Hello, my name is Maureen Madden, and I'm a professor of pediatrics at Rutgers Robert Wood Johnson Medical School and a nurse practitioner at Bristol Myers Squibb Children's Hospital. I'm going to discuss chronic critical illness and medical complexity in the pediatric ICU. I do not have any financial disclosures, and no off-label or investigational use will be discussed. In this presentation about chronically critically ill children and the related medical complexity, we'll begin by defining our terms. This will be followed by a look at factors in epidemiology, including demographics, the patterns of use, and costs associated with these patients. We'll go over outcomes and review the implications for practice, and finally we'll touch on home ventilation. Children with medical complexity are members of a broader group of children with special health care needs. How do we define children with medical complexity? These are children with complex chronic health problems who have substantial health care needs and severe functional limitations, and whose care involves high levels of health resource utilization. This group also includes a subgroup of individuals who are chronically critically ill. There have been multiple attempts to define this group of children, and a familiarity with the related terms can be useful. Mostly, the definition has been based on codes from the International Classification of Diseases because the efforts have heavily relied on database information. We'll take a look at four of these efforts. First, complex chronic conditions, an umbrella term developed by the team at Children's Hospital of Philadelphia, includes 12 major categories and 75 subcategories. It's based on ICD-9 and ICD-10 diagnoses. By design, it does not include all chronic conditions, but only those that are strongly associated with mortality, morbidity, functional limitations, and high resource use. Second, clinical risk group is a proprietary term that was developed by 3M Health Information Systems. It comprises nine major categories and multiple subcategories. However, there's no clear demarcation of medical complexity, so one dominant condition may be combined with other conditions. As a result, it's difficult to draw many conclusions, so this definition is not used as much in critical care literature. Third, chronic condition indicators were developed by the Agency for Healthcare Research and Quality. This effort was based on an evaluation of about 82,000 codes, which were divided into chronic and non-chronic conditions. However, this definition is not specific for children. Lastly, the Pediatric Medical Complexity Algorithm was developed by Seattle Children's Hospital. It uses ICD-9 codes to categorize diseases into three groups, complex chronic disease, non-complex chronic disease, for example, asthma, or non-chronic disease. So, we have those four definitions of medical complexity. Within the group of children with medical complexity, we have those with chronic critical illness. So, how do we define these? Back in 2014, a consensus definition of chronic critical illness in the adult population was proposed. The criteria are an intensive care unit stay of eight days or more and the presence of at least one of five conditions, prolonged acute mechanical ventilation, tracheostomy, sepsis or another severe infection, severe wounds, or multiple organ failure. How can we adapt this to pediatrics, or rather, can we adapt it? Practice patterns in adults are very different from those in the pediatric population. For example, prolonged mechanical ventilation is routinely used for more than a month with premature infants. Tracheostomy is another example. In adult patients, it might be placed within the first couple of weeks, while in pediatrics, it might be delayed much longer. Finally, length of stay in the intensive care unit isn't always independently associated with clinical status. It may reflect institutional practice, that a hospital may not have a unit that's able and equipped to accept a child ventilated via tracheostomy, whereas an adult patient may be able to transfer for a long-term acute care facility, or specific modalities may require that the child stay in the pediatric intensive care unit. And how do we define pediatric chronic critical illness? First, we know that children with medical complexity and pediatric chronic critical illness are not synonymous. We also know that coming up with a definition of pediatric indicators will have implications for prognosis, outcomes, care delivery, training, and many other metrics. Shapiro and colleagues looked at available studies to see how others were defining chronic critical illness, and they found variability in the literature in terms of definitions. For instance, prolonged length of stay can range anywhere from 12 days to 30 days. On top of that, there are additional considerations for pediatrics, such as the underlying medical complexity for current hospitalizations, and the impact of the child's age and developmental status. What did these investigators do? They proposed a single cohesive definition that could be used for both neonates and pediatric patients, and these are the criteria. For neonates, a neonatal intensive care unit stay longer than 28 days post-term, adjusted for age. For those older, a stay in the pediatric intensive care unit lasting longer than 14 days, or a history of prolonged ICU stay, and two or more acute care ICU admissions within the past 12 months. In addition, the child must have an ongoing dependence on one or more forms of technology to sustain vital functions or persistent multiple vital organ system involvement. So these children need to meet one of the first three criteria, as well as having that ongoing dependence. This is a study that looked at about 53,000 admissions to 54 pediatric intensive care units in 2008, using data from the virtual pediatric intensive care unit systems database. The investigators used the complex chronic condition definition we saw earlier. They found that 53% of children admitted to pediatric intensive care units had complex chronic conditions, and 18.5% had non-complex chronic conditions. That means about 75% of these patients had some sort of chronic medical condition. They also found that for most categories, the odds of dying in the unit were greater for children with chronic complex conditions compared with those with non-complex chronic conditions or no chronic condition. All chronic complex condition categories were associated with prolonged length of stay. Another study used data from the Agency for Healthcare Research and Quality Kids Inpatient Database for two periods separated by a span of about 10 years. Overall, the investigators looked at about 70,000 hospitalizations for critical care services and then employed ICD-9 codes to identify comorbid conditions. They reported a number of interesting findings. First, the number of children with comorbid conditions increased from 35% in 1997 to 41% in 2006. When they looked at mortality in those with comorbidity and without, they found that mortality has improved in both groups. However, comorbidity was still associated with nearly an 11% mortality rate compared with about 8% in children without comorbidity. Length of stay for children with comorbidities decreased from 30 days in 1997 to 26 days in 2006. And for those without comorbidities, it decreased by one day. Still, length of stay is about twice as long for children with comorbidities. Lastly, when the investigators looked at resources, they found that hospital charges for children with comorbidities were double the charges for those without comorbidities, and that seemed the same for both periods. As an attempt to weave this information together, this study used data from pediatric health information systems to look at more than 136,000 pediatric intensive care unit admissions from 2010 through 2013. The investigators used the Pediatric Medical Complexity Algorithm, which uses the categories of chronic disease, non-complex chronic disease, and complex chronic disease. They wound up with 54,000 admissions included, and 53% of these children had complex chronic disease. Those patients accounted for 75% of all intensive care unit resources that were included in the study, such as ventilation days, unit costs, extracorporeal membrane oxygenation, and arterial and central venous catheters. Children with a progressive condition accounted for half of the resources, and children with no chronic disease accounted for less than a quarter of all intensive care unit therapies. Looking at outcomes with chronic critical illness, not much information is available. Two studies used data from the BPIC-U database. The first looked at 118,000 children and found a 3.7 total readmission rate, with 38% of those readmissions occurring within 48 hours. And the second study took data from a later time period, looking at about 93,000 children. The investigators found an unplanned readmission in 11% of those children within one year. Also those with multiple complex chronic conditions had an increased risk of readmission. The third study used a different database and looked at about 109,000 cases. Investigators found that non-elective readmissions were associated with longer lengths of stay, younger age, and several chronic and acute conditions. They also found that when the length of stay exceeded 14 days, there was a 36% rate of readmission, compared with a 13.9% rate when the length of stay was just one day. This is another study of outcomes among children with chronic critical illness. This one looked at about 1,000 children from Australia and New Zealand who had a length of stay beyond 28 days. These children accounted for a 1.3% of total admissions, but 23.5% of bed days. For these patients, survival to discharge was 81%, decreasing to 70% at one year and between 64% and 66% at five years. Not surprisingly, predictors of mortality included single ventral physiology and bone marrow transplantation. This is another outcome study, this time from Sweden. It included nearly 3,700 children with about 5,000 admissions to the Pediatric Intensive Care Unit. A fifth of these children had two or more admissions. The researchers found that a single admission was associated with a mortality rate of 3%, while multiple admissions had more than double that rate at nearly 7%. Then they looked at the number of admissions combined with the presence of a complex chronic condition. They found that children with a single admission plus a complex chronic condition had a mortality rate of 7.5%, while those with multiple admissions and a complex chronic condition had a mortality rate of more than double at 16%. So what are the implications of this? A survey of 51 stakeholders looked at ways to reduce excess hospitalizations. Perceived causes of prolonged intensive care unit stays and excess readmissions included inadequate communication between the healthcare team and the family, gaps in home services and durable medical equipment, inconsistent parent support, and policies that limit patient options. Overall, there is this perception that some patients are just stuck in the unit. They have very long or recurrent stays, so they're almost living in the intensive care unit. There are also staff challenges in dealing with these children. These include changing expectations about the work staff members are doing, the challenges of dealing with moral dilemmas on a regular basis, the impact on the sense of accomplishment, and even effects related to individual patient characteristics. We use chronic ventilation as an example to explore some of these challenges. Listed here are some of the issues we need to think about, such as epidemiology, home ventilation, how to get the patient home, challenges, alternatives, and next steps, which would include guidelines, nutrition, home visits, adult transition, and other aspects of care. We'll just briefly touch on epidemiology because of the variability in the literature with regard to prevalence and outcomes. There's also variation between the U.S. healthcare system and other systems. Some studies included both invasive and non-invasive ventilation, and indications for ventilation have changed since some studies were conducted. For instance, 15 or 20 years ago, when many of these studies were conducted, tracheostomy and ventilation were not considered for children with spinal muscular atrophy. Now it's quite common, so it's very difficult to interpret the data accurately. There are two studies we have to draw on. The first is from Utah. If we look at this single center study, it appears as though the prevalence did not change much from 1996 to 2004, increasing slightly from 5 to 6.3 per 100,000. That might be because the population in Utah is fairly homogeneous. About a quarter of the patients had chronic lung disease, and again, this was a single center study. Second study is from the United Kingdom. Over a 10-year period, the number of children needing ventilation increased significantly. This study combined invasive and non-invasive ventilation. We see that respiratory patients comprised about a third of this population. This study used the Agency for Healthcare Research and Quality Kids Inpatient Database to compare children with chronic complex disease to those with long-term mechanical ventilation as a special group. The investigators looked at discharges from 2006, and almost 8,000 discharges with long-term mechanical ventilation were identified nationwide. They found that those receiving long-term mechanical ventilation had higher mortality rates, longer lengths of stay, and more emergency room admissions than children not receiving ventilation. They also found that from 2000 to 2006, the number of long-term ventilation discharges increased by 55%. Not surprisingly, there was a concurrent increase in aggregate hospital charges of 70%. Most of these discharges were patients younger than four years, and about 50% of the aggregate charges were those who were less than a year old. What about outcomes? How did these children do? In a single-center study of 109 patients from Los Angeles, researchers found that the 12-month incidence of non-elective readmission was 40%. Nearly half of those non-elective readmissions occurred within three months of discharge, and the most common causes of readmission were pneumonia and tracheitis. In the Utah group, the total mortality rate was 17%. As of 2007, about 39% of the 1996 cohort had been liberated from ventilation, compared with only 9% of the 2004 cohort. In a paper from the Netherlands, the mortality rate was about 22%. About two-thirds of these children continued mechanical ventilation, while only 6% were weaned off. In a group of 165 neonates from Minnesota, the five-year survival was 89%. About two-thirds had developmental delays, and about three-quarters had more than two comorbidities. In this group, decannulation was nearly 90% in neonates weighing less than 1,000 grams, and about 71% in those weighing more, but a quarter of these children did not come off the ventilator. The reason for reviewing these different studies is to show that even within different patient populations, the numbers are about the same. This Birmingham retrospective study looked at a group of 91 children. Less than a third had chronic lung disease. The five-year survival was about 76% for those with chronic lung disease, but about half of them were decannulated. Back to the Los Angeles study, the cumulative incidence of survival was 80%, with progression of the underlying condition accounting for 35% of deaths. The cumulative incidence of liberation was only 24%, and chronic pulmonary disease was independently associated with liberation. If we look at the survival curves for these different groups and populations, we see that most deaths occur within the first five years. That 60-month period is critical in terms of patient survival. After that, the curves tend to stabilize with higher mortality rates associated with specific underlying conditions. For liberation, the overall results are also similar among these studies. Again, most who are liberated are taken off the ventilator within that initial five-year period. Why is five years so critical? What happens by five years of age? Patients grow new alveoli, they become vertical because they can walk, and they've recruited some of the thoracic musculature. When counseling the family of a neonate or infant with chronic lung disease, inform them that this is the sort of trajectory they can expect. If the child is going to be liberated, it will most likely happen within those first five years. Summing up, here's what we can take away from all these studies. Most programs, 30% to 40% of patients have chronic lung disease. A growing population is made up of neuromuscular and central nervous system patients. Chronic lung disease is associated with the highest likelihood of liberation from the ventilator, and if these children can be liberated, most will do so by their fifth birthday. Lastly, the mortality rate in patients on long-term ventilation is 25% to 30%, but of all patients, premature infants with bronchopulmonary dysplasia seem to do best. Now for some practical help in the care of children who transition to the home ventilator. First, it's important to understand that goals in the neonatal or pediatric intensive care unit change when the child is going home. For example, in the pediatric unit, permissive hypercapnia generally is good, adequate ventilation is fine, that is good enough, is good enough. We try to avoid oxygen toxicity and to promote spontaneous effort to facilitate extubation. These are the goals for children in the neonatal or pediatric unit, but for home ventilation, these goals are different. First, we need some wiggle room in case there's an intercurrent illness. We don't want these children close to the edge, so to speak. For instance, if you send a child home with a carbon dioxide level of 55 millimeters of and a cold or other illness develops, suddenly the carbon dioxide level will be 60 or 65 and the child ends up in the emergency department. So the goal is to aim for a level in the 40s or even high 30s with a high of about 50 and some. For saturations, we're aiming for greater than 92%. We also want these children to have enough exercise, but not so much that the somatic growth is compromised. You also want to promote developmental milestones in these children, and these are the reasons why these patients are going home on the ventilator, and these opportunities should be preserved and not squandered. So how does this look in practice? Most home ventilators require the patient to weigh at least five kilograms. Home ventilators are usually more difficult to trigger than a hospital ventilator. Home ventilators are not gas-driven, and they have limitations in both trigger sensitivity and inspiratory time. These children must be comfortable even when awake and active. They cannot require minute-to-minute adjustments in oxygen. For example, the child cannot require pre-oxygenation for daily care activities such as baths or suctioning. They cannot depend on or require frequent adjustments in home ventilation. Because of all this, tidal volumes of 8 to 10 millimeters per kilogram are not uncommon. In fact, this is often what's necessary to accomplish the child's physiologic goals. Discharge is based on three components. First is the patient's medical stability. This might require placing a tracheostomy, transferring the patient to the home ventilator, generally making certain the child is ready for home care. The second component is the education and training of the family in matters such as tracheostomy care, medications, feeding and bathing, ventilator management, emergency response, and travel preparations. The third component is the administrative requirements such as obtaining the necessary equipment, arranging home visits and nursing staff, and securing funding, and the state waiver approval. So what does this look like in terms of a timeline? As an example, educating two trained caregivers requires approximately four to six weeks. Caregivers must complete a 24-hour stay where they are the sole providers of the child's care. A checklist can ensure that all details are addressed before the child's discharged. That list should include matters such as consultations and communication with the primary care provider. Finally, a discharge planning meeting held about a week before discharge can identify any gaps and ensure the transfer is as smooth as possible. This should include input from the nursing staffing and other stakeholders and participants. Other challenges surrounding the family element exist, specifically preparing the parents for the child's care. Those challenges might include transportation, other children who need care while the parents are trained, or parental job demands that interfere with training and ongoing care. There may be other issues such as homelessness, undocumented family members, substance abuse problems, or financial issues that result in roadblocks to the home transition plan. All of these issues need to be considered and addressed for transfer to occur. We need to be compassionate, but at the same time, we must ensure the child has a place to live other than the intensive care unit. There are also administrative and financial issues. For instance, there's a lack of home care nurses, so staffing becomes problematic. Sometimes the child's location is a problem. For example, if the child lives in a rural area or a neighborhood perceived to be unsafe. Only a small number of durable medical equipment companies serve pediatric needs, and that can be quite a barrier to overcome as well. Finally, discharge from the hospital to transitional care is becoming the norm, usually because of social issues in the home or a lack of nursing that prevents the child from receiving the care needed. Finally, there are economic challenges. These include applying for waiver programs, keeping track of all the changes in those programs, and making sure providers are reimbursed correctly and on time. Care must be coordinated across a number of providers and institutions. When the child ages out of a program, that can lead to a new set of problems. For such patients, we need to find a way to transition them to adult providers. There are also issues surrounding special care trusts. Not surprisingly, these problems become even more difficult when the parents are undocumented, which is a growing problem in some areas. In these circumstances, the child may be covered by programs, but the parents may be unwilling to provide information. We now have a set of guidelines for pediatric home ventilation from the American Thoracic Society. These guidelines are comprehensive and forward-thinking, and may be used to advocate with insurance companies on behalf of patients, for instance, when insurers deny coverage for home nursing because the parent is believed to be capable of providing all home care. These guidelines support the need for home nursing when 24-hour care is necessary. What happens when patients' needs increase, but the resources may not? The nursing crisis is a growing problem that can leave children more or less marooned in the intensive care unit, simply because no home nursing staff is available. In a prospective study of children with chronic medical complexity in Minnesota, lack of home care nursing was the most frequent cause of discharge delay. This study also compared the costs of home care to those of hospital care and found that the cost of hospital care for patients waiting for home care was about $205,000, whereas if the child had been cared for in the home, the cost would have been about $33,000, with a difference of $172,000. So it costs about $170,000 more to keep these children in the hospital while waiting for home care to become available. That was with an average wait time of 50 days. In a retrospective review of 72 patients, slightly more than half had discharge delays. Of those, 62% were related to a lack of home nursing. About 10% of hospital charges occurred while waiting for home nursing to become available. What's more, most patients received far fewer hours than desired. In this particular study, most patients got about half the number of hours requested, but many only got a quarter of the hours desired. In the United States, most home nursing care relies on Medicaid, and there's great variability in coverage from one state to another. Payments for home care are often lower than they are for inpatient care. The funds that pay for home care and for hospital care come from different budget areas. There's a need for a specialized workforce, and these providers must be paid appropriately. One of the major reasons for the lack of home nursing care is low pay rates. So what's the solution? Actually, multiple solutions are needed. Solving these problems requires payment reform, training and education for the health care staff and parents, coordination between academic institutions and home health care providers, and innovations such as telehealth. So it's a huge issue that requires different approaches to find a comprehensive and satisfactory solution. So what would an ideal program look like? First, urgent outpatient care would be available for acute problems, no matter where the patient lives. Second, an outpatient provider would be available to address acute and chronic issues of all types, including medical, functional, and psychosocial issues. Third, decision-making would be coordinated among all providers. And finally, plans of care would be proactive and aimed at anticipating health problems and addressing global well-being. All of these challenges are very similar to the challenges that exist in the Pediatric Intensive Care Unit. In both instances, the important outcomes for the parents of a critically ill child include the child's emotional health, the quality of life for the child and the family, and the child's physical functioning and ability to take part in the activities of daily living. As in the Pediatric Intensive Care Unit, numerous processes can promote improvement. These include rehabilitation and recovery, team dynamics and continuity of support, communication among health care providers and family members, family support, and discharge readiness programs. To sum up, it's probably fair to say that children with chronic critical illness are a microcosm of both the PICU population and children with medical complexity. They occupy a space at the intersection of technology dependence and an uncertain prognosis. Ideal systems should mitigate the severity of the chronic condition while meeting the needs of both the family and the child, maximizing function while also minimizing unnecessary utilization. This will open up the Intensive Care Unit for acute patients. And finally, many practical interventions can be implemented, including the ABCDEF bundle, use of a continuity intensivist, and other interventions mentioned earlier.

pediatric ICU

medical complexity

chronic critical illness

home ventilation

healthcare coordination

quality of life