HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): John E. Mayer Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jenkins, K. J. Articles by Mayer, J. E. Search for Related Content PubMed PubMed Citation Articles by Jenkins, K. J. Articles by Mayer, J. E.

Ann Thorac Surg 1998;66:860-869
© 1998 The Society of Thoracic Surgeons

---

Original articles: Cardiovascular

Validation of relative value scale for congenital heart operations

^a Department of Cardiology, Children’s Hospital, Boston, Massachusetts, USA
^b Department of Cardiovascular Surgery, Children’s Hospital, Boston, Massachusetts, USA
^c Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA

Accepted for publication April 2, 1998.

Address reprint requests to Dr Jenkins, Department of Cardiology, Children’s Hospital, 300 Longwood Ave, Boston, MA 02115

	Abstract

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Background. To determine the validity of the newly assigned work relative value unit (RVU) scale for surgical procedures for congenital heart disease, we measured its relationship to length of hospital stay, total hospital charges, and mortality.

Methods. We identified cases by the presence of ICD-9-CM codes in nine statewide, administrative hospital discharge abstract databases for 1992. Computer algorithms were generated to assign RVUs to individual cases. Spearman correlation coefficients between work and practice expense RVUs and median length of hospital stay, total hospital charges, and in-hospital mortality were determined, as well as parameter estimates from linear and logistic regression.

Results. Using data from 5,192 cases involving 34 surgical procedures for congenital heart disease, higher work RVUs were associated with longer lengths of hospital stay (r_s = 0.72, p < 0.0001), higher total hospital charges (r_s = 0.81, p < 0.0001), and higher in-hospital mortality (r_s = 0.45, p = 0.01). A 5-point increase in the relative value scale was associated with an increase in the length of stay by a multiplicative factor of 1.3 (p < 0.0001); total hospital charges by 1.5 (p < 0.0001); and the odds of in-hospital death by 1.9 (p < 0.0001). Findings were similar for practice expense RVUs, as work and practice expense RVUs were highly correlated (r_s = 0.93, p < 0.0001).

Conclusions. The group of work RVUs for surgical procedures for congenital heart defects are reasonable relative measures, on average, of physician work for these procedures, thus supporting the use of this scale to determine physician reimbursement. Practice expense RVUs may not be an independent measure for these procedures.

	Introduction

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
On January 1, 1992, the Health Care Financing Administration (HCFA) adopted a fee schedule based, in part, on a Resource-Based Relative Value Scale (RBRVS) to reform physician reimbursement for Medicare patients [1–5]. Since its adoption, use of the scale or its components has become accepted as a standard methodology for determining professional payment [6, 7]. The intention of the RBRVS is to base compensation primarily on the "work" involved in performing various procedures in an equitable fashion across subspecialties. Physician work is incorporated as a work-based relative value scale, where linear weights (RVUs) are assigned to various procedures. In addition, to account for other required expenses, the Medicare fee schedule also includes scales to account for "practice" and "malpractice" expense. Fees are computed by adjusting each scale component by a factor reflecting geographic cost differences and then multiplying by a conversion factor [5].

The dimensions of work included in the work-based scale are (1) time, (2) mental effort and judgment, (3) technical skill and physical effort, and (4) psychological stress [1, 2, 8]. Work RVUs are assigned to physicians’ services in relation to standard reference procedures designed to link the scale across subspecialties, and are assigned according to the American Medical Association’s (AMA’s) Current Procedural Terminology (CPT) codes.

Because the work-based scale was originally designed for reimbursement for Medicare patients, initial versions did not include values for many pediatric procedures. As use of these scales has become increasingly adopted by other insurers, efforts have been made by the American Academy of Pediatrics and multiple subspecialty groups to develop RVUs for pediatric procedures and to develop CPT codes for pediatric care [9]. The assignment of new RVUs for pediatric and other procedures has occurred under the auspices of the AMA Relative Value Update Committee with subsequent approval by HCFA.

For pediatrics, an early effort in 1993–1994 resulted in the assignment and adoption of new work RVUs to a group of 81 congenital heart operations. The assignment was based on newly revised CPT, version 4 codes [10], and was made using standard methodology based on the results of a national survey of pediatric cardiac surgeons. As a part of the survey, surgeons were given case scenarios of "average" patients undergoing congenital heart operations, as well as a previously assigned RVU for standard reference procedures. The respondents were asked to rate the preoperative, intraoperative, and postoperative work entailed to perform each cardiac procedure. For example, if a surgeon thought that a congenital heart operation required 25% more work than the reference procedure, he or she would assign that procedure an RVU that was 25% higher than the reference RVU. In general, the median RVU value for all respondents was then assigned to the new procedure.

Historically, practice expense (PE) RVUs were derived from the differences between expenses attributable to work and "usual and customary charges" based on Medicare historic charge data. For almost all pediatric cardiac surgical procedures, no historic charge data existed. Practice expense RVUs for these procedures were derived by comparison with "analogous" procedures for which values had been assigned, according to agreements between representatives of The Society of Thoracic Surgeons and HCFA.

The validity of the original RBRVS as a measure of work was based primarily on excellent internal consistency among physicians regarding the ordering of procedures with respect to work [1]. Even before its implementation, the validity of the scale and the methodology by which it was created was questioned for adult cardiac surgery [11]. Since its implementation, the need for additional validation has been suggested, but little work has been performed [8, 12–14]. To begin this process, we suggest that valid work RVUs for congenital heart procedures should show reasonable correlation with other measures that reflect physician time and effort, such as resource consumption. Given the nature of congenital heart disease, RVUs for these procedures should also be correlated to some degree with clinical outcomes such as mortality, because more complicated operations are likely to represent higher risk to the patient. Relative value units that show consistently poor correlation with these measures may suggest "invalid" RVU assignments. We have thus measured the correlation between this portion of the work RVU scale and length of hospital stay, total hospital charges, and in-hospital mortality using administrative data from nine states. Although the theoretical basis for the practice expense scale is less well defined, making a priori hypotheses regarding the expected correlation between PE RVUs and resource consumption or clinical outcomes difficult, we have also measured these correlations.

	Material and methods

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Databases
Hospital discharge data from 1992 for patients 18 years of age and younger were obtained from nine states: California, Colorado, Florida, Illinois, Massachusetts, Maryland, Pennsylvania, Washington, and Wisconsin. These states were chosen in part based on the availability of the data in the public domain, but also to provide a wide geographic sample of states known to have centers of various sizes performing surgical procedures for congenital heart disease in children. These administrative databases included information in standard formats for acute care hospital discharges from all institutions with few exceptions (eg, federal facilities such as Veterans Affairs facilities). Data included routine demographic and administrative information such as age, sex, race, admission type and source, discharge status, length of hospitalization, and total hospital charges (excluding professional fees), as well as all discharge diagnoses and procedure codes assigned to the patient, using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes.

Case selection
All cases with ICD-9-CM codes indicating surgical repair of a congenital heart defect (procedure codes 35.xx, 39.0, or 39.21, excluding 35.41, 35.52, and 35.96) or codes 38.35, 38.45, 38.85, or 39.59 and any diagnostic code for a congenital heart defect (745.xx, 746.xx, 747.0x–4x, 424.0–3, excluding 746.86) were selected. Premature infants (diagnostic codes 765.0x–1x) and cardiac transplants (procedure code 37.5) were eliminated. Infants with patent ductus arteriosus (PDA) as an isolated cardiac defect presumably complicating newborn respiratory disease (identified by diagnostic code 747.0 and procedure code 38.85 and age less than 0.3 years) were also eliminated. To exclude transcatheter interventions, we also eliminated patients whose sole cardiac procedure was atrial septal defect closure (diagnosis code 745.5, procedure codes 35.51, 35.71), ventricular septal defect closure (diagnosis code 745.4, procedure codes 35.53, 35.72), PDA closure (diagnosis code 747.0, procedure code 38.85), atrial septectomy (procedure code 35.42), or vessel repair or occlusion (procedure codes 38.85 or 39.59) with concomitant catheterization codes (37.21–23, 88.42–44, 88.50–59) and no code for cardiopulmonary bypass (39.61).

Assignment of CPTs to cases
Because the administrative data contained ICD-9-CM diagnostic and procedure codes, but not CPT codes, computer algorithms were generated to assign CPTs to individual cases. The ICD-9-CM and CPT coding schemes were similar for most common procedures. However, because of coding differences (eg, specification of surgical detail within the CPT scheme, which was not encoded in the ICD-9-CM codes) some cases could not be assigned to CPTs. Only unambiguous assignments were made; cases that could be assigned to more than one CPT were not assigned to any, except that complex procedures with concomitant PDA ligation, atrial septal defect closure, or atrial septectomy were assigned the CPT of the more complex procedure. The appropriate work and PE RVUs were then used for the analysis. In several instances, for similar procedures, mean values for RVUs were assigned. The ICD-9-CM codes used for CPT, work RVU, and PE RVU assignment are shown in Table 1.

	Results

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Descriptive statistics
Within the nine states, a total of 7,301 cases of surgical procedures for congenital heart defects in children were identified. Current procedural terminology codes, and thus work and PE RVUs, were successfully assigned to 5,192 (71.1%) cases involving 34 different surgical procedures, including most common congenital heart operations. Within the entire surgical series, the median length of stay was 7 days (range, 0 to 352 days); the median total hospital charges were $38,970 (range, $961 to $1,720,216); and the overall mortality rate was 6.1%. Summary data for cases that could be successfully matched to work and PE RVUs are shown in Table 2 according to state and in Table 3 by procedure.

View larger version (20K): [in this window] [in a new window]   Fig 1. Relationships between the work relative value weight (RVU) (A) or the practice expense relative value weight (RVU) (B) for each procedure and the median length of stay for that procedure. A listing of the procedure represented by abbreviations is given in Table 3. The symbol rs represents the Spearman correlation coefficient for the association, with its associated p value.

View larger version (22K): [in this window] [in a new window]   Fig 2. Relationships between the work relative value weight (RVU) (A) or the practice expense relative value weight (RVU) (B) for each procedure and the median total hospital charges for that procedure. A listing of the procedure represented by abbreviations is given in Table 3. The symbol rs represents the Spearman correlation coefficient for the association, with its associated p value.

View larger version (19K): [in this window] [in a new window]   Fig 3. Relationships between the work relative value weight (RVU) (A) or the practice expense relative value weight (RVU) (B) for each procedure and the percentage of patients who died during the hospitalization for that procedure. A listing of the procedure represented by abbreviations is given in Table 3. The symbol rs represents the Spearman correlation coefficient for the association, with its associated p value.

	Comment

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Mechanisms for determining physician reimbursement have received considerable scrutiny during the current health care reform debate. Even health care reform systems based on capitation require methods to divide capitation fees among various health care providers. The decision by HCFA to adopt a resource-based relative value scale—including a multidimensional measure of physician work—as a primary determinant of physician fees for Medicare patients, and the subsequent adoption of the scale by other payors, suggests the appropriateness of basing reimbursement on the time, effort, training, and skill required to perform various procedures. However, development of valid measures of relatively subjective constructs such as physician work, particularly with the potential for self-interest given the intended use of the scale to set individual income, requires considerable effort.

Our findings suggest that the work RVU scale for pediatric cardiac surgical procedures has reasonable internal consistency as a measure of physician work. In a large, population-based analysis using 1992 data from nine states, a substantial portion of the scale showed high correlation with other measures that are likely to reflect physician time and effort, eg, length of hospital stay and total hospital charges, and moderate correlation with an outcome that is likely to reflect "complexity," eg, in-hospital mortality. Surgical procedures that were assigned higher RVUs resulted in longer lengths of hospital stay, higher total hospital charges, and higher in-hospital death rates than cases with lower RVUs.

This study does not address the important question as to whether the magnitude of pediatric cardiac surgical work RVUs are consistent with the magnitude of RVUs for other types of procedures, and thus does not affirm the entire process by which RVUs are currently assigned. Nevertheless, our work does suggest that the survey technique used to assign this group of RVUs, which was among the first to begin the important task of assigning values for pediatrics, resulted in a scale with reasonable, independent validity, despite the considerable differences between procedures performed in adults and children.

Several additional observations are warranted. First, techniques similar to the ones we used might be useful to continue the important task of validation of the many additional RVUs that have been assigned. Clearly, correlation with the measures that we used, length of hospital stay, total hospital charges, and in-hospital mortality, will not be appropriate for many RVUs. Other independent measures that are likely to similarly reflect specific types of physician effort, training, and skill will need to be used for this purpose.

Second, for this group of codes, based on our results, some refinements in work RVU assignment may need to be considered. For example, the three procedures for aortic valve replacement—replacement with a prosthetic valve (RVU 29.17), replacement with a homograft valve (RVU 32.00), or replacement with a pulmonary autograft (RVU 35.00)—had moderately disparate RVUs, but very similar median length of stay (range, 7 to 9 days) and median charges (range, $41,308 to $50,290). Also, three procedures that are technically fairly simple—systemic to pulmonary artery shunt (RVU 20.94), atrial septectomy (RVU 21.49), and Glenn shunt or cavopulmonary anastomosis (RVU 23.00)—but that are usually performed in infants or young children with complex disease, were all assigned very low RVUs. Yet all three had median lengths of stay, median charges, and in-hospital mortality rates that were quite high in comparison. In both of these examples, the discrepancies noted may reflect ways that resource consumption and patient outcomes fail to accurately reflect physician work (in particular, technical difficulty or other intraservice aspects of a surgical procedure), and thus reflect the incompleteness of our measures. For example, total operative time, cardiopulmonary bypass time, and aortic cross-clamp time were not incorporated in these measures, but may be different among the three aortic valve operations. However, it is also possible that the surveyed surgeons had a tendency to overemphasize technical considerations, in comparison with other dimensions of work. If this is the case, then the discrepancies may suggest procedures for which the scale is less valid, which could result in overreimbursement or underreimbursement for actual work performed.

Finally, we have provided data showing the correlation between PE RVUs and resource consumption and mortality. Not surprisingly, given the very high correlation between PE and work RVUs for these procedures, correlations were similar to those for the work scale. In general, however, the correlations were lower for all measures. Given the relatively poor definition of the construct that PE RVUs are designed to reflect and the requirement that codes be assigned according to analogous codes, it is difficult to draw firm conclusions based on these findings as to whether the PE RVUs are behaving as intended in this dataset. Nevertheless, the similarities with work RVUs are striking. It is not clear from our data whether PE and work RVUs are independent measures for these procedures.

Despite these observations, we believe that our findings suggest that the group of work RVUs for surgical procedures for congenital heart disease are reasonable relative measures, on average, of physician work for these procedures and thus support the appropriateness of use of this scale for physician reimbursement. Continued efforts will be necessary across many subspecialties to determine the general validity of the work RVU scale in its entirety and the methodology with which it was created, including cross-specialty linkages, but our results are quite encouraging.

	Acknowledgments

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 
This research was supported in part by a Clinical Investigator Development Award (#HL02936-01), NHLBI (K.J.J.) and by the Kobren Fund (K.G.).

	References

Top Abstract Introduction Material and methods Results Comment Acknowledgments References

 

1. Hsiao W.C., Braun P., Yntema D., Becker E.R. Estimating physicians’ work for a resource-based relative value scale. N Engl J Med 1988;319:835-841.[Abstract]
2. Hsiao WC, Braun P, Becker ER, et al. A national study of resource-based relative value scales for physician services: final report to the Health Care Financing Administration. Harvard School of Public Health, 1988 (Publication no. 18-C-98795/1-03).
3. Medicare program: model fee schedule for physician services; notice. 55 Federal Register 36177–36245.
4. Medicare program: fee schedule for physician services. 56 Federal Register 25792–25883.
5. Medicare program: revisions to payment policies and adjustments to the relative value units under the physician fee schedule for calendar year 1996. 60 Federal Register 63124–63127.
6. Blue Cross and Blue Shield of Massachusetts I: changes in reimbursement policy and reporting structure. 1995.
7. Medicaid RBRVS payment system begins April 1. Tex Med 1992;88:46.
8. Becker E.R., Dunn D., Braun P., Hsiao W.C. Refinement and expansion of the Harvard resource-based relative value scale: the second phase. Am J Public Health 1990;80:799-803.[Abstract/Free Full Text]
9. AAP Division of Physician Payment Systems. AAP pushes pediatric involvement in RBRVS. Am Acad Pediatr News 1996;12:24.
10. Kirschner C.G., Davis S.J., Duffy C., et al. Physician’s current procedural terminology. Chicago: American Medical Association, 1998.
11. Noether M. Results from the Cardiothoracic and Vascular Surgery Work Survey by Abt Associates Inc. Ann Thorac Surg 1991;52:403-407.[Abstract]
12. Morton S.C., Kominski G.F., Kahan J.P. An examination of the resource-based relative value scale cross-specialty linkage method. Med Care 1994;32:25-39.[Medline]
13. Kahan J.P., Morton S.C., Farris H.H., Kominski G.F., Donovan A.J. Panel processes for revising relative values of physician work. A pilot study. Med Care 1994;32:1069-1085.[Medline]
14. Feldman R., Hillson S.D., Wingert T.D. Measuring the dimensions of physician work. Med Care 1994;32:943-957.[Medline]

This article has been cited by other articles:

				O. J. Benavidez, K. Gauvreau, P. D. Nido, E. Bacha, and K. J. Jenkins Complications and Risk Factors for Mortality During Congenital Heart Surgery Admissions Ann. Thorac. Surg., July 1, 2007; 84(1): 147 - 155. [Abstract] [Full Text] [PDF]

				J. A. Connor, K. Gauvreau, and K. J. Jenkins Factors Associated With Increased Resource Utilization for Congenital Heart Disease Pediatrics, September 1, 2005; 116(3): 689 - 695. [Abstract] [Full Text] [PDF]

				J. P. Jacobs, F. G. Lacour-Gayet, M. L. Jacobs, D. R. Clarke, C. I. Tchervenkov, J. W. Gaynor, T. L. Spray, B. Maruszewski, G. Stellin, J. Gould, et al. Initial Application in The STS Congenital Database of Complexity Adjustment to Evaluate Surgical Case Mix and Results Ann. Thorac. Surg., May 1, 2005; 79(5): 1635 - 1649. [Abstract] [Full Text] [PDF]

				J. P. Jacobs, C. Mavroudis, M. L. Jacobs, F. G. Lacour-Gayet, C. I. Tchervenkov, J. W. Gaynor, D. R. Clarke, T. L. Spray, B. Maruszewski, G. Stellin, et al. Lessons learned from the data analysis of the second harvest (1998-2001) of the Society of Thoracic Surgeons (STS) Congenital Heart Surgery Database Eur. J. Cardiothorac. Surg., July 1, 2004; 26(1): 18 - 37. [Abstract] [Full Text] [PDF]

				C. Mavroudis and J. P. Jacobs Congenital heart disease outcome analysis: Methodology and rationale J. Thorac. Cardiovasc. Surg., January 1, 2002; 123(1): 6 - 7. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): John E. Mayer Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jenkins, K. J. Articles by Mayer, J. E. Search for Related Content PubMed PubMed Citation Articles by Jenkins, K. J. Articles by Mayer, J. E.