Background: Fall-related fractures are associated with mortality, morbidity, functional decline, and reduced quality of life in nursing home (NH) residents. Antipsychotic (AP) drugs are used in 25% of NH residents to treat dementia-related behavioral & psychological symptoms. AP drug use is known to be associated with falls and fractures, but these findings are derived from studies using older, typical APs. Newer, atypical APs are reported to have better tolerability and fewer extrapyramidal side effects than older, typical APs. Due to rapid adoption of ? newer, atypical APs in NH settings, fracture patterns during AP use must be re-examined. Large observational data sets can be used to examine events such as fracture that may be relatively rare in short-term clinical trials.
Study Aims : 1) to determine whether fracture rates (hip- and non-hip fracture) are higher among incident APUsers than Non-Users; 2) to determine whether fracture rates are higher among incident AP-Users receiving older, typical APs than among those receiving newer, atypical APs; 3) to perform pair-wise comparisons of fracture rates between the most commonly used AP drugs: haloperidol, risperidone, olanzapine, & quetiapine. Study Methods: This is a retrospective study of Medicaid-eligible NH residents in 5 states (CA, FL, NJ, MO, PA) during 1999, using three linked data sources. Medicaid pharmacy claims will provide AP drug exposure and dose information. Fractures will be identified in Medicare claims data using diagnosis and procedure codes. The Minimum Data Set (MDS) will provide demographic, clinical, and functional factors related to fracture risk. Unique ? facility identifiers for each NH will be included to allow adjustment for clustering effects. Comparison groups will be matched on State, age, gender, cognitive impairment score, and nursing facility, and then followed for hip fracture and non-hip fracture using survival analysis methods. Cox proportional hazards regression models with time-dependent covariates and shared frailty will be employed, adjusting for fracture risk factors as derived from an MDS-based validated fracture prediction model. The impact of AP drug dose will be examined using both the average dose and the last dose prescribed just before fracture. Doses will be expressed as a proportion of the maximal recommended dose for older adults to standardize values across drugs with different dosing ranges. Implications: Improved understanding of fracture patterns among NH residents using current AP drugs may contribute to improved AP prescribing decisions and enhanced quality of care for older adults with dementia. ? ?
| Rigler, Sally K; Shireman, Theresa I; Cook-Wiens, Galen J et al. (2013) Fracture risk in nursing home residents initiating antipsychotic medications. J Am Geriatr Soc 61:715-22 |
| Rigler, S K; Ellerbeck, E; Whittle, J et al. (2011) Comparing methods to identify hip fracture in a nursing home population using Medicare claims. Osteoporos Int 22:57-61 |
