The objective of Intensive Care Medicine, as defined by the international societies of Intensive Care Medicine in the Santander 2012 statement,2 is to provide critically ill patients with medical care tailored to their needs, of high quality and as safe as possible. Intensive Care Medicine is one of the main elements in modern healthcare systems – Intensive Care Units (ICUs) being in increasing demand, and with an important healthcare cost impact. It is believed that in the United States more than half of the population will be admitted to an ICU at some point in life, and that a significant percentage will die in these Units, consuming between 0.5% and 1% of the country's Gross Domestic Product.3,4 More than 5 million patients are admitted annually to ICUs in the United States. The 5 primary ICU admission diagnoses are, in decreasing order: respiratory insufficiency/failure, postoperative management, ischemic heart disorders, sepsis, and heart failure. However, the availability of ICU beds varies greatly across different countries and healthcare systems.5

The management of some critical processes entails major economic and sanitary resource consumption. As an example, in our country (Spain), the annual incidence of severe sepsis is 14.1 cases/10,000 inhabitants, with an overall mortality rate of 33% and a mean hospital stay of 28.9 days. The annual cost of such management (Community of Madrid, Spain) is 70 million €, which represents a significant impact in terms of healthcare resource consumption.6 In 2005, critical care services in the United States were estimated to cost $81.7 billion, or 0.66% of the gross domestic product.3,4

ICU patients are in a critical condition, i.e., with potentially life-threatening organ dysfunctions or disorders. They require continuous monitoring and often need organ functional support measures. Although these patients conform a heterogeneous population, all of them share the need for a higher level of acute care than most hospitalized subjects. Frequently, patients in the ICU require cardiovascular or ventilatory support, invasive monitoring and intensive observation by nursing and medical staff members, with a greater reliance upon technology to keep critically ill patients alive (e.g., mechanical ventilation, hemodialysis, plasmapheresis, extracorporeal membrane oxygenation, etc.). Despite the widespread belief that the mortality rate is inherently high in our specialty, the percentage of patients who survive to discharge from the ICU is 90% in overall terms.7

Scarcity of available ICU beds sometimes leads to rejection of ICU admission8 or to delays in admission to these Units, and that early discharge in turn can lead to a later need for patient re-admission.9 Both circumstances are clearly correlated to an increased patient risk and a poorer prognosis (increased length of stay and mortality in both the ICU and in Hospital), and imply greater associated costs arising from the need to use more complex support techniques, an increased need for vital support, and a longer length of stay in the ICU and in Hospital.3,10,11 As an example, Cardoso et al.11 have reported that each hour of delay in admission of a patient to the ICU is associated with a 1.5% increase in the risk of death in the ICU and a 1% increase in Hospital mortality. Likewise, Sakr et al.12 have found mortality among critically ill patients to be clearly related to the initial evolution of organ failure and the SOFA score at the time of admission to the ICU. If readmission to the ICU proves necessary, there is a four-fold increase in mortality risk, and the length of Hospital stay is moreover doubled.13,14 Furthermore, information is still lacking on the influence of the availability of beds in the ICU upon the decisions referred to patient triage and upon the outcome of patients in which admission has been rejected.15

Not all potentially acute patients are admitted to the ICU; indeed, we can find patients in conventional hospitalization units with diagnoses such as severe sepsis, septic shock, acute organ dysfunctions, and even supposedly specific conditions such as acute respiratory distress syndrome (ARDS).16,17 In addition, we must have a forecast of available ICU beds for high-risk surgical patients who benefit from scheduled admission, since it may result in an improved prognosis.18 More than 50% of all hospitalized patients do not receive optimal treatment before admission to the ICU, and a large percentage of admissions are moreover avoidable.19 Chen et al. have recently reported that 70% of the patients with a predicted mortality rate of over 30% were not admitted to the ICU and received treatment in conventional hospitalization wards.20

In this situation, intensivists, in collaboration with other hospital specialists, must continue working to improve performance: the aim is to be more EFFECTIVE by decreasing morbidity and mortality among our patients; to be more EFFICIENT by reducing the length of stay in the ICU and in Hospital, reducing drug expenditure and the use of consumables (curbing the costs associated with healthcare and improving its health performance); and to improve PERCEIVED QUALITY on the part of the patients, their family, and the healthcare team.21,22

The early clinical care of patients at risk admitted to conventional hospital wards could favorably influence the clinical course and disease prognosis. In effect, if the clinical condition is serious enough, admission to the ICU must be advanced, avoiding unnecessary delays in treatment. Even more importantly, however, if the patient clinical condition allows diagnostic reorientation or intensified therapeutic measures, clinical improvement might be achieved, thereby avoiding the need for admission to the ICU, and also allowing better management of the available healthcare resources (ICU beds).1

In recent years, the concept of early detection has become so ingrained that the European Resuscitation Council recommendations on the management of cardiac arrest presently include as first link in the “survival chain” the introduction of measures aimed at preventing cardiac arrest, by detecting previously occurring pathophysiological alterations. This is a necessary measure, for despite structured organization and quality in management of cardiac arrest, the results obtained are disappointing, with a survival rate to Hospital discharge of about 17% according to the American Heart Association (AHA). In the case of an initial shockable rhythm, this figure increases to 37%, but drops to only 11% in the case of asystolia or pulseless electrical activity.23,24 It has been demonstrated that during the hours before a serious patient event occurs (aggravation or even cardiac arrest), detectable physiological changes have already developed. These altered parameters are common to initial disease in general, since they reflect organ failure, such as heart rate, blood pressure, respiratory frequency, oxygenation, urine output, or level of consciousness.25–29 In other words, most cases of in-hospital cardiac arrest are not a sudden or unpredictable event.

Working methods are thus required to allow the early detection of acute and potentially acute patients in any location within the hospital, with intervention in the early stages of the disease before damage becomes established. Such measures may comprise direct admission to the ICU, or diagnostic and/or therapeutic intervention in the ward where the patient is located, with protocolized close follow-up. Moreover, it is clear that the best results are obtained from collaboration among professionals. It is therefore essential to work with the Ward physician. In this way the medical care provided by intensivists has gradually crossed the physical boundaries of the ICU.

The implementation of this workflow benefits from the continuous training of the healthcare staff involved: intensive care and hospital physicians and nurses. The aim is to strengthen the concept of early detection, and to train in the detection of warning signs and in the initially required actions.26,30 Nurses play a key role, since they spend more time in contact with the patient, and are responsible for taking vital signs.

Different solutions have been proposed to detect patients at risk:

• 1.

  On request systems: such systems involve the definition of alarm or severity criteria that trigger warning to the intensivist. These criteria are usually a combination of clinical and laboratory values, such as blood pressure, heart rhythm alterations, hypoxia or respiratory frequency alterations, decreased level of consciousness, or lowered diuresis.

  Initially, each hospital designed its own alarm systems, based mainly on Australian multiparametric designs, and on the English model using combined and weighted systems that generate a graded response rate according to the score achieved.31,32 Up to 25 alarm criteria have been described. This wide range reflects a lack of full confidence in their usefulness, and moreover does not allow comparison among studies. In fact, the success of the Medical Emergency Teams lies in correctly identifying patients at risk and in the level of monitoring of the healthcare staff involved. The criteria therefore must have sufficient sensitivity, must be simple to obtain, and should not increase the workload. In addition, continuous training of healthcare staff is easier when these criteria are stable and concise. In the year 2010, an alarm system known as ViEWS33 was published. It includes 7 clinical variables with different scores according to their degree of alteration. The sum of all items can be used to scale the type of response to be given in each situation. In 2012, this system was externally validated in a Canadian study34 as an early predictor of hospital mortality. That same year a modified ViEWS system was described, including lactate levels, and affording an improved positive predictive value referred to hospital mortality.35

• 2.

  On request systems designed for specific diseases: The evolution of different diseases such as stroke, acute coronary syndrome, polytraumatism or sepsis, among others, has proved to be dependent on how quickly treatment is started. In this sense, there are specific alarm systems that combine clinical and laboratory data, and provide a decision tree for the initial procedures and the need to notify the ICU.36–39 The most recent case is represented by the septic patient care programs, designed and disseminated in many hospitals following the worldwide initiative of the Surviving Sepsis Campaign. A number of publications in our setting have shown these initiatives to afford greater adherence to clinical practices guidelines, with benefit for this group of patients, a shortened hospital stay, and even lesser mortality.40–42

• 3.

  Prospective detection of patients at risk: This comprises a different approach, with the assessment of patients who, for various reasons, are considered to be at risk of severe worsening. The following criteria can be established:

  • •

    Discharge from the ICU in certain situations, including after a long ICU stay, evolving organ failure, surgical patients with concomitant severe medical conditions, etc.

  • •

    Patients admitted to hospital wards and considered to be “at risk” because of their acute condition, such as those admitted to Emergencies or Emergency Observation Areas.

The issue to be resolved in these patients subjected to prospective evaluation after due screening (according to physical location in the hospital) is precisely the impossibility of covering all inpatients, which should be the ultimate goal. In this context, and thanks to the digitalization of medical records, it is now possible to use electronic systems to detect pre-defined parameters, such as laboratory data, microbiological information, vital signs or nursing notes, with a view to identifying candidate patients.

The program can be complemented with the use of remote vital signs monitoring. This would be useful to intensify the treatment and follow-up of patients who are not in a critical situation requiring admission to the ICU.43 Furthermore, collaboration with other specialists must be included, so that a daily bidirectional patient “session” is established to allow the “hospital” physician (with an internist and/or surgeon as the leading example) to conduct joint assessment of patients which because of their disease, evolution or clinical history, are of particular concern to the attending physician.

The literature already offers the results of over 10 years of experience with strategies aimed at early detection and intervention in critically ill patients. These strategies are called rapid response systems (RRSs), Medical Emergency Teams (METs) or Critical Care Outreach (CCO) systems, usually led by an intensivist. Such teams aim to improve patient outcomes through recognition and intervention before serious deterioration occurs implying cardiac arrest or urgent admission to the ICU. These teams must be trained to assess, diagnose, initiate treatment and decide ICU admission if necessary; the intensivist therefore offers the appropriate professional profile in this sense.26,44–46

The recent systematic review published by Winters et al.47 shows that many hospitals have implemented rapid response systems over the past 15 years to improve recognition of and response to deteriorating patients in the general hospital ward. Moderate-strength evidence suggests that RRSs are associated with reduced rates of cardiorespiratory arrest and mortality. Important components of successful RRSs include criteria and a system for response team notification and activation; a response team; and an administrative and quality improvement component to train staff, collect and analyze event data, provide feedback, coordinate resources, and ensure improvement or maintenance over time. Implementation issues are critical in RRSs, because rates of use are often suboptimal because of various barriers that could be improved.

Previous recommendations and clinical guidelines have always stressed the importance for all hospitals to develop a plan based on their needs, resources and organization. The fundamental aspects are multidisciplinary collaboration, staff training in the recognition and correct interpretation of the severity signs, the leadership of an intensivist, and involvement of the hospital executive bodies.

Given this background, in our center we decided to develop an Intensive Care Medicine management system fundamented upon critically ill patient safety throughout the hospitalization process. It establishes a strategic focus on “early detection of the critically ill patient outside the ICU”, which refers to the identification of patients at risk outside the Unit, and is based on the recognition, diagnostic orientation and early treatment of acute patients, in collaboration with other clinical specialties, and regardless of their location within the Hospital.

At this time it can be affirmed that there is still important work to be done in the detection of severity and early intervention in patients at risk of organ dysfunction. This work must be adapted to the circumstances of each center and must include training in the detection of severity, multidisciplinary work in the global patient clinical process, and the use of technological systems for intervention referred to laboratory parameters and the monitoring of physiological parameters – with effective and efficient use of the information generated. Not only information must be generated, but also efficient management systems for this information must be sought.

The authors declare no conflicts of interest.