Consider this typical Medicare patient. She is a 72-year-old woman with diabetes, hypertension and coronary artery disease. Her blood pressure is controlled, but her cholesterol and blood sugars have been high since she abandoned her diet because of depression over her daughter’s divorce and the unruly grandchildren who now live with her. She has occasional angina, which has been more frequent in recent weeks, has arthritis in her spine and knees, has recently developed vaginal itching, heartburn and a persistently stuffy nose, and she worries about changes in a spot on her skin. Her problems are all common ones and could be readily managed by one primary care physician – or by visits to seven specialists (endocrinology, cardiology, psychiatry, gynecology, gastroenterology, otolaryngology and dermatology).

Medicare will pay for consults with each of the seven specialists, at $117.70 or $216.27 each, for an aggregate cost to Medicare of about $1,200, plus any lab tests, imaging, allergy testing and mental health counseling. On the other hand, Medicare will pay the primary care physician at most $80.18 to $117.70 (and up to 30 percent less if he or she practices in a rural area) for the hour or more needed to evaluate her chronic problems and new concerns. That physician’s overhead for the visit will be about $125, based on conservatively estimated practice costs of $300,000 per physician working 2,400 hours per year.

Seeing all those specialists is expensive in time and access: The patient may omit services for lack of time or transportation to locations outside her area, or if she has Medicaid the state and federal government may pay multiple times for transportation. She may omit treatment for her depression because of the stigma of visiting a mental health clinic. Some of her tests will be duplicated, because the specialists have different clinics and records. She is likely to give at least two of the specialists an incomplete list of her medications, thus opening the way for prescriptions that produce dangerous interactions. These are exactly the quality problems with the health care system that the Institute of Medicine report documents.¹

Appropriate treatment is not fragmented

In this example, an appropriate evaluation would ultimately reveal that the patient has monilial vaginitis resulting from the high blood sugars; allergic rhinitis caused by allergy to the grandchildren’s dog; a benign seborrheic keratosis; and new-onset gastroesophageal reflux, depression and increased angina, all related to stress. Her primary care physician could take the time to discuss her diet, lifestyle and stress. The physician, who knows the patient, could also discuss the conflicts she is having with her daughter over parenting and the behavior of the grandchildren. He or she could counsel the patient in a way appropriate for her culture and environment, prescribe treatment for the vaginitis, and start her on an over-the-counter H₂-receptor blocker, a nasal steroid and an affordable antidepressant, at the same time adjusting her therapies for heart trouble and diabetes. The patient would leave feeling better able to cope with her stress and with plans for a follow-up appointment.

Efficiency and incentives. A single visit at which multiple problems are addressed is more efficient for patients, families and the health care system (one medical record, one claim to be sent and one claim to be paid). The current evaluation and management (E/M) payment system creates incentives to avoid efficiency. We should not expect the system to fix itself; the specialist-heavy, procedure-heavy U.S. health care system will not voluntarily change, and it is doubtful that the AMA or specialty organizations will endorse system improvements that negatively affect members’ own financial interests. That should not deter us, however. The system must be fixed.

The fix need not be a massive system overhaul; in fact, the system already contains the model needed. The doctor who performs two or more surgical services at the same session is paid for each of them, with a multiple-procedure discount applied. What if we could apply a “multiple-service modifier” to E/M services?

Consider how this might work in the case of the patient I described earlier. Since the same history and examination elements can be counted in multiple organ systems, decision making drives the level of service for each of the patient’s problems. According to the E/M documentation guidelines, for established patients, straightforward decision-making is associated with 99212, low complexity with 99213, moderate with 99214 and high with 99215. In this case, then, the primary care physician is providing, at one visit, the services shown in “Applying the ‘multiple-service discount,’” below.

If a “multiple-service” rule for E/M care worked the same way the multiple-procedure rule works for surgery (100 percent for the highest-value problem, 50 percent value for each additional), payment would be based upon 7.45 relative value units (RVUs). Using the Medicare conversion factor in effect for early 2008, it would come to $283.56. – far less than the $1,200 or more it would cost if those problems were managed by multiple physicians, but still reasonable compensation.

The potential for abuse. To prevent abuse of a multiple-service rule, it might be necessary to allow only one CPT code per organ system or to define which conditions within an organ system count as “separate problems.” For example, a physician evaluating a rash and two skin lesions might not be allowed to count each as a separate service. Physicians would need to develop a consensus about which diagnoses or problems belong to which organ systems (e.g., Is hyperlipidemia endocrine or cardiac, or both?). The ICD-9 system has a well-recognized system of classification that could be used as a basis. Just as the Correct Coding Initiative has edits for procedures that cannot be billed separately, it could include edits for diagnoses, to exclude simultaneous E/M claims for hypertension and coronary artery disease, say, since they are in the same organ system, while allowing simultaneous claims for hypertension and sprained ankle because they’re in separate systems.

Effects on providers. This change would reward physicians for addressing multiple problems concurrently. It would tend to increase reimbursement for primary care physicians relative to other specialists, which would help stem the attrition in primary care. It would provide incentives for all physicians to address multiple problems to the extent their training permits. Even limited specialists would have incentives to address other issues brought up during regular visits, instead of generating referrals with their attendant costs and confusion. There would be an explicit, well-documented mechanism for appropriate payment for greater work.

Effects on physicians’ billing and insurance. Claim-generating software would not require major change, except for applying multiple-service modifiers to multiple E/M services, as is done for procedural services. Coders and internal compliance reviewers could work with current rules. Internal compliance would be simpler and more predictable, since there would usually be a link between a single E/M service and a single diagnosis or problem.

Effects on quality initiatives. Currently, many problems managed by primary care and other physicians are not reflected in billing or claim databases, because the effort of coding and documenting them is not worthwhile. A visit may include diabetes follow-up but be billed as abdominal pain, because that was the “major problem.” Since the secondary diagnoses do not influence reimbursement, there is no incentive to provide full information about them. The one-to-one linkages between organ system problems and E/M services would greatly facilitate quality improvement, audits and data mining.

Effects on patients. This change would rapidly increase patient access, decrease use of emergency departments and improve quality of care, particularly with respect to chronic diseases such as diabetes and heart disease. Better compensation for multiproblem management would permit physicians to take appropriate time with patients or to develop care teams, thus improving communication and attention to psychosocial issues, language issues and cultural issues, which is what patients want and need.

Effects on the system as a whole. Especially if commercial insurers, Medicare and Medicaid all adopt this strategy, the change could have positive effects on the health care system that are widespread and fairly rapid. In the long term, this change would decrease duplication of documentation and testing, prevent medication errors, make more efficient use of health care resources, provide better access for patients and decrease costs.

Implementation

Physicians agree that change in the E/M documentation rules is needed, and this proposal addresses one enormous problem with those rules. Citizens want bold action to improve the health care system. Conservatives should like this solution because it is market-driven. Liberals should like it because it emphasizes reimbursement for chronic disease care, payment for managing psychosocial factors involved in racial disparities, access for the underserved and redressing the disparities in physician incomes. The Centers for Medicare and Medicaid Services should like it because it requires no new programs or change in the law, just a payment decision. The change would improve quality of care and access, while reducing racial disparities. The change would correct, not exacerbate, the differential incomes of physicians. This is one change that would fix a lot of problems.