CLINICAL INVESTIGATIONS
Risk Factors Associated with the Occurrence of Fractures in U.S. Nursing Homes: Resident and Facility Characteristics and Prescription Medications
William Spector, PhD, Thomas Shaffer, MHS, D. E. B. Potter, MS, Rosaly Correa-de-Araujo, MD, M. Rhona Limcangco, PhD

OBJECTIVES: To determine whether resident and facility characteristics and prescription medications influence the occurrence of fractures in nursing homes (NHs).

DESIGN: Panel study with 1-year follow-up.

SETTING: A nationally representative sample of NHs from the Medical Expenditure Panel Survey (MEPS).

PARTICIPANTS: Residents aged 65 and older who were in sample NHs on January 1, 1996.

MEASUREMENTS: Health status measures were collected from facility records and abstracted using a computer-assisted personal interview instrument. Fracture and drug data were updated every 4 months to provide a full year of information. Drug data were obtained from monthly medication administration records. The occurrences of fractures were obtained from medical records. Administered medications were classified using the Department of Veterans Affairs medication classification system. Facility characteristics were based on MEPS survey data collected from NH sources.

RESULTS: In 1996, 6% of residents in a NH at the beginning of the year experienced a fracture during their NH stay(s). Resident risk factors included aged 85 and older, admitted from the community, exhibited agitated behaviors, and used both wheelchair and cane or walker. Use of anticonvulsants, antidepressants, opioid analgesics, iron supplements, bisphosphonates, thiazides, and laxatives were associated with fractures. A high certified nurse aide ratio was negatively associated with fractures.

CONCLUSION: The findings indicate that fractures are associated with resident and facility characteristics and prescribing practices. It reaffirms the importance of medication review with special attention on opioid analgesics, antidepressants, and anticonvulsants to reduce the risk of fractures. J Am Geriatr Soc 2007;55(3):327-333.

Decisions to Forgo Hospitalization in Advanced Dementia: A Nationwide Study
Susan L. Mitchell, MD, MPH, Joan M. Teno, MD, MS, Orna Intrator, PhD, Zhanlian Feng, PhD, and Vincent Mor, PhD

OBJECTIVES: To examine the prevalence and factors associated with decisions to forgo hospitalization in nursing home (NH) residents with advanced dementia.

DESIGN: Cross-sectional study.

SETTING: All Medicare- and Medicaid-certified NHs within the 48 contiguous U.S. states.

PARTICIPANTS: NH residents with advanced dementia were identified using Minimum Data Set (MDS) assessments completed close to April 1, 2000 (N=91,521).

MEASUREMENTS: Multilevel, multivariate logistic regression identified factors independently associated with having a do-not-hospitalize (DNH) directive. Independent variables included subject characteristics (MDS), facility factors (On-line Survey of Certification of Automated Records), and hospital referral region (HRR) features (Dartmouth Atlas).

RESULTS: Nationwide, 7.1% (n=6,518) residents with advanced dementia had DNH orders (range 0.7% in Oklahoma to 25.9% in Rhode Island). Resident characteristics associated with having a DNH order were older age, white, living will, durable power of attorney for health care, and total functional dependence. Controlling for these factors, DNH orders were more likely in residents of facilities with the following features: not part of a chain, urban location, special care dementia unit, fewer black residents, nurse practitioner or physician assistant on staff, higher staffing ratios, and location in HRRs with fewer intensive care unit admissions during terminal hospitalizations.

CONCLUSION: Directives to forgo hospitalization for U.S. NH residents with advanced dementia are uncommon and are associated with the organizational features of the facilities caring for them and the intensity of end-of-life care practiced in the region, as well as individual resident characteristics. J Am Geriatr Soc 2007;55(3):432-438.

Implementation and Evaluation of a Nursing Home Fall Management Program
Kimberly Rask, MD, PhD, Patricia A. Parmelee, PhD, Jo A. Taylor, RN, MPH, Diane Green, PhD, Holly Brown, MSN, APRN-BC, Jonathan Hawley, Laura Schild, Harry S. Strothers III, MD, MMM, and Joseph G. Ouslander, MD

OBJECTIVES: To evaluate the feasibility and effectiveness of a falls management program (FMP) for nursing homes (NHs).

DESIGN: A quality improvement project with data collection throughout FMP implementation.
SETTING: NHs in Georgia owned and operated by a single nonprofit organization.

PARTICIPANTS: All residents of participating NHs.

INTERVENTION: A convenience sample of 19 NHs implemented the FMP. The FMP is a multifaceted quality improvement and culture change intervention. Key components included organizational leadership buy-in and support, a designated facility-based falls coordinator and interdisciplinary team, intensive education and training, and ongoing consultation and oversight by advanced practice nurses with expertise in falls management.

MEASUREMENTS: Process-of-care documentation using a detailed 24-item audit tool and fall and physical restraint use rates derived from quality improvement software currently used in all Georgia NHs (MyInnerView).

RESULTS: Care process documentation related to the assessment and management of fall risk improved significantly during implementation of the FMP. Restraint use decreased substantially during the project period, from 7.9% to 4.4% in the intervention NHs (a relative reduction of 44%), and decreased in the nonintervention NHs from 7.0% to 4.9% (a relative reduction of 30%). Fall rates remained stable in the intervention NHs (17.3 falls/100 residents per month at start and 16.4 falls/100 residents per month at end), whereas fall rates increased 26% in the NHs not implementing the FMP (from 15.0 falls/100 residents/per month to 18.9 falls/100 residents per month).

CONCLUSION: Implementation was associated with significantly improved care process documentation and a stable fall rate during a period of substantial reduction in the use of physical restraints. In contrast, fall rates increased in NHs owned by the same organization that did not implement the FMP. The FMP may be a helpful tool for NHs to manage fall risk while attempting to reduce physical restraint use in response to the Centers for Medicare and Medicaid Services quality initiatives. J Am Geriatr Soc 2007;55(3):342-349.

Consequences of an Intervention to Reduce Restrictive Side Rail Use in Nursing Homes
Elizabeth Capezuti, PhD, RN, Laura M. Wagner, PhD, RN, Barbara L. Brush, PhD, RN, Marie Boltz, MSN, RN, Susan Renz, MSN, RN, and Karen A. Talerico, PhD, RN

OBJECTIVES: To examine the effect of an advanced practice nurse (APN) intervention on restrictive side rail usage in four nursing homes and with a sample of 251 residents. A secondary question explored the association between restrictive side rail reduction and bed-related falls.

DESIGN: Pre- and posttest design.

SETTING: Four urban nursing homes.

PARTICIPANTS: All nursing home residents present in the nursing home at three time points (n=710, 719, and 707) and a subset of residents (n=251) with restrictive side rail use at baseline.

INTERVENTION: APN consultation with individual residents and facility-wide education and consultation.

MEASUREMENTS: Direct observation of side rail status, resident and nurse interview for functional status, mobility, cognition, behavioral symptoms, medical record review for demographics and treatment information, and incident reports for fall data.

RESULTS:At the institutional level, one of the four nursing homes significantly reduced restrictive side rail use (P=.01). At the individual participant level, 51.4% (n=130) reduced restrictive side rail use. For the group that reduced restrictive side rails, there was a significantly (P<.001) reduced fall rate (−0.053; 95% confidence interval (CI)=−0.083 to −0.024), whereas the group that continued restrictive side rail did not demonstrate a significantly (P=.17) reduced fall rate (−0.013; 95% CI=−0.056–0.030).

CONCLUSION: An APN consultation model can safely reduce side rail use. Restrictive side rail reduction does not lead to an increase in bed-related falls. Although side rails serve many purposes, routine use of these devices to restrict voluntary movement and prevent falls is not supported. J Am Geriatr Soc 2007;55(3):334-341.