Nursing assistants detect behavior changes in nursing home residents that precede acute illness: development and validation of an illness warning instrument

J Am Geriatr Soc. 2000 Sep;48(9):1086-91. doi: 10.1111/j.1532-5415.2000.tb04784.x.


Background: Acute illness causes considerable morbidity and mortality in nursing home residents but is often difficult to recognize early. Nursing assistants often notice early signs of acute illness but do not methodically document or communicate their observations with medical staff.

Objective: To enhance nursing assistants' observation and documentation of signs of acute illness by developing a validated, standardized instrument for communication with medical staff.

Design: Observational cohort study.

Setting: Urban not-for-profit nursing home.

Subjects and methods: Candidate instrument items were generated in focus group interviews with nursing home staff. Twenty-three nursing assistants completed the instrument on 74 nursing home residents over 4 weeks. Acute illness, the primary outcome, was identified by nurse report and chart review and determined according to preset criteria.

Measurements: Predictive validity was assessed by determining the relationship between instrument responses and development of acute illness within 7 days. Interobserver agreement was calculated between morning and afternoon nursing assistants' responses. Convergent validity was assessed by comparing instrument responses with three standard status indicators.

Results: The instrument consisted of 12 items that assessed behavioral and functional status changes. Residents with an instrument-recorded change were more likely to develop an acute illness within 7 days than those with no change (risk ratio 4.1, 95% confidence interval 2.6, 6.3). A final five-item instrument had a sensitivity of 53% and a specificity of 93% for acute illness. Nursing assistants' documentation of signs of illness preceded chart documentation by an average of 5 days. Interobserver agreement between morning and afternoon nursing assistants was 76%. Correlation of responses with standard indicators of functional, mental, and global status was high.

Conclusions: A new instrument developed for nursing assistants to document behavioral and functional status changes in nursing home residents demonstrates fair sensitivity and high specificity for acute illness. Close monitoring of patients with a positive instrument might avert morbidity and mortality from acute illness by allowing earlier treatment.

Publication types

  • Research Support, U.S. Gov't, Non-P.H.S.
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Activities of Daily Living*
  • Acute Disease / nursing*
  • Acute Disease / psychology
  • Aged
  • Aged, 80 and over
  • Cognition*
  • Documentation / standards
  • Female
  • Focus Groups
  • Geriatric Assessment*
  • Health Status Indicators
  • Humans
  • Male
  • Middle Aged
  • Nursing Assessment / methods*
  • Nursing Assistants*
  • Nursing Homes*
  • Nursing Records / standards
  • Predictive Value of Tests
  • Reproducibility of Results
  • Risk Factors