Study of Individualization and Bias in Nursing Home Fall Prevention Practices.

Journal Article (Journal Article;Multicenter Study)

OBJECTIVES: Little is known about how nursing home staff use resident characteristics to individualize care delivery or whether care is affected by implicit bias. DESIGN: Randomized factorial clinical vignette survey. SETTING: Sixteen nursing homes in North Carolina. PARTICIPANTS: Nursing, rehabilitation, and social services staff (n = 433). MEASUREMENTS: Vignettes describing hypothetical residents were generated from a matrix of clinical and demographic characteristics. Resident age, race and gender were suggested by a photo. Participants completed up to four randomly assigned vignettes (n = 1615), rating the likelihood that 12 fall prevention activities would be used for the resident. Fixed and random effects mixed model analysis examined the impact of vignette resident characteristics and staff characteristics on four intervention categories. RESULTS: Staff reported a higher likelihood of fall prevention activities in all four categories for residents with a prior fall (0.2-0.5 points higher, 10 point scale, P < 0.05), but other risk factors did not affect scores. There was little evidence of individualization; only dementia increased the reported likelihood of environmental modification (0.3, P < 0.001, 95% CI 0.2-0.5). Individualization did not vary with staff licensure category or clinical experience. Registered nurses consistently reported higher likelihoods of all fall prevention activities than did licensed practical nurses, unlicensed staff and other professional staff (1.0-2.7 points, P < 0.001 to 0.005). There was a small degree of implicit racial bias; staff indicated that environmental modification would be less likely to occur in otherwise identical vignettes including a photo of a black rather than a white resident (-0.2 points, 95% CI -0.3 to -0.1). CONCLUSION: Nursing home staff report a standardized approach to fall prevention without individualization. We found a small impact from implicit racial bias that should be further explored.

Full Text

Duke Authors

Cited Authors

  • Colón-Emeric, CS; Corazzini, K; McConnell, E; Pan, W; Toles, M; Hall, R; Batchelor-Murphy, M; Yap, TL; Anderson, AL; Burd, A; Anderson, RA

Published Date

  • April 2017

Published In

Volume / Issue

  • 65 / 4

Start / End Page

  • 815 - 821

PubMed ID

  • 28186618

Pubmed Central ID

  • PMC5397322

Electronic International Standard Serial Number (EISSN)

  • 1532-5415

Digital Object Identifier (DOI)

  • 10.1111/jgs.14675


  • eng

Conference Location

  • United States