The potential for harm from alarm fatigue in single-room NICUs

Machine learning in critical care: state-of-the-art and a sepsis case study

BACKGROUND

Like other scientific fields, such as cosmology, high-energy physics, or even the life sciences, medicine and healthcare face the challenge of an extremely quick transformation into data-driven sciences. This challenge entails the daunting task of extracting usable knowledge from these data using algorithmic methods. In the medical context this may for instance realized through the design of medical decision support systems for diagnosis, prognosis and patient management. The intensive care unit (ICU), and by extension the whole area of critical care, is becoming one of the most data-driven clinical environments.

RESULTS

The increasing availability of complex and heterogeneous data at the point of patient attention in critical care environments makes the development of fresh approaches to data analysis almost compulsory. Computational Intelligence (CI) and Machine Learning (ML) methods can provide such approaches and have already shown their usefulness in addressing problems in this context. The current study has a dual goal: it is first a review of the state-of-the-art on the use and application of such methods in the field of critical care. Such review is presented from the viewpoint of the different subfields of critical care, but also from the viewpoint of the different available ML and CI techniques. The second goal is presenting a collection of results that illustrate the breath of possibilities opened by ML and CI methods using a single problem, the investigation of septic shock at the ICU.

CONCLUSION

We have presented a structured state-of-the-art that illustrates the broad-ranging ways in which ML and CI methods can make a difference in problems affecting the manifold areas of critical care. The potential of ML and CI has been illustrated in detail through an example concerning the sepsis pathology. The new definitions of sepsis and the relevance of using the systemic inflammatory response syndrome (SIRS) in its diagnosis have been considered. Conditional independence models have been used to address this problem, showing that SIRS depends on both organ dysfunction measured through the Sequential Organ Failure (SOFA) score and the ICU outcome, thus concluding that SIRS should still be considered in the study of the pathophysiology of Sepsis. Current assessment of the risk of dead at the ICU lacks specificity. ML and CI techniques are shown to improve the assessment using both indicators already in place and other clinical variables that are routinely measured. Kernel methods in particular are shown to provide the best performance balance while being amenable to representation through graphical models, which increases their interpretability and, with it, their likelihood to be accepted in medical practice.

Does the architectural layout of a NICU affect alarm pressure? A comparative clinical audit of a single-family room and an open bay area NICU using a retrospective study design

OBJECTIVES

To determine differences in alarm pressure between two otherwise comparable neonatal intensive care units (NICUs) differing in architectural layout-one of a single-family room (SFR) design and the other of an open bay area (OBA) design.

DESIGN

Retrospective audit of more than 2000 patient days from each NICU cataloguing the differences in the number and duration of alarms for critical and alerting alarms, as well as the interaction of clinicians with the patient monitor.

SETTING

Two level 3 NICUs.

RESULTS

A total of more than 150 000 critical and 1.2 million alerting alarms were acquired from the two NICUs. The number of audible alarms and the associated noise pollution varied considerably with the OBA NICU generating 44% more alarms per infant per day even though the SFR NICU generated 2.5 as many critical desaturation alarms per infant per day.

CONCLUSION

Differences in the architectural layout of NICUs and the consequent differences in delays, thresholds and distribution systems for alarms are associated with differences in alarm pressure.

A Strategy to Reduce Critical Cardiorespiratory Alarms due to Intermittent Enteral Feeding of Preterm Neonates in Intensive Care

Background

Many preterm infants require enteral feeding as they cannot coordinate sucking, swallowing, and breathing. In enteral feeding, milk feeds are delivered through a small feeding tube passed via the nose or mouth into the stomach. Intermittent milk feeds may either be administered using a syringe to gently push milk into the infant’s stomach (push feed) or milk can be poured into a syringe attached to the tube and allowed to drip in by gravity (gravity feed). This practice of enteral feeding is common in neonatal intensive care units. There is, however, no evidence in the literature to recommend the use of one method of feeding over the other.

Objective

The aim of this study was to investigate which of the two methods of feeding is physiologically better tolerated by infants, as measured by the incidence of critical cardiorespiratory alarms during and immediately after feeding.

Methods

We conducted a prospectively designed observational study with records of all feeding episodes in infants of gestational age less than 30 weeks at birth and with a minimum enteral intake of 100 mL/kg/day. In total, 2140 enteral feeding episodes were noted from 25 infants over 308 infant-days with records for several characteristics of the infants (eg, gestational age), feeding (eg, the position of infants), and of nursing-care events before feeding (eg, diapering). Logistic regression with mixed effects was used to model cardiorespiratory alarms for the push and gravity methods of feeding.

Results

After adjustments were made for all confounding variables, the position of infants was found to be statistically significant in changing the outcome of critical alarms for the two methods of feeding (P=.02). For infants in the lateral position, push feeds led to 40% more instances of one or more critical cardiorespiratory alarms in comparison with the gravity method. Both methods of feeding created a statistically comparable number of alarms for infants in the prone position.

Conclusions

This study provides objective data that may assist in optimizing enteral feeding protocols for premature infants. The incidence of critical cardiorespiratory alarms for infants in the lateral position can be lowered by the use of gravity instead of push feeding. No differences were observed between the two types of feeding when infants were in the prone position.

Pattern discovery in critical alarms originating from neonates under intensive care

Patient monitoring generates a large number of alarms, the vast majority of which are false. Excessive non-actionable medical alarms lead to alarm fatigue, a well-recognized patient safety issue. While multiple approaches to reduce alarm fatigue have been explored, patterns in alarming and inter-alarm relationships, as they manifest in the clinical workspace, are largely a black-box and hamper research efforts towards reducing alarms. The aim of this study is to detect opportunities to safely reduce alarm pressure, by developing techniques to identify, capture and visualize patterns in alarms. Nearly 500 000 critical medical alarms were acquired from a neonatal intensive care unit over a 20 month period. Heuristic techniques were developed to extract the inter-alarm relationships. These included identifying the presence of alarm clusters, patterns of transition from one alarm category to another, temporal associations amongst alarms and determination of prevalent sequences in which alarms manifest. Desaturation, bradycardia and apnea constituted 86% of all alarms and demonstrated distinctive periodic increases in the number of alarms that were synchronized with nursing care and enteral feeding. By inhibiting further alarms of a category for a short duration of time (30 s/60 s), non-actionable physiological alarms could be reduced by 20%. The patterns of transition from one alarm category to another and the time duration between such transitions revealed the presence of close temporal associations and multiparametric derangement. Examination of the prevalent alarm sequences reveals that while many sequences comprised of multiple alarms, nearly 65% of the sequences were isolated instances of alarms and are potentially irreducible. Patterns in alarming, as they manifest in the clinical workspace were identified and visualized. This information can be exploited to investigate strategies for reducing alarms.