Computational Barthel Index: an automated tool for assessing and predicting activities of daily living among nursing home patients

Abstract

Background Assessment of functional ability, including activities of daily living (ADLs), is a manual process completed by skilled health professionals. In the presented research, an automated decision support tool, the Computational Barthel Index Tool (CBIT), was constructed that can automatically assess and predict probabilities of current and future ADLs based on patients’ medical history. Methods The data used to construct the tool include the demographic information, inpatient and outpatient diagnosis codes, and reported disabilities of 181,213 residents of the Department of Veterans Affairs’ (VA) Community Living Centers. Supervised machine learning methods were applied to construct the CBIT. Temporal information about times from the first and the most recent occurrence of diagnoses was encoded. Ten-fold cross-validation was used to tune hyperparameters, and independent test sets were used to evaluate models using AUC, accuracy, recall and precision. Random forest achieved the best model quality. Models were calibrated using isotonic regression. Results The unabridged version of CBIT uses 578 patient characteristics and achieved average AUC of 0.94 (0.93–0.95), accuracy of 0.90 (0.89–0.91), precision of 0.91 (0.89–0.92), and recall of 0.90 (0.84–0.95) when re-evaluating patients. CBIT is also capable of predicting ADLs up to one year ahead, with accuracy decreasing over time, giving average AUC of 0.77 (0.73–0.79), accuracy of 0.73 (0.69–0.80), precision of 0.74 (0.66–0.81), and recall of 0.69 (0.34–0.96). A simplified version of CBIT with 50 top patient characteristics reached performance that does not significantly differ from full CBIT. Conclusion Discharge planners, disability application reviewers and clinicians evaluating comparative effectiveness of treatments can use CBIT to assess and predict information on functional status of patients.

Algorithmic harms and digital ageism in the use of surveillance technologies in nursing homes

“…Such technologies may also involve the application of natural language processing (chatbots, and social robots), or machine learning techniques to process continuously collected information about the movements, location, activities, and physiological state of older adults in their environment (Chandonnet, 2021;Orlov, 2021). There is also increasing interest in expanding the role of artificial intelligence (AI) in nursing homes to process information collected by a full range of technologies, including CCTV cameras, virtual assistants, and electronic health records, and to develop predictive algorithms and automated decision-making systems about care (Wojtusiak et al, 2021;Khan et al, 2022;Zhu et al, 2022). The push to adopt surveillance technologies in nursing homes is thus largely driven by their imagined future benefit for prevention and quality improvement through more timely identification of adverse events and personalization of care in the context of widespread staffing shortages.…”

Detecting disabilities in everyday life: evidence from a geriatric assessment

“…Using the 14 explanatory variables, selected under λ min , functional limitations in everyday life were identified with high accuracy (AUC = 0.91). Therefore, the potential of the joint use of the data-driven selected variables to capture ADL disability was supported by these higher detection accuracies compared to previous literature [ 7 – 9 , 34 , 46 ] when no prior observed ADL was taken into account [ 4 ].…”

“…These can be categorized into the four health domains of mobility (TUG, TMT, 1 TUG ,1 TMT ), cognitive function (MMSE, 1 MC ), nutritional status (MNA) and urinary incontinence (ICS, ICD 4 ). Wojtusiak et al, despite relying on a different methodological approach and using large-scale rather than clinical data, used a very similar variable choice, i.e., emphasizing the role of cognitive functions, age and urinary incontinence for ADL [ 4 ]. Despite broad agreement, these results show that different variables are required for targeted ADL detection depending on the research goal and clinical setting.…”

“…Previous research has adopted three major approaches to approximate ADL [ 4 ]. A first strand of literature used sensor data from wearable devices or smart home environments to obtain information on functional limitations [ 5 , 6 ].…”

“…A second line of research relied on large-scale data and machine learning algorithms to predict ADL. For example, Wojtusiak et al used patient records of approximately 200,000 patients in the US and identified 578 patient characteristics to explain ADL [ 4 ]. In addition to the fact that comparable datasets are not easily available, such a large number of explanatory variables is hardly suitable for daily clinical use, particularly in the outpatient sector.…”