One of the most urgent targets for improved efficiencies in inpatient hospital care is critical care services. Between 2000 and 2005, critical care medicine beds increased in the United States by 6.5% (from 88,252 to 93,955), and occupancy rates increased by 4.5%.1 In that period, critical care costs per day increased by 30.4% (from $2698 to $3518). In 2005, critical care medicine accounted for 13.4% of hospital costs in the United States and consumed 0.66% of gross domestic product. Patients requiring mechanical ventilation are among the largest consumers of critical care resources, and hospitals often experience financial losses in providing care for them. As many as 2.8% of hospitalized patients in the United States received mechanical ventilation in 2005, representing 2.7 episodes of mechanical ventilation per 1000 population.2 Estimated national costs were $27 billion. This chapter reviews the economic implications of mechanical ventilation. Basic principles of health economics are reviewed to provide a framework for interpreting health economic analyses related to mechanical ventilation. Actual costs of mechanical ventilation are addressed, followed by a discussion of whether mechanical ventilation is cost-effective. Finally, strategies for cost containment are reviewed.

The goal of health economics is to ascertain the highest level of efficiency in providing health care.3,4 A key assumption in this field is that health resources are a finite commodity. In such a system, a series of questions should be answered regarding any new or current medical intervention:

• Is the intervention effective relative to other available therapies?
  
• How much does it cost relative to other available therapies?
  
• From whose perspective are the costs being considered?
  
• How widely will the intervention be utilized?

Measured approaches to answering these questions allow health care systems to select medical therapies based upon evidence rather than assumptions, commercial marketing, or bias. Economic analysis has become a standard component of decision making for health systems in countries such as the United Kingdom or Australia, where health care policymaking is centralized on a national level. In countries such as the United States, delivery of health care is much less regulated, and many practitioners and most patients have unbounded access to any available therapies. Few physicians in the United States, however, are able to practice without significant awareness of the resource implications of their decision making. One goal of recent health care reform efforts in the United States is cost control, and it is highly likely that third-party payers will increase efforts to balance available services with more efficient delivery. The formation of Accountable Care Organizations, in which providers are incentivized to organize care delivery in a way that improves quality and outcomes and reduces costs, is mandated in the recent Affordable Care Act.5 Other delivery reform initiatives include pay-for-performance measures,6 expansion of medical homes, bundled payments, and value-based purchasing. Consequently, a basic understanding of how the efficiency of various health care practices is defined is becoming essential to the practicing clinician. The following section outlines some of the common definitions and methodologies employed in health economic analysis.

Costs

When considering the cost of therapies, the unit price of a drug or piece of equipment is often the basis of discussion in the clinical setting, especially if the analysis is being performed in the interest of the physician or hospital. Payers may have a broader view, especially if they are responsible for health care costs after hospital discharge. Patients and society view the cost of illness from an even larger perspective that includes a longer time horizon. Ideally, economic analyses determine the true costs that are accrued from the beginning of a patient’s disease to the long-term outcome for the patient. These costs are categorized as direct medical costs, direct nonmedical costs, and indirect costs.3,4,7

Direct Medical Costs

Direct medical costs are expenses for the provider of a service. These can include variable costs such as physician labor, drug costs, or use of diagnostic or therapeutic equipment.8 Variable costs are not generated unless the service is provided. An important subgroup is direct variable costs, which are defined as variable costs excluding staff salary and equipment costs.9 Nursing labor and equipment are usually considered fixed costs, as nurse salaries must be paid regardless of bed occupancy, unless staff are laid off. Another form of fixed costs that is more removed from medical practice is overhead costs, which include general operations of the hospital or clinic such as utilities, leases, capital improvement, insurance, and administrative overhead or taxes.

Direct Nonmedical Costs

Direct nonmedical costs include costs incurred by the patient that are not related to care provided by the physician or hospital. These costs may include emergency transportation to the hospital, transportation and lodging for the patients’ family, and domestic help and rehabilitative services after discharge. Although these costs are difficult to measure, they can have an important impact on the patient and their perception of the benefits of care.

Indirect Costs

Indirect costs include the overall financial burden of illness to the patient. This can include loss of wages and benefits as a consequence of missed work or loss of earnings and unpaid care provided by family members. This, too, is difficult to measure, but it has substantial impact on the well-being of the patient and family.10–13 Indirect costs for the patient can ultimately affect direct costs for a health system. If economic burdens on the patient decrease access to resources, affect medical adherence, and result in incomplete recovery, then recurrence of hospitalization or even critical illness can result.

Measuring Costs

One of the greatest challenges in health economic analysis is accurate measurement of relevant costs. Many hospitals have adopted sophisticated cost-accounting systems that assign a specific cost to each service based on hospital expenses.8 Expenses can be determined from either the hospital’s acquisition costs or the service-specific relative-value units based upon the Center for Medicare and Medicaid Services Resource-Based Relative Value Scale. Overhead costs are assessed internally and assigned to each department.

Most hospitals increase the patient charge for many services above acquisition costs and expenses to help pay for services that are otherwise inadequately reimbursed. Because this practice can vary according to different payment plans, hospital charges are often an inappropriate surrogate for hospital costs.14 Large studies involving patients under a single-payer such as Medicare can adjust charges using a standard cost-to-charge ratio to gain acceptable estimates of costs, but this is less reliable with more heterogeneous groups of patients and payers.

As nursing time, monitoring costs, physician costs and certain laboratory and radiology ordering practices are linked to a patient’s presence in an intensive care unit (ICU), ICU length of stay is considered by many to be a reliable and convenient surrogate for ICU costs. In one frequently cited model that is based on the relative costs per day of hospital admission, 4.5 units are assigned to the first day of each ICU stay, 2.5 units for each additional ICU day, and 1 unit for each non-ICU day after the first ICU discharge.15

For comparisons between institutions, length of stay can be confounded by differences in ICU admission criteria, nurse-to-patient ratios, presence of intermediate care or “stepdown” units, and transfers to other acute care facilities. To standardize resource use between different ICU settings, instruments such as the Therapeutic Intervention Scoring System (TISS) can be used.16 The TISS instrument assigns weights to seventy-six ICU interventions based upon the severity of illness associated with the need for each intervention. Intensity of ICU care can then be assessed by comparing the added weights or TISS scores between intervention groups, and costs can be standardized by applying a specific cost to each service.

Cost data for economic evaluations can come from multiple sources, but the most common approach is to collect cost data prospectively during clinical trials.17,18 Utilizing cost data from a single, large clinical trial has the advantage of uniform methodology, but it may not be generalizable if the trial involved highly selected patients managed in ways that are not standard to the typical clinical setting. When clinical trial data are not available, cost and outcome data can be derived from cohort studies. With this approach, sensitivity analyses to account for clinical variability and potential biases in the cohort studies are particularly important. A combined approach that begins with data from a large clinical trial and adds multiple scenarios that account for clinical variability is useful.

Types of Economic Analyses

Economic analyses function to provide more than just a description of the costs of a therapy. Their role is to compare costs between therapies relative to the benefits or efficacies of those therapies. Table 67-1 describes the four primary types of economic analyses.3,4,17 Cost-effectiveness analysis, the most useful approach to economic analysis, is described in more detail below.

Cost-Effectiveness Analyses

Cost-effectiveness analyses use standard clinical measures such as life-years gained or quality of life units rather than monetary units to assess benefit. Outcomes are expressed as a ratio of cost to measure of benefit, for example, cost per life-year gained. Cost-effectiveness analyses are easy to understand for clinicians and have the advantage of not having to assign a monetary value to outcomes. Consequently, they are the most common type of economic evaluation performed. They can, however, only compare the costs of therapies that have a common clinical outcome.

Cost-Utility Analyses

An important and common subgroup of cost-effectiveness analysis is cost-utility analysis, which is performed when therapies are likely to have an impact on quality of life as well as mortality. The level of well-being for a given health state (utility, rated from 0 for worst health or death to 1 for best health) is multiplied by the amount of time spent in that health state. The resulting index is called a quality-adjusted life-year (QALY). For example, a therapy that results in a health state valued at 0.5 for 2 years would yield 1 QALY. Utilities can be measured by interviewing participants of a clinical trial, or predetermined values measured in similar groups of patients can be utilized. Results of a cost-utility analysis are expressed as cost per QALY. Results of selected cost-utility analyses related to mechanical ventilation and ICU care are listed in Table 67-2.

The advantages of cost-effectiveness analyses are that costs and benefits of different therapies involving different types of patients can be compared, and outcomes beyond survival are factored in. The disadvantage is that many clinicians have difficulty understanding the complexity of the studies, and the models can involve numerous assumptions that vary across patient populations. Therefore, strict guidelines19 must be followed in conducting the studies to ensure transparency in how utilities are assigned, and rigorous testing in the form of sensitivity analyses should be performed to determine how variability in clinical factors could affect results of the models.

Cost-effectiveness analyses can aid decision making by providing data on how much therapies cost relative to the outcomes that are achieved, particularly if one therapy has greater benefits than the alternative, but the costs are greater as well. Cost-effectiveness analysis is not a cost-containment tool. Instead, it should be considered a method to improve value. The analyses may not always lead to decreased costs for a health care system, especially when underutilized services that increase cost but improve value are identified and encouraged.20 In most Western health systems, an incremental cost per QALY of less than $20,000 is strong evidence for adoption of a therapy. Incremental costs per QALY of between $50,000 and $100,000 provide moderate evidence for adoption. If the incremental cost per QALY is greater than $100,000, the therapy may not be considered cost-effective, although more recent recommendations raise that boundary to a range of $183,000 to $264,000.21 These boundaries of cost-effectiveness are somewhat arbitrary however. Ultimate decision making must take into account other factors such as seriousness of the health condition, availability of alternatives, and number of patients who would receive the therapy (total budgetary impact).3,22

The average total cost of hospitalization for a critically ill patient ranges from $14,135 to $32,253, depending on the study methods and patient population.23,24 This is nearly three times the cost of a hospitalized patient managed on the medical or surgical floor. Two-thirds of the costs associated with critically ill patients are accrued during their stay in the ICU. For those whose length of stay in the ICU is more than 5 days, as much as 80% of their hospital costs are accrued in the ICU. Hospital costs for ICU patients who require mechanical ventilation is significantly higher than for nonventilated patients in the ICU (e.g., $47,158 vs. $23,707 in one study).23 It is important, however, to understand that the difference in cost between ventilated and nonventilated patients is a factor of higher illness severity rather than the cost of providing mechanical ventilation.25

Maintaining a mechanical ventilator accounts for less than 5% of direct ICU costs for a ventilated patient.9 A ventilator includes a one-time cost of $20,000 to $45,000 for the hospital plus nominal maintenance charges, and it is used for a number of years. A respiratory therapists’ time ranges from $70 to $130 per day depending on local salary and staffing levels, and administrative costs are a smaller factor. Rather than the mechanical ventilator itself, the major contributor to variable costs for a ventilated patient is the nursing effort assigned to the patient (Table 67-3).26,27 The nurse-to-patient ratio for acutely ill, mechanically ventilated patients is usually 1:1 or 1:2.

One study assessed the fraction of total costs attributable to variable and direct-variable costs for mechanically ventilated patients at an urban teaching center.9 The average total cost for each patient was $69,472. Costs were highest during the first 2 days of intensive care and decreased significantly thereafter (Fig. 67-1). Only 18.4% were direct-variable costs (costs not attributable to overhead, staff, or equipment). Direct-variable costs were highest for the blood bank (44%) and pharmacy (48%), and lower for radiology (8%) and respiratory care (3%).

Is Mechanical Ventilation Cost-Effective?

Thorough economic analyses of mechanical ventilation have been limited because of difficulties in applying adequate study methods in this complex patient population. One high-quality study, however, addressed the question of cost-effectiveness of mechanical ventilation in severely ill patients who required mechanical ventilation for pneumonia or acute respiratory distress syndrome (ARDS).28 In the analysis, patients were stratified based upon likelihood of 2-month survival (Table 67-4). The incremental cost per QALY of providing mechanical ventilation to these patients was $29,000 for low-risk patients (>70% estimated survival), $44,000 for medium-risk patients (51% to 70% estimated survival), and $110,000 for high-risk patients (≤50% estimated survival). Sensitivity analyses that increased mortality and costs to twice the baseline estimates resulted in incremental costs per QALY that were still less than $80,000 for low-risk and medium-risk patients. In another cost-utility analysis from France,27 cost-utility ratios for ICU care of patients with acute respiratory failure were estimated to be $11,970 for patients without chronic lung disease and $14,365 for patients with chronic lung disease (see Table 67-2). This analysis did not include costs of care following discharge, so it is relevant only from the hospital’s perspective.