Potential conflict of interest: Nothing to report.
As a more consultative but less procedurally oriented specialty, Hepatology has been considered a financial liability in some academic centers. However, no actual data exist on the relative contribution of a Hepatology practice. The purpose of this study was to evaluate the direct and indirect (i.e., downstream effect) charges generated by a Hepatology section in comparison with a Gastroenterology section. Using a computerized database, retrospective cohorts of new outpatient consultations and initial admissions seen by the Hepatology and Gastroenterology sections over a 3-month period were created. The cohorts were followed for 12 months. Charges generated directly to the section (direct charges) and to the hospital system (indirect charges) were calculated. Each cohort consisted of 179 patients. The Hepatology patients generated $5,851,463 in overall charges for the hospital, compared with $2,273,339 for the Gastroenterology cohort. Only 3.6% of the Hepatology charges were direct, compared with 15.9% of the Gastroenterology charges. For every $1 billed by Hepatology, the hospital system generated an additional $26.95 in charges ($51.03 for the orthotopic liver transplantation patients, and $14.26 for the non–orthotopic liver transplantation patients). For every $1 billed by Gastroenterology, the hospital system generated an additional $5.31 in charges. Similar inpatient collection rates were seen between the two groups (27.7% for hepatology and 33.6% for gastroenterology). In conclusion, although Hepatology generates only a small amount of direct charges, it accounts for a very substantial amount of indirect or downstream billing for an academic medical center. This study validates the importance of a hospital's support for a Hepatology section, especially in a center performing orthotopic liver transplantation. (HEPATOLOGY 2005;41:968–975.)
A financial discrepancy has existed between procedural and nonprocedural fields of medical care in large academic institutions. Procedural fields have generated more direct income for an institution and have often been rewarded with larger budgets and benefits. Medical fields not heavily based in procedures have generally been reimbursed at a lower level. The extent to which nonprocedural fields are recognized for their contributions varies widely. Although less direct revenue may be generated, these specialties may be a substantial source of downstream (i.e., indirect) revenue for an institution on the basis of patient admissions and additional ancillary services including radiology, laboratory, and pathology. In addition, consultations generated by these specialties may result in the performance of procedures that create additional revenue.
These concepts were originally explored by Schneeweiss et al. in 1989.1 They reviewed the contributions made by the family practice clinic at the University of Washington. They found that for every $1 billed by the family practice section, an additional $6.40 in charges was generated by the hospital system. On an individual physician basis, each primary care physician annually generated more than $1,000,000 in charges for the hospital and the subspecialists. In a similar analysis at Oregon Health Sciences University, it was shown that for every $1 billed by the primary care clinics, an additional $7.23 in charges was generated by the hospital system.2 Results of these analyses suggested that supporting a primary care clinic is financially beneficial for the hospital system. Despite a growing emphasis on the profitability of individual departments and individual physician productivity, little research since these studies has tackled the issue of quantifying downstream financial contributions.
Hepatology is a subsubspecialty of internal medicine. Although a certificate of qualification for transplant hepatology by the American Board of Internal Medicine is expected to be implemented in 2006,3 currently no separate board certification examinations are available for hepatology. However, most academic centers require additional training in liver diseases for hepatology faculty. In addition, liver transplant centers require at least one hepatologist to meet criteria for the United Network for Organ Sharing certification.4 The actual practice of hepatology and the billing structure in academic centers is extremely variable. There are centers with distinct sections of hepatology, others where hepatology is merged with gastroenterology, and still others where hepatology is under the auspices of transplantation surgery.
Most clinical hepatologists are board certified in gastroenterology and perform endoscopic procedures, but generally do so to a lesser extent than their gastroenterology colleagues. For this reason, hospital systems have considered hepatologists analogous to the nonprocedural physicians who spend more time on the cognitive aspects of medicine. Thus, hepatology can be viewed as a financial liability for an academic medical center because of the lower direct revenue.
The burden of liver disease in the United States, especially related to hepatitis C, is expected to increase over the next 20 years. Projections have estimated that hepatitis C virus–related mortality will double or triple over the next 2 decades.5–10 Although the National Institutes of Health funding for liver disease research has tripled over the last decade,11 there has been no obvious increase in funding for clinical hepatology. Some institutions have substantially reduced funding for hepatology practices as a way to reduce expenditures without fully considering the indirect financial effect to the institution.
Articles have been published on the general aspects of economics relating to hepatologists,12–14 but no data exist that specifically address the financial impact of clinical hepatology. The main objective of this study was to examine the financial impact of a clinical hepatology practice relative to a gastroenterology practice on a large academic medical center. Both the direct and indirect (downstream) charges from each practice were examined.
RUMC, Rush University Medical Center; OLT, orthotopic liver transplantation; FTE, full-time equivalent.
Patients and Methods
Rush University Medical Center (RUMC) is an 824-bed tertiary academic medical center in Chicago, Illinois. RUMC was the largest adult liver transplantation program in Chicago in 2003 with 97 adult orthotopic liver transplantations (OLTs) performed.
Hepatology is a separate section in the Department of Internal Medicine at RUMC. During the study period, the hepatology section consisted of three full-time equivalent (FTE) positions (two full-time physicians and two half-time physicians). All of the physicians performed liver biopsy and limited endoscopy (upper and lower endoscopies, but no endoscopic retrograde cholangiopancreatography). The gastroenterology section consisted of 7.5 FTE persons, all of whom performed endoscopy (ranging from upper and lower endoscopy to endoscopic retrograde cholangiopancreatography and endoscopic ultrasound) but no liver biopsies.
The hospital's computerized database was used for compilation of the two retrospective cohorts. The 3-month study entry period was from January 1, 2003, to March 31, 2003. Patients were followed for 12 months from their initial date of entry. All new patients seen by the hepatology section during the study period were entered into the study database. An equal number of consecutive patients meeting the entry criteria seen by the gastroenterology section during the study period were used as the comparison cohort. New patients were classified into one of three mutually exclusive types, including initial outpatient consultations (sent by a referring physician), initial outpatient visits (self-referrals), and initial admissions or transfers to the inpatient hepatology or gastroenterology service, respectively.
Inpatient and outpatient charges were collected for the cohorts during the 12-month study period. Direct charges included inpatient physician admission charges, daily inpatient visit charges, inpatient professional procedure charges, outpatient visit charges, and outpatient professional procedure charges. Indirect charges included all hospital-related billing defined as inpatient hospitalization charges, ancillary charges (including radiology, laboratory, and pathology), and procedural facility charges.
New inpatient consultations were excluded from entry into the study because it would have been impossible to accurately determine the relative contribution of the hepatologist or gastroenterologist versus that of the primary physician. Charges generated from the cohort not related to hepatological or gastroenterological issues were excluded. Patients entered into research studies were excluded because it was not possible to track exact charges, and these patients often received additional (non–standard of care) work-ups related to the requirements of their particular protocol. Admissions, evaluations, ancillary testing, and procedures performed at facilities other than RUMC were not included in this evaluation, even if ordered by an RUMC physician. Charges for outpatient medications were not included in this study.
Actual inpatient hospital charges were obtained from the computerized hospital financial database for each admission. Inpatient and outpatient professional charges for hepatology and gastroenterology (including physician visits and procedural charges) were obtained and standardized at current 2004 billing charges. Outpatient hospital and professional charges were also entered manually into the study database using 2004 billing charges. Actual payments were available only for the inpatient hospitalizations per patient in aggregate.
Patients seen for hepatological issues were classified further as OLT or non-OLT patients because of the substantial differences in treatments and resource use.
The protocol for this study was approved by the hospital's institutional review board.
Using the hospital's computer database, 179 new hepatology patients were identified who met entry criteria. Eight patients entered the study as new inpatient admissions, 147 patients entered the study as new outpatient consultations, and 24 patients entered the study as new outpatient visits. A total of 684 patients meeting the entry criteria were seen by the gastroenterology service during the same 3-month period. The first consecutive 179 patients were matched to the hepatology cohort (to include 8 new admissions and 171 new outpatients).
The total charges generated for the entire cohort are outlined in Table 1. The total charges for the hepatology cohort were $5,851,463, or $32,690 per patient (median, $2,718; range, $370–$783,080). The total charges for the gastroenterology cohort were $2,273,339, or $12,700 per patient (median, $4,649; range, $255–$296,645).
Table 1. Total Charges, Hepatology and Gastroenterology Cases, 2003-2004
All Hepatology (n = 179)
OLT Hepatology (n = 9)
Non-OLT Hepatology (n = 170)
Gastroenterology (n = 179)
Total charges per patient, $
Aggregate charges for cohort, $
Professional and visit fee
Professional visit fee
Professional procedure fee
Aggregate direct and indirect charges for cohort
Total charges, $
Direct practice charges, $
Total charges, %
Indirect (downstream) hospital charges, $
Percent of total charges,
Ancillary (laboratory, radiology, and pathology), $
Forty-one of the 179 hepatology patients (22.9%) were admitted during the 12-month study period. Hospital charges for all admissions, excluding hepatology professional fees for visits and procedures, totaled $5,121,092, or $124,905 in inpatient hospital charges per patient admission. The hepatology inpatient professional fees added $67,764, for a total inpatient charge of $5,188,856, or $126,557 in total inpatient charges per patient admission. In comparison, 23 (12.8%) of the 179 gastroenterology patients were admitted to the hospital during the study period. Total inpatient hospital charges for this group were $1,327,372, or $57,712 per admitted patient. The addition of the gastroenterology inpatient professional charges ($46,208) increased the total inpatient charges to $1,373,580, or $59,721 per admitted patient.
The total outpatient charges generated by the hepatology patients were $662,607, or $3,702 per patient. The total outpatient charges for the gastroenterology patients were $899,760, or $5,027 per patient.
Hepatology generated $209,352 in direct charges ($1,170 per patient) to their section, whereas gastroenterology generated $360,290 ($2,013 per patient) to their section. Table 1 and Figs. 1 and 2 show the distribution of the direct and indirect charges.
Nine patients in the hepatology cohort underwent liver transplantation during the study period. Table 1 shows the distribution of charges between the OLT and non-OLT patients. The total charges for the OLT patients were $3,759,849, or 64.3% of the overall hepatology charges. Charges per patient for OLT were more than 30 times the total charges for non-OLT patients ($417,761 per patient for OLT vs. $12,304 per patient for non-OLT).
Figures 3, 4, 5, 6 show the allocation of the relative indirect charges. For every $1 of direct billing by the hepatology section, an additional $26.95 of indirect billing was generated for the hospital system. For the OLT patients, this increased to $51.03 of hospital charges per $1 billed by hepatology. For the non-OLT patients, the hospital was able to generate $14.26 in charges for every $1 billed by hepatology. For every $1 of direct billing by gastroenterology, an additional $5.31 of indirect billing was generated for the hospital system.
For the 41 hospitalized hepatology patients, $1,418,542 (27.7%) of the total inpatient hospital charges of $5,121,092 was collected. For the 23 hospitalized gastroenterology patients, $445,997 (33.6%) of the total hospital charges of $1,327,372 was collected.
From a hospital system standpoint, there is no actual benchmark to define how much a hepatology section needs to generate to be considered sufficiently profitable. Given the concept of earning revenue for the section versus the entire system, a specific dollar amount would be extremely difficult to calculate. Although this is only a descriptive study, it does provide information that is useful to both hepatology practices and academic medical centers.
This study was limited to new patients entering the hospital system as outpatient evaluations or inpatient admissions or transfers. This cohort was chosen to capture a specific patient population that could be distinctly followed. By limiting the study to this specific subset of patients, the results represent only a fraction of the total charges that were generated by the respective hepatology and gastroenterology sections. Charges generated from the cohort were included only if the patients received care directly related to their hepatology or gastroenterology issues. Other charges not evaluated included those generated by work done at other hospitals and clinics in the RUMC system, clinical and basic science research, and outpatient pharmacy.
Charges generated from the care of previously established patients were not included in the study for two reasons. First, the rate of liver transplantation of established outpatients varied over the study period, with a significant increase in late 2003. It was believed that this could sway the statistics unfairly in favor of hepatology. Second, an endoscopic procedure is often performed soon after the initial consultation in gastroenterology patients. Thus, it was believed that previously established gastroenterology patients (most of whom had already undergone endoscopy) would likely contribute minimal revenue to the hospital system and could also influence the results in favor of hepatology.
This study focused on charges generated per patient for all care provided in a 12-month period. Because of the ambiguity in identifying whether downstream charges after new inpatient consultations were attributed to the hepatologist or gastroenterologist versus a primary care physician or other provider, the study did not include new inpatient consultations, and therefore, total charges reported should not be interpreted as total charges for the section. New inpatient consultations would have added a considerable amount of billing charges to the study. Gastroenterology relies on inpatient consultations (often resulting in procedures) to generate a major portion of its revenue. Hepatology also relies on inpatient consultations for the generation of procedures and the introduction of new potential liver transplantation recipients.
In this cohort study, hepatology patients generated 2.6 times greater overall charges than the gastroenterology patients. Most charges from both groups came from inpatient hospital fees (including inpatient facility charges and ancillary charges). Hepatology had a higher percentage of patients admitted (22.9% vs. 12.8%) during the study period, and each admission generated more charges for the hospital system ($124,905 vs. $57,712 per admitted patient).
Compared with gastroenterology patients, hepatology patients generated five times more indirect charges per $1 billed by the section (OLT patients almost 10 times, non-OLT patients almost three times). As outlined in Figs. 3, 4, 5, 6, most of these differences were attributable to greater total inpatient hospitalization charges for the hepatology patients because of both higher rates of hospitalization (22.9% of the patients in hepatology vs. 12.8% in gastroenterology) and greater charges per patient for those who were hospitalized ($124,905 per patient in hepatology vs. $57,712 per patient in gastroenterology). However, even if the inpatient charges were eliminated, hepatology still contributed 52.8% ($2.49 vs. $1.63) more indirect outpatient billing to the system for every $1 of direct charges. Most of this additional outpatient revenue was attributable to a more than five times increase in outpatient laboratory and radiology charges.
Despite the fact that new hepatology patients generated considerably higher indirect revenue for the medical center, the section brought in little direct revenue to itself (see Fig. 7). Only 3.6% of the overall billing qualified as direct charges for hepatology. The gastroenterology section recorded 15.9% of their overall charges as direct charges. The main difference was a substantially larger amount of procedural revenue generated by gastroenterology.
A liver transplantation program requires hepatologists to provide patient care and to fulfill the United Network for Organ Sharing requirements. Our data confirm that transplantation hepatology generates substantial revenue for the hospital system. Even though the transplantation patients made up only 5.0% of the hepatology cohort group, they were responsible for 64.3% of the overall charges. The vast majority of these charges were attributable to hospital facility and ancillary charges, with only 1.9% considered direct hepatology charges.
Whether our data would apply to academic medical centers with less active (or inactive) OLT programs may be questioned. In this regard, if the nine OLT patients were eliminated from the study, the mean charges per patient for the non-OLT hepatology patients ($12,304) were quite similar to the mean charges for the gastroenterology patients ($12,700).
The purpose of the study was to look at the benefits of a hepatology practice to a large academic medical center. The goal was not to compare total production with that of a gastroenterology section, but rather to use gastroenterology as a benchmark of an established, profitable, procedurally-based specialty with which hepatology often is compared. Given the size of the gastroenterology section at our institution (7.5 FTE positions vs. 3 FTE positions), gastroenterology physicians were able to see many more patients and generate substantially more charges than hepatology physicians over the study period. If all new eligible gastroenterology patients had been used for the same study period, a total of 684 (rather than the 179) patients would have been seen for new inpatient or outpatient evaluations. Applying the average charges per patient in the gastroenterology study cohort of $12,700 to the 684 patients seen during the study period, gastroenterology's total charges generated by all eligible patients seen during the study period would have been $8,686,800. Looking at these projected charges on a FTE basis, each gastroenterologist (n = 7.5) would have generated $1,158,260 in charges for the system, whereas each hepatologist (n = 3) would have generated $1,903,462 in charges for the system.
Several potential limitations exist with this study. Given the fact that hepatology practices differ between institutions (i.e., as a separate section, merged with gastroenterology, as part of transplantation surgery, or combinations of these), our data may not be applicable to all hepatology practices.
This study used billed charges rather than revenue collected. One obvious limitation to the study is that charges may not accurately reflect reimbursements. The exact reimbursement rate would depend on the patient population, insurance coverage, and insurance contracts. The exact amount of revenue would obviously be lower than the billing charges. One concern of the study would be that the two cohorts would show dramatically different collection-to-charge ratios. The charge-based results were validated by examining the reimbursement rates based on the inpatient hospitalization charges. In fact, the two sections had relatively comparable payment collection rates of 27.7% for hepatology and 33.6% for gastroenterology.
In conclusion, our data support the financial importance of a hepatology section with regard to revenue production for an academic hospital system. Even though direct charges per patient were considerably higher for gastroenterology, hepatology generated a substantially larger amount of indirect or downstream revenue for the overall hospital system because of increased inpatient hospital stays and laboratory, radiology, and pathology services. Patients undergoing liver transplantation had the largest financial impact on the hospital system. Given the dramatic increases expected in liver disease, cirrhosis, liver cancer, and transplantation over the next 2 decades, our data suggest that a long-term investment in clinical hepatology is potentially advantageous to an academic medical center.
Even though our data clearly suggest that hepatology should be compensated and supported by the institution, where does this money come from? One obvious problem in today's medical environment is the concept of a fixed pool of resources. An academic medical system must weigh its priorities and decide on an appropriate distribution of funds. Unfortunately, no perfect system seems to exist. A team or equal partnership approach can be used with equal distribution of revenue to all departments. This would favor the gatekeepers of the system (such as primary care physicians or hepatologists) who earn more indirect revenue, but would be a disincentive to those who produce more direct revenue, such as surgeons and procedural-based physicians. However, a system that rewards departments based solely on direct revenue would disadvantage the less procedurally-oriented disciplines. The academic system must recognize that profitable ventures such as surgery and endoscopic procedures require interdepartmental collaboration and depend, to a large extent, on referrals from other nonprocedural specialties. As an example demonstrated in Table 1, transplantation patients generate the largest amount of charges for the hospital system, but the least direct revenue to the hepatology section. As established by the United Network for Organ Sharing guidelines, liver transplantation cannot occur without hepatologists. Thus, there must be a system in place to support hepatology and to maintain patient care for liver transplantations. It is beyond the scope of this analysis to suggest an exact system of financial distribution for every academic center. Hopefully, academic hepatologists and the academic medical centers themselves can use the results of this study to negotiate a fair, system-based compensation method. Such compensation may involve funding and support from several sources, including the hospital system administration, the department of medicine, and the department of (transplantation) surgery.