Let Sleeping Patients Lie, avoiding unnecessary overnight vitals monitoring using a clinically based deep-learning model

Safer and more efficient vital signs monitoring protocols to identify the deteriorating patients in the general hospital ward: an observational study

Background

The frequency at which patients should have their vital signs (e.g. blood pressure, pulse, oxygen saturation) measured on hospital wards is currently unknown. Current National Health Service monitoring protocols are based on expert opinion but supported by little empirical evidence. The challenge is finding the balance between insufficient monitoring (risking missing early signs of deterioration and delays in treatment) and over-observation of stable patients (wasting resources needed in other aspects of care).

Objective

Provide an evidence-based approach to creating monitoring protocols based on a patient's risk of deterioration and link these to nursing workload and economic impact.

Design

Our study consisted of two parts: (1) an observational study of nursing staff to ascertain the time to perform vital sign observations; and (2) a retrospective study of historic data on patient admissions exploring the relationships between National Early Warning Score and risk of outcome over time. These were underpinned by opinions and experiences from stakeholders.

Setting and participants

Observational study: observed nursing staff on 16 randomly selected adult general wards at four acute National Health Service hospitals. Retrospective study: extracted, linked and analysed routinely collected data from two large National Health Service acute trusts; data from over 400,000 patient admissions and 9,000,000 vital sign observations.

Results

Observational study found a variety of practices, with two hospitals having registered nurses take the majority of vital sign observations and two favouring healthcare assistants or student nurses. However, whoever took the observations spent roughly the same length of time. The average was 5:01 minutes per observation over a 'round', including time to locate and prepare the equipment and travel to the patient area. Retrospective study created survival models predicting the risk of outcomes over time since the patient was last observed. For low-risk patients, there was little difference in risk between 4 hours and 24 hours post observation.

Conclusions

We explored several different scenarios with our stakeholders (clinicians and patients), based on how 'risk' could be managed in different ways. Vital sign observations are often done more frequently than necessary from a bald assessment of the patient's risk, and we show that a maximum threshold of risk could theoretically be achieved with less resource. Existing resources could therefore be redeployed within a changed protocol to achieve better outcomes for some patients without compromising the safety of the rest. Our work supports the approach of the current monitoring protocol, whereby patients' National Early Warning Score 2 guides observation frequency. Existing practice is to observe higher-risk patients more frequently and our findings have shown that this is objectively justified. It is worth noting that important nurse-patient interactions take place during vital sign monitoring and should not be eliminated under new monitoring processes. Our study contributes to the existing evidence on how vital sign observations should be scheduled. However, ultimately, it is for the relevant professionals to decide how our work should be used.

Study registration

This study is registered as ISRCTN10863045.

Funding

This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: 17/05/03) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 6. See the NIHR Funding and Awards website for further award information.

Development of a dynamic prediction model for unplanned ICU admission and mortality in hospitalized patients

Frequent assessment of the severity of illness for hospitalized patients is essential in clinical settings to prevent outcomes such as in-hospital mortality and unplanned admission to the intensive care unit (ICU). Classical severity scores have been developed typically using relatively few patient features. Recently, deep learning-based models demonstrated better individualized risk assessments compared to classic risk scores, thanks to the use of aggregated and more heterogeneous data sources for dynamic risk prediction. We investigated to what extent deep learning methods can capture patterns of longitudinal change in health status using time-stamped data from electronic health records. We developed a deep learning model based on embedded text from multiple data sources and recurrent neural networks to predict the risk of the composite outcome of unplanned ICU transfer and in-hospital death. The risk was assessed at regular intervals during the admission for different prediction windows. Input data included medical history, biochemical measurements, and clinical notes from a total of 852,620 patients admitted to non-intensive care units in 12 hospitals in Denmark's Capital Region and Region Zealand during 2011-2016 (with a total of 2,241,849 admissions). We subsequently explained the model using the Shapley algorithm, which provides the contribution of each feature to the model outcome. The best model used all data modalities with an assessment rate of 6 hours, a prediction window of 14 days and an area under the receiver operating characteristic curve of 0.898. The discrimination and calibration obtained with this model make it a viable clinical support tool to detect patients at higher risk of clinical deterioration, providing clinicians insights into both actionable and non-actionable patient features.

Society of Anesthesia and Sleep Medicine Position Paper on Patient Sleep During Hospitalization

This article addresses the issue of patient sleep during hospitalization, which the Society of Anesthesia and Sleep Medicine believes merits wider consideration by health authorities than it has received to date. Adequate sleep is fundamental to health and well-being, and insufficiencies in its duration, quality, or timing have adverse effects that are acutely evident. These include cardiovascular dysfunction, impaired ventilatory function, cognitive impairment, increased pain perception, psychomotor disturbance (including increased fall risk), psychological disturbance (including anxiety and depression), metabolic dysfunction (including increased insulin resistance and catabolic propensity), and immune dysfunction and proinflammatory effects (increasing infection risk and pain generation). All these changes negatively impact health status and are counterproductive to recovery from illness and operation. Hospitalization challenges sleep in a variety of ways. These challenges include environmental factors such as noise, bright light, and overnight awakenings for observations, interventions, and transfers; physiological factors such as pain, dyspnea, bowel or urinary dysfunction, or discomfort from therapeutic devices; psychological factors such as stress and anxiety; care-related factors including medications or medication withdrawal; and preexisting sleep disorders that may not be recognized or adequately managed. Many of these challenges appear readily addressable. The key to doing so is to give sleep greater priority, with attention directed at ensuring that patients' sleep needs are recognized and met, both within the hospital and beyond. Requirements include staff education, creation of protocols to enhance the prospect of sleep needs being addressed, and improvement in hospital design to mitigate environmental disturbances. Hospitals and health care providers have a duty to provide, to the greatest extent possible, appropriate preconditions for healing. Accumulating evidence suggests that these preconditions include adequate patient sleep duration and quality. The Society of Anesthesia and Sleep Medicine calls for systematic changes in the approach of hospital leadership and staff to this issue. Measures required include incorporation of optimization of patient sleep into the objectives of perioperative and general patient care guidelines. These steps should be complemented by further research into the impact of hospitalization on sleep, the effects of poor sleep on health outcomes after hospitalization, and assessment of interventions to improve it.

Potential for remote vital sign monitoring to improve hospital patient sleep: A feasibility study

BACKGROUND

Even low-acuity patients suffer from disrupted sleep in the hospital in part due to routine overnight vital sign (VS) checks. When invasive monitoring is not needed, vital sign monitoring devices (VSMDs) similar to consumer-grade health monitors may play a role in promoting sleep, which can aid healing and recovery.

METHODS

We provided one VSMD to neuroscience ward patients during their hospital stays and used surveys to assess patient and nurse attitudes toward the device and the impact of the device on patient comfort. We also compared VSMD-streamed vS data to nurse-recorded vS data in the chart to evaluate the consistency of data streaming and data concordance between the device and nurse-collected vital sign values.

FINDINGS

21 patients and 15 nurses enrolled. Overall, patients and nurses responded positively to the device and patients preferred wearing the device to receiving manual vital checks overnight. The most common device-related cause of sleep disruption per patients was device weight (29%). Device vS were concordant with nurse vS on average but there was significant variance in agreement between nurse and device values.

INTERPRETATION

Patients and nurses feel positively about the use of VSMDs and their use in the hospital. The device we tested may be limited in its sleep promotion by its weight and patient comfort assessment. Further research is needed to assess the precision of the device in measuring vital signs when used in a clinical setting. Future studies should compare VSMD models and assess their impacts on patient sleep in the absence of manual vS checks overnight.

FUNDING

Funding provided by the Sara & Evan Williams Foundation Endowed Neurohospitalist Chair at UCSF.

Prediction of intrapartum fever using continuously monitored vital signs and heart rate variability

Objectives

Neonatal early onset sepsis (EOS), bacterial infection during the first seven days of life, is difficult to diagnose because presenting signs are non-specific, but early diagnosis before birth can direct life-saving treatment for mother and baby. Specifically, maternal fever during labor from placental infection is the strongest predictor of EOS. Alterations in maternal heart rate variability (HRV) may precede development of intrapartum fever, enabling incipient EOS detection. The objective of this work was to build a predictive model for intrapartum fever.

Methods

Continuously measured temperature, heart rate, and beat-to-beat RR intervals were obtained from wireless sensors on women (n = 141) in labor; traditional manual vital signs were taken every 3-6 hours. Validated measures of HRV were calculated in moving 5-minute windows of RR intervals: standard deviation of normal-to-normal intervals (SDNN) and root mean square of successive differences (RMSSD) between normal heartbeats.

Results

Fever (>38.0 °C) was detected by manual or continuous measurements in 48 women. Compared to afebrile mothers, average SDNN and RMSSD in febrile mothers decreased significantly (p < 0.001) at 2 and 3 hours before fever onset, respectively. This observed HRV divergence and raw recorded vitals were applied to a logistic regression model at various time horizons, up to 4-5 hours before fever onset. Model performance increased with decreasing time horizons, and a model built using continuous vital signs as input variables consistently outperformed a model built from episodic vital signs.

Conclusions

HRV-based predictive models could identify mothers at risk for fever and infants at risk for EOS, guiding maternal antibiotic prophylaxis and neonatal monitoring.

Patient-nurse agreement on inpatient sleep and sleep disturbing factors

Background

Sleep is vital for recovery during hospital stay. Many sleep-promoting interventions have been investigated in the past. Nurses seem to overestimate their patients sleep and their perspective is needed for these interventions to be successfully implemented.

Objectives

To assess the patient's and nurse's agreement on the patient's sleep and factors disturbing sleep.

Methods

The instruments used included 1) five Richard-Campbell Sleep Questionnaire (RCSQ) items plus a rating of nighttime noise and 2) the Consensus Sleep Diary (CSD). The mean of the five RCSQ items comprised a total score, which reflects sleep quality. Once a week, unannounced, nurses and patients were asked to fill in questionnaires concerning last night's sleep. Neither nurses nor patients knew the others' ratings. Patient-nurse agreement was evaluated by using median differences and Bland-Altman plots. Reliability was evaluated by using intraclass correlation coefficients.

Results

Fifty-five paired patient-nurse assessments have been completed. For all RCSQ subitems, nurses' scores were higher (indicating “better” sleep) than patients’ scores, with a significantly higher rating for sleep depth (median [IQR], 70 [40] vs 50 [40], P = .012). The Bland-Altman plots for the RSCQ Total Score (r = 0.0593, P = .008) revealed a significant amount of variation (bias). The intra-class correlation coefficient (ICC) indicated poor reliability for all 7 measures (range −0.278 – 0.435). Nurses were relatively overestimating their own role in causing sleep disturbances and underestimating patient-related factors.

Conclusions

Nurses tend to overestimate patients’ sleep quality as well as their own role in causing sleep disturbances.

•

There is a relatively poor agreement between patient self-reported and nurse proxy-reported sleep quality.

•

Nurses tend to overestimate inpatients' sleep quality.

•

Nurses overestimate their own role in causing sleep disturbances.

The role of artificial intelligence in enhancing clinical nursing care: A scoping review

AIM

To present an overview of how artificial intelligence has been used to improve clinical nursing care.

BACKGROUND

Artificial intelligence has been reshaping the healthcare industry but little is known about its applicability in enhancing nursing care.

EVALUATION

A scoping review was conducted. Seven electronic databases (CINAHL, Cochrane Library, EMBASE, IEEE Xplore, PubMed, Scopus, and Web of Science) were searched from 1 January 2010 till 20 December 2020. Grey literature and reference lists of included articles were also searched.

KEY ISSUES

Thirty-seven studies encapsulating the use of artificial intelligence in improving clinical nursing care were included in this review. Six use cases were identified - documentation, formulating nursing diagnoses, formulating nursing care plans, patient monitoring, patient care prediction such as falls prediction (most common) and wound management. Various techniques of machine learning and classification were used for predictive analyses and to improve nurses' preparedness and management of patients' conditions CONCLUSION: This review highlighted the potential of artificial intelligence in improving the quality of nursing care. However, more randomized controlled trials in real-life healthcare settings should be conducted to enhance the rigor of evidence.

IMPLICATIONS FOR NURSING MANAGEMENT

Education in the application of artificial intelligence should be promoted to empower nurses to lead technological transformations and not passively trail behind others.

AI Tools Poised to Improve Patient Health Care

Technologies to provide early predictions of breast cancer risk, to identify which hospital patients actually should have their vital signs monitored overnight and which should be left to their restorative sleep, and to swiftly identify rare infant diseases are all joining the ranks of approaches that are powered by artificial intelligence (AI).

Changing Hospital Care For Older Adults: The Case for Geriatric Hospitals in the United States

Hospital care of frail older adults is far from optimal. Although some geriatric models of care have been shown to improve outcomes, the effect size is small and models are difficult to fully implement, sustain and replicate. The two root causes for these shortcomings are competing interests (high revenue generating diseases, procedures and surgeries) and current hospital cultures (for example a culture of safety that emphasizes bed alarms and immobility rather than frequent ambulation). Geriatric hospitals would be hospitals completely dedicated to the care of frail older patients, a group which is most vulnerable to the negative consequences of a hospitalization. They would differ from a typical adult hospital because they could implement evidence based principles of successful geriatric models of care on a hospital wide basis, which would make them sustainable and allow for scaling up of proven outcomes. Innovative structural designs, unachievable in a typical adult hospital, would enhance mobility while maintaining safety. Financial viability and stability would be a challenge but should be feasible, likely through affiliation with larger health care systems with other hospitals because of cost savings associated with geriatric models of care (decreased length of stay, increased likelihood of discharge home, without increasing costs).

Melatonin and melatonin-receptor agonists to prevent delirium in hospitalized older adults: An umbrella review

BACKGROUND

Alterations in circadian rhythm play an important role in the development of delirium. In this umbrella review, we examined the efficacy of melatonin and ramelteon for delirium prevention in hospitalized older adults.

METHODS

Umbrella review methodology from the Joanna Briggs Institute guided the review process. Only meta-analyses were included. Risk of bias was evaluated using the AMSTAR-2 checklist.

RESULTS

Three meta-analyses were included in this review. The quality of studies was low-to-moderate. Two meta-analyses reported a significant reduction in delirium using melatonin or ramelteon (pooled OR and 95% confidence intervals ranged from 0.41 [0.19-0.86] to 0.63 [0.46-0.87]). Melatonergics significantly reduced delirium on medical units (OR = 0.25, 95% CI 0.07-0.88) but not surgical units (OR = 0.62, 0.16-2.43). Heterogenity was high, with I² ranging from 72.14% to 84%.

CONCLUSIONS

Melatonergics appear to prevent delirium among hospitalized older adults, particularly those on medical units. Based on these results, providers may consider using melatonergics as complements to high-quality multicomponent delirium prevention.

Sleep is essential to health: an American Academy of Sleep Medicine position statement

CITATION

Sleep is a biological necessity, and insufficient sleep and untreated sleep disorders are detrimental for health, well-being, and public safety. Healthy People 2030 includes several sleep-related objectives with the goal to improve health, productivity, well-being, quality of life, and safety by helping people get enough sleep. In addition to adequate sleep duration, healthy sleep requires good quality, appropriate timing, regularity, and the absence of sleep disorders. It is the position of the American Academy of Sleep Medicine (AASM) that sleep is essential to health. There is a significant need for greater emphasis on sleep health in education, clinical practice, inpatient and long-term care, public health promotion, and the workplace. More sleep and circadian research is needed to further elucidate the importance of sleep for public health and the contributions of insufficient sleep to health disparities.

CITATION

Ramar K, Malhotra RK, Carden KA, et al. Sleep is essential to health: an American Academy of Sleep Medicine position statement. J Clin Sleep Med. 2021;17(10):2115-2119.