Distribution of Variable vs Fixed Costs of Hospital Care

The dramatic rise in health care costs over recent decades has resulted in expenditures of more than $1 trillion per year in the United States. Health care now consumes 14% of the gross domestic product.¹ The majority of these expenditures represent payments by third-party payers to health care providers, such as hospitals.^{2 - 3} As a result, many programs for reducing cost have focused on restricting patient access to medical services and minimizing use of expensive tests and treatments.^{4 - 7} Unfortunately, this approach has not worked as expected. We hypothesize that there are factors in addition to health resource consumption rates that contribute to escalating costs.^{6 ,8 - 9}

Hospitals purchase goods such as buildings, equipment, labor, pharmaceuticals, and medical supplies for conversion into patient care services. The total cost of medical service includes all hospital expenditures, including facility operating costs. Hospital costs can be categorized into 2 types: fixed and variable. Fixed costs are not saved by a hospital over the short term if a particular service is not supplied. Examples of fixed cost elements include buildings, equipment, and some salaried labor costs. For instance, a computed tomographic (CT) scan is thought of as an expensive test and a source of significant cost savings if it is not performed. However, the scanner and space have already been rented or paid for, and the technician receives a salary that must be paid whether any individual patient receives a CT scan or not. If the radiologist who interprets the test is also receiving a salary, the additional cost to the hospital of doing the test is minimal—the price of radiographic film, paper, and contrast. Therefore, fixed costs do not change over the short term with changes in output.^{10 - 12}

Variable costs are those that do change with output and can be saved by the hospital if a service is not provided.^{12 - 13} Examples include medication, test reagents, and disposable supplies. Over the short term, reductions in output of a particular resource save primarily variable costs for a hospital. Cost defined as fixed or variable can change depending on the time frame chosen. Many costs defined as fixed over the short term, such as salaried labor, can be variable over the long term if staff is downsized.¹² Furthermore, if a hospital purchases medical services such as CT scans from an outside supplier, the entire CT cost is variable for that hospital.

Another cost function is marginal cost. The overall change in total cost to the institution as a result of a change in output is defined by the marginal cost function. This indicates the additional cost of treating the next patient or increment of patients.^{12 ,14 - 16} Marginal cost is usually represented as a smooth continuous function.¹² However, in reality, the marginal cost of health care most likely increases in incremental steps as output rises. For example, the marginal cost of a single CT scan is the film, contrast media, paper, and radiologist fee (if the radiologist is paid per scan). The marginal cost for 100 additional CT scans per month will also include the salary of an additional technician and cost of an extra scanner if the current staff and scanner are operating at maximum capacity. Marginal cost is a key concept because hospitals hoping to save average total cost by reducing services will generally save only marginal cost.¹⁴

Defining the cost of providing medical care also depends on whose point of view is assumed.^{11 ,15 ,17 - 18} Third-party payers focus on reimbursement, which they pay to hospitals.¹⁰ The hospital bill is a reflection of the hospital's total cost, but is a variable cost to third-party payers. If a medical service is not provided or is not reimbursed, third-party payers save the amount of that bill payment.^{15 ,19} In contrast, if the hospital does not provide the service, it does not receive reimbursement, yet must still cover fixed costs for that service.^{10 ,20} The overall effect for the hospital over the short term is to increase the fixed cost per work unit or to decrease hospital efficiency.^{6 ,13} From the physician and patient point of view, cost has traditionally been defined as pain, suffering, death, or lost productivity, and benefit as accurate diagnosis, relief of suffering, or cure. The concept of "total cost to society" would integrate all these different perspectives to arrive at an unbiased evaluation of a cost-benefit ratio.^{21 - 24}

Individual hospitals vary in the relationship between fixed and variable costs, making it difficult to design and implement cost-cutting programs.^{4 ,25 - 29} For example, at a public teaching hospital where all physicians are receiving a salary, physician costs are essentially fixed over the short term. In contrast, the majority of cost for physicians who are paid fee-for-service is variable. Similarly, a hospital that must pay an on-call technician specifically for performing a test at night has a higher variable cost for that test than one with in-house salaried technicians at night.¹⁴ Therefore, ideal cost-saving strategies could vary greatly for different hospital types.

The purposes of this article are to delineate overall fixed and variable costs for medical care at our institution and discuss the health policy implications of these results.

This study was performed at Cook County Hospital in Chicago, Ill, an 886-bed urban public teaching hospital. In 1993 (the year for which costs were determined), there were nearly 114,000 emergency department visits, 40,000 hospital admissions, 240,000 inpatient days, and more than 500,000 outpatient clinic visits. Costs were computed using the 1993 annual expenditure report of the Cook County Health Bureau. All costs were defined as fixed or variable. Fixed costs included capital, employee salaries, benefits, building maintenance, and utilities. Variable costs included health care worker supplies such as gloves, patient care supplies, paper, food, radiographic film, laboratory reagents, glassware and medications, with their delivery systems (such as intravenous catheters or bottles).

Fixed costs were further characterized as either service or support. Service centers were defined as those health care providers who directly care for patients such as physicians and nurses or departments such as radiology, laboratory, or pharmacy who provide diagnostic or other measurable service to individual patients.

Support center costs were of 2 types. The first type was nonsalary expenditures by the hospital for outside products (eg, cleaning supplies, utilities, or health insurance benefits) that contribute to the function of the hospital and its employees. The second type of support center cost was the salaries of employees in departments such as housekeeping, payroll, and administration. They perform functions throughout the facility for the entire organization.

The total fixed cost for medical service includes the salaries of employees working in the service centers providing the medical service plus all allocated support costs. To fully capture the cost of medical services, all support center costs were allocated completely to the service centers. There are a number of methods for determining how much of the electric bill should be attributed to the cost of a chest radiograph or how much of the payroll department cost should be attributed to the pharmacy.^{30 - 31} The multiple distribution method was used to capture the costs exchanged between the support center departments more accurately and to avoid underestimating medical service costs.^{30 - 31} The basis for allocation of each support center cost was that recommended by the American Hospital Association.³² For example, housekeeping costs were allocated to each service and support center based on each center's square footage. Likewise, payroll costs were allocated to each service and support center relative to each center's total number of full-time equivalent employees. Adjustments were made to reflect disproportionate support consumption by specific departments. For instance, the emergency department, which operates 24 hours per day, received proportionally more housekeeping and administrative attention than less active areas.

Table 1 shows the hospital fixed-cost structure before the allocation of support center cost to service centers. First, nonsalary expenditures were allocated directly to all hospital support and service centers based on the number of square feet or employees in each department. Next, each support center cost (including nonsalary allocations from step 1) was allocated to all of the other support and service centers based on square footage or employees. At the end of this first round of support center cost allocation, each support center now contained only allocated costs it had received from other support centers. These residual support center costs were again allocated in a second round to all support and service centers. This process was repeated a total of 4 times. Each round resulted in more of the total support center cost being allocated to service centers and smaller residuals remaining in each support center. Once all support costs were completely allocated to the service centers, the resulting service center cost totals were summed to confirm the accuracy of the multiple distribution process and to measure any discrepancy between the budgets before and after allocation. Table 2 shows the total budget after support allocation.

Table 1 outlines the hospital fixed-cost structure before the support center costs were allocated to the service centers. The total budget for 1993 was $429.2 million. The fixed portion of this budget totaled $360.3 million (84%) and the variable portion was $68.8 million (16%). Overall, 31.5% of the $429.2 million total was for support and 52.4% was direct salary for service center department personnel. Employee salaries totaled $251.3 million (58.5%). An additional $90.3 million in employee benefits and in seasonal, overtime, and part-time employee costs brought total labor expenditures to $341.6 million or nearly 80% of the total.

Table 2 summarizes the total budget after allocation of support costs. The fixed portion was $360.0 million and the total came to $428.8 million. The discrepancy between budgets before and after the allocation process was less than 0.1% of the original total. The majority of hospital cost (80.8%) was related to patient treatment areas such as hospital wards, clinics, procedure laboratories, or the emergency department. Less than 20% of total expenditures were for the major ancillary service departments: radiology, laboratory, and pharmacy. Within the major categories of service the variable percentages for patient treatment areas, radiology, laboratory, and pharmacy were 9%, 14%, 17%, and 75%, respectively.

We found that the majority of hospital cost (84%) was for labor, equipment, building space, and maintenance. These costs are fixed over the short term at our institution. This low rate of variable cost has important implications. From the hospital perspective, decreased output without organizational adjustment leads to technical inefficiency.⁶ The cost per patient treated goes up because fewer patients must share the operating costs of the facility. We postulate that this fact has contributed to the inability of the medical system to save money with reduced use of hospitals. Unless reduced consumption allows hospitals to reduce staff, space, or avoid capital expenditures, restriction of service has minimal influence on the long-term fixed portion of total cost. For example, there has been interest recently in diverting patients to ambulatory settings such as clinics and observation units as a cost-saving alternative to hospital admission.^{33 - 37} However, the resultant savings are small until inpatient beds are closed, staff dismissed, and physical resources liquidated. Moreover, initiating care in new health care settings may even temporarily increase total expenditures if fixed costs in the hospital are not reduced in the face of new fixed costs incurred with ambulatory care.^{10 ,38}

National hospital trends support this contention. From 1990 to 1995, the average cost per patient day in community hospitals rose from $687 to $968.³⁹ This is a 21% increase after adjustment for the general inflation rate.⁴⁰ Over this same time interval, the total number of inpatient days nationwide decreased faster than did the total number of hospital beds (10% vs 6%).³⁹ Reduced occupancy in the setting of high fixed cost will increase cost per inpatient day. In addition, there are alternative or contributing reasons for increased cost per day. When the reduction in total number of patient days is adjusted for population growth, the use of inpatient days decreased by 15% per capita.^{1 ,39} If the pattern of disease severity remained the same while the population expanded, then inpatients would have a higher severity of illness as smaller proportions of the total are hospitalized.^{13 ,38} Average length of stay has also decreased by 10%.³⁹ If the use of medical resources is unchanged for each hospitalization, then the intensity of services used per day will increase as the length of stay is reduced.^{3 ,9} Both greater severity of illness among hospitalized patients and reduced length of stay could have contributed to increased cost per day.

At the same time hospital inpatient use was decreasing, annual hospital outpatient visits simultaneously increased from 322 million to 414 million, an increase of 29%.³⁹ Diversion of hospital patients to outpatient settings faster than the infrastructure has adapted to the change is 1 more reason for increased health care expenditures.^{6 ,38 - 39 ,41}

Economic and political pressure is causing hospitals, physicians, and patients to abandon their traditional individual viewpoints and areas of expertise. We believe that understanding the relativity of cost for different decisionmakers can promote improved outcomes at each level. For instance, physicians are encouraged to make patient care decisions based on total cost of care while ideal resource use involves comparing marginal or even variable cost of service to potential patient benefit.^{14 ,23 ,42 - 45} Over the short term, physicians could have their most immediate impact by reducing costs for consumables with high variable cost such as pharmaceutical agents. Moreover, poor decisions (such as choosing an expensive new antibiotic when a less costly one is just as effective) are obscured by the overall high total cost of hospital care.

Hospital administrators face 2 distinct challenges. The first (defined in Figure 1 as technical efficiency) is to reduce the cost per unit of individual medical services.^{46 - 49} The second, defined as economic efficiency, is to maximize reimbursements relative to expenditures for services provided.^{46 ,50} These tasks could be in conflict as health policy induces hospitals to restrict patient service, whereas optimizing technical efficiency requires maximal use of fixed-cost resources.^{30 ,46 ,48 - 49} Better deployment of the labor force would improve technical efficiency,^{46 ,49 ,51} but for the facility's economic survival, staff are required to expend energy on proper documentation for hospital billing and regulatory requirements.^{8 ,20 ,38 ,46 ,52} At any given output of service, technical efficiency is further hindered because current third-party reimbursement strategies include the element of denied charges. The understandable response of hospitals to this pressure is to improve marketing, cost accounting, and billing procedures, further driving up fixed costs.^{6 ,26 ,28 ,52 - 54} Supporting evidence comes from Woolhandler and Himmelstein^{26 ,55} who reported increased hospital administrative costs from 1983 through 1987 and again from 1990 through 1994.

A notable finding of this study was that the variable proportion of cost at our facility (16%) was lower than the 25% to 42% often quoted for hospital care.^{14 ,56} The main reason is that different methods of categorizing cost were used. Most important, all employees at our institution, including physicians, were salaried. Most consider labor a variable cost.^{57 - 58} We defined labor costs as fixed because they could not be reduced on a short-term basis but only by long-term restructuring. There is some justification for considering hospital labor costs relatively fixed.⁵² Nationwide, the total number of hospital employees increased from 3.4 million in 1990 to 3.7 million in 1995, while inpatient days decreased by 10%.³⁹ It could be that more employees were performing outpatient procedures or that the intensity of service is greater per hospital day. An alternate reason is that reduction of hospital staff occurred more slowly than the decrease in output. The intense specialization of health care workers may make it more difficult to downsize than in the traditional firm. For instance, pediatric nurses cannot promptly adapt to adult cardiac care units.

Next, Cook County Hospital has a high patient volume, allowing economies of scale. We negotiate low rates for consumables by large volume purchasing. Last, this publicly funded hospital works with a restricted budget. Expenditures for supplies such as pharmaceuticals are kept as low as possible. Physicians are expected to order medications available on the formulary and more expensive patent medications are purchased only if clear benefit or special need is demonstrated. A further limitation in generalizing these results to other institutions is that Cook County Hospital is a large urban public teaching hospital. Our relatively low variable cost highlights the need for more data on the cost differences between hospitals. However, our ability to obtain detailed hospital cost information may have been facilitated by the fact that it is a public rather than private institution.

Our findings suggest that delays in reducing inpatient capacity while rapidly diverting patients to new ambulatory settings contributes to increased total health care expenditures.¹⁰ Downsizing and closing of hospitals is the logical solution but will be complicated by changes in our national demography. While total inpatient days have decreased and capacity is being reduced to match the need, this situation will eventually reverse as the population rapidly ages.¹ Forecasting the ideal capacity of the hospital system along with continued development of ambulatory alternatives could offset the need for new infrastructure as the demand for hospital space increases in the future.

Many of the proposed solutions for controlling health expenditures—such as regulating prices, capitation, and managed care—do not directly address technical efficiency from the hospital perspective.^{13 ,20 ,34 ,47 ,50 - 52 ,59 - 64} The volume of service provided by our national health care system justifies the development of specific methods of reducing both fixed and variable costs at the health care provider level.³⁴ Even small savings per patient encounter would translate into sizable amounts on a national scale. The concept of variable vs fixed cost has important implications for controlling health care expenditures.