I. COSTS AND BURDEN OF CRITICAL CARE
A. Caring for the critically ill or injured is a complex and resource-intensive endeavor. There are roughly 6,000 intensive care units in the United States, representing 10% of the total hospital beds but a disproportionate share of hospital operating costs—roughly one-third. Today’s ICUs are highly specialized units dedicated to care for very specific populations of patients at risk for or experiencing overt, critical organ dysfunction. Many hospitals, typically in larger centers, further subdivide patients into burn or neurosurgery subspecialty ICU’s. This increased specialization, coupled with growing demand for intensive care services, presents challenges to hospitals seeking to control costs. Applying a family- and patient-oriented, value-based framework to intensive care helps practitioners focus on optimizing outcomes while simultaneously reducing costs.
B. Detailed analysis indicates that critical care represents 13.4% of hospital charges, 4.1% of US health care expenditures, and 0.66% of gross domestic product. This is due in part to high overhead costs—on average, an ICU bed costs $3,500 to $4,000 per day—together with the expense of high-intensity nurse and therapist staffing. Moreover, critical care expenditures in the United States are projected to increase as a proportion of total health care costs as the population ages, the prevalence of chronic disease increases, and technological advances help people live longer following serious illness and injury. In 2000, for example, roughly 12% of the US population was age 65 years or older; by 2050, persons age 65 years and older are projected to comprise 21% of the US population. More people are expected to live to advanced ages, with the percentage of US adults age 85 and older projected to increase from less than 1% in 2000 to 5% in 2050. This growth in the elderly population will fuel demand for intensive care services, further straining resources. In addition, health care economists suggest technological advances combined with greater access to health insurance increased demand for critical care services in recent years, a trend that is likely to continue. New technology includes a broader array of pharmaceuticals, devices, and new medical procedures (most of which tend to be expensive), as well as a deeper understanding of the clinical process of care. These advances in aggregate have substantially improved outcomes for patients, but drive up demand for critical care services and costs, accordingly. Thus, for most hospitals with shrinking margins in a difficult policy environment, critical care represents a potential high-impact opportunity for focusing on efficiencies and improving care.
C. Evidence suggests that the supply of intensivists, critical care nurses, respiratory therapists, and other clinical staff required to effectively manage critically ill patients is not keeping pace with the demand for critical care services, putting increasing pressure on critical care providers. Some estimate a 35% shortage of intensivists by 2020, with even greater shortages in specialty areas such as surgical and pediatric critical care, or those with dual-board certification in emergency medicine and critical care, or palliative medicine and critical care. Yet, ICU clinicians and administrators have several potential levers to improve the quality and efficiency of ICU care and therefore reduce costs, such as preventing hospital-acquired infections, reducing length of stay, and ensuring proper staffing of the ICU with intensivists.
D. All of these factors combine to create a challenging matrix for providing high-value, patient- and family-centered care to critically ill patients. Measuring the value of the care provided in our ICUs is an essential first step in addressing these challenges and can help identify opportunities to create efficiencies and improve clinical outcomes.
II. THE VALUE OF CRITICAL CARE EXPENDITURES
A. A new paradigm, “the value proposition,” has gained traction as the US health care system tries to balance the tension found in delivering the highest quality care at the lowest possible cost. Value in health care is defined as the outcomes experienced by the patient (typically a quality metric) divided by the cost of delivering those outcomes/metrics. Value can therefore be maximized by either improving outcomes and/or reducing costs. Measuring hospital costs is challenging because health care costs are much less transparent than that in other industries. The prices charged by providers and the amounts reimbursed by insurers often do not reflect the actual cost of the care provided. In addition, high-margin treatments and procedures are used to subsidize other areas where reimbursements do not cover the costs. Determining costs for patients in the ICU is particularly challenging because of the interdisciplinary nature of the care provided and the heterogeneity of the ICU patient population. The value measurement framework includes equal emphasis on achieving goals for both quality outcomes and costs. The value of medical care can be improved either by reducing costs or improving quality. Ideally, both scenarios would be met to maximize value. There are many opportunities to apply this lens to critical care and drive improvement on both fronts.
B. Quality Care: Traditionally, measures of quality for critical care were focused solely on the ICU portion of the patient’s hospital stay, using metrics (like length of stay) that are easy to measure but less important to patients and families (functional status, long-term survival, and quality of life). The new value framework provides a structure for measuring the impact and value of critical care services over the long term, using patient- and family-centered outcomes. One popular approach groups health outcomes into three tiers: health status achieved or retained, process of recovery, and sustainability of health (Table 39.1).
This approach provides a framework for measuring value for critically ill patients that includes a continued focus on survival but also calls attention to duration of recovery and ultimate health obtained.
1. Measures of outcomes that drive up overall cost: hospital-acquired infection rates, pressure ulcer rates, falls and injury rates, readmissions to the ICU, and readmissions to the hospital postdischarge (within specified time periods, typically 30 days)
2. Measures of overall quality that are value-producing: adherence to protocols for managing specific diseases such as sepsis, early mobilization, monitoring for delirium, etc.
| ||
Tier | Domain | Measure |
1. Health Status Achieved or Maintained | • In-hospital mortality • Postdischarge mortality (30 d, 6 mo, 1 y) | |
• Discharge location (LTACH, SNF, Home) • Ability to live independently • Ability to return to work • Related readmissions • For end-of-life patients, ability to access palliative care and hospice services in accordance with patient/family goals | ||
2. Process of Recovery | • Critical care unit length of stay • Overall hospital length of stay • Time to return of full functionality (reanimation) | |
• Hospital-acquired infections (e.g., CLABSI, CAUTI, MRSA, C. difficile, SSI) • Incidence of pressure ulcers • Delirium | ||
3. Sustainability of Health | • Quality of life years • Return to full functionality (reanimation) | |
• Long-term cognitive deficits • Depression/anxiety/other mental health diagnoses |
Adapted from Porter ME, Teisberg EO. Redefining health care. Cambridge, MA: Harvard Business School Press, 2006.
C. Costs: There are many different approaches to measuring health care costs, each with advantages and disadvantages. Many analyses of costs are completed retrospectively using administrative (billing) data. This approach is often used because data are readily available and in a structured format that lends itself to analysis. However, hospital billing data are complex, with many assumptions built in regarding direct, indirect, variable, and fixed costs. Direct costs are directly related to patient care (labor and supplies); indirect costs refer to shared resources that are allocated to individual patients on the basis of a set of assumptions (e.g., salaries of administrators). Fixed costs are those that do not change with the volume of patients; these tend to represent the operational costs of the ICU, such as clinician salaries, or the purchase of specialized equipment. Variable costs will change with the volume of patients, such as the cost of mechanical ventilation or the cost of lab tests. Conventional wisdom is that roughly 80% of costs in the ICU are fixed costs, although new thinking on costing methodology for the health care industry is challenging that notion. Alternatively, costs can be measured prospectively using a framework of Time-Driven Activity-Based Costing (TDABC), which involves tracking the amount of time it takes each clinician to conduct patient care tasks and then calculating the costs of the labor and supplies on the basis of the actual amount of time spent on each task. This approach is much more accurate in terms of determining the actual costs for a specific procedure or patient with a particular clinical condition because it views every resource as variable and fungible. However, TDABC is a more labor-intensive method that involves process mapping and time studies, which may not be practical in all situations.
1. The approach to measuring costs will depend on the question of interest and the amount of time and resources available for the analysis. Administrative data are best suited to address basic questions of utilization (e.g., number of central line days, number of CT scans) and the associated labor and supply costs. This approach is a simple way to determine whether key metrics around resource utilization are trending in the desired direction and is therefore often used in ongoing monitoring. Yet it is often challenging to disentangle the total costs of a particular patient’s ICU stay using administrative data. Physician billing data are often located in a separate data system. Nursing costs are typically included in “room and board” charges, with no ability to track the actual costs of the time providers spend with patients.
2. TDABC, on the other hand, provides a more robust and reliable analysis of the actual labor and supply costs for patients with particular medical conditions or procedures. This approach is most helpful in answering questions about capacity, or measuring the cost savings of a particular intervention, such as a quality improvement project. TDABC is also very useful in determining the potential cost savings from substitutions; for example, substituting a nurse practitioner for an MD for a particular patient care task. However, TDABC requires process maps, time studies, and observational data collection, all of which are time-consuming and add cost to the analysis.
3. By understanding the tradeoffs between these two costing strategies—administrative data versus TDABC—ICU leaders can choose the approach that best suits the question. If using administrative data, we recommend focusing efforts on measuring direct costs that clinicians can readily affect, both from a cost and quality approach. Some examples include measures of utilization such as ventilator days, catheter days, utilization of high-cost drugs, ICU length of stay, and total average length of stay. When using TDABC, it is important to scope the project so it is achievable and easily replicated in other environments.
III. INFLUENCE OF POLICY ENVIRONMENT
A. Health care reform is redefining the way we think about and measure quality and costs in the ICU. Under a traditional fee-for-service model, ICU care was viewed as a distinct and separate part of the patient’s hospital stay. As reimbursement policy evolves from fee-for-service to bundled payments for episodes of care and population health management, the care provided in the ICU must be measured in a broader context. In addition, pay-for-performance programs are increasingly focused on outcomes and costs rather than clinical process of care, consistent with the shift to patient- and family-centered care. This shift in financial incentives puts additional pressure on ICUs to improve the quality and efficiency of care provided. Although this policy environment is challenging, critical care units that effectively improve performance on key metrics will cultivate a competitive advantage while also providing high-value care to the sickest patients.
In the following pay-for-performance programs, ICU care will have either a direct or indirect effect on hospital performance:
1. Value-based purchasing (VBP): VBP is a CMS annual pay-for-performance program incenting optimized process of clinical care, patient experience, outcomes, and efficiency. Under this program, for example, hospitals have an opportunity to gain or lose up to 1.75% of their base DRG payments in fiscal year 2016 (increasing to 2% in FY17). Although process of care measures were initially 70% of the VBP program, by fiscal year 2016 these measures will represent only 10% of the dollars at risk. Outcome and efficiency measures, in contrast, will comprise 65% of the dollars at risk in VBP by fiscal year 2016. Outcome measures in the VBP program include many ICU-related phenotypes, such as mortality from heart failure, AMI and pneumonia, hospital-acquired infections (CLABSI and CAUTI), the AHRQ composite patient safety indicator (PSI-90), and surgical-site infections. The efficiency score is based on Medicare Spend Per Beneficiary, a risk-adjusted measure of total expenditures for 3 days prior to and 30 days after an inpatient hospital stay.
2. Hospital-acquired condition reduction program (HAC): The CMS annual HAC reduction program is a payment penalty assessed on hospitals with poor performance (7th to 10th decile) on hospital-acquired infection rates; the penalty is currently 1% of base DRG payments. In the first year of the program, metrics included AHRQ PSI-90 (a composite score of several Patient Safety Indicators), CLABSI, and CAUTI (ICU rates only). In future years, the program will include surgical-site infection rates for hysterectomy and colon surgery, MRSA bacteremia, and C. difficile rates. Hospitals that work to reduce the rate of these potentially preventable, hospital-acquired conditions in the ICUs will have a competitive advantage in this performance program.
3. Readmissions penalty program: The CMS Readmissions Reduction Program represents another effort to incent efficiency and improve the value of care by reducing readmissions within 30 days of discharge for selected conditions and procedures. Up to 3% of the hospital’s base DRG payments are at risk in this program. Conditions and procedures currently included in the program include AMI, heart failure, pneumonia, COPD, and hip and knee arthroplasty (additional conditions and procedures will be added in the future). It is not uncommon for patients with these diagnoses and procedures to spend some portion of their inpatient stay in the ICU. Thus, care provided in the ICU has the potential to influence the trajectory of recovery and therefore the likelihood of readmission within 30 days.
4. Medicaid and commercial P4P programs: State Medicaid programs and commercial insurers often model their incentive programs on Medicare programs, with similar focus on improving efficiency, reducing hospital-acquired conditions, and improving value. Hence, pressure will likely increase on those who provide critical care to demonstrate evidence of sufficient value and return on the invested resources.
B. Interventions that Affect Value: The 5 Do’s and 5 Don’ts of Critical Care
1. A new term, post–intensive care unit syndrome or PICS, was coined recently to refer to phenotypes associated with survival after severe, debilitating, critical illness. PICS describes new or worsening problems for patients and/or their families during recovery from critical illness, including physical, cognitive, and/or mental health deterioration. Two recent, complementary, national campaigns illustrate how performance improvement interventions can be used to improve the value of care provided in ICUs and to potentially avoid the ravages of PICS. For example, both the American Association of Critical Care Nursing and the Society of Critical Care Medicine promote the ABCDE Bundle, which couples five best-practice protocols into a coherent approach to care redesign: coordination of spontaneous awakening and spontaneous breathing trials, delirium assessment and management, and early mobilization. Use of this bundle in a study of medical and surgical patients was associated with less time on the ventilator, less delirium, and less immobility in the ICU. The cost-saving implications are therefore significant (and should be compelling to hospital administrators responsible for performance improvement). Similarly, the American Board of Internal Medicine, partnering with the Critical Care Societies Collaborative, identified by consensus five ICU care areas in which to “choose wisely”: the ICU Choosing Wisely Campaign ( Table 39.2). These items are intended to educate and to prompt discussions between ICU clinicians, patients, and their families to evaluate critically use of therapies that may not be helpful (and may even be harmful) under certain conditions. Lack of benefit and/or avoiding harm both increase quality and decrease costs, supporting the value proposition.
| ||
1. Don’t order diagnostic tests at regular intervals (such as every day), but rather in response to specific clinical questions.
2. Don’t transfuse red blood cells in hemodynamically stable, nonbleeding ICU patients with a hemoglobin concentration greater than 7 g/dL.
3. Don’t use parenteral nutrition in adequately nourished critically ill patients within the first 7 d of an ICU stay.
4. Don’t deeply sedate mechanically ventilated patients without a specific indication and without daily attempts to lighten sedation.
5. Don’t continue life support for patients at high risk for death or severely impaired functional recovery without offering patients and their families the alternative of care focused entirely on comfort.