Background and importance Fall incidents are common among nursing home patients. Different tools have been developed in the prevention of fall incidents but with unsatisfactory results.

Aim and objectives To develop (part I) and validate (part II) a clinical rule (CR) that can predict a fall risk in nursing home patients.

Material and methods The study was conducted in two parts. In part I, the variables which could lead to an increased risk of falls were determined and implemented in the predictive clinical rule. Subsequently, data from a retrospective cohort study were used to validate the developed clinical rule.

Multiple linear regression analysis was conducted to identify the fall risk variables in part I. With these, a predictive fall risk algorithm was developed where the overall prediction quality was assessed using the area under the receiver operating characteristic curve (AUROC), and a cut-off value was determined for the predicted risk ensuring a sensitivity ≥0.85. This prediction model and cut-off value were externally validated in part II.

Results A total of 1668 (824 in part I, 844 in part II) nursing home patients were included in the study. Eleven fall risk variables were identified in part I. The externally validated AUROC of the prediction model, obtained in part II, was 0.603 (95% CI 0.565–0.641) with a sensitivity of 83.41% (95% CI 79.44–86.76%) and a specificity of 27.25% (95% CI 23.11–31.81%).

Conclusion and relevance Medication data and patient characteristics were not sufficient to develop a successful clinical rule with a high sensitivity and specificity to predict the risk of falls.

References and/or acknowledgements No conflict of interest.