Development of the ADFICE_IT Models for Predicting Falls and Recurrent Falls in Community-Dwelling Older Adults: Pooled Analyses of European Cohorts With Special Attention to Medication

Abstract

Background Use of fall prevention strategies requires detection of high-risk patients. Our goal was to develop prediction models for falls and recurrent falls in community-dwelling older adults and to improve upon previous models by using a large, pooled sample and by considering a wide range of candidate predictors, including medications. Methods Harmonized data from two Dutch (LASA, B-PROOF) and one German cohort (ActiFE Ulm) of adults aged ≥ 65 years were used to fit two logistic regression models: one for predicting any fall and another for predicting recurrent falls over one year. Model generalizability was assessed using internal-external cross-validation. Results Data of 5722 participants were included in the analyses, of whom 1868 (34.7%) endured at least one fall and 702 (13.8%) endured a recurrent fall. Positive predictors for any fall were: educational status, depression, verbal fluency, functional limitations, falls history, and use of antiepileptics and drugs for urinary frequency and incontinence; negative predictors were: body mass index (BMI), grip strength, systolic blood pressure, and smoking. Positive predictors for recurrent falls were: educational status, visual impairment, functional limitations, urinary incontinence, falls history, and use of anti-Parkinson drugs, antihistamines, and drugs for urinary frequency and incontinence; BMI was a negative predictor. The average C-statistic value was 0.65 for the model for any fall and 0.70 for the model for recurrent falls. Conclusions Compared with previous models, the model for recurrent falls performed favorably while the model for any fall performed similarly. Validation and optimization of the models in other populations is warranted.

Development of the ADFICE_IT clinical decision support system to assist deprescribing of fall-risk increasing drugs: A user-centered design approach

“…Regarding the knowledge base of the system, the final set of clinical rules covering 22 classes of FRIDs, with specific deprescribing advice based on diagnoses, lab values, and concurrent medications, including the final fall-risk prediction model were integrated into the CDSS. The final prediction model consisted of the following 14 predictors: educational status, depression, body mass index, grip strength, gait speed, number of functional limitations, systolic blood pressure, at least one fall in the previous 12 months, at least two falls in the previous 12 months, fear of falling, smoking status, use of calcium channel blockers, use of antiepileptics, and use of drugs for urinary frequency and incontinence (see Van de Loo et al for detailed results) [35].…”

“…The aim of this phase was to cultivate a robust (theoretical) understanding about how to develop the CDSS components of the AD F ICE_IT intervention [30], which consisted of the (1) development of the user interface, (2) development of the clinical knowledge base, (3) development of the prediction model, and (4) development of the software. A detailed description of the methods for the systematic literature review, European online survey, modified Delphi study, and development of the fall-risk prediction model have been published elsewhere by the research team [32–35].…”

“…To enable the generation of a patient’s risk of falling within 12 months, a fall-risk prediction model was developed by Van de Loo et al to fit the context of this study using a harmonized dataset ( n = 5722) of two Dutch cohorts and a German cohort study of community-dwelling older adults (65+) [35]. This prediction model comprises a part of the CDSS’s knowledge base.…”

Predicting mortality risk after a fall in older adults using health care spending patterns: a population-based cohort study

“…The model was internally validated and exhibited potential clinical benefit, high discrimination and good calibration for predicting 1-year mortality. However, in this representative, nationwide cohort of over 30,000 older individuals, the prediction of our secondary outcome, an endpoint with ongoing clinical interest, namely recurrent falls [ 41–43 ], was at best modest.…”

Effects of a clinical decision support system and patient portal for preventing medication-related falls in older fallers: Protocol of a cluster randomized controlled trial with embedded process and economic evaluations (ADFICE_IT)

“…The design and the development of the ADFICE_IT intervention was guided by the Medical Research Council (MRC) Framework for Complex Interventions (46). In the preparation phase of the MRC framework, we developed a prediction model for estimating a patient's risk of falling (47).…”