose calculated on a much larger and representative group. Liu and co-workers identified dose categories with a cutoff of 0·5 and 1·5 times the defined daily dosage (DDD). Data on TCAs lend support to this approach, but the same does not hold true for SSRIs. Unlike TCAs, in fact, dosage adjustment of SSRIs seems to be unnecessary in the elderly.2Leonard BE Pharmacological differences of serotonin reuptake inhibitors and possible clinical relevance.Drugs. 1992; 43: 3-9Crossref PubMed Scopus (114) Google Scholar Consequently, an overly low cutoff might have inflated the number of persons taking intermediate or high doses—as the small proportion (11%) of patients taking SSRIs at low dose suggests is the case. This imbalance would have inevitably resulted in a dilution of the risk associated with SSRIs given at higher doses, obscuring a possible dose relation. Finally, since practice is to start slow and low in new users of SSRIs (22% vs 9% for TCAs of either class), a high proportion might have been prescribed a low dose. If so, a trend towards higher risk among new users might also have contributed to the inability to detect a dose dependency. In this respect, and in consideration of the time to steady-state plasma concentrations, it would have been instructive had the investigators reported the number of days between SSRIs prescription and index date. With the SAGE database3Bernabei R, Gambassi G, Lapane KL, et al. Characteristics of the SAGE database: a new resource for research on outcomes in long-term care. J Gerontol A Med Sci (in press).Google Scholar, we undertook a similar study among elderly patients living in every nursing home of five US states. Comparing 8851 cases with 35086 matched controls, we noted that both TCAs and SSRIs were associated with a similarly increased (30–40%) risk of femur fracture.4Lapane KL, Gambassi G, Hume A, Sgadari A, Mor V. Which antidepressants increase the risk of femur fracture in long term care? American Geriatric Society and American Federation for Aging Research, Annual Meeting, 1998: 142.Google Scholar By contrast with Liu and colleagues' findings, we documented a strong dose relation of the effect of SSRIs. Adjusted odds ratio was 1·6 (95% CI 1·4–1·9) at low dose (< 1·0 DDD) and 2·9 (95% CI 1·3–6·6) at high dose (>1·0 DDD). Moreover, we found substantial differences among the SSRIs. Although fluoxetine (n=689) was not associated with a significant increased risk of fracture, paroxetine (n=298) nearly doubled it, and sertraline (n=618) conferred an intermediate risk. The anecdotal notion of SSRIs being safer agents relative to TCAs needs to be revisited, especially in the clinical management of high-risk individuals.5Ruthazer R Lipsitz LA Antidepressants and falls among elderly people in long-term care.Am J Public Health. 1993; 83: 746-749Crossref PubMed Scopus (94) Google Scholar We have to elucidate possible differences between individual SSRI agents, and to clarify the relative importance of dosage. Elderly patients, use of antidepressants, and hip fractureAuthors' reply Full-Text PDF