medications are associated with a significant risk of mortality and morbidity in older patients with dementia. Guidelines and expert opinion now recommend that, when treating neuropsychiatric symptoms associated with dementia, clinicians use first nonpharmacological interventions; when patients do not respond to these nonpharmacological interventions, clinicians should carefully weigh the potential risks and benefits of antipsychotics and alternative pharmacological treatment (e.g., selective serotonin reuptake inhibitors [SSRIs], cognitive enhancers, or anticonvulsants); if they decide to use an antipsychotic, they should obtain consent from the surrogate decision maker after having informed him or her of the potential risks associated with the use of antipsychotics and they should carefully document the discussion; finally, they should discontinue antipsychotics in non-responders and systematically attempt to discontinue antipsychotics after a few months in responders. Four articles in the current issue 4e7 illustrate the challenges faced by clinicians when engaging in this process and by the field of geriatric psychiatry in general when trying to collect better evidence on which to base clinical algorithms and guidelines. The article by Langballe et al. 4 reports on a longitudinal cohort analysis of 26,940 Norwegian outpatients with dementia who were prescribed cognitive enhancers and either an antipsychotic or other psychotropic medications (antidepressants, benzodiazepines, lithium, or anticonvulsants) for up to seven years. Compared with other psychotropic drugs, antipsychoticswereassociatedwithaboutadoublingofthe risk of mortality both in the short term (during the first 30 days after initiating treatment) and in the long term (during the 2- to 7-year follow-up period). The analysis adjusted for age, sex, doses, and—as a proxy for comorbid physical conditions—drugs used for the treatment of cancer, diabetes, and cardiovascular, lung, or musculoskeletal diseases. Congruent with several previous analyses, 8 haloperidol was associated with a higher mortality risk than risperidone and the risk associated with risperidone was not different from the risk associated with olanzapine or quetiapine. As in previous studies, Langballe et al. 4 found a very high relative increase in the risk of mortality associated with antipsychotic use; the corresponding absolute increase, however, was quite low. During the first 30 days of use, the adjusted hazard ratio (HR) for the use of antipsychotics versus other psychotropic drugs was 2.1, corresponding to about a doubling (110% increase) of the risk. The absolute risk of dying, however, was 1.13% (93 deaths among the 8,214 outpatients treated with an antipsychotic) versus 0.42% (79 deaths among the 18,726 outpatients treated with other psychotropic drugs), yielding an (unadjusted) absolute increase of only 0.71% (1.13%e0.42%). Similarly for the 2- to 7-year follow-up period, the adjusted HR was 1.7 (about a 70% increase) but the
Read full abstract