A predictive model for fall risk in hospitalized adults: A case–control study

Aim To develop and validate a predictive model for falls in hospitalized adult clinical and surgical patients, assessing intrinsic (i.e. patient-related) and extrinsic factors (i.e. care process-related). Background To identify factors predictive of falls and enable appropriate management of fall risk it is necessary to understand patient and environmental factors, along with care delivery processes. Design A matched case-control study. Methods This study was conducted in the medical and surgical wards of a Brazilian teaching hospital. The sample included 536 patients, with data collected in 2013-2014. Data analysis included descriptive statistics and conditional logistic regression. Cases of patients aged 18 years or older who fell while hospitalized were included. One patient who did not fall during hospitalization, matched by sex, ward and admission date, was selected as a control for each included case. Results The SAK Fall Scale (Severo-Almeida-Kuchenbecker) was developed and validated. The scale includes seven variables: disorientation/confusion, frequent urination, walking limitations, lack of caregiver, postoperative status, previous falls and number of medications administered within 72 hr prior to the fall. This scale showed acceptable predictive accuracy. Conclusions The newly developed SAK Fall Scale includes five intrinsic and two extrinsic variables and differs from other predictive scales for falls. The findings of this study are broad and the scale, which is easy to apply, can be used worldwide by nurses in health services. In advanced practice, the testing of a new model for fall risk contributes to preventive interventions and thus has an impact on patient safety.