EUVIMEDEUVIMED entdecken
Crossref

A Mixed Effects Model Analysis for Inpatient Falls Using Health Record Data From 72 Hospitals

Cynthia M LaFond iD, Ana Christina Perez Moreno, Vallire Hooper, Susan Okuno-Jones, Mary Sitterding

Journal of Clinical Nursing · 2026

DOI: 10.1111/jocn.70204 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Clinical Nursing, 2026. Verfasst von Cynthia M LaFond und 4 weiteren Autor:innen. Beteiligte Einrichtung: Ascension, St. Louis, Missouri, USA. Hauptthemen laut MEDLINE: Accidental Falls, Inpatients, Hospitals. Die Arbeit nennt 55 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aim/Design This retrospective cohort study evaluated the relationship between patient falls, Morse Fall Scale (MFS) items, patient demographics, length of stay and hospital site.

Methods

Data were acquired from 72 hospitals in a health system. Logistic regression models were conducted including MFS items, demographics, length of stay, and interaction terms. The final mixed effects logistic regression model included significant patient‐level covariates as fixed effects and hospital site as a random effect.

Results

6531 of 978,920 total admissions included a patient fall. Four MFS items (fall history, secondary diagnosis, gait weak/impaired, mental status—overestimates/forgets limitations) and three demographic items (male gender, increased age, longer length of stay) were associated with increased likelihood of falling. Two MFS items (ambulatory aids, intravenous therapy/lock) and Hispanic ethnicity were associated with decreased risk of falling. An interaction effect was present between male gender and mental status. Males who overestimate/forget limitations had 3.16 times higher odds of falling than females oriented to their own ability. The proportion of variance in falls between hospitals was 0.23 and the median odds ratio (MOR) 1.57.

Conclusion

This study uniquely assessed fall risk at the level of the patient and hospital, using data from nearly 1 million admissions at 72 hospitals. Controlling for patient characteristics, results demonstrate variability in fall risk among hospitals. Research informing hospital differences as well as gender and racial/ethnic differences in falls is needed to identify appropriate interventions. Implications for Patient Care As hospitals increasingly adopt risk‐directed fall prevention, assessment tools should be re‐evaluated for clinical utility and corresponding prevention practices. The MFS may be enhanced by removing intravenous lock as a risk and screening for additional risks such as medications and medical equipment. Quality improvement efforts must also consider the hospital's environment and processes that may further contribute to fall risk. Reporting Method Authors adhered to STROBE guidelines for reporting. Patient Contribution No Patient or Public Contribution.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Accidental FallsInpatientsHospitalsHumansMaleFemaleRetrospective StudiesAgedMiddle AgedAged, 80 and overRisk Factors

Stichwörter der Autor:innen

Morse Fall Scaleaccidental falls/prevention and controlfall risk assessment or risk factorshospitalsnursing carequality in health care

Autor:innen und Institutionen

  • Ascension, St. Louis, Missouri, USA.

ORCID-Profile: Cynthia M LaFond

Zitieren

APA

LaFond, C. M., Moreno, A. C. P., Hooper, V., Okuno-Jones, S., & Sitterding, M. (2026). A Mixed Effects Model Analysis for Inpatient Falls Using Health Record Data From 72 Hospitals. Journal of Clinical Nursing. https://doi.org/10.1111/jocn.70204

Vancouver

LaFond CM, Moreno ACP, Hooper V, Okuno-Jones S, Sitterding M. A Mixed Effects Model Analysis for Inpatient Falls Using Health Record Data From 72 Hospitals. Journal of Clinical Nursing. 2026. doi:10.1111/jocn.70204

BibTeX
@article{LaFond2026,
  author = {LaFond, Cynthia M and Moreno, Ana Christina Perez and Hooper, Vallire and Okuno-Jones, Susan and Sitterding, Mary},
  title = {A Mixed Effects Model Analysis for Inpatient Falls Using Health Record Data From 72 Hospitals},
  journal = {Journal of Clinical Nursing},
  year = {2026},
  doi = {10.1111/jocn.70204}
}
Kostenlos starten →