Nearly 8 years has passed since I had the opportunity, along with Jamie Burns, JD, to write an article in this publication entitled “Price, Volume, and Productivity.”1 The intent of the 1996 article was to elucidate three variables that, with attention and management, could benefit the day-to-day operation of a radiology group ranging from methodologies to calculating capitation to human resource issues. The article was based on experience gained from considerable years of private radiology practice, a myriad of individual practice business consultations, close work with the Radiology Business Management Association (RBMA), and several years of service as the chair of the American College of Radiology (ACR) Task Force on Managed Care. In the latter role, I was fortunate to have the opportunity to work with the most visionary minds in our profession as applied to the business of radiology and fully supported by all the resources needed from the ACR to develop an extensive educational program.

It was at about that same time that the siren song of health care policy lured me from full-time practice to the business world in the form of the development of a radiology benefit management (RBM) company. My challenge as chief medical officer was to assist in the development of a program of utilization management of diagnostic imaging that was scientifically valid, represented best practice, and was not capitation based. The early RBM initiatives were in direct response to increasing pressure from both payor groups and purchasers of health care to lend some rationality to what was perceived as a benefit out of control. Purchasers noted that while sharp increases in imaging cost were evident, there was little or no scientific evidence to indicate, beyond anecdotal reports, that an offsetting benefit could be demonstrated in their overall medical cost or health of the population. An enlightened ACR leadership has recognized the need for such evidence-based information in their creation of the American College of Radiology Imaging Network, ACRIN, a group dedicated to the scientific study and research of new and existing technology.3

Together with our competitors, the total number of lives under some form of radiology management now exceeds 25 million and it is safe to say that radiology benefit management has come of age. For the most part, each of the RBM companies has a greater day-to-day impact on the ordering physician, but imaging providers are also affected by the introduction of these programs. It is with this additional experience that I revisit Price, Volume, and Productivity (with the addition of quality) from the parallax view. As a rendering provider, you may find the following observations helpful, though admittedly  somewhat controversial, in your current operations as well as for future planning.


The discussion of price should be prefaced with a comment on the capability of current payor claims systems. Older or legacy claims systems as well as today’s state-of-the-art systems should not be underestimated in the quality and depth of information that they can provide regarding the micro- and macro-management of the delivery of health care.4 The referenced article from the National Blue-Cross Blue-Shield Association (BCBSA) indicates a year-over-year increase in the cost of diagnostic imaging to approximately between 20% and 30%.4 NIA data show similar increases in the unmanaged setting with wide regional variations. When working with health plans, the application of a utilization and price management program can flatten this increase and in some instances modestly reverse the trend. To date, the primary impact of RBM has been to moderate “volume” and not unit price, the exception being plans that employ “redirection” to lower-cost providers.

Unit price, or reimbursement for similar services, is a complex issue, with considerable regional variation. For purposes of planning, unit price, with few exceptions, is likely to remain stable. In recent years, increases and/or significant decreases in allowable unit price reimbursement have been minimal. (An exception is the potential Medicare decrease of 4.5% in physician reimbursement for 2004.) An important differentiation must, however be made between “allowable” fee schedules and real cost. True cost to payors can be significantly decreased in many communities when payors choose to redirect studies as part of a prior-consultation strategy. With payors taking advantage of competition between freestanding or independent diagnostic testing facilities (IDTFs) and hospital-based delivery, the sanctity of traditional referral patterns is changing. This puts the strictly “hospital-based” radiology groups at a significant disadvantage because while the radiologists may be more than willing to negotiate their fees, they are hamstrung by hospital “technical component” reimbursement rates often negotiated as a percent of charges, sometimes three times as much as an equal-quality IDTF. Unfortunately, that gives the hospital-based radiologist a diminishing margin with which to negotiate. This is compounded by the fact that the traditional 33/66% (professional to technical) split has diminished to ~21% as noted in Figure 1.5.

Therefore, in many communities where the hospital technical rate is as much as triple that of an equal or superior-quality IDTF, any cost benefit that would occur as the result of offers to discount professional component is seriously mitigated. This represents a noteworthy change from the early to mid 90s when many nonhospital facilities were using relatively poor-quality equipment and payors were directing patients to hospital facilities. Note that an additional and notable exception to unit price stability will be the inevitable move toward “graduated payment” for multiple examination encounters.

Graduated payment is the practice of serially discounting rates for contiguous examinations performed during the same encounter. For example, full reimbursement for a chest CT, 50% for an abdomen, and 25% in the case of an additional pelvic CT. This is the paradoxical result of the rapid advances in technology advertised by both vendors and competitive radiology groups. Even the most humble of insurance carriers understand the remarkable increase in throughput afforded by multi-detector scanners. Most readers are, or should be, aware of the resources that large payors expend on independent “technical assessment,” including meta-analysis of the clinical value and appropriate pricing of the deployment of new technology and new uses of existing science. Several large payors are employing this payment logic now and many will follow. Though of modest comfort, this is likely to be limited to the technical component of the imaging charge(s) as there are economies of scale related to resource consumption, but there will still be (in the example cited) three studies to interpret. Figure 2 (page 20) demonstrates the percentage of claims typically submitted as “combination” studies should you choose to calculate the potential extent that deployment of this form of logic may have on your revenue stream.5

If revenue is the equivalent of billing and pricing to the radiologist or facility, then payment is the equivalent of expense to the payor or purchaser. Given the increasingly competitive environment and inevitable changes in reimbursement methodologies, there will be an eventual equilibration and stabilization of unit price and cost. Absent volume increases, that will leave little wiggle room to increase revenue and may result in the temptation to engage in “creative coding.”

Creative coding walks a fine line between abuse and fraud. Rather than pontificate on the ethics of this aspect of revenue enhancement, know that it is transparent to payors. Initiating a simple claims query for instances when MRI of the brain and MRI of the orbits are conjointly billed is routine. Figure 3 (page 21) is an illustration of the typical result of such a query.5

The claims submission experience in Figure 3 is reported as a ratio between claims for MRI of the brain (70551, 70552, 70553) alone and coupled claims for Brain MRI and Orbits (70540, 70542, 70543). Group X bills jointly more than 30% of the time while the average is 5% and the most conservative group is 0%. Since this degree of variation is common and difficult to explain, a call routinely is placed to the outlier groups. In the case of Group X, the practice representative said, “If we comment on the orbits, we will bill the additional code, and if we are not paid, we will not comment on their status.” While billing approaches like this are in the minority, they are more  common than you would imagine, and they put their business and business managers at extreme risk. If this type of abuse is discovered, insurance companies will usually seek financial reconciliation and eject the group from the network, but if fully pursued and fraud is proven, they are subject to a $10,000 fine per event.


The rapid introduction of clinically credible new imaging technology along with new uses of existing technology lend near certainty to increased (and appropriate) utilization across populations. As an imaging provider, you must anticipate the extent of new technology, competition within the profession in your community, and competition from outside the profession as you plan for the future.

The extent of utilization growth is somewhat debatable. Data from the ACR suggests that, with the US population growing at 1% per year, imaging procedures are growing at 4.5% per year, and relative value units at 6% per year.6 These numbers represent a dramatically different picture than those found in the National Imaging Associates (NIA), Hackensack, NJ, data warehouse. NIA routinely collects year-over-year data from our client health plans (managed) as well as potential clients (unmanaged), and a review of the trend in volume of services based on more than 150 million encounters indicates a significantly different trend. NIA data substantiates the Blue-Cross Blue-Shield Association estimate of a 30% trend (increase) in an unmanaged population.4 “Unmanaged” does not necessarily mean a conventional insurance plan or PPO, as most so-called managed care organizations do not yet employ some form of RBM and must be considered unmanaged for the purposes of imaging growth (Figure 4, page 24).

Most payors and purchasers understand, and will accept, that diagnostic imaging cost and utilization will rise but will not tolerate 20% to 30%  increases in cost and utilization. The precipitous rise in diagnostic imaging cost and utilization is, quite simply, no longer “below the radar.” In addition to the BCBSA initiative, public payor initiatives substantiate the higher trends as elucidated in a recent monograph prepared by the Massachusetts Office of Health and Human Services.7 NIA clients expect that with effective management the year-over-year trend will be in the low single digits.

Competition for this increase in volume requires regional strategies. Figure 5 (page 24) demonstrates the distribution of imaging services according to site and profession.

This distribution is relatively consistent at a regional (multi-state) level but can significantly vary within local communities.  Realize that a practice’s share is dependent on all the variables that make up the competition in a local market area. There is a plethora of information and/or consultants that can be of service in this regard. All too often, however, aggressive competition may border on illegal inducements or de facto fee splitting.

Illegal inducements are creative methods to curry favor with potential referring physicians that can be interpreted as buying business. The gift of a plasma screen television for the holidays in exchange for referrals is blatantly wrong but pales by comparison to the recent trend of “trading” “no-charge” interpretations of PCP office-generated films in exchange for “advanced” imaging referrals to a hospital or IDTF.  Good-quality IDTFs represent a reasonable method to meet and beat the competition, but payors are closely monitoring (radiologist) staffing patterns of the facilities.

Staffing an IDTF with a radiologist should not be a controversial issue. In 1998 NIA instituted a privileging program for one of our western Pennsylvania clients. We maintained, with local radiologist support, that if discretionary protocols were to be considered and/or contrast material was likely to be necessary, then a radiologist should be present on-site. However, NIA found that radiologist support was not unanimous across the board and the issue went to the state society. It seemed that running a “trap line” and gaining volume without radiologist expense was more important to some than the need for a radiologist on-site. NIA, on behalf of our clients, continues to disagree. From a professional standpoint, it is difficult to maintain that radiologists should control diagnostic imaging when you do not feel that you have to be present. Consider the enrollee who visits an IDTF on Monday and receives a nonenhanced brain MRI. The interpreting radiologist visits the facility on Tuesday and decides that the enrollee needs an enhanced examination. The patient then receives the most disconcerting call of his or her life to return for more images…and the payor cost doubles. Is this good-quality service? When convenient, take the time to review a few of the contracts held with national insurers. At least two national insurers require a minimum of half-time, on-site presence of a diagnostic imaging specialist, not a PCP that has an office conveniently located next door. How many IDTFs would be estimated to be in material breach?

Combination study protocols represent another questionable method to increase volume. Mentioned earlier in the discussion of cost, these are not legal violations but can border on “abuse.” These are “house rules” that range from “I will never perform an abdomen CT without a pelvis” to “All MRAs require an associated MRI.” While some sincerely believe in the clinical value of their “mandated” approach, it is common to find that no one else in the community feels the same. This is discerned easily during the preconsultation or preauthorization process and is routinely reported to the health plan. In fairness, this is not confined to radiologist imaging. The so-called “drive-by” renal arteriogram done routinely by the occasional cardiologist when performing a cardiac catheterization represents such an approach. It takes but a simple query to review all claims where these two codes appear together. It is usually one or a single group of cardiologists in a community or in a large medical service area. Their rationale can best be left to the reader’s imagination. While on the subject of other specialties performing imaging procedures, no discussion of volume would be complete without at least a mention of “turf” issues.

Turf issues, or nonradiologists performing imaging procedures, is, in my estimation, the single biggest threat to the cohesion of the specialty of diagnostic imaging/angiography and interventional (see Figure 6, page 60). I can categorically state that the erosion of the arm’s-length distance between the request and the provider is bad health care policy and can lead only to increased and inappropriate self-referral and quality compromises that are inherent in such a construct. Is it truly inevitable that orthopedic surgeons will all own in-office MRI units? Must you acquiesce? As a profession, radiologists should consider the benefit of enabling nonradiologist imaging by offering “overreads” versus the near-complete loss of musculoskeletal imaging to another specialty.

Radiologist self-referral is a fact. Call it consultation or caution; a significant number of studies are performed at the written suggestion of a radiologist. For 2 months, NIA added the following question to all advanced imaging reviews: “Is this study being requested at the written suggestion of a radiologist?” The tally was a little over 20%. Some are well intentioned, some can be considered “churning,” and some are indicative of an interpreter that cannot live in a binary world, but all predicate against the allusion that radiologists are not in control of utilization.


This story has not changed much in the past 8 years, regardless of which side of the table you view it. If the erosion of the breadth or scope of diagnostic imaging is the number one challenge to the profession, then human resource issues rank second. We have all seen good groups break apart because of so-called “productivity” arguments. In my opinion, when you start counting RVUs, your group needs serious help. Quite simply, the RVU allocation as it relates to diagnostic imaging is monumentally out of date. How many of you would rather read a brain MRI than a mammogram? Probably most, and in doing so would amass many more RVUs than your partners who are not only reading very difficult cases but putting themselves at extraordinary legal risk. Further, there must be some recognition of the fact that some outstanding radiologists would rather consult with physicians in the reading room than “melt the stacks.” These people are gifts.  They are your goalies. They protect your office so you can melt the stacks and in that regard they are extraordinarily productive. It is certainly difficult to tolerate a laggard or an insecure professional, but my suggestion is that you do not use RVU generation to prove it. You will know when someone is incompatible and you should seek professional help to rectify the problem.  Resources are available in the independent sector as well as from the RBMA and the ACR for help in these matters.


W. Edwards Deming, PhD, an accepted authority on quality, lists 14 points that are essential to quality management.8  Deming’s point number three extols the fact that, in the best of circumstances, quality need not be measured as it should be incorporated in the product. A loose translation would be that in all circumstances, unintended variation is bad quality. Readers may be more familiar with the many articles of John E. Wennberg, MD, MPH, on the Deming principles as applied to medicine, specifically small area variation analysis.9 This concept, perhaps more than any other driving force, has placed RBM in the focus of health care purchasers, payors, and managers. When NIA uses age and sex as an accepted proxy for severity, we find as much as a 3X utilization multiple across similar risk adjusted populations.10 When health care analysts see this degree of variation, their immediate and logical questions are what is the clinical value and to what degree is the use of this service discretionary? Since imaging examinations that employ ionizing radiation are potentially harmful to the population, inappropriate overutilization is poor quality. Beyond this are adverse effects that range from false positives to patient anxiety.  Cross-population quality differentials are not limited to utilization issues alone.   Experience documented by NIA data analysis indicates three additional variables in quality even as applied to appropriate examinations: the technical quality of the study, the quality of the interpretation, and the quality of the dictated report.

Technical quality is easily understood. Compromises in this aspect of imaging are likewise easily discovered at the time of prior consultation. It is common for a requesting physician to opine on the various centers in his or her community. Once validated, these are reported to the health plan and the facility may be subject to termination as a participating provider.

Interpretive ability is similar to technical quality compromises but, in a group setting, more difficult to manage.  Again, the typical requesting physician accessing the NIA call center for a repeat examination will comment on the fact that the initial study was of poor quality. These circumstances are collated, validated, and, if substantiated, are reported to the health plan. The difficulty in this process is that the compromised interpreter may be only one of a larger group of radiologists. Typically, in these instances data will be presented to the group president or chair with a request for a corrective action plan so as not to penalize the entire group.

The quality of the dictated report is seldom considered and sometimes the subject of humor, but is dead serious. All too often a technically excellent study is interpreted by a highly skilled radiologist but in translation to paper becomes clinically confusing or useless. The “hedge” is not the flower of radiology. At best, it can be noncommittal to the point of worthlessness and at worst it may send referring physicians down false and potentially harmful pathways. Question: do you honestly believe that referring physicians are impressed or seriously helped by an exhaustive differential, comments such as “definite possibility,” or “clinical correlation is recommended?” This may at first consideration seem a petty issue, but our NIA physician reviewers receive complaints daily about just such “in the trenches” problems.


Diagnostic imaging and its interventional cohort remains one of the most integral, challenging, and intellectual medical specialties in today’s health care system. Its use is widespread and growing, and because of its value, it is coveted by other specialties. Its rapid changes are inevitably accompanied by clinical misunderstanding resulting in both intentional and unintentional abuse. These factors along with others discussed earlier in this document assure that RBM is a reality that is here to stay. While your ultimate accountability is to your patients, the role of those engaged in RBM is to assist you in performing referring physician consultation, to serve as a unifying force in the profession, and to afford accountability to your health plan payors and purchasers of care.

Thomas G. Dehn, MD, is chief medical officer for National Imaging Associates, Hackensack, NJ, and based in Rancho Cordova, Calif. Since 1995, NIA has managed the use of diagnostic imaging delivered to nearly 14 million lives, and currently processes nearly 200,000 requests for advanced imaging procedures per month. He welcomes constructive criticism.


  1. Dehn T, Burns J. Price, volume and productivity. Decisions in Axis Imaging News. 1996;9(2):49-52,74.
  2. Medicon, now operating as American Imaging Management (AIM).
  3. Hillman B. The American College of Radiology Imaging Network, ACRIN; research educational opportunities for academic radiology. Academic Rad. 2002;5:561-2.
  4. Keeping Medical Technology Affordable. Chicago: Blue-Cross Blue-Shield Association; October 2003.
  5. National Imaging Associates Data Warehouse, Hackensack, NJ. Updated November 4, 2003.
  6. Sunshine J. The radiologist shortage. Paper presented at: 2003 Economics of Diagnostic Imaging Symposium; October 23, 2003; Arlington, Va.
  7. Massachusetts Office of Health and Human Services, Division of Health Care Finance and Policy. Diagnostic imaging: a new cost driver. HEALTHPOINT. Boston: Massachusetts Office of Health and Human Services; April 2003:No. 27.
  8. Deming WE. Out of Crisis. Cambridge, UK: Cambridge University Press; 1982.
  9. Wennberg JE. Small area analysis and the medical care outcome problem. In: Sechrest L, Perrin E, Bunker J, eds. AHCPR Conference Proceedings. Bethesda, Md: Public Health Service; 1990:177-206. AHCPR publication 90-3454.
  10. Daley J. Validity of risk-adjustment methods. In: Risk Adjustment for Measuring Healthcare Outcomes. 2nd ed. Chicago: Healthcare Administration Press; 1997.