Prevalence and Predictors of Antidepressant Prescribing in Nursing Home Residents in the United States Swapna U. Karkare, MS; Sandipan Bhattacharjee, MS; Pravin Kamble, MS; and Rajender Aparasu, MPharm, PhD Department of Clinical Sciences and Administration, College of Pharmacy, University of Houston, Texas Medical Center, Houston, Texas ABSTRACT Background: Late-life depression is a common psychiatric disorder associated with increased morbidity and mor- tality. Depression is often under-detected and undertreated in elderly nursing home residents. Objectives: The aim of this study was to examine the prevalence of antidepressant drug use and to identify the factors associated with its use among elderly nursing home residents. Methods: The study involved the analysis of a nationally representative sample of prescription and resident files from the 2004 National Nursing Home Survey (NNHS). The study sample included all elderly nursing home residents 65 years of age. The analysis focused on prescribing from any antidepressant class, including selective serotonin reuptake inhibitors (SSRIs), tricyclic antidepressants (TCAs), serotonin modulators, serotonin-norepinephrine reuptake inhib- itors (SNRIs), monoamine oxidase inhibitors (MAOIs), and others. Descriptive weighted analysis was performed to examine antidepressant use prevalence patterns in elderly nursing home residents. Multiple logistic regression analysis within the conceptual framework of Anderson’s behavioral model was used to examine the predisposing, enabling, and need characteristics associated with antidepressant use. Results: According to the 2004 NNHS, overall prevalence of antidepressant use among elderly nursing home residents was 46.22% (95% CI, 45.16 – 47.27). Most antidepressant users were 85 years of age (49.7%), female (75.7%), non-Hispanic (96.4%), and white (91.1%). The most prescribed class of antidepressants was SSRIs (31.09%; 95% CI, 30.12–32.07), followed by serotonin modulators (4.65%; 95% CI, 4.22–5.09), SNRIs (2.78%; 95% CI, 2.45–3.12), TCAs (2.34%; 95% CI, 2.03–2.65), and MAOIs (0.01%; 95% CI, 0.00 – 0.03). Citalopram (12.92%; 95% CI, 12.21–13.63) was the most prescribed individual antidepressant, followed by mirtazapine (10.19%; 95% CI, 9.55–10.84). Among the predisposing characteristics, age, race, and marital status were significantly associated with antidepressant use. The odds of receiving an antidepressant were lower for those aged 85 years and those who were unmarried elderly residents, when compared with their counterparts; whites were more likely to receive an antidepressant than nonwhites. Enabling factors such as Medicaid and bed capacity significantly predicted antidepressant use. Having Medicaid was positively associated with antidepressant prescription, whereas an increase in the total number of beds decreased the probability of an antidepressant prescription. Among need characteristics, the likelihood of antidepressant prescription use decreased with increased dependence in decision-making ability and out-of-bed mobility. The presence of depressed mood indicators and a history of falls/fractures increased the likelihood of antidepressant prescription use. The odds of receiving an antidepressant increased with diagnosis of depression but decreased with diagnosis of anxiety. Conclusion: Nearly half of elderly nursing home residents received antidepressants. In addition to need factors, predisposing and enabling factors played an important role in influencing the use of antidepressants in elderly nursing home residents. (Am J Geriatr Pharmacother. 2011;9:109 –119) © 2011 Elsevier HS Journals, Inc. All rights reserved. Key words: antidepressants, nursing home, elderly, depression. Accepted for publication March 16, 2011. doi:10.1016/j.amjopharm.2011.03.001 © 2011 Elsevier HS Journals, Inc. All rights reserved. 1543-5946/$ - see front matter S.U. Karkare et al. The American Journal of Geriatric Pharmacotherapy Volume 9 ● Number 2 April 2011 109