The differential costs of employing support personnel: an approach to improving efficiency and fiscal performance.

ABSTRACT

Recognizing the need to control costs, including the wages and salaries of support personnel, this paper develops a generic model that evaluates the differential costs assigned to employees that perform support functions. The differential cost construction measures the variation in the wages and salaries of support personnel when related expenses are treated as a variable rather than a fixed cost. The model partitions the differential cost into portions that are attributable to the relative intensity of using support personnel, rates of compensation, volume of service and the efficiency of medical personnel who provide direct patient care. Based on an Excel spreadsheet, the results indicate that, other factors remaining constant, a modest increase in the intensity of employing support personnel and the implementation of policies that link the use of these employees to the volume of direct patient care result in substantial cost savings, lower cash disbursements and an improvement in profitability.

*********

The importance of monitoring, evaluating and controlling the costs assigned to members of the support staff stems from at least three considerations. First, when viewed from the perspective of society, an improvement in the process of monitoring and controlling spending on support functions, such as dietary, housekeeping, medical records and laundry may release resources that might be allocated to the provision of direct patient care. Hence, a more intense use of support personnel may enable the health service organization and society to avoid opportunity costs that assume the form of foregone health services.

Second, when viewed from the perspective of proprietary organizations, the goal of maximizing profitability requires a focus on the general problem of controlling operating expenses. Given that the costs of labor contribute to operating expenses, the profitability of the organization and the net cash flow derived from operations might be improved by adopting more stringent methods of monitoring, evaluating and controlling the amounts recognized as an expense of employing support personnel.

The third factor emanates from the external environment of the typical health service organization. The penetration of managed care organizations in American health markets and the increased dependence on prospective payment systems to finance the use of health care transfers financial risk from the insurer to the provider. As a consequence, health care providers must monitor and control the fiscal risks to which their organizations are exposed. As summarized by Boles and Glenn (1986), the increased dependence on regulated or negotiated rates of compensation to finance the use of health services requires the provider to control revenue risk, financial leverage and operating leverage.

As is well known, operating leverage is a source of risk that emanates from the relative responsiveness of cost to variation in operating activity. If price exceeds the variable cost per unit, changes in the rate of activity result in proportionate changes in total revenue and total variable costs, an outcome that imposes no risk on the health organization. On the other hand, fixed costs are invariant with respect to changes in the rate of operating activity and, as volume and revenue fall, the ability of the health organization to honor fixed obligations may be impaired. Accordingly, the adoption of strategies that enable the organization to transform fixed costs into variable expenses by adjusting resource consumption in response to variation in volume may increase profitability, improve the net cash flow derived from operations and reduce the fiscal risks that are attributable to operating leverage.

The relative importance of fixed expenses in the cost structure of a large urban teaching hospital was examined by Roberts et al. (1999). The results indicated that approximately 84 percent of the expenses were fixed and 31.5 percent of these costs were related to support expenses. Further, Roberts et al. also reported that 52 percent of the costs were traced to the salaries of personnel who provide direct patient care. The authors concluded that the efficiency of operations and the effectiveness of efforts to control costs are contingent in part on the adoption of practices that transform fixed costs into variable expenses by adjusting resource consumption to differences in patient volume. The focus of this paper is on an approach that enables the health service organization to monitor the consumption of support personnel and adjust their use to reflect variation in the provision of direct patient care.

OBJECTIVES

Recognizing the importance of improving the intensity of using support personnel, this paper has two objectives. The first is to develop a generic model that might be used to evaluate the difference in the costs of treating compensation paid to support personnel as a variable rather than a fixed expense. The approach proposed in this paper partitions the difference in spending on support functions into portions that are attributable to the relative intensity of using related personnel, rate of compensation, the relative efficiency of using providers of direct patient care, and the volume of service. In addition, the model depicts the joint or interactive effects of the main components that combine to determine the differential costs and cash disbursements associated with performing support functions.

Employing Excel, the second objective is to develop an interactive spreadsheet that demonstrates the precision of the approach. The results derived from the spreadsheet identify the components that contribute most to financial leverage and to cash disbursements that are potentially avoidable. Accordingly, the paper concludes with a discussion of the policies or practices that might be adopted to reduce fiscal risk, increase profitability and improve the net cash flow of the organization.

THE MODEL

The focus of the model is on a differential cost that is defined as the difference in the expenses that results from adopting one of two scenarios. In the first, it is assumed, heuristically, that support personnel are scheduled so as to satisfy peak demand for service and that the compensation paid to the support staff is regarded as a fixed expense. In the second scenario, it is assumed that the expenses of employing support personnel are regarded as a variable cost. Specifically, the model partitions the differential costs assigned to support personnel into portions that are attributable to: 1) the relative intensity of using members of the support staff; 2) deviations in the rate of compensation; 3) the relative efficiency of using personnel who provide direct patient care; 4) the volume of direct patient service and 5) a set of interactive effects.

Essentially two sets of costs are excluded from the analysis. The first consists of those costs that relate to items such as utilities, rent and communication. The second set stems from a need to maintain a staffing pattern that is required to sustain operating activity at a minimum level. It should be noted that the mix of personnel required to sustain activity at a minimum level is regarded as invariant with respect to changes in volume or the use of personnel who provide direct patient care. Accordingly, these expenses are not the focus of the model. For purposes of illustration, suppose the fixed cost associated with the need to sustain a minimum rate of activity is the same in the both scenarios. As a consequence, these expenses exert no influence on the magnitude of the differential cost that is examined by the model.

In the following, let the subscript j correspond to one of several occupational categories. Similarly, the indices v and f represent the scenario in which compensation is regarded as a variable and fixed expense respectively. The method of evaluating variation in the expenses assigned to occupational category j is based on the differential cost that is defined by

(1) DIFF([COST.sub.j]) - COMP([COST.sub.f]) - COMP([COST.sub.vj])

As indicated by equation 1, the differential cost is simply the difference in labor expenses resulting from a scenario in which the compensation paid to support personnel is regarded as fixed relative to a situation in which wage costs are regarded as variable.

The variation in payments to employees assigned to occupational category j is given by

(1.1) DIFF([COST.sub.j]) = [RATE.sub.fj] * [SUPPORTHOURS.sub.fj] - [RATE.sub.vj ] * [SUPPORTHOURS.sub.vj] (1.1) where [RATE.sub.fj] and [RATE.sub.vj] are the rates of compensating personnel assigned to category j.

Let the index k represent an individual assigned to one of several occupational categories and H correspond to the number of hours that the typical employee is expected to commit to market activity during the year. The average amount of compensation per hour associated with the variable cost scenario is obtained by

[RATE.sub.vj] - [SIGMA] [COMPENSATION.sub.vkj]/([n.sub.vj] * H)

In this case, the notation [n.sub.vj] refers to the variable cost scenario and the number of personnel assigned to the jth occupational category. After substituting the index f for v, the rate of compensation associated with the fixed cost scenario may be defined by

[RATE.sub.vj] - [SIGMA] [COMPENSATION.sub.fkj]/([n.sub.fj] * H)

In most cases, it is possible to ensure that [RATE.sub.fj] exceeds [RATE.sub.vj]. Specifically, the treatment of compensation of support personnel as a variable expense induces management to increase the use of part-time employees, a practice that may lower, the costs associated with fringe benefits, the average wage rate and the total compensation paid to employees assigned to a given occupational category.

It can be shown that the differential cost of employing support staff may be expressed in the form

(1.2) DIFF([COST.sub.j]) - ([RATE.sub.fj] - [RATE.sub.vj]) * [SUPPORTHOURS.sub.vj] + [RATE.sub.vj] * ([SUPPORTHOURS.sub.fj] - [SUPPORTHOURS.sub.vi]) + ([RATE.sub.fj] - [RATE.sub.vj]) * ([SUPPORTHOURS.sub.fj] - [SUPPORTHOURS.sub.vj])

In equation 1.2, the first term, identified hereafter as DIFF([RATE.sub.j]) measures the portion of the differential cost that is attributable to differences between the rates of compensation associated with the variable and fixed cost scenarios. Measured in hours, the terms [SUPPORTHOURS.sub.vj] and [SUPPORTHOURS.sub.fj] correspond to the use of support staff assigned to category j. Accordingly, the second of the three components that combine to determine the differential costs corresponds to the portion of the differential cost that is attributable to the variation in the use of support staff associated with the variable and fixed scenarios. The final term is the portion of the differential cost that results from the interaction of variation in the rate of compensation and the use of support staff, grouped by occupational category.

The model is predicated on the proposition that the use of support personnel is directly related to the demand for direct patient care. It also is assumed that the professional complement, comprised of physicians, nurses, physician assistants or nurse assistants is adjusted for variation in the volume of care, the flow of patients or the patient census. Hence, excluding the mix of employees required to sustain activity at a minimum level, the volume of direct patient care or the amount of resources committed to the provision of direct patient care is used to determine the use of the support staff. As a consequence, the difference represented by [SUPPORTHOURS.sub.fj] - [SUPPORTHOURS.sub.vj] appearing in equation 1.2 might be partitioned into components that represent the relative intensity of using support personnel, and the relative efficiency of using personnel who provide direct patient care. As indicated below, a portion of the variation in the use of support personnel consists of an interactive term that depicts the conjoint effects of the intensity of using support personnel and the amount of time committed to direct patient care.

Adopting the approach suggested by Finkler (1994), Finkler and Ward (1999), Broyles et al. (1998), the intensity of using administrators or support personnel might be measured by a ratio depicting the annual number of hours of support staff relative to the annual number of hours committed by the professional staff to the provision of direct patient care. In the following, the terms [RATIO.sub.fj] and [RATIO.sub.vj] correspond to mix of support personnel per hour of personnel who provide direct patient care. Measured by the total number of annual hours committed to market activity, let [HOURSPROV.sub.f] and [HOURSPROV.sub.v] correspond to the use of medical professionals to provide direct patient care. With the focus on the variable cost scenario, the intensity of using support staff assigned to one of several occupational groups is defined by

[RATIO.sub.vj] - [SUPPORTHOURS.sub.vj]/[HOURSPROV.sub.v]

Similarly, substituting the index f for the subscript v, the intensity of using support personnel that is associated with the fixed cost scenario is given by

[RATIO.sub.fj] - [SUPPORTHOURS.sub.fj /[HOURSPROV.sub.f]

If the professional complement, comprised of physicians, nurses, physician assistants or nurse assistants, is adjusted for variation in the flow of patients, the volume of patient care or the census, the ratio also measures the intensity of using support personnel relative to variation in operating activity. Further, the ratios also measure indirectly the efficiency of the support staff relative to the volume of care provided. Adopting this notation, the variation in the use of support personnel may be expressed in the general form

(2) ([SUPPORTHOURS.sub.fj] - [SUPPORTHOURS.sub.vj]) = ([RATIO.sub.fj] - [RATIO.sub.vj]) * [HOURSPROV.sub.v] + [RATIO.sub.vj] * ([HOURSPROV.sub.f]- [HOURSPROV.sub.v]) + ([RATIO.sub.fj] - [RATIO.sub.vj]) * ([HOURSPROV.sub.f]- [HOURSPROV.sub.v])

With respect to the first term of equation 2, note that, if value of [RATIO.sub.fj] exceeds [RATIO.sub.vj], the product ([RATIO.sub.fj] - [RATIO.sub.vj]) * [HOURSPROV.sub.v] measures the additional hours of using support personnel that result from treating related expenses as a fixed rather than a variable cost. In a similar fashion, the term [RATIO.sub.vj] * ([HOURSPROV.sub.f][HOURSPROV.sub.v]) corresponds to the portion of the differential resource use that is attributable to the relative number of hours committed to the provision of direct patient care. The final term of equation 2 represents the portion of the difference in the use of support staff, measured in hours, that is traceable to variation in the relative intensity of employing personnel and variation in the relative use of those who provide direct patient care.

As indicated by equation 2, the model is based on the proposition that operating leverage and total fiscal risk are reduced by policies or practices that determine the use of personnel who perform support functions by the volume of care provided by the health service organization. The differential use of physicians and others who provide patient care might be expressed in the form

(3) ([HOURSPROV.sub.f] - [HOURSPROV.sub.v]) = (HOURS/[UNIT.sub.f]- HOURS/[UNIT.sub.v]) * UNIT[S.sub.v] + HOURS/[UNIT.sub.v] * ([UNITS.sub.f]- UNIT[S.sub.v]) + ([HOURSAUNIT.sub.f] - HOURS/[UNIT.sub.v]) * ([UNITS.sub.f] - [UNITS.sub.v])

Suppose that the hours per unit of service associated with the fixed cost scenario are greater than the hours per unit derived for the variable cost scenario. Measured in hours, the first term of equation 3 measures the additional use of physicians and other personnel that is attributable to the relative efficiency of the process by which direct patient care is provided. The second term is the portion of ([PROVHOURS.sub.f] - [PROVHOURS.sub.v]) that is traced to differences in the volume of care while the third term identifies the portion of the variation in the use of physicians and other professional personnel that results from the interaction between variation in productivity and volume.

After substituting appropriately, equation 2 may be expressed in the form

(4) ([SUPPORTHOURS.sub.rj]- [SUPPORTHOURS.sub.vj]) = ([RATIO.sub.fj]- [RATIO.sub.vj]) * [HOURSPROV.sub.v] + [RATIO.sub.vj] * [(HOURS/[UNIT.sub.f]- HOURS/[UNIT.sub.v]) * UNIT[S.sub.v]] + [RATIO.sub.vj] [HOURS/[UNIT.sub.v] * (UNIT[S.sub.f]- UNIT[S.sub.v])] + [RATIO.sub.vj] [(HOURS/[UNIT.sub.f]HOURS/[UNIT.sub.v])

* ([UNITS.sub.f] - [UNITS.sub.v])] + [([RATIO.sub.fj]- [RATIO.sub.vj]) * [(HOURS/[UNIT.sub.f]- HOURS/[UNIT.sub.v]) * UNIT[S.sub.v]] + HOURS/[UNIT.sub.v] * (UNIT[S.sub.f]- UNIT[S.sub.v]) + [(HOURS/[UNIT.sub.t] - HOURS/[UNIT.sub.v]) * ([UNITS.sub.f] - [UNITS.sub.v])]}

As before, the first term of equation 4 measures the portion of HOUR[S.sub.fj] - HOUR[S.sub.vj] that is attributable to the relative intensity of using support personnel. Measured in hours, the second is the portion of the variation in the use of those who perform support functions those results from the relative efficiency of physicians and other professional personnel who provide direct patient care. The third term measures the difference in the use of support personnel that is traced to variation in the volume of direct patient care. The fourth term measures the portion of [SUPPORTHOURS.sub.fj] - [SUPPORTHOURS.sub.vj] that results from the interactive effects of the relative efficiency of those who provide direct patient care and the volume of service. The final term measures the interactive effects of the relative intensity of using support personnel, the relative efficiency of physicians and others who provide direct patient care and the volume of service.

The discussion indicates that the differential costs of employing support personnel, as defined by equation 1, is equivalent to

(5) DIFF([COST.sub.j]) = DIFF([RATE.sub.j]) + DIFF([INTENSITY.sub.j]) + DIFF([EFFICIENCY.sub.j]) + DIFF([VOLUME.sub.j]) + DIFF(INTERACTIV[E.sub.j])

As described previously, the terms of equation 4 may be summarized as follows.

(5.1) DIFF([RATE.sub.j]) = ([RATE.sub.fj] - [RATE.sub.vj]) * [SUPPORTHOURS.sub.vj]

(5.2) DIFF([INTENSITY.sub.1]) = [RATE.sub.vj]I([RATIO.sub.fj] - [RATIO.sub.vj]) * [HOURSPROV.sub.v]]

(5.3) DIFF([EFFICIENCY.sub.j]) = [RATE.sub.vj] * [RATIO.sub.vj] [(HOURS/UNITf-HOURS/UNIT,,) * UNITSvl

(5.4) DIFF([VOLUME.sub.j]) = [RATE.sub.vj] * [RATIO.sub.vj][HOURS/[UNIT.sub.v], * (UNIT[S.sub.f]-UNIT[S.sub.v])] and

(5.5) DIFF(INTERACTIV[E.sub.j]) = ([RATE.sub.fj]- [RATE.sub.vj]) * ([SUPPORTHOURS.sub.fj]-[SUPPORTHOURS.sub.vj]) + [RATE.sub.vj][ ([RATIO.sub.fj.] - [RATIO.sub.vj]) * ([HOURSPROV.sub.f ]- [HOURSPROV.sub.v])] + [RATE.sub.vj] * RATIOvj[(HOURS/[UNIT.sub.f] - HOURS/[UNIT.sub.v]) * (UNIT[S.sub.f]- UNIT[S.sub.v])] (5.5)

In this formulation, the term DIFF([RATE.sub.j]) is the portion of the differential cost that is attributable to variation in the rate of compensating those who are assigned to the jth support category. The term DIFF([INTENSITY.sub.j]) is the portion of the differential cost that is attributable to the relative intensity of using support staff per hour of direct patient care while the term DIFF(EFFICIENC[Y.sub.j]) is the portion of the differential cost that is attributable to the relative efficiency of using physicians and other professional personnel to provide direct patient care. Similarly, the term DIFF(VOLUM[E.sub.j]) is the portion of the differential cost that results from variation in the volume of direct patient care. Finally, the term DIFF(INTERACTIV[E.sub.j]) is the mutual or interactive effect of differences in the relative intensity of using support personnel, related rates of compensation, the relative efficiency of personnel who provide direct patient services and the volume of care.

AN ILLUSTRATION

In this section, the model described previously is applied to a set of data depicting the operations of a health service organization that provides only ambulatory care. The information was obtained originally from the records of a managed care organization and, in order to ensure the anonymity of the provider and to avoid the possibility of compromising the proprietary nature of the data, the parameters of the model were subsequently modified. Accordingly, the organization and the data on which the illustration is based must be regarded as hypothetical in nature. Limited to five occupational categories, the purpose of the discussion is to use Excel to construct an interactive spreadsheet that calculates the components that combine to determine the differential costs of employing those assigned to each of several occupational categories.

Summary data for each of the five occupational categories are listed in Exhibit 1. Consistent with equation 1, the differential cost associated with each group or occupational category was obtained by comparing the expense associated with the fixed cost scenario with the corresponding cost associated with the variable cost scenario. In summary of these calculations, it is important to note that the total of the cost differences to employees assigned to the five occupational categories amounted to $726,752.40.

Exhibit 2 summarizes the data that pertain to the fixed and variable cost scenarios. With respect to the variable cost scenario, it is assumed that the health service organization provides 220,000 visits. With the focus on the fixed cost scenario, it is assumed that total volume, to include peak demand, amounts to 240,000 visits, suggesting that the difference UNIT[S.sub.f]--UNIT[S.sub.v] is 20,000 visits. As shown in the third column of the exhibit, the amount of time per visit, expressed in hours, was obtained by dividing the number of minutes per visit by 60. The results of these calculations suggest that the difference HOURS/UNI[T.sub.f]--HOURS/UNI[T.sub.v] is 3 minutes or 0.05 of an hour. The summary presented in the exhibit also indicates that, when compared to the fixed cost scenario, the variable cost scenario requires 17,000 fewer hours of support personnel.

The basic data that define the variable cost scenario are listed in Exhibit 3. The annual compensation for the typical individual assigned to each of the occupational categories is listed in the first column of the exhibit. The ratios, RATI[O.sub.vj], that appear in the second column of the exhibit were obtained by applying equation 3.1 to the data. For purposes of illustration, it is assumed that the product of 40 hours per week and 52 weeks or 2080 hours represents the standard work year. Accordingly, the amount of compensation per hour associated with the variable cost scenario appear in the fourth column and were obtained by dividing 2080 hours into the annual wage expense listed in column 1 of the exhibit. The data depicting the fixed cost scenario were calculated in a similar fashion and are summarized in Exhibit 4.

The analysis of the components that combine to determine the differential cost presented in Exhibit 5 is based on the data listed in Exhibits 3 and 4. In particular, the calculations are an extension of the Excel spreadsheet that lists the summary information presented in Exhibit 1. The method of determining the amounts assigned to the main and interactive components that combine to determine the differential cost is illustrated in Panel A of Exhibit 5 while the joint effects that determine the magnitude of the interactive term appear in Panel B of the exhibit.

As indicated, a portion of the differential cost is attributable to differences between the rate of compensation calculated for the fixed and variable scenarios. Variation in rates of pay for a given occupational group may result from the increased use of part-time employees and a decline in the costs associated with fringe benefits and overtime pay. Based on the differences in the computed values of RAT[E.sub.fj] and RAT[F.sub.vj], an application of equation 5.1 to the hypothetical information yielded the results appearing in the first column of Panel A in Exhibit 5. As these results suggest, the portion of the differential cost that is attributable to differences in the rates of pay, grouped by occupational category, amounts to $97,175.48 or approximately 13 percent of the total.

An inspection of the values computed for RATI[O.sub.fj] and RATI[O.sub.vj] indicates that, relative to the fixed cost scenario, support personnel were used more intensely when the related expenses are regarded as a variable cost. An application of equation 5.2 to the hypothetical data indicates that the portion of the differential cost that resulted from differences between RATI[O.sub.fj] and RATI[O.sub.vj] amounted to $ ! 86,947.12. When compared to the additional costs of $726,752.40, the calculations presented in Panel A of Exhibit 5 suggest that the sum of the terms DIFF(INTENSIT[Y.sub.j]) represent approximately 26 percent of the total differential cost. Accordingly, the model not only indicates the additional costs, foregone profitability and related cash disbursements that might have been avoided by a more diligent process of monitoring and evaluating the use of related personnel. The results derived from the model also alert the health service organization to the presence of a problem that requires the implementation of remedial policies designed to adjust the level of employing support personnel to the volume of direct patient care.

An inspection of Exhibit 2 also indicates that the time associated with a typical visit required an additional 3 minutes when expenses are regarded as a fixed rather than a variable cost. After applying equation 5.3 to the data, the model indicates that the difference HOURS/UNI[T.sub.f]--HOURS/UNI[T.sub.v] contributed $217,487.98 to the differential cost, an amount that also represents approximately 30 percent of the total variation in the costs of employing support personnel. In this case, the additional time committed to the average visit might be a reflection of one or more considerations. For example, deficiencies in the scheduled arrival or departure of patients may contribute to the commitment of more time per visit than planned. Holding the number of visits constant, the provision of a disproportionate number of visits that requires more time than the value assigned to HOURS/UNI[T.sub.v] also may result in a greater number of hours per unit of service. In this situation, it may be desirable to assemble data depicting volume and related time in accordance with the Physicians Fee & Coding Guide (HCCA, 1994).

As indicated, a portion of the differential costs of employing support personnel is attributable to the deviation in the number of visits. As indicated in Exhibit 2, the volume of care assigned to the fixed cost scenario exceeds the amount of service that is assumed for the variable cost scenario by 20,000 visits. After applying equation 5.4 to the hypothetical data, the results indicate that a failure to adjust support personnel to the lower volume of service results in additional costs of $98,858.17, an amount that represents approximately 14 percent of the total differential cost.

The precision of the model is verified by noting that the sum of the components, to include the interactive effects calculated in Panel B of Exhibit 5 is $726,752.40, an amount that is identical to the total differential cost presented in Exhibit 1. Further, as shown in Panel C of Exhibit 5, the results of the model may be expressed as a proportion of the total differential cost. An inspection of Panel C reveals that a relatively large proportion of the difference in labor cost is attributable to relative intensity of using employees who perform support functions, and in particular to the relative intensity of using individuals assigned to occupational categories A and C. In addition, the data also indicate that a large percentage of the differential cost is attributable to the additional time committed to the provision of a unit of service. Accordingly, the results enable the health service organization to identify areas in the process of providing direct patient care that require investigation and perhaps the implementation of remedial policies or practices.

Finally, the relative distribution enables the health service organization to identify, with precision, the occupational categories that contribute the largest amounts to the differential cost. In terms of the illustration, the results indicate that the intensity of employing support personnel assigned to occupational category A results in approximately 17 percent of the differential cost of employing support personnel. Similarly, the relative magnitude of DIFF(EFFICIENC[Y.sub.j]) calculated for employees assigned to occupational categories C and E also indicates a need for investigation and perhaps the implementation of remedial action.

LIMITATIONS AND DISCUSSION

In summary, the model is based on the proposition that support personnel should be employed as efficiently or as intensely as possible and that the expense assigned to these occupational groups should be linked directly to the rate of providing direct patient care. When viewed from the perspective of society, an improvement in the process of monitoring and controlling spending on support functions may release resources that might be allocated to the provision of additional patient care. Hence, a more intense use of support personnel may enable the health service organization and society to avoid opportunity costs that assume the form of foregone health services. Further, when viewed from the perspective of proprietary organizations, the goal of maximizing profitability requires a focus on the general problem of controlling operating expenses. As is well known, salaries and labor related expenses represent the lion's share of operating costs. Hence, the profitability of the organization might be increased by improved methods of controlling or lowering the amounts recognized as an operating expense of employing support personnel.

As indicated, the most important parameter of the model is the ratio of support personnel to the complement of personnel who provide direct patient care. If the ratio is "too high", resources are used less efficiently than possible. Recognizing that the services of labor are either used efficiently or foregone, methods that improve the intensity of using support personnel not only lower expenses and cash disbursements but also improve the profitability and liquidity of the health organization. Conversely, a ratio that is "too low" implies that the complement of support personnel may be inadequate, an outcome that might reduce the frequency or the effectiveness with which support functions are performed. As a consequence, it is possible to argue that the ratio of support staff to the complement of personnel responsible for direct patient care is a standard of performance that enables the administrator to develop a target or a desired level of employing support staff.

The value assigned to the term RATI[O.sub.vj] might be based on the time required to perform the activities or functions that are precipitated by the volume and mix of patient care or the number of personnel who provide direct patient care (Chan 1993; Udpa 1996; Canby 1995). For example, an initial or return visit requires the health service organization to initiate or revise the patient's medical record, to schedule the provision of prescribed ancillary care, to establish or revise financial records and to process the accounting information that is derived from the delivery of service. In addition, the provision of direct patient care requires the maintenance and calibration of equipment, a task that also requires the use of support personnel. Each of the functions or activities consumes the services of employees and forms the basis for estimating an optimal relation between the use of support personnel and the volume of operating activity or the number of personnel who provide direct patient care.

The results of the model also enable the organization to control or reduce risk when viewed from an ex post perspective. In particular, each of the components that determine the differential cost might be regarded as controllable or uncontrollable. A controllable deviation is defined as a portion of the differential cost that is attributable to factors which are controlled, to a significant extent, by an individual or group associated with the health service organization. For example, the amount derived for DIFF(RAT[E.sub.j]) is contingent on the decisions of senior administrators to adjust rates of compensation earned by current employees or to hire additional individuals whose salary or wage exceeds the average pay of those currently assigned to a given occupational group.

Similarly, the adoption and implementation of policies that control DIFF(INTENSITY) or ensure the efficient use of the support staff assigned to each of the occupational categories might lower the amount of differential cost assigned to each of the groups. After accommodating the potential need to provide care to emergent cases, the organization might improve the process of scheduling activity so as to ensure that the services of labor are used optimally. As described previously, the results derived for DIFF(INTENSIT[Y.sub.j]) also may indicate a need or an opportunity to consolidate the functions currently assigned to two or more employees or occupational categories.

The efficiency of using employees assigned to occupational category j might be increased by adopting one or more policies designed to increase the latitude with which staffing patterns are established. First, the health service organization might implement a training program that ensures individuals are able to perform multiple functions. It is possible to argue that, an increase in the scope of functions performed by each employee enhances the flexibility available when establishing or revising the staffing pattern. Accordingly, a well designed training program improves the ability of the organization to adjust the use of support personnel in response to changes in volume and to transform a fixed expense into a variable cost.

The organization also might rely on a pool of temporary or part-time employees as a source of staff during periods in which the need to provide support services increases. However, the effectiveness of a decision to rely on a pool of part time or temporary employees to adjust the use of support personnel is contingent on conditions in the local labor market. In particular, a labor market that is characterized by a shortage of personnel clearly reduces the availability of potential employees and the ability of the organization to adjust staffing patterns. It is possible to argue that temporary or part time employees may lack the organizational commitment that is required for the delivery of quality care or a consistent compliance with regulatory requirements such as those specified by HIPPA.

It also is important to note that a reliance on a pool of temporary or part time employees to increase the availability of personnel who provide direct patient care may compromise the quality of care. In addition to those reasons cited previously, it is well recognized that the proficiency of the provider is related positively to the frequency of providing a given procedure or performing a given function. In this case, part time or temporary employees are likely to provide the service or perform the procedure with a lesser frequency than their full time counterparts, implying that, on balance, the quality of direct patient care might suffer.

As described in this paper, the magnitudes derived for DIFF(EFFICIENC[Y.sub.j]) are dependent on the hours of professional staff per unit of service. A ratio that is "too high" indicates that nurses, nurse assistants, physician assistants, technicians and, to a lesser extent, physicians were used inefficiently or less efficiently than possible. On the other hand, a ratio that is "too low" may indicate that human resources are inadequate, an outcome that might compromise the quality of care. Hence, the ratio depicting the number of hours of professional personnel per unit of service is a standard of performance that represents a balance between the need to use resources as efficiently as possible and the need to provide quality care. In turn, the ratio might be used to develop the desired or target complement of medical personnel.

The professional staff and, hence, the support complement is contingent, in part, on not only the composition of required services, but also the number and mix of patients, to include the diagnostic distribution and the severity of cases within each of the diagnostic conditions. Accordingly, the primary responsibility for controlling the portion of the differential cost assigned to DIFF(EFFICIENC[Y.sub.j]) resides with the medical staff and those who schedule the provision of service.

Apart from initial visits associated with a given episode of illness, the amount assigned to DIFF(VOLUM[E.sub.j]) represents the primary responsibility of the medical staff. Accordingly, both DIFF(EFFICIENC[Y.sub.j]) and DIFF(VOLUM[E.sub.j]) represent a portion of the differential cost that is controlled, to a significant extent by an individual or group or associated with the health organization. In turn, both might be regarded as controllable portions of the differential cost that might be monitored, evaluated and controlled so as to ensure a favorable fiscal outcome.

Consider next the use of the model in the process of monitoring, evaluating and controlling activity. As presented in Panels A and C of Exhibit 5, the results derived from the model enable the health service organization to identify problem areas with specificity. First, the column totals depict the portion or percentage of the differential cost that is attributable to the source or the cause of the deviation. On the other hand, the row totals depict the portion or percentage of the total difference that is assigned to each of the objects of expense. Finally, values appearing in the cells identify the variance in spending, grouped both by object of expense and source or cause. Hence, the results of the model enable the organization to identify, with relative precision, not only the source of the differential cost, but also those occupational categories in which additional expenses exert the most deleterious influence on fiscal performance.

Finally, the model might be used to perform sensitivity analysis, a practice that might lower risk when viewed from an ex ante perspective. In order to ensure their fiscal viability, most health organizations require an accurate estimate of cost prior to negotiating fixed or prospective rates of compensation. As described in Exhibit 5, the model might be used to perform a sensitivity analysis that indicates the change in cost resulting from a sequential variation in one of the parameters, holding all others constant. For example, the model instantaneously calculates the change in cost that results from variation in the intensity of using the support staff, the efficiency of providing care, the volume of service and rates of compensation. When viewed from an ex ante perspective, accurate projections of cost and differences in the expenses associated with alternate scenarios may improve the ability of management to negotiate rates of payment that are congruent with the fiscal needs of the health organization. Hence, the model presented in this paper enhances the ability of the organization to negotiate more favorable rates of payment, improve the use of support personnel, lower fixed costs, reduce fiscal risk and improve fiscal performance.

EXHIBIT 1

THE DIFFERENTIAL COST OF USING SUPPORT PERSONNEL

OCCUPATIONAL   FIXED SCENARIO   VARIABLE SCENARIO

Category       Wage Expense     Wage Expense        Differential Cost

A              $249,230.77       $125,600.96         $123,629.84
B                96,923.08         50,240.38           46,682.69
C               597,115.38        333,173.08          263,942.31
D               310,154.85        195,012.02          115,142.83
E               560,769.23        383,413.46          177,355.77
Total        $1,814,192.31       $1,087,439.90         $726,752.40

EXHIBIT 2

THE BASIS DATA: DIRECT PATIENT CARE

                    Number of   Minutes     Hours       Hours of
                    Visits      Per Visit   Per Visit   Direct Care

Variable Scenario   220,000     15          0.25        55,000
Fixed Scenario      240,000     18          0.30        72,000
Difference           20,000      3          0.05        17,000

EXHIBIT 3

THE BASIC DATA: THE VARIABLE COST SCENARIO

Occupational   Annual Wage    Ratio   Hours    Rate    Total
Category       Expense                                 Expense

A              $47,000        0.10     5,500   22.84   $125,600.96
B               38,000        0.05     2,750   18.27     50,240.38
C               31,500        0.40    22,000   15.14    333,173.08
D               29,500        0.25    13,750   14.18    195,012.02
E               29,000        0.50    27,500   13.94    383,413.46

EXHIBIT 4

THE BASIC DATA: THE FIXED COST SCENARIO

Occupational   Annual Wage    Ratio   Hours    Rate     Total
Category       Expense                                  Expense

A              $48,000        0.15    10,800   $23.08   $249,230.77
B               40,000        0.07     5,040    19.23     96,923.08
C               37,500        0.46    33,120    18.03    597,115.38
D               32,000        0.28    20,160    15.38    310,154.85
E               30,000        0.54    38,880    14.42   $560,769.23

EXHIBIT 5
THE ANALYSIS OF THE DIFFERENTIAL COST

PANEL A: THE MAIN AND INTERACTIVE COMPONENTS OF THE DIFFERENTLAL COST

Occupational
  Category          Rate    Intensity    Efficiency      Volume

     A          $1,322.12    62,800.48    25,120.19    11,418.27
     B           2,644.23    20,096.15    10,048.08     4,567.31
     C          63,461.54    49,975.96    66,634.62    30,288.46
     D          16,526.44    23,401.44    39,002.40    17,728.37
     E          13,221.15    30,673.08    76,682.69    34,855.77
   Total       $97,175.48   186,947.12   217,487.98    98,858.17

Occupational
  Category     Interactive       Total

     A           22,968.75   123,629.81
     B            9,326.92    46,682.69
     C           53,581.73   263,942.31
     D           18,483.17   115,141.83
     E           21,923.08   177,355.77
   Total        126,283.65   726,752.40

PANEL B. THE INTERACTIVE COMPONENTS OF THE DIFFERENTIAL COST

                   Rate,       Ratio,
Occupational     Support     Hours of    Efficiency,
  Category         Hours         Care       Volume        Total

     A          $1,274.04    19,411.06     2,283.65    22,968.75
     B           2,201.92     6,211.54       913.46     9,326.92
     C          32,076.92    15,447.12     6,057.69    53,581.73
     D           7,704.33     7,233.17     3,545.67    18,483.17
     E           5,471.15     9,480.77     6,971.15    21,923.08
   Total       $48,728.37    57,783.65    19,771.63   126,283.65

PANEL C. RELATIVE DISTRIBUTION OF THE COMPONENTS THAT DETERMINE THE
DIFFERENTIAL COST

Occupational
  Category          Rate    Intensity    Efficiency      Volume

     A             $0.002        0.086        0.035        0.016
     B              0.004        0.028        0.014        0.006
     C              0.087        0.069        0.092        0.042
     D              0.023        0.032        0.054        0.024
     E              0.018        0.042        0.106        0.048
   Total           $0.134        0.257        0.299        0.136

Occupational
  Category     Interactive       Total

     A               0.032        0.170
     B               0.013        0.064
     C               0.074        0.363
     D               0.025        0.158
     E               0.030        0.244
   Total             0.174        1.000

REFERENCES

Boles, K.E., & Glenn, J.K. (1986). What accounting leaves out of financial Statements? Hospital and Health Services Administration, March/April, 8-27.

Broyles, R.W., Brandt E.N., & Biard-Holmes, D. (1998). A practical method of adjusting for risk in the prospective costs of capitated systems. Health Care Management Review, 23(2), 63-75.

Canby, J.B. (1995). Applying activity-based costing to healthcare settings. Healthcare Financial Management, February, 51.

Chan, Y. (1993). Improving hospital cost accounting with activity-based costing. Health Care Management Review, 18(1), 71-77.

Finkler, S.A., & Ward DM. (1999). Issues in Cost Accounting for Health Care Organizations, Aspen, Gaithersburg, MD.

Finkler, S.A. (1994). Essentials of Cost Accounting for Health Care Organizations, Aspen, Gaithersburg, MD.

Health Care Consultants of America. (1994). Physicians Fee and Coding Guide, Augusta, GA.

Roberts, R.R., Frutos, P.W., Ciavarella, G.G., Gussow, L.M., Mensah, E.K., Kampe, L.M., Straus, H.E., Joseph, G., & Rydman, R.J. (1999). Distribution of variable vs. fixed costs of hospital care. Journal of the American Medical Association, 281 (7), 644-49.

Udpa, S. (1996). Activity-based costing for hospitals. Healthcare Management Review, 21 (3), 87-89.

Robert W. Broyles University of Oklahoma

Amir Khaliq University of Oklahoma

Madeline J. Robertson University of Oklahoma

Lutchmie Narine University of North Carolina at Charlotte