Standardize the sequence of work or activities as well as the physical layout of the process. For mechanical ventilation as any other process, there exists initiation, maintenance, and termination phases. Each of these nodes of activity need to be carefully scrutinized. Standard physical layout of a cardiac intensive care unit (CICU) room ensures that the appropriate mechanical ventilator and supplies are always present for a scheduled or emergency admission. Electronic order sets for each phase of mechanical ventilation help ensure standard work (see below) for each aspect of the process. In terms of mechanical ventilation, although the maintenance aspect of ventilator support typically comprises the largest duration of time, intubation and extubation actually represent the most dangerous time for critically ill patients, particularly those with cardiovascular disease [6–8]. Accordingly, as part of the process flow of mechanical ventilation in the CICU, it would be prudent to include team huddles at both initiation and termination of mechanical ventilation to consider worst case scenarios and how to respond to them [9].

Implement standard work for all elements of the process with explicit methods. This requirement is considered in detail in another portion of this chapter. However, with respect to flow, standard work will define the expected clinical pathway or trajectory for the process. There can be no measured improvement without standard work because in its absence nuisance variables typically obscure any meaningful process change [10]. In developing standard work it is useful to scrutinize the work where it is actually performed and the way it is actually done. Standard work provides the infrastructure for process flow and continuous improvement. Along the process flow, standard work should be evidence-based as possible and consensus-based where evidence is not available. In the case of mechanical ventilation, standard work can be composed for emergent, hypoxemic, acute respiratory failure and various management aspects of pediatric acute respiratory distress syndrome as well as management of the post-operative surgical patient [11]. It should be stressed that in distinction to a reliability method, standard work involves a time component as will be illustrated for mechanical ventilation. Decision support tools mandating scheduled assessments provide this time component [12].

Keep pace with customer (patient) demand. Within the current context this concept refers to recognition of physiologic prompting by the patient that will enable to pull off the flow process. Administratively this may also refer to administrative load-levelling. For example, a surgical schedule that is balanced and extends over the course of the entire week ensuring that the CICU is not overloaded some days of the week and underloaded during other days of the week.

Ensure flexibility of the system. This characteristic relates to responding to variation in demand. In the case of mechanical ventilation, proper process flow would dictate continuous, deliberate weaning around the clock (24/7/365). In the case of mechanical ventilation, flexibility also particularly relates to dynamics of the patient’s cardiovascular status, analgesic and sedative needs, nutritional status and input, as well as fluid balance. All of these will influence the process flow along the trajectory of a patient’s mechanical ventilation course.

Pull the flow. In the quality improvement literature this characteristic refers to the notion of pulling the flow rather than pushing it [13]. Specifically for mechanical ventilation, the practice of scheduled assessments to facilitate early weaning of mechanical ventilation results in such a pull system [14]. Frequently, the need for mechanical ventilation restrains the patient to the CICU. Typically, when there is no longer a need for mechanical ventilation, the patient can be discharged to the general ward or even to home. Weaning mechanical ventilation typically encumbers the majority of time that the patient spends on mechanical ventilation. In a pull system, instead of weaning the ventilation with some haphazard schedule, weaning is accomplished according to the physiology of the patient that actually drives the weaning process [12, 15–18]. In this scenario, weaning is accomplished when the patient is ready and at all times of the day. Ideally, there is some form of instantaneous feedback regarding the consequences of a particular pull on the process flow. In the case of mechanical ventilation, the CICU provides multiple monitoring signals of the success or not of a particular weaning step. Computerized decision support tools to pull the mechanical ventilation weaning process may prove to be useful in the future. Such tools utilize so-called time-based kanban signalling for assessment and possible weaning interventions. For mechanical ventilation, input variables may include ventilatory rate, PaCO₂ or end tidal carbon dioxide, and SpO₂ or PaO_2. Output variables that may be suggested include changes in peak inspiratory pressure, tidal volume, positive end expiratory pressure or fraction of inspired oxygen concentration. Such decision support tools prompt proactive, scheduled pull of the process flow, in this case weaning of mechanical ventilation. Such process flow tools have been shown to decrease the duration of both mechanical ventilation and intensive care unit (ICU) length of stay [12, 15–18].