Previous examinations of administrative costs have assumed that these expenditures have added costs without benefit, a position that we believe has not been fully substantiated. In this study, we examine the change in hospital administrative costs compared with changes in other areas of the hospital budget for an 8-year period (1983 to 1990) and also examine the change in expenditures for patient care departments compared with nonpatient care departments and departments with a higher number of regulatory requirements compared with those having fewer regulatory requirements. We also investigated whether hospitals with many patients may be able to more efficiently meet these demands than smaller hospitals, by examining the growth of administrative costs by hospital size.

Methods

Top Methods Results Discussion Author & Article Info References

Hospitals Surveyed

Hospitals holding contracts with Blue Cross of Western Pennsylvania (BCWP) are required to submit annual Standard Provider Operating Budgets. Permission to examine these budgets for the years 1983 to 1990 was requested from all non-Veterans Affairs BCWP-participating hospitals. Seventy hospitals agreed to this request (responding hospitals) and 30 did not (nonresponding hospitals).

The size of responding hospitals was categorized by the total annual in-hospital patient admissions in 1983, obtained from the operating budget. The size of nonresponding hospitals was categorized by total in-hospital patient admissions as listed in the American Hospital Association Directory [4]. Hospitals were also classified according to whether they provided general medical care or specialty care and whether they were short-term or long-term care providers [4]. The location of each hospital was classified according to location within or outside the metropolitan Pittsburgh area, as defined by the American Hospital Association Pennsylvania state map [4]. Finally, all hospitals were classified by teaching affiliation status according to criteria developed by the Pennsylvania Health Care Cost Containment Council [5].

Data Categories

Annual operating expenditures were obtained from the BCWP Standard Provider Operating Budget, which represented hospital estimates of anticipated expenditures for the next budgeted fiscal year. Departmental expenditure estimates included salaries (except for nurses and salaried physicians), capital costs, and operating expenses.

We categorized the data in three ways to isolate the change in administrative costs from changes in other expenditures. The first grouping was based on the six budget categories listed in the provider standard operating budgets: 1) administration; 2) service departments; 3) ancillary services; 4) professional care; 5) miscellaneous expenditures; and 6) the total hospital budget (Table 1). Administrative departments included general administration, accounting and fiscal services, admitting, data processing, nonpatient telephones, and purchasing-receiving. Service departments were those not directly involved in the provision of clinical care, excluding those listed in the administration budget category. Ancillary services included departments providing nonprofessional clinical services. Professional care consisted of salaries and wages for services provided by nurses and salaried physicians. Miscellaneous expenditures included nonreimbursable expenditures (for example, parking, gift shop, volunteer services) and other expenses (for example, interest, malpractice premiums, alcohol detoxification programs). The total hospital budget was obtained by summing all budget categories. The BCWP standard operating budget groupings do not perfectly separate administrative costs from other costs. For example, service departments include nursing administration and medical records, both of which are administrative functions. Utilization review (listed under miscellaneous expenditures) also included substantial administrative costs. However, most administrative expenditures were concentrated in the administration category. We did a sensitivity analysis where we redefined administrative cost to include all of those departments we believed had largely administrative functions. In addition to general administration and nursing administration, these included medical records, personnel, medical affairs, utilization review, and malpractice expenses.

In the second method of isolating changes in administrative costs Table 2, we compared expenditures from hospital departments involved in direct patient care (ancillary services and professional care) with other budget categories that generally represented nonpatient care services (administration, service departments, and miscellaneous expenditures). Again these budget groupings were not perfect. For example, skilled nursing and home health, both listed under "miscellaneous," represent patient care expenditures. However, most patient care services were concentrated in the patient care categories.

In the third method of isolating changes in administrative costs, we examined annual expenditures of several departments that, in a previous study [6], had been shown to have either a larger or a smaller proportion of their resources devoted to compliance with regulatory activities. Departments with greater regulatory requirements were defined as those that spend more than 40% of their resources on compliance with regulatory requirements from both public agencies and third party payers, and department with fewer regulatory requirements were defined as those that spend less than 10% of their resources on regulatory compliance. These estimates were based on a 1978 report of the Hospital Association of New York that identified the amount of resources devoted by hospital departments to regulatory compliance. Data for specific departments with greater and fewer regulatory requirements were incomplete for some responding hospitals. Thus, this analysis was based on departments for which at least 85% of responding hospitals reported expenditure data. The departments with greater regulatory requirements in this analysis included administration, medical records, and nursing administration. The departments with fewer regulatory requirements in this analysis included laundry and linen, central sterile, and dietary (Table 3).

Data Analysis

Characteristics of responder and nonresponder hospitals were examined for differences using the chi-square analysis, Wilcoxon rank-sum test, and the Fisher exact test.

The percentage change in budget category operating expenditures from 1983 to 1990 was calculated. Departments with greater and fewer regulatory requirements were summed separately, and total expenditure growth was compared. The percentage change in operating expenditures was calculated for hospitals of different sizes, classified by the annual number of admissions. The three categories of hospital size were determined using an analysis of the distribution of responding hospitals. All costs are reported in nominal dollars because there was no reason to believe that administrative costs were affected by inflation differently than were nonadministrative costs.

Results

Top Methods Results Discussion Author & Article Info References

Of the 100 hospitals surveyed, 70 responded. No statistical differences occurred among the responding and nonresponding hospitals Table 4, according to teaching status, location, number of admissions per year, acuity of hospital care, or type of hospital.

Figure 1 shows the results of the first analysis to isolate changes in administrative costs using standard BCWP budget categories. Administrative costs increased 90% during the 8-year study period compared with an increase in the total hospital budget of 45%; whereas expenditures for service departments increased only 29%; ancillary services, 30%; professional care, 52%; and miscellaneous expenditures, 70%. Administrative expenditures represented 10.6% of the hospital budget in 1983 (mean costs of $2 915 161 per hospital) and 13.9% in 1990 (mean cost of $6 371 043 per hospital), with a peak of 15.3% in 1989 (mean cost of $5 693 839 per hospital, Figure 2. When the sensitivity analysis was done with a broader definition of administrative activities, these costs increased 107% during the 8-year study period, which represented 14.6% of the hospital budget in 1983 and 20% of the budget in 1990. Figure 3 shows the breakdown of administrative costs by hospital size, where there were no discernible trends.

View larger version (16K): [in this window] [in a new window]   Figure 1. Cumulative percentage increase in average expenditures by budget category. Increase was from 1983 to 1990.

View larger version (10K): [in this window] [in a new window]   Figure 2. Percentage of administrative budget compared with total hospital budget, by year.

View larger version (16K): [in this window] [in a new window]   Figure 3. Percentage of administrative costs compared with total hospital budget by hospital size. Costs were from 1983 to 1990.

Figure 4 shows the results of the second analysis that compared expenditures for direct patient care with all other expenditures (nonpatient care activities). The cumulative percentage increase in expenditures for departments with patient care responsibilities did not differ from those providing nonclinical services, an increase of 44% and 46%, respectively. Patient care expenditures were approximately 49% of the total budget for both 1983 and 1990. In hospitals with more than 15 000 admissions per year, patient care expenditures increased 97% between 1983 and 1990, compared with 75% and 71% for medium-size and small-size hospitals, respectively.

View larger version (13K): [in this window] [in a new window]   Figure 4. Cumulative percentage increase in patient care and nonpatient care expenditures. Increase was from 1983 to 1990.

In the third analysis of costs, the budgets of departments we defined as having a greater number of regulatory requirements increased by 84% during the study period, compared with just 5% for the departments with fewer requirements (Figure 5). Departments with greater regulatory requirements comprised 14% of the total budget in 1983 and 17% in 1990, whereas departments with fewer requirements decreased as a percentage of the total budget from 8% in 1983 to 6% in 1990. There were no statistical differences in the rate of change of departmental budgets with greater regulatory or fewer regulatory requirements among hospitals of different sizes.

View larger version (13K): [in this window] [in a new window]   Figure 5. Cumulative percentage increase in expenditures for departments with a greater number of regulatory requirements and those with fewer requirements. Increase was from 1983 to 1990.

Discussion

Top Methods Results Discussion Author & Article Info References

From 1983 through 1990, estimates of hospital administrative costs in Western Pennsylvania increased faster than did ancillary services, service departments, professional care, miscellaneous expenses, or the total hospital budget. This finding was confirmed in both our first definition of administrative activities as well as in our sensitivity analysis that allowed for a stricter definition of administrative duties. A similar trend was seen among departments with a greater number of regulatory obligations compared with departments with fewer regulatory requirements. These findings suggest that administrative expenditures outpaced other areas of the hospital budget during the study period. However, overall expenditures in departments with direct patient care responsibilities did not increase appreciably faster than in departments that did not provide clinical services, possibly reflecting the fact that administrative costs may be increasing equally in both patient care departments and nonpatient care departments.

Several possible explanations could account for our findings. First, increasing regulation and external administrative requirements may be responsible for the increasing cost of hospital administration. Government and private insurers have increased the paperwork required for reimbursement, and these demands have fallen disproportionately to the administrative cost centers of hospitals. The growth of managed care in many hospital market areas, for example, may have increased the resources necessary to comply with third party payors' demands for increased monitoring of medical activities and shorter lengths of stay. Second, in an effort to provide practicing physicians with valuable clinical information, hospitals are intensifying their quality assurance and quality improvement activities, which may require an infusion of new administrative personnel and data systems, although this may improve quality of care as well. We could not measure the effect of increasing administrative costs on quality of care.

Although administrative costs represent only a fraction of total expenditures, their rapid growth may be worrisome if hospitals are unable to borrow from patient care resources without having a negative impact on quality. Such a scenario may explain why hospitals having more stringent certificate-of-need programs and regulatory mandates have been shown to have higher mortality rates [7]. Decreased hospital utilization and increased age-adjusted mortality rates have also coincided with the institution of the prospective payment system and other cost-containment measures [8].

However, more recent reports [9] show that no change occurred in mortality rate among Medicare patients after the implementation of prospective payment, suggesting that patient outcomes have not worsened. Indeed, increased administrative costs may reflect increased requirements for personnel and paperwork, without a substantial impact on quality of care. In addition, it is possible that nonclinical expenditures have improved hospital functioning and that better quality of care has led to shorter lengths of stay and improved patient outcomes.

There are several limitations to our study. First, we were unable to make clear distinctions between administrative and nonadministrative tasks. For example, because nursing administration and medical records were classified as service departments, they were included in the administrative cost estimates for the second and third approaches to isolating administrative costs but were not included in the first. Utilization review departments were categorized as a miscellaneous expenditure in the first analysis and as a nonpatient care expenditure in the second analysis. Such problems with classification may explain the large increase in expenditures among miscellaneous services, which included utilization review and malpractice costs, both of which have some overlap with administrative services. Second, we were unable to validate the accuracy of the figures provided in the BCWP-estimated hospital budgets. Third, our analysis of departments with greater and fewer regulatory requirements relies on categories developed in an analysis done 15 years ago. Departmental responsibilities and functions may have greatly changed in the intervening years.

Nonetheless, the trend toward greater administrative costs deserves further attention. Specifically, future research should focus on whether these expenditures increase costs without benefit or whether they actually improve patient care.

The opinions expressed in this paper are those of the authors and do not necessarily represent those of the Shadyside Hospital Foundation or the Robert Wood Johnson Foundation. All omissions and errors are those of the authors.

Abbreviation

BCWP: Blue Cross of Western Pennsylvania