 |
|
HEALTH CARE REFORM
Is It Time to Eliminate Consultation Codes?An Analysis of Impact and Rationale
Joel I. Shalowitz, MD, MBA
Arch Intern Med. 2010;170(1):(doi:10.1001/archinternmed.2009.446).
ABSTRACT
| |
Background As issues of health care cost escalation and parity of payment between primary care and other physicians have become more important, one proposal has been to eliminate consultation codes. Little is known about the current payment accuracy or financial impact of such a change.
Methods To assess the impact of consultation code elimination, 2 assessments were conducted. First, from June 1, 2008, to July 1, 2009, 500 consecutive referrals from primary care physicians to other specialists were reviewed and matched with claims for accuracy of coding and billing. Second, to evaluate the financial impact of this change, year 2007 data on outpatient consultations from the Centers for Medicare and Medicaid Services were reviewed.
Results Of the 500 claims reviewed, 466 were appropriate for analysis. Overall, the coding error rate was 32.4%. When the requesting physician ordered a consultation, the error rate was 5.5%; however, with lower paid referral requests, the error rate was 78.0%. Changing ambulatory consultation codes to those for new patient visits would save Medicare $534.5 million per year.
Conclusions Consultation codes are being billed erroneously at a high rate. Furthermore, the differential cost to Medicare of these codes over those for new patient evaluation and management codes is over half a billion dollars per year. With the growing needs for cost savings as well as encouraging payment parity for cognitive services for primary care physicians, it is time these codes are reevaluated.
INTRODUCTION
In 1987, Hsiao and colleagues1 published their landmark article explaining the Resource-Based Relative Value Scale (RBRVS). This new system promised to remedy the payment inequities between cognitive and procedural specialists and set fees based on resource use rather than arbitrary physician-determined charges. While intended to replace a compensation system for Medicare Part B providers, just 2 years after its implementation it was quickly becoming the benchmark for most private and public physician payments.2 Two decades later, payment parity between primary care physicians and other specialties has not occurred, and in many cases, is worse.3 Furthermore, because of rising Part B expenditures, the Centers for Medicare and Medicaid Services (CMS) have recommended across-the-board cuts, which would further exacerbate the problem. Although a 1.1% increase (rather than a 20% decrease) was applied in 2009, it was another stopgap measure to delay significant payment decreases mandated by current law.4
One of the sources of inequities (and confusion) that has received attention is the difference between payments for consultation codes and new patient visits. The 2000 version of Current Procedural Terminology (CPT)5 clarified the requirements for a consultation. The specialist must receive a written or verbal request from another health care provider to see a patient. The request must be documented in the patient's medical record by both the requestor and the consultant. While the purpose of the consultation is to render an opinion, the consultant may initiate diagnostic and/or therapeutic services at an initial or subsequent visit. Furthermore, the consultant may assume care of the patient as a result of the initial consultation. The patient need not be new to the specialist for a consultation code to be billed if it has been more than 3 years since the last visit for any reason or the patient was sent for evaluation of a new problem, regardless of the intervening time. Until recently, a formal written report of the findings needed to be sent to the referring source. At present, any follow-up written communication, such as a copy of the consultant's patient encounter record, will suffice.
If, however, the patient self-refers to a specialist, that health care provider must code a new patient visit. Also, if a health care provider has a patient with a known diagnosis and sends that patient to a specialist for the purpose of turning over care, the visit is considered a referral and not a consultation; a new patient visit should be billed for this type of encounter as well.
According to the CMS's most recent Comprehensive Error Rate Testing program report (November 2007), the following 3 of the consultation codes were among the top 10 improperly paid: 99245 (19.1%), 99244 (17.5%), and 9243 (9.6%).6 While this CMS review was conducted to assess the appropriateness of the coding level, the reasons for consultation have not been recently evaluated for accuracy.
Furthermore, although CMS recently issued recommendations to eliminate consultation codes, there is no recent published analysis of the scope of this issue or the financial implications of this reform proposal. The objectives of this article are to describe the reasons for these codes, analyze the accuracy of their use, project the financial repercussions of consultation code elimination, and suggest recommendations about their continuance.
METHODS
The first analysis was conducted to assess the accuracy of consultation coding. Data were gathered prospectively by the author over a 14-month period (June 2008 to July 2009) from 500 consecutive written requests for ambulatory consultation. These requests were made using a standardized form and originated from a primary care group of 20 physicians (11 internists and 9 pediatricians). The data on the form included the reason for the request, whether the patient was new to the specialist, and whether the request was for a consultation or the patient was being referred for the specialist to assume care. The written reason for the patient to see the specialist was matched to the bill submitted by that physician for those specified services. Criteria for classification were those previously described. In compliance with the Health Insurance Portability and Accountability Act, patient-specific data were excluded from analysis. The eligible population from which these requests were drawn was approximately 4000 patients enrolled in 1 of 3, non-Medicare capitated plans. Excluded from analysis were requests in which the reason was unclear. Furthermore, ophthalmology visits were excluded because of the availability to use unique evaluation and management (E&M) codes for that specialty, ie, of the form 92xxx, as opposed to 99xxx.5
In a second analysis, 2007 CMS Medical Provider Analysis and Review (MEDPAR)7 data covering frequency of coding and charges were used to calculate the effect of eliminating outpatient consultation codes and replacing them with new patient visit codes. (MEPAR data did not allow assessment of inpatient consultation codes.) The calculation of this effect, as displayed in Table 1, was performed with the use of 2007 data as follows:
|
|
|
|
Table 1. Cost Differences Between Outpatient Consultation (OC) and New Patient (NP) Codesa
|
|
|
(Total Charges for Consultation Code, $/Number of Charges for That Code) – (Total Charges for E&M Code at the Same Level as the Consultation Code, $/Number of Charges for That Code).
To calculate the total financial impact of these charges, the differences for each service level were multiplied by the number of consultations at that level and then summed.
RESULTS
The results of the first analysis are listed in Table 2. Of the 500 claims reviewed, 466 were appropriate for analysis. While the error rate for both consultations and referrals was 32.4%, the results indicate that the higher-paid consultation code requests are billed more inaccurately. Although this analysis may have sampling limitations because of group size and geographic location (north and northwest suburbs of Chicago, Illinois), it nevertheless represents the private insurance impact of primary care referrals. Furthermore, there is no ethical reason to assume that consultant billing patterns differ for Medicare and private pay patients.
|
|
|
|
Table 2. Accuracy of Billing for Outpatient Consultation Codes
|
|
|
An analysis of the cost impact of eliminating these codes is found in Table 1. When 2007 CMS volume data are used, if all new ambulatory consultation codes were paid at the same rate as new primary care visits, eg, CPT code 99242 visits would be paid the same as CPT code 99202 visits, Medicare would save approximately $534.5 million annually. While this amount is a small percentage of the total $58.7 billion Medicare spent in 2007 for "physician fee schedule services,"8 the health care system savings would be much larger if private insurance also adopted this measure. Similarly, further savings might be achieved with the elimination of inpatient consultation codes.
COMMENT
The aforementioned data demonstrate both a high billing error rate for referrals as well as potentially large financial consequences of eliminating consultation codes. One can speculate that this error rate is due to physicians' disagreement with the CMS definition of consultation and/or the fact that these codes have a higher reimbursement rate. Nevertheless, given that any proposed changes will be opposed by those most financially affected, a discussion of their rationale and current status will illuminate the discussions about change.
The establishment and persistence of ambulatory consultation codes are due to 4 factors. First, the original RBRVS calculation was to include the opportunity costs of additional specialty training. When this portion was eliminated (in favor of a component to account for malpractice expense), other methods, such as increased work units, were used to add a specialty differential. Second, after 1992, the CMS set 3 different multipliers for primary care, surgical specialties, and "other nonsurgical" specialties (such as radiology), thus adding another potential source for extra specialty payments. However, this method changed in 1997 with passage of the Balanced Budget Act; starting in 1998, a single conversion factor applied to all specialties. Third, work relative value units (RVUs) are reevaluated every 5 years by the Relative Value Scale Update Committee (RUC) of the American Medical Association (AMA). The 29 member committee includes 23 members nominated by specialty societies. As stated by Bodenheimer et al,3(p303) "Although primary care physicians provide about half of Medicare physician visits, primary care makes up only 15% of RUC's voting members." Because the CMS accepts 95% of RUC recommendations,9 it should therefore come as no surprise that while many nonprimary care specialty RVU changes have been made over the past decade, "office visit RVUs did not increase from 1995 to 2005."3 The higher consultation code payments are, indeed, due to the higher work RVUs assigned to those services. As Betsy Nicoletti, MS, CPC, head of Medical Practice Consulting commented
In its proposal, CMS notes that the AMA's CPT staff and CMS policy staff have disagreed for years on the distinction between when it's a consult and when it is a transfer of care. The agency blames the AMA's lack of clarity in its consult definition as a cause. . . .10
Finally, the extra payment for consultation codes was justified by a requirement that the consulting physician provide a formal written letter to the health care provider who referred the patient. As previously mentioned, this requirement no longer exists.
One measure that would start to remedy the payment differences between primary care physicians and other specialists is the "America's Affordable Health Choices Act of 2009,"11 introduced in the House of Representatives on July 14, 2009. The Act proposes to replace the 21.5% physician payment cut scheduled for 2010 with a 1% pay boost as a transitional update. It further proposed that, starting in 2011, the single fee schedule conversion factor would be replaced by 2 conversion factors for different categories of services: E&M and preventive services would be in one category and all other services (regardless of specialty) would be in the other one. The first category would have an annual target growth rate of 2%, while the latter group would be subject to an annual target growth rate of 1%. Payment formulas would be adjusted annually for each group depending on each one's expenditures compared with target growth rate estimates.
In a separate action to achieve cost savings, on July 1, 2009, CMS stated that it is
. . . proposing to stop making payment for consultation codes. . . . Practitioners will use existing E&M service codes when providing these services instead. Resulting savings would be redistributed to increase payments for the existing E&M services.12
Such changes apply to both inpatient and office/outpatient services beginning January 1, 2010, with the only exception being limited telehealth services. The proposed savings are supposed to be budget neutral by increasing the work RVUs
for new and established office visits by approximately 6% to reflect the elimination of the office consultation codes in the work RVUs for initial hospital and facility visits by approximately 2% to reflect the elimination of the facility consultation codes.13
Since both primary care and nonprimary care physicians use the same E&M codes and because the MEDPAR data (referenced in Table 1) does not differentiate the volume of services provided by specialty, how much of this redistributed income will go to primary care physicians is unclear. If such reallocation is the goal, CMS could link the physician's specialty to the national provider identifier or add a primary code to the CMS 1500 form. (CMS accepted comments on the proposed rule until August 31, 2009, and will respond to all comments in a final rule to be issued by November 1, 2009.)
As this country debates how to supply health care to all, we face an increasing primary care shortage.14 If we expand coverage, this shortage will worsen. For example, after universal coverage was implemented,
(t)he wait to see primary care doctors in Massachusetts has grown to as long as 100 days, while the number of practices accepting new patients has dipped in the past four years, with care the scarcest in some rural areas. . . . 15
Higher payment for consultation codes, while not adding a significant percentage to the overall Medicare budget, sends a signal that primary care cognitive services are not valued equally with such services provided by other specialties. At a time when we want to encourage new physicians to consider primary care and support current practitioners, this differential sends a dissonant message. Furthermore, as patients are increasingly responsible for out-of-pocket payments, it is difficult to explain to them why consultant physicians are paid so much more than their primary care physicians for the same or less time spent with them. To optimize payment for E&M work and encourage the growth of primary care, it is time for the reconsideration of consultation codes.
AUTHOR INFORMATION
Correspondence: Joel I. Shalowitz, MD, MBA, Health Industry Management, Kellogg School of Management, Northwestern University, 2001 Sheridan Rd, Evanston, IL 60208-2007 (j-shalowitz{at}kellogg.northwestern.edu).
Accepted for Publication: October 12, 2009.
Published Online: November 9, 2009 (doi:10.1001/archinternmed.2009.446). This article was corrected on November 19, 2009.
Financial Disclosure: None reported.
Author Affiliation: Health Industry Management, Kellogg School of Management, Northwestern University, Evanston, Illinois.
REFERENCES
| |
1. Hsiao WC, Braun P, Becker ER, Thomas SR. The Resource-Based Relative Value Scale: toward the development of an alternative physician payment system. JAMA. 1987;258(6):799-802.
FREE FULL TEXT
2. McCormack L, Burge RT, Ammering CJ, Mitchell JB. The diffusion of RBRVS [abstract]. AHSR FHSR Annu Meet Abstr Book. 1994;11:130-131.
3. Bodenheimer T, Berenson RA, Rudolf P. The primary care-specialty income gap: why it matters. Ann Intern Med. 2007;146(4):301-306.
FREE FULL TEXT
4. US Government Accountability Office. Medicare Physician Services: Utilization Trends Indicate Sustained Beneficiary Access with High and Growing Levels of Service in Some Areas of the Nation. Washington, DC: US Government Accountability Office; August 2009. Publication No. GAO-09-559.5. American Medical Association. Current Procedural Terminology: CPT 2000. Chicago, IL: American Medical Association; 1999.6. DecisionHealth. Benchmark of the week: Top 10 incorrectly coded services, May 2008. Part B News. May 26, 2008;22(21):5.
7. Centers for Medicare & Medicaid Services. Medicare Provider Analysis and Review (MEDPAR) File. http://www.cms.hhs.gov/IdentifiableDataFiles/05_MedicareProviderAnalysisandReviewFile.asp. Accessed October 20, 2009.8. The Boards of Trustees, Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. 2008 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. Washington, DC: US Department of Health & Human Services, Centers for Medicare & Medicaid Services; 2008.9. American Medical Association. RBRVS: Resource-Based Relative Value Scale. http://www.ama-assn.org/go/rbrvs. Accessed August 2009.10. DecisionHealth. Proposal to dump consults linked to years of confusion, eased documentation rules. Part B News. July 13, 2009;23(27):1.
11. Committee on Energy and Commerce Web site. http://energycommerce.house.gov. Accessed October 20, 2009.12. Centers for Medicare and Medicaid Services Web site. 2009. http://www.CMS.gov. Accessed August 2009.13. Medicare program: payment policies under the physician fee schedule and other revisions to part b (CY 2010). Fed Regist. 2009;749(132):33520-33825. http://www.regulations.gov/search/Regs/home.html#documentDetail?R=09000064809f0082. Document ID: CMS-2009-0058-0003. Accessed August 2009.
14. Kuehn BM. Reports warn of primary care shortages. JAMA. 2008;300(16):1872, 1874-1875.
PUBMED
15. Kowalczyk L. Across Mass, wait to see doctors grows; access to care, insurance law cited for delays. Boston Globe. September 22, 2008. http://www.boston.com/news/local/massachusetts/articles/2008/09/22/across_mass_wait_to_see_doctors_grows/. Accessed October 29, 2009.
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
|
