Development and clinical application of a computer-aided real-time feedback system for detecting in-bed physical activities

Highlights

• The computer-aided real-time bed activities feedback system can be used to provide real-time detection and determination of changes in patients’ posture and state, and provide feedback of information about postural changes.

• The computer-aided real-time bed activities feedback system demonstrates considerably satisfactory validity and test-retest reliability for in-bed posture changes determination.

• The computer-aided real-time bed activities feedback system's real-time posture changes determination is conducive to issuing warnings in real time against dangerous postures of long-term bedridden patients or the disabled in the future, so as to prevent falls or accidental injuries from happening.

Abstract

Background and objective

Being bedridden long-term can cause deterioration in patients’ physiological function and performance, limiting daily activities and increasing the incidence of falls and other accidental injuries. Little research has been carried out in designing effective detecting systems to monitor the posture and status of bedridden patients and to provide accurate real-time feedback on posture. The purposes of this research were to develop a computer-aided system for real-time detection of physical activities in bed and to validate the system's validity and test–retest reliability in determining eight postures: motion leftward/rightward, turning over leftward/rightward, getting up leftward/rightward, and getting off the bed leftward/rightward.

Methods

The in-bed physical activity detecting system consists mainly of a clinical sickbed, signal amplifier, a data acquisition (DAQ) system, and operating software for computing and determining postural changes associated with four load cell sensing components. Thirty healthy subjects (15 males and 15 females, mean age = 27.8 ± 5.3 years) participated in the study. All subjects were asked to execute eight in-bed activities in a random order and to participate in an evaluation of the test–retest reliability of the results 14 days later. Spearman's rank correlation coefficient was used to compare the system's determinations of postural states with researchers’ recordings of postural changes. The test–retest reliability of the system's ability to determine postures was analyzed using the interclass correlation coefficient ICC(3,1).

Results

The system was found to exhibit high validity and accuracy (r = 0.928, p < 0.001; accuracy rate: 87.9%) in determining in-bed displacement, turning over, sitting up, and getting off the bed. The system was particularly accurate in detecting motion rightward (90%), turning over leftward (83%), sitting up leftward or rightward (87–93%), and getting off the bed (100%). The test–retest reliability ICC(3,1) value was 0.968 (p < 0.001).

Conclusions

The system developed in this study exhibits satisfactory validity and reliability in detecting changes in-bed body postures and can be beneficial in assisting caregivers and clinical nursing staff in detecting the in-bed physical activities of bedridden patients and in developing fall prevention warning systems.

Introduction

Many frail inpatients have the concomitant problem of low mobility, which can result in the morbid phenomenon of being bedridden for long periods of time. This phenomenon occurs for approximately one-third of all inpatients [1], of whom approximately 23%–33% are seniors [2], [3]. Whether for frail bedridden seniors or young inpatients, being bedridden for long periods of time can seriously impact the capacity for physical activities, and this impact becomes progressively exacerbated as the number of days in the hospital increases [4]. In addition to such factors as a patient's age or disease, other contributors to a patient's functional decline in daily life may include being required to lie in bed after special medical treatment [5], [6], [7]. For instance, a patient's movement may be restricted or the patient may be treated with a fixation device to prevent injury after an operation [1], [6], [8]. Studies have shown that even when patients have their disease under control and have resumed some activities, even demonstrating the ability to walk, the duration of confinement to bed still accounts for more than 60% of the typical duration of hospitalization [9], [10]. Early-stage research has shown that being bedridden for long periods of time can not only lead to deterioration in patients’ physiological functions but also affect functional performance in daily life [9], even increasing the number of days in the hospital or in long-term care institutions, and raising the risk of death and the incidence of relevant complications and accidental injuries [8], [11], [12], [13]. Furthermore, bedridden patients are susceptible to cognitive disorders, dizziness, and physical weakness, among other effects on their physiological functions, due to drug side effects [12], [13], resulting in insufficient blood supply to the head following rapid postural changes or dizziness caused by balance dysfunction of the inner ear. These effects may result in accidental injuries if the patient falls from the bed to the floor [14].

An inpatient is most likely to fall and become injured at the bedside, mainly because of the unstable position of the body's center of gravity when the inpatient attempts to rise from the bed and walk or grasp the edge of the bed to stand up [15]. Between approximately 13 and 20% of inpatients suffer at least one fall [14]. Studies indicate that such falls result in death in 13% of patients and physical injuries, such as fractures, joint dislocations, lacerations, and bruising, in 25% of patients [14]. Falls may also cause psychological trepidation and impact functionality in daily life; hence, patients may reduce their participation in daily socializing activities for fear of falls [16], [17], [18], [19], increasing the burden on family, medical personnel, and the medical care service system [20], as well as raising medical expenditures [12], [14], [21], [22]. Therefore, many clinical institutions have equipped their sickbeds with protective handrails or hand grips to prevent patients from accidentally injuring themselves as a result of falling from the bedside [23]. However, research has shown that even equipment with bedside handrails remains incapable of significantly reducing the incidence of falls [23]. For this reason, clinical caregivers need to develop strategies to monitor changes in patients’ in-bed activities and prevent bedside fall accidents.

There are many clinical methods for monitoring inpatients’ in-bed activities and postural changes and for preventing bedside falls, such as case records, anti-fall bracelets, and regular ward rounds [10]. However, studies have shown that these methods have many limitations, including deficiencies in recording patients’ activities whereby transient bedside activities by patients are overlooked or recorded erroneously [10]. As the care resources for the elderly become scarcer because of the aging population trends, relevant studies have also noted that a small number of clinical caregivers who take care of numerous elderly inpatients, especially caregivers who are particularly busy at night, are more likely to overlook dangerous accidents in which elderly patients fall when getting out of bed or slide down to the floor from the edge of the bed [24]. Therefore, relevant studies have recommended the use of a detecting system with real-time sensing of posture changes and providing monitoring information to warn caregivers in advance of patients’ potentially dangerous actions. Such a system should be adjustable to the status in different individual cases [14], [23], [24].

The advantages of integrating sensors to monitor patients’ changes between different postures include continuous monitoring, no need for in-person observation, and objective measurement records [10]. For this reason, early studies have integrated image-sensing components and analytical techniques to determine bedridden patients’ in-bed activities, providing caregivers with monitoring information about dangerous actions or postural changes in individual cases [25]. However, because of concerns over invasion of individual privacy, real-time image monitoring is not recommended for use with clinical patients who are bedridden [26]. In recent years, relevant research has developed lightweight, wearable measurement tools to record and analyze information about individual actions in real time and thereby determine postural changes in or off the bed, such as lying down, sitting up, getting off the bed and standing up, etc. [10], [27], [28], [29], [30], [31], [32]. There have even been studies in which wireless communications technology has been used to transmit data acquired through measurement to a computer, allowing caregivers to carry out follow-up analysis and long-term monitoring [10], [22], [24], [31].

Nonetheless, there may be some limitations to the clinical application of relevant sensing components in posture change determination and detection, including interference with patients’ posture-changing processes, possible compression and injury due to oversized sensing components, and the potential for accidental falls when a patient tries to pick up sensing components that have fallen off [24]. These limitations would affect a patient's daily physical activities [25]. Furthermore, excessive variance in the measurement values of falls by the sensing components [33], errors in the posture change determination formula, oversensitive threshold settings, excessively low sampling frequencies, and other factors may also affect the system's ability to measure postural changes accurately and produce anti-fall warnings effectively [25], [32], [33], [34], [35]. For instance, the accuracy of chest-worn detecting systems in monitoring postural changes is in the range of 72–89.3% [11], [22], [25]. Furthermore, most measuring systems do not base their actual measurements on real hospital sickbeds [36], and errors in operation or placement of the sensing components can also lead to incorrect determination of posture changes [16] and affect the accuracy of postural change detection [22].

Therefore, some studies have employed a metal bed surface with pressure-sensing components for detecting postural changes. However, this type of equipment has shortcomings, including patient discomfort, reduced sensing component lifespan, and interference with data measurement resulting from frequent compression of sensing components due to postural changes [16], [24]. Research has also shown that wearable sensing components cannot be used for very long durations to monitor in-bed posture changes, as battery replacement is required approximately every 7–10 days [22], [36], whereas only monitoring systems or instruments such as computer-aided monitoring systems that can measure patients’ physical postural changes over long periods of time can effectively prevent bedridden patients from suffering accidental injuries [12].

The objectives of this research were to develop a computer-aided real-time system for detecting in-bed physical activities and to establish tests of the validity and reliability of systematic measurement of in-bed postural changes, to highlight the value of clinical application of the detection of in-bed postural changes and warnings against falls for frail bedridden patients.