Clinical Outcomes of Passive Sensors in Remote Monitoring: A Systematic Review
1
School of Clinical Medicine, University of Cambridge, Cambridge CB2 0SP, UK
2
Department of General Surgery, Luton & Dunstable Hospital, Luton LU4 0DZ, UK
3
Department of Surgery, Royal Free Hospital, London NW3 2QG, UK
4
Department of Surgery and Cancer, Imperial College London, St. Mary’s Hospital, London W2 1NY, UK
*
Author to whom correspondence should be addressed.
Sensors 2025, 25(11), 3285; https://doi.org/10.3390/s25113285
Submission received: 6 April 2025
/
Revised: 16 May 2025
/
Accepted: 22 May 2025
/
Published: 23 May 2025
(This article belongs to the Special Issue Applications and Design of Digital Systems for Sensors and Sensing Technologies)
Abstract
Remote monitoring technologies have transformed healthcare delivery by enabling the in-home management of chronic conditions, improving patient autonomy, and supporting clinical oversight. Passive sensing, a subset of remote monitoring, facilitates unobtrusive, real-time data collection without active user engagement. Leveraging devices such as smartphones, wearables, and smart home sensors, these technologies offer advantages over traditional self-reports and intermittent evaluations by capturing behavioural, physiological, and environmental metrics. This systematic review evaluates the clinical utility of passive sensing technologies used in remote monitoring, with a specific emphasis on their impact on clinical outcomes and feasibility in real-world healthcare settings. A PRISMA-guided search identified 26 studies addressing conditions such as Parkinson’s disease, dementia, cancer, cardiopulmonary disorders, and musculoskeletal issues. Findings demonstrated significant correlations between sensor-derived metrics and clinical assessments, validating their potential as digital biomarkers. These technologies demonstrated feasibility and ecological validity in capturing continuous, real-world health data and offer a unified framework for enhancing patient care through three main applications: monitoring chronic disease progression, detecting acute health deterioration, and supporting therapeutic interventions. For example, these technologies successfully identified gait speed changes in Parkinson’s disease, tracked symptom fluctuations in cancer patients, and provided real-time alerts for acute events such as heart failure decompensation. Challenges included long-term adherence, scalability, data integration, security, and ownership. Future research should prioritise validation across diverse settings, long-term impact assessment, and integration into clinical workflows to maximise their utility.
1. Introduction
The growing burden of chronic diseases, an ageing global population, and rising healthcare costs have intensified the need for scalable, cost-effective healthcare solutions [1]. Remote monitoring has emerged as a pivotal strategy, shifting care beyond traditional clinical settings and into patients’ homes [2]. Among these innovations, passive sensing technologies have gained traction, offering continuous, real-world health data without requiring active user engagement [3].
Passive remote integrates seamlessly into daily life through wearables, smartphones, smart home systems (e.g., Amazon Alexa, Google Nest), and ambient sensors, to collect continuous, real-world data [4,5,6]. Unlike self-reporting methods, which require user engagement and are susceptible to recall bias, passive sensing enables non-intrusive, continuous health monitoring. This is particularly beneficial for populations with cognitive impairments or conditions that hinder compliance [7,8].
These technologies—spanning Internet of Things (IoT) devices and smart home sensors—offer a rich spectrum of insights by monitoring physiological parameters, environmental conditions, and behavioural patterns [9]. Real-time or near real-time data from passive sensors holds the potential to transform healthcare by enabling early detection of clinical deterioration, facilitating timely interventions, and ultimately improving patient outcomes [3,10]. Additionally, sensor-derived data can produce digital biomarkers, providing longitudinal insights into disease progression and patient activity [7,11]. Such continuous, unobtrusive monitoring outperforms clinic-based assessments and self-reports, which are limited to intermittent snapshots and prone to subjectivity. When combined with predictive analytics and machine learning, passive sensing has the potential to shift healthcare toward a more proactive, prevention-focused model [12].
In recent years, a growing body of evidence has emerged around various remote monitoring interventions. Initial reports suggested limited clinical impact of wearable devices on individuals in the community with chronic diseases [5], but a recent analysis of digital sensor alerting systems in remote monitoring supported their use, having found a reduction in hospitalisation and in all-cause mortality [10]. Similarly, remote patient monitoring (RPM) interventions resulted in improved patient safety and functional status, and reduced cost-related outcomes such as risk of hospital readmission and length of stay [3]. Moreover, RPM was shown to reduce acute care use for patients with cardiovascular disease and COPD [13].
Despite the rapid advancement of passive sensing technologies, there is limited comprehensive evidence assessing their impact on clinical outcomes. Studies have focused on the role of smartphones-based passive sensing, with a paucity of evidence for other modalities [4,14,15]. Furthermore, there has been no analysis of the role of passive sensing. As such, the aim for this study was to synthesise current evidence on passive sensing in remote monitoring, evaluating its impact on clinical outcomes and identifying key gaps in implementation and effectiveness.
2. Materials and Methods
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines [16]. The review was registered at the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42024511755).
This review sought to answer the following questions:
Which passive sensing technologies are used in remote monitoring?
What is the impact of remote monitoring using passive sensing technologies on clinical and healthcare related outcomes?
A systematic search, with expert librarian support, was performed using the following electronic databases: Ovid in Medline, EMBASE, health management information consortium (HMIC), and PsycINFO. The search was undertaken using a list of key terms combined with the appropriate Boolean operators. Key terms included “remote”, “home”, “sens*”, “monitor*”, and “passive”. The complete search strategy is available in Appendix A. Further studies not captured by the search were identified through bibliometric cross-referencing. Grey literature was additionally searched.
All identified studies were uploaded to Covidence (Melbourne, Australia, https://app.covidence.org/sign_in, accessed on 1 August 2024), a Cochrane supported systematic review package tool [17]. Initial screening was conducted by two investigators (ER and SZ) to determine if the eligibility criteria were met. Discrepancies were resolved by discussion. Studies meeting the inclusion criteria underwent full-text screening; supplemental references were scrutinised for additional relevant articles.
Studies published detailing passive sensing technologies in the context of remote monitoring were included. The last search was performed in April 2024 with no language restrictions being placed.
Abstracts, conference articles, opinion pieces, editorials, case studies, reviews, and meta-analyses were excluded from the final review. Studies with inadequate published data relating to the outcome measures were additionally excluded.
The primary outcome measure was the impact of passive sensing technologies on clinical and healthcare-related outcomes, including early detection of health deterioration, functional monitoring, and reduction in hospitalizations. Secondary outcome measures included feasibility, usability, and adherence to passive sensing devices, as well as participant-reported challenges such as privacy concerns and technical difficulties.
All included study characteristics were extracted independently by two investigators with consensus achieved. Disagreement between the two reviewers was resolved by discussion. All full text reports of studies identified as potentially eligible after title and abstract review were obtained for further review.
Non-randomised trials were assessed using the Newcastle–Ottawa scale [18]. It comprises three variables: (i) patient selection; (ii) comparability of study groups; and (iii) assessment of outcomes. Scores range from 0 to 9: scores ≤ 3 were considered low quality, between 4 and 6 moderate quality, and ≥7 high quality. Randomised controlled trials were assessed using the Cochrane risk of bias 2 tool [19]. This assessed five possible sources of bias: (i) randomization, (ii) deviations from intended interventions, (iii) missing outcome data, (iv) measurement of outcomes, and (v) selection of the reported result. Quality assessment was performed by one reviewer (ER) and validated by a second (SZ).
3. Results
3.1. Study Characteristics
A total of 276 results were retrieved through the literature search, of which 53 were removed due to duplications and for not meeting our eligibility criteria. Full text review was performed for 103 articles resulting in the inclusion of 26 articles in our review. The included studies were conducted in diverse global settings, including the United States (12 studies), Europe (8 studies across the United Kingdom, Germany, the Netherlands, and Switzerland), Asia (4 studies in Singapore, Japan, and China), and Australia (2 studies). The study characteristics are shown in Table 1. A PRISMA flow diagram can be seen in Figure 1.
3.2. Types of Passive Sensing Technology
A variety of passive sensing technologies were employed for remote monitoring (Table 2). Wrist-worn devices were the most commonly used wearables, capturing physical activity, sleep patterns, vital signs, and functional measures such as tremor severity and shoulder range of motion. Smartphones were also utilised, extending monitoring capabilities beyond activity and gait to include cognitive function, mood variability, and speech characteristics, thereby providing a more holistic assessment of patient well-being. Measures of physical activity and sleep relied principally on accelerometer and gyroscope data within smartphones and wearables.
In-home sensor systems offered additional, continuous monitoring. These included infrared cameras, infrared motion sensors, and radio-based sensors, which detected daily activity patterns, room transitions, and sleep behaviours. When integrated with door contact sensors and bed sensors, these systems provided spatiotemporally precise movement data, enhancing the characterisation of home-based activity. Infrared sensors sampled at lower frequencies (e.g., 0.5 Hz) and detected larger motions and entrances into rooms. Radio-based sensors, such as the Emerald device, transmitted low power wireless signals and analysed their reflections from nearby humans or objects using machine learning a to monitor sleep, gait speed, location, and activity. In addition to behavioural monitoring, in-home sensors also facilitated physiological assessments. Bed sensors and radio-based systems enabled the continuous tracking of respiratory rate, with deviations from baseline serving as potential indicators of clinical deterioration. Bed sensors placed underneath mattresses used highly sensitive quasi-piezoelectric films capable of measuring even slight pressure differences as produced by the beating heart (ballistocardiography).
3.3. Patient Populations
3.3.1. Conditions Studied
The practice of remote monitoring using passive sensors occurred most commonly (six studies) in the setting of Parkinson’s disease (PD) [20,21,22,23,24,25]. Remote monitoring of older adults [8,32,33], patients with mild cognitive impairment [29,30,31], cancer patients [35,36,37], patients with musculoskeletal problems [41,42,43], and patients with cardiopulmonary conditions [38,39,40] was explored in three studies each. Dementia [27,28] was monitored in two studies. Patients with depression [34], multiple sclerosis (MS) [26], and older adults recovering from traumatic brain injury (TBI) [7] were monitored in one study each.
3.3.2. Parkinson’s Disease and Multiple Sclerosis
Six studies on Parkinson’s disease [20,21,22,23,24,25] and one on multiple sclerosis [26] evaluated passive sensing for remote monitoring (Table 1). Technologies included smartwatches [25], smartphones [23,26], hybrid platforms [20,21], and in-home radio-wave sensors [22,24]. PD studies aimed to capture tremor, gait, and activity metrics, with compliance ranging from 81% for daily smartwatch wear [20] to 61% test session completion over six months [23], indicating feasibility across short and extended periods.
Algorithms processed sensor data into clinically meaningful measures, with several studies validating these against standard assessments. This involved both the conversion of raw sensor data into a clinical measure (e.g., tremor score) and the use of machine learning to categorise sensor data into activity patterns. Fay-Karmon et al. and Gatsios et al. found significant correlations between sensor-derived tremor scores and MDS-UPDRS clinician-rated measures [20,21], while both Kabelac et al. and Liu et al. demonstrated that radio-based in-home gait speed measurements aligned with MDS-UPDRS scores [22,24]. In MS, Lam et al. reported that smartphone keystroke dynamics correlated with clinical motor and cognitive tests, underscoring the potential of passive sensing for tracking disease-specific outcomes [26].
Sensor data distinguished patient and control groups, with PD participants displaying lower gait activity and slower transitions than controls [23,24], and medication-induced symptom fluctuations in PD were detectable with high temporal sensitivity [20,24]. However, predictive power for individual clinical scores remains limited, as shown by Schalkamp et al. [25], highlighting the need for further algorithm refinement.
3.3.3. Dementia, Mild Cognitive Impairment, Older Adults, and Depression
Ten studies evaluated passive sensing for remote monitoring in patients with dementia [27,28], mild cognitive impairment (MCI) [29,30,31], older adults [7,8,32,33], and those with depression [34] (Table 1). Passive sensors, primarily in-home motion sensors, enabled tracking of daily activities, sleep, and vital signs, highlighting their potential in monitoring cognitive and physical decline, acute health deterioration, and functional independence.
Passive monitoring effectively captured activity patterns and cognitive function in dementia and MCI populations. Studies in MCI found subtle but informative differences in daily routines. Wu et al. observed that individuals with MCI had distinct diurnal patterns, such as increased bathroom and kitchen use compared to controls [31], while Muurling et al. noted that independent life space activity decreased among cohabiting MCI individuals [29]. Rawtaer et al. also identified trends where MCI participants had more sleep interruptions and more frequent medication lapses, suggesting changes in daily routines that passive sensing could monitor [30].
In dementia, Gaugler et al. found that activity deviations detected by sensors were communicated to caregivers, maintaining real-time awareness of patients’ well-being [27]. Similarly, Obayashi et al. showed improvements in functioning and independence among dementia patients monitored by passive sensors, with one participant even reducing reliance on night-time assistance [28]. These findings underscore the potential of passive sensors to promote independence and detect early signs of cognitive deterioration.
In older adults, passive sensing provided early detection of acute health issues and insights into health deterioration. Saner et al. reported on patients with heart failure, where episodes of decompensation were linked to reduced activity, increased heart rate, and disrupted sleep patterns [33]. Passive sensors also detected specific health events, such as pneumonia and worsening sleep disturbances [7], facilitating timely interventions. Finch et al. observed that participants with home monitoring had fewer emergency visits and hospital admissions, though cost reductions in healthcare use were not statistically significant [32].
In a study of individuals with depression, Leightley et al. noted a reduction in sleep duration during COVID-19 lockdowns, showing passive sensing’s capability to capture behavioural changes tied to mental health [34]. This study highlighted how passive sensing can monitor mood-related metrics over time, potentially aiding in the management of depressive disorders by detecting early shifts in sleep and activity patterns. Long-term continuous monitoring of such parameters may offer valuable insights and patterns of mental health conditions throughout a person’s life.
3.3.4. Cancer
Three studies explored passive sensing for remote monitoring in cancer patients and survivors, using wrist-worn accelerometers, smartphones, and Fitbit devices to assess physical function and symptom burden [35,36,37] (Table 1). Two studies relied on smartphone and Fitbit data [36,37] while one utilised a wrist-worn accelerometer [35].
Low et al. examined the validity of smartphone and wearable sensors to monitor physical function in older cancer survivors, finding that greater walking duration and speed, and less fragmentation, were significantly correlated with self-reported physical function and performance measures [37]. Dadhania et al., studying high-grade glioma (HGG) patients, found similar results; daily walking time correlated positively with quality of life and physical functioning. Machine learning algorithms further identified that HGG patients exhibited half the activity levels of healthy controls, experienced longer sleep durations, and showed marked declines in moderate activity levels during radiotherapy and disease progression [35].
In a study of pancreatic cancer patients, Low et al. used machine-learning models with mobile sensor data to predict symptom burden, demonstrating higher accuracy than non-sensor-based models. Key predictors of next-day symptom burden included features related to physical activity, heart rate, and location, underscoring the value of passive sensing in anticipating symptom fluctuations [36].
3.3.5. Cardiopulmonary Health
Three studies explored passive sensing for cardiopulmonary health, focusing on heart failure, hypertension, and COVID-19 recovery [38,39,40] (Table 1). Sensors included bed-based systems [38], ingestible sensors [39], and in-home radio-wave technology to capture data on respiratory rates, heart rates, body weight, and medication adherence [40].
Harrington et al. used a bed sensor to monitor heart failure patients, measuring body weight, respiratory signals, and heart rate. Validation against overnight sleep studies demonstrated close agreement with traditional respirometry and ECG, and the bed sensor effectively differentiated between central and obstructive sleep apnea. Long-term monitoring indicated that respiratory and heart rates returned to pre-surgical baselines over the course of two months, highlighting the sensor’s utility in tracking recovery [38].
Noble et al. assessed a digital health feedback system using ingestible sensors to monitor medication adherence in hypertensive patients. Over a two-week period, the system, combined with pharmacist feedback, led to a mean systolic blood pressure reduction of 7.9 mmHg, with individualised root causes of hypertension identified for each patient [39]. This approach underscores the potential of passive sensing to improve medication adherence and personalised care in hypertension management.
Zhang et al. employed a radio-wave-based in-home sensor to monitor respiratory and activity patterns in COVID-19 patients over extended periods. For one patient, a sudden respiratory rate increase coincided with hospitalisation, while stable respiratory rates in other patients indicated smooth recoveries [40]. These results demonstrate the feasibility of passive sensors in detecting early signs of deterioration during COVID-19 recovery.
3.3.6. Musculoskeletal Health
Three studies assessed passive sensing in musculoskeletal health, focusing on gait quality, shoulder motion, and sit-to-stand performance [41,42,43] (Table 1). These studies utilised algorithms to convert raw sensor data into clinically relevant metrics, such as gait-quality [43], three-dimensional shoulder motion [41], and performance in the five-times sit-to-stand (5 × STS) test [42].
In patients with adhesive capsulitis, Chen et al. demonstrated that motion sensor-assisted home rehabilitation significantly improved range of motion (ROM), pain levels, and exercise adherence compared to standard home exercises, underscoring the efficacy of sensor-assisted rehabilitation [41]. Perraudin et al. used wrist-worn accelerometers to evaluate the 5 × STS test in arthritis patients over four weeks, finding that test duration was significantly linked to pain and stiffness intensity reported in patient questionnaires, suggesting that passive sensing can capture relevant functional variations in arthritis [42].
Ribeiro-Castro et al. analysed gait quality in hip and knee arthroplasty patients using smartphone-derived walking metrics, finding that early, consistent walking sessions during the first 30 days post-surgery were positively associated with improved gait speed at 90 days [43]. This highlights the potential for passive sensing to monitor recovery trajectories and optimise early post-operative activity in musculoskeletal patients.
3.4. Participant Feedback
Participant feedback from various studies highlighted the perceived feasibility, benefits, and challenges of passive monitoring for patients and caregivers [8,21,22,27,28,30,33,39]. Feedback revealed both positive experiences and notable concerns, providing valuable insight into the acceptability and usability of these technologies.
Some participants described passive monitoring as unobtrusive, with little impact on their daily routines [22]. However, others reported the sensors to be inconvenient or disruptive, especially when devices required direct interaction [8,30]. Preferences leaned towards non-interactive sensors, particularly for older adults and those with chronic conditions [33], indicating that usability depends heavily on device design.
Caregiver responses were mixed; some found remote monitoring helpful for managing care when living separately from patients, reducing concerns and enabling timely interventions [28,30]. In contrast, caregivers living with or providing full-time in-person care perceived less benefit from monitoring and occasionally viewed it as an additional burden [27]. Increased caregiver burden was, however, positively associated with the likelihood of monitoring use, as noted by Gatsios et al. [21], suggesting that remote monitoring may be particularly valuable in high-burden situations.
3.5. Risk of Bias Assessment
Two randomised trials were evaluated with the Cochrane risk of bias 2 tool and were deemed to be at low risk [21,27]. The quality of 24 observational studies was assessed using the Newcastle–Ottawa Scale, with scores ranging from 3 to 6, indicating a moderate to high risk of bias. Two studies achieved the highest score of 6 [23,32], whilst three studies received the lowest score of 3 [38,39,43].
Exposure ascertainment was consistently well conducted across studies. However, external validity was frequently limited due to inadequate cohort representativeness. Many studies also lacked a control group, preventing assessment of non-exposed cohorts and comparability. Additionally, for studies examining feasibility, sensor validity, or the detection of health changes, it was often challenging to confirm the absence of the outcome of interest at baseline.
Outcome assessment emerged as a methodological strength, with most studies demonstrating rigorous evaluation and adequate follow-up procedures. However, follow-up duration varied widely, and in several studies, insufficient follow-up time contributed to lower overall quality scores.
Overall, while a few studies demonstrated robust design and low risk of bias, the majority exhibited moderate quality with specific weaknesses in patient selection and confounder control.
4. Discussion
This review synthesises evidence exploring the clinical impact of passive sensing technologies in remote monitoring across diverse patient populations, including neurodegenerative, musculoskeletal, cognitive, and cardiopulmonary conditions. Passive sensing, through mobile devices (e.g., wearables and smartphones) and in-home systems (e.g., infrared and radio-wave sensors), demonstrated feasibility and ecological validity in capturing continuous, real-world health data. These technologies offer a unified framework for enhancing patient care through three main applications: monitoring chronic disease progression, detecting acute health deterioration, and supporting therapeutic interventions. This review demonstrates that passive sensing technologies not only correlate with established clinical assessments but also support novel, non-invasive digital biomarkers that could reshape monitoring strategies across health systems.
The scope of application spans various healthcare needs. In chronic conditions such as Parkinson’s disease and cognitive impairments, these technologies enable precise, longitudinal tracking of disease markers (e.g., tremor scores, gait speed, and daily activity patterns), facilitating timely adjustments in care. For acute health events, such as heart failure decompensation or respiratory decline during recovery from COVID-19, real-time alerts enable rapid interventions. In rehabilitation and recovery, sensor-derived feedback supports post-surgical gait improvements and adherence to physiotherapy, enhancing therapeutic outcomes.
Continuous, longitudinal monitoring enables real-time assessment of patients with chronic conditions or vulnerabilities, surpassing the limitations of infrequent clinic visits. Advances in consumer technologies, particularly smartphones, have democratised access to low-cost, high-quality sensors. Monitoring may focus on specific diseases, such as tremor assessment in PD [20], or broader functional metrics, such as sleep patterns, activity levels, or room usage in older adults [8] or those with mild cognitive impairment [30]. Strong correlations between sensor-derived metrics (e.g., tremor scores, gait speed, and sleep patterns) and established clinical assessments, underscore their potential as digital biomarkers for disease monitoring and management. Sensor-derived digital biomarkers offer unparalleled sensitivity for detecting early functional decline, facilitating disease management, and informing research [44,45]. Although many sensors have demonstrated consistent, accurate outputs under controlled conditions in preliminary testing, future research should incorporate reliability metrics in the applied setting to strengthen the validity of sensor outputs in real-world healthcare use.
Passive sensing also enables the detection of acute, clinically significant events, akin to a virtual ward model [46]. Unlike routine monitoring, this approach focuses on real-time assessment of physiological changes—such as respiratory rate (RR), heart rate (HR), and weight—coupled with digital alert systems for immediate intervention. For instance, in heart failure patients, bed sensors identified respiratory and heart rate changes predictive of decompensation [33]. Similarly, radio-wave sensors detected early respiratory deterioration in COVID-19 patients, enabling timely hospital admission [40]. These early interventions have the potential to reduce emergency visits and hospitalizations [32], though further studies are needed to evaluate scalability. A critical challenge remains in distinguishing meaningful signals from background noise. Integrating data from multiple sensor types (e.g., RR, HR, sleep patterns, physical activity) into reliable and actionable outputs is essential.
Furthermore, passive sensors can support therapeutic interventions by providing actionable feedback. For example, motion sensors improved adherence to physiotherapy exercises for shoulder rehabilitation [41] while ingestible sensors enhanced medication adherence in hypertensive patients [39]. Real-time feedback fosters a goal-oriented approach to treatment, improving patient engagement and outcomes.
This review yielded several technically significant insights. First, we identified strong, statistically validated correlations between passive sensor outputs and clinical reference standards, supporting their role as digital biomarkers. Second, multiple studies demonstrated the capacity of passive sensing platforms to deliver real-time, high-resolution data in naturalistic environments. Third, we noted the increasing integration of machine learning models, which translated raw sensor data into symptom prediction, activity clustering, and physiological state monitoring. These findings provide a foundation for future work developing standard protocols, validating predictive models, and integrating passive sensing into clinical workflows. Specifically, longitudinal studies can now build on our synthesised dataset to train and test algorithms aimed at early detection, risk stratification, and recovery monitoring.
While this review underscores the transformative potential of passive sensing technologies in remote monitoring, several limitations must be acknowledged. Firstly, the studies included in this analysis exhibit heterogeneity, encompassing diverse sensing modalities, study designs, and outcome measures. This variability complicates the synthesis of findings and limits the generalizability of the results, but for the purpose of this review, allowed for a comprehensive overview of passive sensing. For example, wearable devices predominated in some studies, focusing on gait and sleep metrics, while others relied on in-home sensors to monitor activity and physiological changes. The lack of standardisation in methodologies highlights the need for unified protocols to facilitate cross-study comparisons. Secondly, the evidence base remains constrained by the predominance of small-scale pilot studies with short follow-up durations. These exploratory investigations, while valuable for feasibility assessments, provide limited insight into the long-term clinical impact of passive sensing. Moreover, most studies were conducted in high-resource settings, raising concerns about the applicability of these findings to low-resource or underserved populations, where such technologies could have significant impact.
In addition, challenges related to usability and adherence were frequently reported, particularly with wearable devices. Despite the promise of passive sensing as a low-burden intervention, issues such as device discomfort, the need for technological proficiency, and concerns about privacy emerged as barriers to sustained use. These limitations underscore the importance of integrating patient and caregiver perspectives into the design and implementation of these technologies to improve their acceptability and usability in real-world contexts. While some studies included an analysis of patient and caregiver perspectives, future research would be enhanced by actively involving patients in the design and development of studies.
Furthermore, privacy and data security remain unresolved challenges that hinder widespread adoption. Data storage practices in longitudinal monitoring were rarely reported, highlighting an area that requires further attention and improvement. Participant discomfort with continuous monitoring, in some cases leading to altered behaviours, was a recurring theme across studies. The absence of consistent regulatory frameworks governing data protection and ethical use poses a significant barrier to the scalability of passive sensing technologies in routine clinical care. Lastly, the reliance on algorithm-driven analyses introduces inherent challenges. Although machine learning has proven effective in deriving clinically meaningful insights from sensor data, the predictive accuracy and reliability of these algorithms remain inconsistent. Many models are still in developmental stages and require further validation in diverse and real-world settings to ensure their robustness and clinical utility.
Further research, for clinicians, should focus on integrating passive sensing technologies into standard clinical workflows. This includes developing robust clinical pathways that leverage sensor-derived data to inform decision-making and enable timely interventions. Training programmes should begin to incorporate the interpretation and application of sensor outputs, ensuring that healthcare professionals can utilise these data to enhance patient care. Furthermore, clinical trials should explore the utility of passive sensors across broader patient populations and in diverse healthcare settings, particularly in resource-limited environments where access to traditional monitoring is constrained.
For academics, advancement in algorithm development and validation should be prioritised. Existing studies demonstrate the feasibility of using passive sensing to detect subtle changes in physiological and behavioural metrics, but challenges remain in achieving predictive accuracy and reliability. Future research should refine machine learning models to improve the detection of clinically significant events and symptom trajectories. One area for future development is the use of algorithms to predict symptom burden based on sensor data. For example, for cancer patients, sensors were able to predict next-day symptom burden [36]. Additionally, there is a need for longitudinal studies to evaluate the long-term impact of passive sensing on health outcomes, patient adherence, and healthcare utilisation. Collaborative efforts between academia and industry can accelerate the development of innovative sensors tailored to specific clinical needs, including multimodal devices capable of integrating physiological, environmental, and behavioural data.
Moreover, the implementation of passive sensing technologies requires clear regulatory frameworks to ensure data privacy, security, and ethical use. Addressing patient and caregiver concerns about surveillance and data handling is paramount to fostering trust and encouraging adoption. Policymakers must also consider the economic implications of these technologies, including cost-effectiveness analyses and reimbursement models that support their integration into public and private healthcare systems. Pilot programmes can evaluate the scalability of passive sensing in real-world settings and provide evidence for broader policy adoption.
In conclusion, passive sensing technologies offer a transformative approach to remote monitoring, enabling continuous, unobtrusive tracking of health metrics across diverse populations. This review underscores their potential to enhance disease management, detect early deterioration, and personalise care. However, challenges in adherence, scalability, and privacy highlight the need for further research to optimise implementation. By addressing these barriers, passive sensors could redefine clinical monitoring and support in both high- and low-resource settings.
Author Contributions
F.M.I. conceptualised the idea; E.R. drafted the manuscript. E.R. and S.Z. independently screened and reviewed all included articles; E.R. and S.Z. graded the quality of included studies. F.M.I., A.A. and M.A. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
No primary data were generated in this study. Data available on request.
Conflicts of Interest
The authors declare no conflicts of interest.
Appendix A
Full Search Strategy
Full search strategy: Hillingdon Hospital Library and Knowledge Service
“Impact of passive sensing in remote monitoring—medline search”:
Ovid MEDLINE(R) ALL <1946 to 8 April 2024>
1 ((remote* or home*) adj3 (sens* or monitor*)).tw. 30,847
2 ((remote* or home*) adj3 (sens* or monitor*)).kf. 6849
3 exp Remote Sensing Technology/5244
4 1 or 2 or 3 35,793
5 passive*.tw. 143,279
6 passive*.kf. 6814
7 5 or 6 144,869
8 4 and 7 682
9 limit 8 to (English language and yr=“2014–Current” and “all adult (19 plus years)”) 64
“Impact of passive sensing in remote monitoring—embase search”
Embase <1974 to 8 April 2024>
1 ((remote* or home*) adj3 (sens* or monitor*)).tw. 37,716
2 ((remote* or home*) adj3 (sens* or monitor*)).kf. 8115
3 exp *remote sensing/7144
4 1 or 2 or 3 42,258
5 passive*.tw. 166,882
6 passive*.kf. 12,529
7 exp *passive surveillance/71
8 5 or 6 or 7 169,585
9 4 and 8 696
10 limit 9 to (english language and yr=“2014–Current” and (adult <18 to 64 years> or aged <65+ years>)) 227
“Impact of passive sensing in remote monitoring—emcare search”
Ovid Emcare <1995 to 2024 Week 13>
1 ((remote* or home*) adj3 (sens* or monitor*)).tw. 8712
2 ((remote* or home*) adj3 (sens* or monitor*)).kf. 1944
3 exp *remote sensing/907
4 1 or 2 or 3 9592
5 passive*.tw. 38,019
6 passive*.kf. 2569
7 exp *passive surveillance/12
8 5 or 6 or 7 38,574
9 4 and 8 159
10 limit 9 to (english language and yr=“2014–Current” and (adult <18 to 64 years> or aged <65+ years>)) 44
Table A1.
CINAHL via Ebscohost:
| S1 | TI ((remote* or home*) n3 (sens* or monitor*)) OR AB ((remote* or home*) n3 (sens* or monitor*)) | 6699 |
| S2 | TI passive* OR AB passive* | 21,467 |
| S3 | S1 AND S2 | 55 |
| S6 | S1 AND S2 (Limiters—Publication Date: 1 January 2014–31 December 2024 Narrow by Subject Age:—all adult Narrow by Language:—english Search modes—Boolean/Phrase) | 21 |
Table A2.
BNI via ProQuest.
| S1 | tiab((remote* or home*) near/3 (sens* or monitor*)) | 797 |
| S2 | MAINSUBJECT.EXACT(“Remote sensing”) | 29 |
| S3 | [S1] OR [S2] | 817 |
| S4 | tiab(passive*) | 1974 |
| S5 | [S3] AND [S4] | 6 |
| S6 | ([S3] AND [S4]) AND (yr(2016) OR yr(2022)) | 2 |
| S7 | ([S3] AND [S4]) AND (la.exact(“ENG”) AND (yr(2016) OR yr(2022))) | 2 |
Table A3.
Psycinfo via ProQuest:
| S8 | tiab((remote* or home*) near/3 (sens* or monitor*)) | 2309 |
| S9 | tiab(passive*) | 31,580 |
| S10 | [S9] AND [S8] | 64 |
| S11 | ([S9] AND [S8]) AND pd(1 January 2014–9 April 2024) | 48 |
| S12 | ([S9] AND [S8]) AND (la.exact(“ENG”) AND pd(1 January 2014–9 April 2024)) | 48 |
| S13 | ([S9] AND [S8]) AND (la.exact(“ENG”) AND subject.exact(“Adulthood (18 yrs & older)”) AND pd(1 January 2014–9 April 2024)) | 28 |
Cochrane Library:
ID Search Hits
#1 (((remote* or home*) near/3 (sens* or monitor*))): ti,ab,kw (Word variations have been searched) 4387
#2 MeSH descriptor: [Remote Sensing Technology] explode all trees 75
#3 #1 or #2 4390
#4 (passive*): ti,ab,kw (Word variations have been searched) 12,372
#5 (adult* or elderl* or man or men or woman or women): ti,ab,kw (Word variations have been searched) 1,006,996
#6 MeSH descriptor: [Adult] explode all trees 613,547
#7 #5 or #6 1,096,496
#8 #3 and #4 and #7 with Cochrane Library publication date Between Jan 2014 and May 2024 29
Total after deduplication in Refworks and manually screening-out remaining duplicates: 276.
Total after screening-out paediatric papers: 272.
Grey literature search via a custom search engine (created by Gavin Moore and Andrew Doyle, University Hospital Coventry and Warwickshire)
References
- Prince, M.J.; Wu, F.; Guo, Y.; Robledo, L.M.G.; O’Donnell, M.; Sullivan, R.; Yusuf, S. The Burden of Disease in Older People and Implications for Health Policy and Practice. Lancet 2015, 385, 549–562. [Google Scholar] [CrossRef] [PubMed]
- Hood, C.; Sikka, N.; Van, C.M.; Mossburg, S. Remote Patient Monitoring. Available online: https://psnet.ahrq.gov/perspective/remote-patient-monitoring (accessed on 14 September 2024).
- Tan, S.Y.; Sumner, J.; Wang, Y.; Wenjun Yip, A. A Systematic Review of the Impacts of Remote Patient Monitoring (RPM) Interventions on Safety, Adherence, Quality-of-Life and Cost-Related Outcomes. npj Digit. Med. 2024, 7, 192. [Google Scholar] [CrossRef]
- Cornet, V.P.; Holden, R.J. Systematic Review of Smartphone-Based Passive Sensing for Health and Wellbeing. J. Biomed. Inform. 2018, 77, 120–132. [Google Scholar] [CrossRef]
- Noah, B.; Keller, M.S.; Mosadeghi, S.; Stein, L.; Johl, S.; Delshad, S.; Tashjian, V.C.; Lew, D.; Kwan, J.T.; Jusufagic, A.; et al. Impact of Remote Patient Monitoring on Clinical Outcomes: An Updated Meta-Analysis of Randomized Controlled Trials. npj Digit. Med. 2018, 1, 20172. [Google Scholar] [CrossRef] [PubMed]
- Enshaeifar, S.; Zoha, A.; Skillman, S.; Markides, A.; Acton, S.T.; Elsaleh, T.; Kenny, M.; Rostill, H.; Nilforooshan, R.; Barnaghi, P. Machine Learning Methods for Detecting Urinary Tract Infection and Analysing Daily Living Activities in People with Dementia. PLoS ONE 2019, 14, e0209909. [Google Scholar] [CrossRef]
- Parkinson, M.E.; Doherty, R.; Curtis, F.; Soreq, E.; Lai, H.H.L.; Serban, A.-I.; Dani, M.; Fertleman, M.; Barnaghi, P.; Sharp, D.J.; et al. Using Home Monitoring Technology to Study the Effects of Traumatic Brain Injury in Older Multimorbid Adults. Ann. Clin. Transl. Neurol. 2023, 10, 1688–1694. [Google Scholar] [CrossRef] [PubMed]
- Lach, H.W.; Lorenz, R.A.; Palmer, J.L.; Koedbangkham, J.; Noimontree, W. Home Monitoring to Track Activity and Sleep Patterns Among Older Adults: A Feasibility Study. CIN 2019, 37, 628–637. [Google Scholar] [CrossRef]
- Reeder, B.; Meyer, E.; Lazar, A.; Chaudhuri, S.; Thompson, H.J.; Demiris, G. Framing the Evidence for Health Smart Homes and Home-Based Consumer Health Technologies as a Public Health Intervention for Independent Aging: A Systematic Review. Int. J. Med. Inf. 2013, 82, 565–579. [Google Scholar] [CrossRef]
- Iqbal, F.M.; Lam, K.; Joshi, M.; Khan, S.; Ashrafian, H.; Darzi, A. Clinical Outcomes of Digital Sensor Alerting Systems in Remote Monitoring: A Systematic Review and Meta-Analysis. npj Digit. Med. 2021, 4, 7. [Google Scholar] [CrossRef]
- Kaye, J.A.; Maxwell, S.A.; Mattek, N.; Hayes, T.L.; Dodge, H.; Pavel, M.; Jimison, H.B.; Wild, K.; Boise, L.; Zitzelberger, T.A. Intelligent Systems for Assessing Aging Changes: Home-Based, Unobtrusive, and Continuous Assessment of Aging. J. Gerontol. Ser. B 2011, 66B, i180–i190. [Google Scholar] [CrossRef]
- Capstick, A.; Palermo, F.; Zakka, K.; Fletcher-Lloyd, N.; Walsh, C.; Cui, T.; Kouchaki, S.; Jackson, R.; Tran, M.; Crone, M.; et al. Digital Remote Monitoring for Screening and Early Detection of Urinary Tract Infections. npj Digit. Med. 2024, 7, 11. [Google Scholar] [CrossRef]
- Taylor, M.L.; Thomas, E.E.; Snoswell, C.L.; Smith, A.C.; Caffery, L.J. Does Remote Patient Monitoring Reduce Acute Care Use? A Systematic Review. BMJ Open 2021, 11, e040232. [Google Scholar] [CrossRef] [PubMed]
- Stuijt, D.G.; Radanovic, I.; Kos, M.; Schoones, J.W.; Stuurman, F.E.; Exadaktylos, V.; Bins, A.D.; Bosch, J.J.; van Oijen, M.G.H. Smartphone-Based Passive Sensing in Monitoring Patients with Cancer: A Systematic Review. JCO Clin. Cancer Inform. 2023, 7, e2300141. [Google Scholar] [CrossRef]
- Trifan, A.; Oliveira, M.; Oliveira, J.L. Passive Sensing of Health Outcomes Through Smartphones: Systematic Review of Current Solutions and Possible Limitations. JMIR mHealth uHealth 2019, 7, e12649. [Google Scholar] [CrossRef]
- Moher, D.; Shamseer, L.; Clarke, M.; Ghersi, D.; Liberati, A.; Petticrew, M.; Shekelle, P.; Stewart, L.A. PRISMA-P Group Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 Statement. Syst. Rev. 2015, 4, 1. [Google Scholar] [CrossRef] [PubMed]
- Covidence. Veritas Health Innovation Covidence Systematic Review Software. 2019. Available online: https://app.covidence.org/sign_in (accessed on 1 August 2024).
- Stang, A. Critical Evaluation of the Newcastle-Ottawa Scale for the Assessment of the Quality of Nonrandomized Studies in Meta-Analyses. Eur. J. Epidemiol. 2010, 25, 603–605. [Google Scholar] [CrossRef]
- Sterne, J.A.C.; Savović, J.; Page, M.J.; Elbers, R.G.; Blencowe, N.S.; Boutron, I.; Cates, C.J.; Cheng, H.-Y.; Corbett, M.S.; Eldridge, S.M.; et al. RoB 2: A Revised Tool for Assessing Risk of Bias in Randomised Trials. BMJ 2019, 366, l4898. [Google Scholar] [CrossRef]
- Fay-Karmon, T.; Galor, N.; Heimler, B.; Zilka, A.; Bartsch, R.P.; Plotnik, M.; Hassin-Baer, S. Home-Based Monitoring of Persons with Advanced Parkinson’s Disease Using Smartwatch-Smartphone Technology. Sci. Rep. 2024, 14, 9. [Google Scholar] [CrossRef] [PubMed]
- Gatsios, D.; Antonini, A.; Gentile, G.; Marcante, A.; Pellicano, C.; Macchiusi, L.; Assogna, F.; Spalletta, G.; Gage, H.; Touray, M.; et al. Feasibility and Utility of mHealth for the Remote Monitoring of Parkinson Disease: Ancillary Study of the PD_manager Randomized Controlled Trial. JMIR mHealth uHealth 2020, 8, e16414. [Google Scholar] [CrossRef]
- Kabelac, Z.; Tarolli, C.G.; Snyder, C.; Feldman, B.; Glidden, A.; Hsu, C.-Y.; Hristov, R.; Dorsey, E.R.; Katabi, D. Passive Monitoring at Home: A Pilot Study in Parkinson Disease. Digit. Biomark. 2019, 3, 22–30. [Google Scholar] [CrossRef]
- Lipsmeier, F.; Taylor, K.I.; Kilchenmann, T.; Wolf, D.; Scotland, A.; Schjodt-Eriksen, J.; Cheng, W.-Y.; Fernandez-Garcia, I.; Siebourg-Polster, J.; Jin, L.; et al. Evaluation of Smartphone-Based Testing to Generate Exploratory Outcome Measures in a Phase 1 Parkinson’s Disease Clinical Trial. Mov. Disord. 2018, 33, 1287–1297. [Google Scholar] [CrossRef] [PubMed]
- Liu, Y.; Zhang, G.; Tarolli, C.G.; Hristov, R.; Jensen-Roberts, S.; Waddell, E.M.; Myers, T.L.; Pawlik, M.E.; Soto, J.M.; Wilson, R.M.; et al. Monitoring Gait at Home with Radio Waves in Parkinson’s Disease: A Marker of Severity, Progression, and Medication Response. Sci. Transl. Med. 2022, 14, eadc9669. [Google Scholar] [CrossRef] [PubMed]
- Schalkamp, A.-K.; Harrison, N.A.; Peall, K.J.; Sandor, C. Digital Outcome Measures from Smartwatch Data Relate to Non-Motor Features of Parkinson’s Disease. npj Park. Dis. 2024, 10, 110. [Google Scholar] [CrossRef]
- Lam, K.-H.; Twose, J.; Lissenberg-Witte, B.; Licitra, G.; Meijer, K.; Uitdehaag, B.; De Groot, V.; Killestein, J. The Use of Smartphone Keystroke Dynamics to Passively Monitor Upper Limb and Cognitive Function in Multiple Sclerosis: Longitudinal Analysis. J. Med. Internet Res. 2022, 24, e37614. [Google Scholar] [CrossRef] [PubMed]
- Gaugler, J.E.; Zmora, R.; Mitchell, L.L.; Finlay, J.M.; Peterson, C.M.; McCarron, H.; Jutkowitz, E. Six-Month Effectiveness of Remote Activity Monitoring for Persons Living with Dementia and Their Family Caregivers: An Experimental Mixed Methods Study. Gerontologist 2019, 59, 78–89. [Google Scholar] [CrossRef]
- Obayashi, K.; Kodate, N.; Ishii, Y.; Masuyama, S. Assistive Technologies and Aging in Place for People with Dementia and Disabilities: A Proof-of-Concept Study with in-Home Passive Remote Monitoring with Interactive Communication Functions. Disabil. Rehabil. Assist. Technol. 2023, 19, 2341–2354. [Google Scholar] [CrossRef]
- Muurling, M.; Au-Yeung, W.-T.M.; Beattie, Z.; Wu, C.-Y.; Dodge, H.; Rodrigues, N.K.; Gothard, S.; Silbert, L.C.; Barnes, L.L.; Steele, J.S.; et al. Differences in Life Space Activity Patterns Between Older Adults with Mild Cognitive Impairment Living Alone or as a Couple: Cohort Study Using Passive Activity Sensing. JMIR Aging 2023, 6, e45876. [Google Scholar] [CrossRef]
- Rawtaer, I.; Mahendran, R.; Kua, E.H.; Tan, H.P.; Tan, H.X.; Lee, T.-S.; Ng, T.P. Early Detection of Mild Cognitive Impairment with In-Home Sensors to Monitor Behavior Patterns in Community-Dwelling Senior Citizens in Singapore: Cross-Sectional Feasibility Study. J. Med. Internet Res. 2020, 22, e16854. [Google Scholar] [CrossRef]
- Wu, C.-Y.; Dodge, H.H.; Gothard, S.; Mattek, N.; Wright, K.; Barnes, L.L.; Silbert, L.C.; Lim, M.M.; Kaye, J.A.; Beattie, Z. Unobtrusive Sensing Technology Detects Ecologically Valid Spatiotemporal Patterns of Daily Routines Distinctive to Persons with Mild Cognitive Impairment. J. Gerontol. Ser. A 2022, 77, 2077–2084. [Google Scholar] [CrossRef]
- Finch, M.; Griffin, K.; Pacala, J.T. Reduced Healthcare Use and Apparent Savings with Passive Home Monitoring Technology: A Pilot Study. J. Am. Geriatr. Soc. 2017, 65, 1301–1305. [Google Scholar] [CrossRef]
- Saner, H.; Schütz, N.; Botros, A.; Urwyler, P.; Buluschek, P.; du Pasquier, G.; Nef, T. Potential of Ambient Sensor Systems for Early Detection of Health Problems in Older Adults. Front. Cardiovasc. Med. 2020, 7, 110. [Google Scholar] [CrossRef]
- Leightley, D.; Lavelle, G.; White, K.M.; Sun, S.; Matcham, F.; Ivan, A.; Oetzmann, C.; Penninx, B.W.J.H.; Lamers, F.; Siddi, S.; et al. Investigating the Impact of COVID-19 Lockdown on Adults with a Recent History of Recurrent Major Depressive Disorder: A Multi-Centre Study Using Remote Measurement Technology. BMC Psychiatry 2021, 21, 435. [Google Scholar] [CrossRef] [PubMed]
- Dadhania, S.; Pakzad-Shahabi, L.; Mistry, S.; Williams, M. Triaxial Accelerometer-Measured Physical Activity and Functional Behaviours among People with High Grade Glioma: The BrainWear Study. PLoS ONE 2023, 18, e0285399. [Google Scholar] [CrossRef]
- Low, C.A.; Li, M.; Vega, J.; Durica, K.C.; Ferreira, D.; Tam, V.; Hogg, M.; Zeh Iii, H.; Doryab, A.; Dey, A.K. Digital Biomarkers of Symptom Burden Self-Reported by Perioperative Patients Undergoing Pancreatic Surgery: Prospective Longitudinal Study. JMIR Cancer 2021, 7, e27975. [Google Scholar] [CrossRef] [PubMed]
- Low, C.A.; Bartel, C.; Fedor, J.; Durica, K.C.; Marchetti, G.; Rosso, A.L.; Campbell, G. Associations between Performance-Based and Patient-Reported Physical Functioning and Real-World Mobile Sensor Metrics in Older Cancer Survivors: A Pilot Study. J. Geriatr. Oncol. 2024, 15, 101708. [Google Scholar] [CrossRef]
- Harrington, N.; Bui, Q.M.; Wei, Z.; Hernandez-Pacheco, B.; DeYoung, P.N.; Wassell, A.; Duwaik, B.; Desai, A.S.; Bhatt, D.L.; Agnihotri, P.; et al. Passive Longitudinal Weight and Cardiopulmonary Monitoring in the Home Bed. Sci. Rep. 2021, 11, 24376. [Google Scholar] [CrossRef] [PubMed]
- Noble, K.; Brown, K.; Medina, M.; Alvarez, F.; Young, J.; Leadley, S.; Kim, Y.; DiCarlo, L. Medication Adherence and Activity Patterns Underlying Uncontrolled Hypertension: Assessment and Recommendations by Practicing Pharmacists Using Digital Health Care. J. Am. Pharm. Assoc. 2016, 56, 310–315. [Google Scholar] [CrossRef]
- Zhang, G.; Vahia, I.V.; Liu, Y.; Yang, Y.; May, R.; Cray, H.V.; McGrory, W.; Katabi, D. Contactless In-Home Monitoring of the Long-Term Respiratory and Behavioral Phenotypes in Older Adults with COVID-19: A Case Series. Front. Psychiatry 2021, 12, 754169. [Google Scholar] [CrossRef]
- Chen, Y.-P.; Lin, C.-Y.; Tsai, M.-J.; Chuang, T.-Y.; Lee, O.K.-S. Wearable Motion Sensor Device to Facilitate Rehabilitation in Patients with Shoulder Adhesive Capsulitis: Pilot Study to Assess Feasibility. J. Med. Internet Res. 2020, 22, e17032. [Google Scholar] [CrossRef]
- Perraudin, C.G.M.; Illiano, V.P.; Calvo, F.; O’Hare, E.; Donnelly, S.C.; Mullan, R.H.; Sander, O.; Caulfield, B.; Dorn, J.F. Observational Study of a Wearable Sensor and Smartphone Application Supporting Unsupervised Exercises to Assess Pain and Stiffness. Digit. Biomark. 2018, 2, 106–125. [Google Scholar] [CrossRef]
- Ribeiro-Castro, A.L.; Surmacz, K.; Aguilera-Canon, M.C.; Anderson, M.B.; Van Andel, D.; Redfern, R.E.; Cook, C.E. Early Post-Operative Walking Bouts Are Associated with Improved Gait Speed and Symmetry at 90 Days. Gait Posture 2024, 107, 130–135. [Google Scholar] [CrossRef] [PubMed]
- Piau, A.; Wild, K.; Mattek, N.; Kaye, J. Current State of Digital Biomarker Technologies for Real-Life, Home-Based Monitoring of Cognitive Function for Mild Cognitive Impairment to Mild Alzheimer Disease and Implications for Clinical Care: Systematic Review. J. Med. Internet Res. 2019, 21, e12785. [Google Scholar] [CrossRef] [PubMed]
- Spira, A.P.; Kaufmann, C.N.; Kasper, J.D.; Ohayon, M.M.; Rebok, G.W.; Skidmore, E.; Parisi, J.M.; Reynolds, C.F. Association between Insomnia Symptoms and Functional Status in U.S. Older Adults. J. Gerontol. Ser. B 2014, 69 (Suppl. S1), S35–S41. [Google Scholar] [CrossRef] [PubMed]
- Norman, G.; Bennett, P.; Vardy, E.R.L.C. Virtual Wards: A Rapid Evidence Synthesis and Implications for the Care of Older People. Age Ageing 2023, 52, afac319. [Google Scholar] [CrossRef]
Figure 1.
PRISMA flow diagram for included studies (Covidence).
Table 1.
Characteristics of included studies.
| Reference | Condition or Disease | Sample Size (N) | Length of Monitoring | Sensor(s)/Feedback to Participants | Outcome Measures | Algorithms/Machine Learning | Compliance and Acceptability/Feasibility | Outcomes |
|---|---|---|---|---|---|---|---|---|
| Fay-Karmon et al., 2024 [20] | Parkinson’s disease | 21 | 14 days | Smartwatch | Compliance (smartwatch usage). Smartwatch algorithm scores for tremor, dyskinesia and activity. Clinic recorded MDS-UPDRS, Daily and in-clinic motor tasks, daily symptom diary and medication intake report. | Smartwatch data as input to two algorithms: free living and motor tasks algorithm (Intel Pharma Analytics). | Compliance: 81% wearing the smartwatch for more than 12 h daily. | Validity: Significant correlation between SWA scores for tremor and both clinician assessment and MDS-UPDRS scores for tremor. Patterns: Patients categorised into four groups according to individual patterns of tremor fluctuations. |
| Gatsios et al., 2020 [21] | Parkinson’s disease | 132 | 14 days | Smartphone, wristband, insoles | Compliance (protocol completion, median usage). Measures of motor and non-motor function: NMSS, UPDRS, EQ-5D-5L, PDQ-8. | An algorithm derived from a single sensor on the wrist for detecting tremor in PD patients (PD_manager). | Compliance: 85% participants completed the protocol. Median use was 11.57 days over a 14-day period. | Validity: Significant correlation between tremor score and tremor evaluation by clinicians. Significant difference between tremor score between no-tremor (UPDRS = 0) and tremor (UPDRS > 1) groups. |
| Kabelac et al., 2019 [22] | Parkinson’s disease | 7 | 353 days | Radio-wave based in-home monitoring (Emerald device) | Activity: Gait speed, time in bed per day, circadian schedule (number and duration of nighttime awakenings, intraday rest-activity rhythms). Baseline MDS-UPDRS. | N/A | Validity: Significant correlation between device-derived gait speed and MDP-UDRS scores. | |
| Lipsmeier et al., 2018 [23] | Parkinson’s disease | 79 (35 controls) | 6 months | Smartphone based active and passive sensing | Compliance (test session completion). Activity: algorithm determined turning speed, sit to stand transitions, proportion of gait-related activity. MDP-UPDRS during screening. Active smartphone motor testing. | Machine-learning algorithm used to classify accelerometer data into labelled activities (e.g., walking, climbing stairs, standing, and sitting). | Compliance: 61% of possible test sessions over 6 months. | Validity: All active and passive monitoring features significantly discriminated PD from controls. PD participants had less time performing gait related activities, fewer sit to stand transitions and slower 90 degree turns than controls. |
| Liu et al., 2022 [24] | Parkinson’s disease | 50 (16 controls) | 1 year (only first year included) | Radio-wave based in-home monitoring | Activity: In-home gait speed. Clinical assessment at baseline, 6 months and 12 months: MDS-UPDRS, H&Y stage, TUG, Intraday ON-OFF state from Hauser diary. | Machine learning algorithms to analyse the reflected radio signals and extracted walking movements and trajectories through the home. | Validity: Significant correlation between in-home gait speed and MDS-UPDRS and H & Y stage. Individuals with PD had about 23% lower in-home gait speed than the control cohort The decline in gait speed over one year was greater for the PD group (−0.026 m/s) compared to the control group (−0.015 m/s). Patterns: intraday gait speed oscillates according to medication administration and variation in intraday gait speed is highly correlated with patient perception of disease impact on daily function. | |
| Schalkamp et al., 2024 [25] | Parkinson’s disease | 149 | 485 days | Smartwatch | Activity: Weekly averages of: sleep (total time, REM time, NREM time, WASO, awakenings, sleep efficiency), physical activity (step count, walking minutes), and vital signs (pulse rate, mean RMSSD, median RMSSD, RMSSD variance). Clinic assessments: UPDRS, MoCA, LNS, HVLT, GDS, STAI, ESS, SCOPA, RBDSQ. | N/A | Validity: Digital measures capturing weekly averages of physical measures are concordant with clinical assessments of multiple non motor symptoms (cognitive, autonomic and daily functioning). Prediction: Digital weekly averages could not predict the scores of standardised non motor symptom questionnaires. | |
| Lam et al., 2022 [26] | Multiple sclerosis | 102 | 1 year | Smartphone | Smartphone keystroke dynamics: fine motor cluster and cognition score clusters. Standard measures: nine-hole peg test (arm function), symbol digit modalities test (information processing speed). | Keystroke features were processed to form fine motor and cognition score clusters. | Validity: The fine motor score cluster was significantly associated with the nine-hole peg test. The cognition score cluster was significantly associated with the symbol digit modalities test. Smartphone keystroke dynamics were longitudinally associated with multiple sclerosis outcomes. | |
| Gaugler et al., 2019 [27] | Dementia | 132 (64 remote monitoring, 68 usual care control) | 6 months | Home activity monitors. Feedback: Significant and persistent deviations from baseline are communicated to family caregivers. | Mean caregiver self efficacy, sense of competence and distress. | Data are analysed using algorithms. Baseline activity pattern is established. | No difference in caregiver mean self efficacy, sense of competence and distress at 6 months with remote monitoring compared to control. | |
| Obayashi et al., 2023 [28] | Dementia | 5 | 12 weeks | Home passive infrared sensors, biometric sensors. Feedback: Alerts from the system are communicated to terminals held by care providers. | Activity: movement and sleep patterns. Weekly measures of ICF (seven measures): communication, movement, self-care, domestic activities, interpersonal activities, performing tasks (money and leisure). Weekly interRAI measurements (care measure). | N/A | Improvements in weekly measures of the International Classification of Functioning, Disability and Health (ICF) score. One participant no longer required an in-person helper at night. | |
| Muurling et al., 2023 [29] | Mild cognitive impairment | 129 homes with people living alone, 52 homes with people living as a couple | Mean of 719 days | Home passive infrared sensors, door contact sensors. | Activity: In 2-person vs. 1-person homes and MCI vs. non-MCI homes: TOOH, ILSA, kitchen, bathroom, bedroom and living room use. | Algorithms to calculate TOOH and ILSA. | Activity: People living together have a shorter TOOH, a longer ILSA, and longer room use, but only ILSA was affected by MCI status. | |
| Rawtaer et al., 2020 [30] | Mild cognitive impairment | 49 (21 controls) | 2 months | Multimodal sensor system: home passive infrared motion sensors, door contact sensor, proximity beacon tags, sensor-equipped medication box, faucet usage sensor, bed sensor, wearable sensor, smart plug. Feedback: If the system detected no movement for 8 h, an alert was sent to the caregiver. | Acceptability of sensor system Activity: Daily step count, heart rate, Sleep duration, number of sleep interruptions, time away from home daily, television use. Frequency of forgetting medication, wallet and keys. | N/A | Acceptability: Sensor system was acceptable to 80% (40/49). | Activity: MCI participants were less active, had more sleep interruptions per night and forgot their medications more frequently than their cognitively healthy counterparts, although none reached significance. |
| Wu et al., 2022 [31] | Mild cognitive impairment | 161 older adults (26 with mild cognitive impairment) | 1 month | Home passive infrared sensors, door contact sensors. | Activity: trips to bathroom, bedroom, kitchen and living room. Latent trajectory models used to identify distinct diurnal patterns of room activities. | Algorithm used to extract room activities and out-of-home activities from the motion sensors. | Activity: No difference in the total number of trips to rooms at a day level between MCI and non-MCI participants. However, MCI participants were more likely to be in the high bathroom and kitchen usage groups based on diurnal patterns of activity | |
| Finch et al., 2017 [32] | Older adults | 268 (74 with passive monitoring, 2 control groups) | 12 months | Home passive monitors. Feedback: Trending health alerts and real-time alerts and an emergency response pendant connecting the participant to an emergency call centre. | Activity: time in bed, toilet use, opening/closing of refrigerator, front door opening. Healthcare costs per member per month (inpatient, ED, LTC, SNF, ambulance, home care, outpatient, pharmacy, other, total). Healthcare use per member per month (acute hospitalisations, hospital days, ED visits, LTC admissions, LTC days, SNF admissions, SNF days, home care visits, outpatient visits). | Algorithms identified trends, which automatically trigger health alerts. | Healthcare costs: mean total per member per month costs were lower for the intervention group. Costs for inpatient services and ED visits were at least 10% lower in the intervention group. Home care and outpatient costs were not different. No differences were statistically significant for cost. Healthcare use: the intervention group had fewer acute hospitalisations, ED visits, long-term care admissions and days, and SNF admissions. | |
| Lach et al., 2019 [8] | Older adults | 5 | 3 months | Home motion sensors, bed sensor, chair sensor, door sensor, wearable sensor. | Feasibility (subjective). Activity: daily activity, sleep time and sleep efficiency, kitchen activity, bathroom activity, total activity, entry/exit, time in bed. | N/A | Validity: Sleep and activity data were similar across actigraphy and sensor measures. | |
| Saner et al., 2020 [33] | Older adults | 20 | 1 to 2 years | Multimodal sensor system: home passive infrared sensor, door sensor, bed sensor, wearable sensor. | Feasibility (completion). Activity: motion in the home, entry/exit, room use, sleep duration and disruption. Vital signs. Health deterioration: heart failure decompensation, palpitations, falls, pneumonia, urinary infection, pulmonary embolism. Health-related data collected from weekly visits: EQ-5D-3L questionnaire, information about accidents, information about lifestyle changes. | Sleep data extracted using algorithms. | Feasibility: 20/24 completed the study (2 died, 2 suffered health deterioration). 242,232 person hours were recorded. | Health deterioration: Episodes of heart failure decompensation were characterised by decreased activity, increased respiratory and heart rates and decreased sleep quality. Patient complaints of intermittent palpitations were confirmed by heart rate increases measured by the bed sensor. |
| Parkinson et al., 2023 [7] | Older adults (traumatic brain injury) | 3 | 6 months | Home passive infrared sensor, pneumatic bed mat. | Activity: bedroom, bathroom and kitchen activity, sleep patterns and disruption. Health deterioration: pneumonia, worsening sleep disturbance. | N/A | Health deterioration/recovery: an abnormally high overnight activity correlated with an episode of pneumonia in one patient, whereas consistent sleep cycles mapped a return to work for another patient recovering from traumatic brain injury. | |
| Leightley et al., 2021 [34] | Depression | 252 | 9 months | Smartphone, wearable sensor (RADAR-MDD) | Sleep duration. PHQ-8 Depressive symptom questionnaire and Rosenberg self-esteem scale questionnaire. | N/A | Mean sleep duration decreased significantly between during- and post-lockdown periods. No evidence of change in depressive symptoms or self-esteem before, during, or post lockdown. | |
| Dadhania et al., 2023 [35] | High grade glioma | 42 | Up to 14 weeks | Wearable sensor (wrist worn accelerometer) | Activity: mean daily activity, time spent performing functional activities (walking, light tasks, moderate activity, sedentary, sleep). PRO questionnaires: EORTC QLQ-C30, QLQ BN20, MoCA, MFI. Clinical data: MRI imaging, histopathology, radiotherapy and chemotherapy treatment plans. | Raw data classified into 5 accelerometer-predicted functional behaviours using a two-stage machine learning model. | Validity: Mean acceleration and time spent walking daily correlated positively with health quality of life and physical functioning scores and inversely with fatigue score. | |
| Low et al., 2021 [36] | Pancreatic cancer | 44 | At least 2 weeks prior to surgery and 60 days after postoperative discharge. | Smartphone with AWARE app, wearable sensor. | Activity: duration of exertional and nonexertional activity and sleep. Patient-reported symptoms each morning: pain, fatigue, sleep disturbance, trouble concentrating, feeling sad, feeling anxious, shortness of breath, numbness, nausea, diarrhoea or constipation. | Models used sensor data to predict whether the next day would be a high or low symptom burden day. | Prediction: Models using LightGBM were able to correctly predict whether the next day would be a high symptom day with 73.5% accuracy, surpassing baseline models. LightGBM models predicting next-day diarrhoea (79.0% accuracy), fatigue (75.8% accuracy), and pain (79.6% accuracy) performed similarly. | |
| Low et al., 2024 [37] | Older cancer survivors | 40 | 4 weeks | Smartphone, wearable sensor. | Activity: daily step count, peak gait cadence, activity fragmentation, time at home, total distance travelled, geographic mobility. PRO: PROMIS-PF. Performance-based physical function: SPPB, TUG. | N/A | Validity: Wearable device features reflecting greater amount and speed and lower fragmentation of walking in daily life were significantly related to better patient-reported function and performance-based physical function. Smartphone features reflecting more geographic mobility were related to better performance-based physical function but not patient-reported function. | |
| Harrington et al., 2021 [38] | Heart failure | Feasibility in 8 patients, long term monitoring of 1 patient. | 3 months | Bed sensor (BedScales) | Overnight sleep study for validation of respiratory monitoring, heart rate monitoring and sleep study and apnoea monitoring. Long term monitoring of a heart failure patient: RR, HR, weight, presence or absence in bed. | Total body weight, respiratory waveforms, ballistocardiography quantified from force sensor data. Algorithms separate signals when the bed is shared by a partner or pet. | Validity: Close quantitative agreement of respiratory rate with measurements of belt respirometer. Heart rate monitoring showed quantitative agreement with ECG recordings. In overnight sleep studies, Bedscales could discriminate central and obstructive apneas. Health deterioration/recovery: RR and HR decreased to presurgical levels over the course of two months. | |
| Noble et al., 2016 [39] | Hypertension | 39 | 2 weeks | Ingestible sensor, wearable ‘patch’ sensor Feedback: Digital health feedback system: measurement of medication adherence using an ingestible sensor to inform tailored advice regarding medication or lifestyle modifications. | Serial blood pressure recordings | N/A | The mean change in SBP over the 2-week evaluation period was −7.9 ± 22.1; the mean change in DBP was −2.8 ± 12.9. | |
| Zhang et al., 2021 [40] | COVID-19 | 3 | 327 days | Radio-wave based in-home monitoring (Emerald device) | Activity: Gait speed, sleep patterns (sleep efficiency and total time in bed), home activity patterns. Respiratory signals (RR). COVID-19 testing. | The sensor transmits wireless signals and analyses their reflections from nearby humans and inanimate objects using machine learning to infer physiological and behavioural markers. | Health deterioration/recovery: RR of patient 1 shows a smooth recovery, whereas patient 2 had a sudden increase in RR on day of hospitalisation. Asymptomatic patient 3 had a stable RR for the whole recovery period. | |
| Chen et al., 2020 [41] | Adhesive capsulitis | 15 | 3 months | Wearable sensors (sternum, upper arm and wrist) Feedback: Physiotherapists can view the latest shoulder ROM, exercise completion rates, can assign exercises for and directly communicate with patients through the mobile app. | Sensor measured shoulder mobility. Patient reported daily exercise completion rate. Clinic measurements of shoulder active and passive ROM by two examiners PRO: qDASH (shoulder function) and VAS (pain) | Raw data from each sensor was converted into a 3-dimensional (3D) motion of the shoulder structure | Compliance: those in the motion sensor–assisted rehabilitation group had a significantly higher patient-reported exercise completion rate. | Validity: Good to excellent reliability of the motion sensor device for measuring shoulder ROM compared to measurements of two physicians. Activity: Patients in the motion sensor-assisted rehabilitation group had significantly better ROM improvement, pain reduction and higher exercise completion rates than the home-based exercise group after three months of rehabilitation |
| Perraudin et al., 2018 [42] | Arthritis | 45 (15 controls) | 4 weeks | Wearable sensor (wrist worn accelerometer) | Adherence to STS tests 5 × STS test duration (3 times weekly) PRO questionnaires: 8 items on pain and stiffness | Algorithm to detect 5 × STS tests | Compliance: 56% of tests performed. | Validity: 5 × STS test duration was significantly associated with pain and stiffness intensity reported by patient-reported outcome questionnaires. |
| Ribeiro-Castro et al., 2024 [43] | Joint arthroplasty | 2255 | 90 days | Smartphone | Activity: Gait quality metrics within 30 days of surgery. Gait speed at 90 days post surgery. | Accelerometer data from smartphone was converted into gait quality metrics | Activity: Gait speed at 90 days was significantly associated with 30-day session length in the TKA and THA cohorts Across procedure types, more evenly distributed walking bouts during the early post-operative period were positively associated with gait speed at 90 days. |
Properties of included studies and their main outcomes. (5 × STS test: five-times sit-to-stand test; DBP: diastolic blood pressure, ED: emergency department; EORTC QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-3/5L: EuroQol 5-Dimension 3/5-Level; ESS: Epworth Sleepiness Score; GDS: Geriatric Depression score; HC: healthy cognition; H & Y: Hoehn and Yahr stage; HVLT: Hopkins Verbal Learning Test; ICF: International Classification of Functioning; Disability and Health; ILSA: Independent Life Space Activity; InterRAI: standardised assessments; LightGBM: Light Gradient Boosting Machine; LTC: long-term care; MCI: mild cognitive impairment; MDS-UPDRS: Movement Disorder Society—Unified Parkinson’s Disease Rating Scale; MoCA: Montreal Cognitive Assessment; NMSS: Nonmotor Symptoms Scale; NREM: Non Rapid Eye Movement; PDQ-8: Parkinson’s Disease Questionnaire-8; PRO: Patient-Reported Outcomes; PROMIS-PF: Patient-Reported Outcomes Measurement Information System Physical Functioning; qDASH: Quick Disabilities of the Arm, Shoulder, and Hand questionnaire; RBDSQ: REM Sleep Behaviour Disorder Screening Questionnaire; REM: Rapid Eye Movement; RMSSD: Root Mean Squared Successive Differences (heartbeat); ROM: Range of Motion; RR: respiratory rate; SBP: systolic blood pressure; SCOPA: Scale for Outcomes in Parkinson’s Disease for Autonomic Symptoms; SNF: skilled nursing facility; SPPB: Short Physical Performance Battery; STAI: State Trait-Anxiety Inventory; SWA: Smartwatch Algorithm; TOOH: Time Out of Home; TUG: Timed Up and Go; VAS: Pain Severity Visual Analog Scale; WASO: Wake After Sleep Onset; WMS-III LNS: Letter-Number Sequencing Test).
Table 2.
Types of passive sensors and their outputs.
| Sensor Context | Type of Sensor | Sensor Data/Outcome Measure | Studies |
|---|---|---|---|
| Wearable | Wrist/Arm worn sensors (smartwatch, Fitbit, AX3 accelerometer, armband sensors) | Physical activity | [20,21,25,35] |
| Step count | [25,30,36,37] | ||
| Gait speed | [37] | ||
| Functional behaviours | [35] | ||
| Tremor/Dyskinesia | [20,21] | ||
| Shoulder range of motion | [41] | ||
| Five times sit to stand test duration | [42] | ||
| Sleep duration | [8,21,25,34,36] | ||
| Sleep efficiency | [8,21,25] | ||
| Heart rate | [25,30,33,34,36] | ||
| Respiratory rate | [33] | ||
| Skin temperature/galvanic skin response | [33] | ||
| Patch sensor (torso) | Step count | [39] | |
| Sleep duration and interruption | [39] | ||
| Insoles | Weight bearing, balance and motion | [21] | |
| Gait | [21] | ||
| Ingestible | Ingestible sensor | Medication adherence | [39] |
| Portable | Smartphone | Activity recognition | [23,36] |
| Gait features | [43] | ||
| Speech patterns | [21,34] | ||
| Mood | [21,34] | ||
| Cognitive function | [21,34] | ||
| Keystroke dynamics | [26] | ||
| Application use | [36] | ||
| Location/Distance travelled | [36,37] | ||
| Light | [36] | ||
| Screen features | [36] | ||
| In-home | Home activity monitor | Activity patterns | [8,27,32] |
| Infrared camera | Activity patterns | [28] | |
| Sleep patterns | [28] | ||
| Living conditions | [28] | ||
| Infrared sensor | Room use/activity | [7,29,31,33] | |
| Time out of home | [29,30] | ||
| Independent life space activity | [29] | ||
| Radio-signals sensor | Room use/activity | [40] | |
| Sleep patterns | [22] | ||
| Sleep efficiency | [40] | ||
| Gait speed | [22,24,40] | ||
| Respiratory rate | [40] | ||
| Biometric sensor | Body temperature | [28] | |
| Heart rate | [28] | ||
| Respiratory rate | [28] | ||
| Door contact sensor | Time out of home | [29,30] | |
| Entry/exit to home | [30,33] | ||
| Room use | [31] | ||
| Refrigerator use | [33] | ||
| Bed sensor | Sleep duration | [7,8,30,33] | |
| Sleep interruption | [7,30] | ||
| Sleep onset delay | [33] | ||
| Movements in bed | [7,33] | ||
| Body weight | [38] | ||
| Heart rate | [7,33,38] | ||
| Respiratory rate | [7,33] | ||
| Sleep apneas | [38] | ||
| Chair sensor | Time on chair | [8] | |
| Proximity beacon | Frequency of forgetting wallet/keys | [30] | |
| Medication box | Frequency of forgetting medication | [30] | |
| Smart plug | Television use | [30] |
Table 3.
Risk of Bias assessment using Newcastle–Ottawa quality assessment scale for observational studies (★ = adequate, ☆ = not done/inadequate).
Table 3.
Risk of Bias assessment using Newcastle–Ottawa quality assessment scale for observational studies (★ = adequate, ☆ = not done/inadequate).
| Study | Patient Selection | Comparability | Outcome | Total | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Representativeness of the Exposed Cohort | Selection of the Non-Exposed Cohort | Ascertainment of Exposure | Outcome of Interest Was Not Present at Start of the Study | Controls for Most Important Factor | Controls for Any Additional Factor | Assessment of Outcome | Follow-Up Length | Adequacy of Follow-Up Cohorts | ||
| Chen et al., 2020 [41] | ☆ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 4 |
| Dadhania et al., 2023 [35] | ☆ | ☆ | ★ | ☆ | ★ | ★ | ★ | ☆ | ★ | 5 |
| Fay-Karmon et al., 2024 [20] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ☆ | ★ | 4 |
| Finch et al., 2017 [32] | ★ | ★ | ★ | ☆ | ★ | ☆ | ★ | ★ | ☆ | 6 |
| Harrington et al., 2021 [38] | ☆ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ☆ | 3 |
| Kabelac et al., 2019 [22] | ☆ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 4 |
| Lach et al., 2019 [8] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ☆ | 4 |
| Lam et al., 2022 [26] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 5 |
| Leightley et al., 2021 [34] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ☆ | 4 |
| Lipsmeier et al., 2018 [23] | ☆ | ★ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 6 |
| Liu et al., 2022 [24] | ☆ | ★ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ☆ | 4 |
| Low et al., 2021 [36] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 5 |
| Low et al., 2024 [37] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 5 |
| Muurling et al., 2023 [29] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ☆ | ★ | 4 |
| Noble et al., 2016 [39] | ☆ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ☆ | ★ | 3 |
| Obayashi et al., 2023 [28] | ☆ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 4 |
| Parkinson et al., 2023 [7] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ☆ | ☆ | 3 |
| Perraudin et al., 2018 [42] | ☆ | ☆ | ★ | ☆ | ★ | ☆ | ★ | ★ | ☆ | 4 |
| Rawtaer et al., 2020 [30] | ★ | ★ | ★ | ☆ | ★ | ☆ | ★ | ☆ | ☆ | 5 |
| Ribeiro-Castro et al., 2024 [43] | ★ | ☆ | ☆ | ☆ | ☆ | ☆ | ★ | ★ | ☆ | 3 |
| Saner et al., 2020 [33] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 5 |
| Schalkamp et al., 2024 [25] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ☆ | 4 |
| Wu et al., 2022 [31] | ★ | ☆ | ★ | ☆ | ★ | ★ | ★ | ☆ | ☆ | 5 |
| Zhang et al., 2021 [40] | ★ | ☆ | ★ | ☆ | ☆ | ☆ | ★ | ★ | ★ | 5 |
Table 4.
Cochrane Risk of Bias 2 assessment for randomised controlled studies (D1: Bias arising from the randomisation process; D2: Bias due to deviations from intended interventions; D3: Bias due to missing outcome data; D4 Bias in measurement of the outcome; D5 Bias in the selection of the reported result).
Table 4.
Cochrane Risk of Bias 2 assessment for randomised controlled studies (D1: Bias arising from the randomisation process; D2: Bias due to deviations from intended interventions; D3: Bias due to missing outcome data; D4 Bias in measurement of the outcome; D5 Bias in the selection of the reported result).
| Study | D1 | D2 | D3 | D4 | D5 | Overall |
|---|---|---|---|---|---|---|
| Gatsios et al., 2020 [21] | Low risk | Low risk | Low risk | Low risk | Low risk | Low risk |
| Gaugler et al., 2019 [27] | Low risk | Low risk | Low risk | Low risk | Low risk | Low risk |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Rama, E.; Zuberi, S.; Aly, M.; Askari, A.; Iqbal, F.M. Clinical Outcomes of Passive Sensors in Remote Monitoring: A Systematic Review. Sensors 2025, 25, 3285. https://doi.org/10.3390/s25113285
AMA Style
Rama E, Zuberi S, Aly M, Askari A, Iqbal FM. Clinical Outcomes of Passive Sensors in Remote Monitoring: A Systematic Review. Sensors. 2025; 25(11):3285. https://doi.org/10.3390/s25113285
Chicago/Turabian StyleRama, Essam, Sharukh Zuberi, Mohamed Aly, Alan Askari, and Fahad M. Iqbal. 2025. "Clinical Outcomes of Passive Sensors in Remote Monitoring: A Systematic Review" Sensors 25, no. 11: 3285. https://doi.org/10.3390/s25113285
APA StyleRama, E., Zuberi, S., Aly, M., Askari, A., & Iqbal, F. M. (2025). Clinical Outcomes of Passive Sensors in Remote Monitoring: A Systematic Review. Sensors, 25(11), 3285. https://doi.org/10.3390/s25113285
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
