John C. Steinmann, D.O.; Charles Edwards II, M.D.; Thomas Eickmann, M.D.; Angela Carlson, MHA
Surgeon ownership in medical device distribution, otherwise commonly referred to as physician owned distribution (POD) is a novel model that has gained considerable popularity in recent years. Best described, this model represents a stocking distributorship model whereby the surgeons invest in and take title to a large inventory of implants and often instruments and contract with hospitals to sell products from that inventory. The model makes effective use of volume purchasing, effective negotiation and competition among manufacturers of like quality products.
Proponents suggest that this model introduces market forces that are largely absent in the traditional commissioned model, leading to substantial reduction in healthcare costs. One published report (Steinmann, Burton, Hopkins, & Skubic, 2009) found that surgeon ownership in medical device distribution led to 36% first year savings for like implants including both spine and total joint devices. This reduction in healthcare costs should help to protect the financial viability of local hospitals resulting in sustained access for patients to many valuable surgical services. Proponents of the model will also argue that surgeons offer far greater value as the distributor than a non-surgeon distributor due to their ability to value new technology and negotiate pricing more effectively.
Those segments of the industry that oppose the proliferation of this model suggest that the model will incentivize overutilization and the use of substandard products. While the issue of product selection will be separately studied, the issue of utilization is a very important concern to society. The performance of surgery on patients who may not meet appropriate indications will lead to patient harm and increased expense to the healthcare system.
There exist strong arguments for and against the potential influence that surgeon ownership in medical device distribution might have on utilization but, to date, no study has investigated this concern. The purpose of this study is to investigate the influence that establishment of a surgeon owned distribution model has on surgical treatment decisions.
Four surgeon owned orthopedic and spine implant distributorship companies that met inclusion criteria were studied. Inclusion criteria included: surgeon equity ownership in a stocking medical device distributorship, minimum of one year of distributorship operations by December 31, 2010, and sufficient and accurate surgical performance data preceding the onset of the distributorship for a minimum of one year. CPT (current procedural terminology) codes were reported by the surgeon’s clinical practice billing department, and accuracy of the data was confirmed by the surgeon. Of the possible ten distributorships considered, only four met
the ownership and operations criteria. Six distributorship were excluded due to distributorship start dates after May 2010, and therefore had insufficient data to be included under the parameters of this study.
Practice volume was measured by the number of new, established, and consultative patient visits (PV) captured for Current Procedural Technology (CPT) codes: 99201 -99205, 99211-99215, 99241-99245. This same data was then gathered prospectively.
Surgical volume (SV) was measured by the volume of primary total knee and hip replacements, and instrumented posterior lumbar surgical cases performed.
Surgical volume includes one
hundred percent of the procedures for the CPT code reported, regardless of the source of implants used (from the surgeon owned distribution company or not). Capturing representative spine data is problematic as most spine CPT codes do not distinguish between the primary surgical code and additional levels. Data for SV was, therefore, limited to only those CPT codes that clearly represent a “decision for surgery”. Surgical case volume CPT codes included:
Instrumented Posterior Lumbar
CPT billing codes were collected for the 12 calendar month interval preceding the initiation of the distributorship and for each 12 calendar month interval thereafter. The twelve calendar months prior to the establishment of the distributorship is defined as the baseline for comparison.
This data was collected as part of each distributorships application to the American Association of Surgeon Distributors (AASD).
Four surgeon owned distribution companies meeting the inclusion criteria completed data on both patient visits and surgical volumes(Table 1). The number of surgeons in the distributorship by year was determined by the date which the surgeon began participating in the distributorship. Surgeons who had at least nine months of participation in a calendar year were included in that calendar year. Surgeons with less than 3 months participation were included in the following year for the purposes of Table 1.
The SV/PV ratio could be viewed as a “decision for surgery” index. Figure 1 identifies that between 2005 and 2010 there was remarkable consistency among surgeons regardless of practice type (group or solo) and a trend identifying a decrease in the likelihood to recommend surgery (44.4/1000 to 40.5/1000). This data suggests that the decision for surgery is independent and not influenced by these surgeons participation in a medical device distributorship.
Inherent in the doctor patient relationship is the ethical covenant that the physician’s decision making is guided solely by what is in the patient’s best interest. Surgeon ownership in medical device distribution introduces the potential conflict of interest that the surgeon might be incentivized to recommend surgery inappropriately due to their potential to financially benefit from the sale of implants through their ownership interest in the distribution company. Even the perception of conflict in interest is potentially damaging to the doctor patient relationship. This study is very important as no published study to date has investigated the influence of surgeon ownership of medical device distribution on actual physician practice patterns and specifically, the surgeon’s decision to treat surgically.
Figure 1 identifies that surgical volume remained consistent across all distributorships and varied only with changes in patient volume. Surgical volume and patient volume increased as more surgeons are added to the study group yet the ratio of surgical procedures to patient volume remained stable.
Although general comparison can be made to previously published data, substantive comparison is limited by differences in data source and sample size. Previous reports are based on hospital claims data (Deyo, Mirza, Martin, Kreuter, Goodman, & Jarvik, 2010; 303(13)) (Weinstein, Lurie, Olson, Bronner, & Fisher, 2006 ) and most are limited to Medicare enrollee information.
This study utilized professional procedure codes rather than hospital claims. Furthermore, the patient data included all payor types. Studies restricted to Medicare will not be representative of the overall spinal fusion rates in the United States since spinal fusions are more likely to be performed on younger patients (Merrill & Elixhauser, July 2007).
To our knowledge this is the first study to report total joint replacement or spinal fusion rates relative to practice profile metric such as patient volume. This comparison may provide valuable data on a surgeon’s decision for surgery within a given population. Only forty percent of the original groups were included in this study, due to insufficient data.
Given the consistency of the decision for surgery metric, we would not anticipate alternate findings in a broader sample set, however, a broader data set would be desirable in future studies.
Conflicts of interest already exist in many areas of the surgeon/patient relationship. When a surgeon recommends a surgical procedure, the surgeon is financially remunerated for their performance of the procedure. While surgeon ownership in implant distribution adds to this financial conflict of interest, it is important to note that remuneration from distributions from the implant distribution company are a fraction of the remuneration provided by the surgeons professional fee and as such, the surgeons ownership in implant distribution should not be considered to present a new or substantial influence.
Potential conflicts of interest, when identified, should be managed with full transparency to patients, hospitals and surgical colleagues. Furthermore, it is advisable for all distributorships to track surgical volume and to adopt a method for dealing with a surgeon who might show an increase in utilization that is not explained by a corresponding increase in reliable practice patterns. The American Association of Surgeon Distributors (AASD) has published standards governing the legal and ethical application of this model and have established mechanisms for such tracking and auditing of implant utilization.
Patients and society need to be satisfied that surgeon ownership interest in a distribution company does not compromise their dedication to serving the patient’s best interest. The current study demonstrates that for a select sample of surgeons, their commitment to ethical and professional standards were not compromised by their ownership interest in implant distribution. This result is especially meaningful as the surgeons were unaware that this investigation would be carried out until after the data collection periods for the control and trial years were completed.
The authors of this study support the safeguards established by the AASD that serve to ensure that the POD model is ethically implemented and maintained. These include full transparency and disclosure to the patient, hospitals and to colleagues. In addition, such standards should ensure the surgeon distributorship remains the lowest average cost vendor to the hospital and can demonstrate and document a rigorous process to insure product quality assessments prior to purchase of products for the distributorship.
This study demonstrates that, in this group of surgeons, all who have demonstrated proper intent through transparency, cost savings and endorsement of the AASD standards, ownership in the distribution of medical devices does not seem to influence surgical practice patterns and hence, does not corrupt the sacred doctor patient relationship. To encourage ongoing ethical and legal operation of a surgeon invested medical device distributorship, full transparency and independent ongoing assessments of practice patterns are recommended.
Deyo, R., Mirza, S., Martin, B., Kreuter, W., Goodman, D., & Jarvik, J. (2010; 303(13)). Trends, major medical complications, and charges associated with surgery for lumbar spinal stenosis in older adults. Journal of the American Medical Association, 1259-1265. Kurtz, S., On, K., Lau, E., Mowat, F., & Halpern, M. (2007 (89)). Projections of Primary and Revision Hip and Knee Arthroplast in the United States from 2005 to 2030. Journal of Bone and Joint Surgery, 780-785. Merrill, C., & Elixhauser, A. (July 2007). Statistical Brief # 34 -Hospital Stays Involving Muscoloskeletal Procedures: 1997 – 2005. Rockville, Maryland: U. S.Agency for Healthcare Research and Quality: Healthcare Cost and Utilization Project. Steinmann, J., Burton, P., Hopkins, G., & Skubic, J. (2009). Surgeon Ownership in Medical Device Distribution: Economic Analysis of An Existing Model. American Academy of Orthopedic Surgeons, (p. Scientific Exhibit). Weinstein, J. N., Lurie, J. D., Olson, P., Bronner, K., & Fisher, E. (2006 ). United States’ Trends and Regional Variations in Lumbar Spine Surgery: 1992 – 2003. Spine, 31(23).