Terrible triad injury of the elbow

 

                             Dr. KS Dhillon

Introduction

The terrible triad injury of the elbow was originally described by Hotchkiss in 1996. It constitutes a highly unstable form of fracture-dislocation of the elbow. It consists of elbow dislocation with concomitant radial head or neck and coronoid process fractures [1,2,3]. 

This injury pattern is designated as terrible because of historically poor outcomes and high complication rates. The elbow is well known as one of the most stable joints of the body. The complex anatomical structure and higher functional requirements make treating the elbow injuries more difficult [4].

Owing to its complex anatomical structure, even isolated elbow dislocations without bony fragmentation involve substantial soft tissue injury with capsular and ligamentous disruption. In complex elbow dislocations, there

are fractures of one or more major bony stabilizers. Fractures can involve the radial head, coronoid process, or olecranon. These fractures can destabilize the dislocation and nearly always need operative intervention to restore functional anatomic alignment and joint stability [5].

Controversies remain regarding the appropriate treatment algorithm for these injuries despite clinical and operative advancements and an increased understanding of pathoanatomy and elbow biomechanics. Successful evaluation and treatment require detailed knowledge of the elbow anatomy, functional importance, and its contribution to elbow stability [5].

Anatomy

The elbow joint (fig 1) consists of three sub-joints, namely the humeroradial, humeroulnar, and superior radioulnar joints. The joints are enveloped by a common joint capsule. The humerus, radius, ulna, and related capsules, and ligaments make up these sub-joints. These joints allow elbow flexion and extension, and forearm pronation and supination. The coronoid process is a triangular-shaped protrusion at the proximal ulna and it plays a major role in keeping the elbow stable. Apart from bony structures, several ligaments also contribute to elbow stability. These include the medial collateral ligament complex (MCLC) and the lateral collateral ligament complex (LCLC). The MCLC is composed of three small ligaments namely the anterior medial collateral ligament, the posterior medial collateral ligament, and Cooper's ligament. The LCLC is made up of four small ligaments: the lateral ulnar collateral ligament, the lateral radial collateral ligament, the annular ligament, and the accessory lateral ligament. The stability of the elbow largely depends on the functions of the radial head, coronoid process of the ulna, LCLC, and the anterior medial collateral ligament.

Fig 1- elbow joint

Etiology

About 60% of complex dislocations are caused by a fall from standing height [6]. Considerable force is required to sustain a complex dislocation. Falling on an extended arm that precludes valgus, axial, and posterolateral rotational forces, producing a posterolateral dislocation, is often the mechanism of insult.[1]

Fracture dislocations of the elbow tend to occur in distinct patterns depending on the mechanism of injury. Elbow extension with forearm supination and added valgus stress puts the most strain on the ulnohumeral joint, radial head, and MCL respectively. This causes a posterolateral rotational instability pattern of fracture-dislocation which includes posterior dislocation with a radial head fracture and the terrible triad injury with an added coronoid fracture. An axial load to the elbow in extension and a varus stress will cause compression injury to the medial side of the elbow leading to coronoid fractures, and tension forces acting laterally causing LCL rupture [7]. With the elbow in more flexion, this pattern can also cause a fracture of the olecranon. The outcome from this mechanism of injury is termed as varus posteromedial rotational instability. A direct blow to the posterior aspect of the flexed elbow can cause an anterior dislocation with an olecranon fracture. A hyperextension injury can also cause anterior dislocation with an olecranon fracture.

Epidemiology

Radial head fractures constitute between 20% to 30% of all adult elbow fractures [1]. Eighty-five percent of radial head fractures occur between the ages of 30 to 60 years. The mean occurrence is at age 45 years [8]. 

Coronoid fractures constitute 10% to 15% of elbow injuries. The elbow is the second most commonly dislocated joint, although it is one of the most stable joints in the body. About 20% of dislocations are associated with a fracture [6,9,10].

Pathophysiology

The elbow is made up of three sub-joints namely the humeroradial, humeroulnar, and superior radioulnar joints. The subjoints are made up of the humerus, radius, ulna, and related capsuloligamentous structures [11,12]. 

The radial head is an important restraint to posterolateral rotatory instability and it also acts as a secondary valgus stabilizer. In a normal elbow, the radial radiocapitellar articulation contributes minimally to valgus stability. However, in the event of MCL or coronoid injury, the radial head acts as the primary stabilizer to valgus stresses. It also prevents elbow subluxation[10]. Radial head fractures can be associated with episodic elbow instability, mechanical block to elbow motion, and injury to the distal radioulnar joint and/or the interosseous membrane (Essex-Lopresti) [13].

The coronoid process of the ulna provides ulnohumeral stability anteriorly and a varus buttress while resisting posterior subluxation [11].

The lateral collateral ligament (LCL) and medial collateral ligament (MCL) are the main capsuloligamentous stabilizers of the elbow. The MCL is the main stabilizer of valgus movements. It consists of the anterior bundle, posterior bundle, and transverse ligament. The robust anterior bundle is most important for stability [11]. Cavaderic studies have shown that fracture dislocations of the elbow are most likely to occur between 15 degrees of extension and 30 degrees of flexion, where the MCL is the least effective [14]. The lateral collateral ligament is the primary restraint to posterolateral rotatory instability. It contains four components namely the lateral ulnar collateral ligament, radial collateral ligament, annular ligament, and accessory (posterior) collateral ligament. The lateral ulnar collateral ligament is most important for stability [11]. 

The humeroulnar joint is the primary contributor to elbow stability, with its highly constrained articulation. The anteromedial facet resists varus movements and the muscles crossing the elbow joint contribute dynamically. The osseous and ligamentous structures afford static stability. 

In the terrible triad, the structures of the elbow fail from lateral to medial as the forearm supinates and is loaded. There is disruption of the lateral collateral ligament first, then the anterior capsule, and finally the medial collateral ligament [6]. The pattern of disruption from lateral to anterior/posterior and then medial is commonly referred to as the Horri circle [15].

History and Physical Examination

The initial evaluation should proceed according to the Advanced Trauma Life Support (ATLS) protocol. Concomitant fractures, dislocations, and injuries throughout the ipsilateral extremity have to be excluded.

Distal radioulnar joint (DRUJ) tenderness may represent an interosseous ligament disruption and there may be concurrent Essex-Lopresti injury [13]. 

Fracture-dislocations of the elbow will present with swelling, pain, and deformity. There will be limitation of movements of the elbow [6][4].

A thorough neurovascular examination should be carried out. The ulnar nerve is most vulnerable to injury. Brachial artery injury, although rare, can occur and lead to ischemia and compartment syndrome [16].

Elbow stability is tested by doing a posterolateral drawer and posterolateral pivot shift tests, and varus/valgus instability stress testing. The DRUJ examination is done by palpating over the wrist for tenderness and translation of more than 50% in the sagittal plane. The interosseous membrane is palpated for tenderness. A radial pull test is done at the time of surgery and if there is more than 3 mm translation, there should be a concern for longitudinal forearm instability (Essex-Lopresti) [13]. When the patient presents in a delayed or recurrent fashion, the examiner should assess elbow flexion/extension, forearm rotation, and nerve function.

Evaluation

Anteroposterior and lateral radiographs are done for diagnosis. Additional shoulder, wrist, and hand imaging is done if the injury is suspected in other joints of the ipsilateral limb. Radiographs will help to evaluate the concentricity of humeroulnar and radiocapitellar joints. Lateral films will help in detecting coronoid fractures. Most injuries can be diagnosed with plain radiographs. A computed tomography (CT) scan is always obtained for patients with the terrible triad to identify fracture patterns, comminution, and displacement, that may not be evident on plain radiographs. Reconstructed CT scans are useful to better evaluate the injury pattern and assist with preoperative planning [17].

The CT scan will show coronoid, radial head, or olecranon injuries missed on initial radiographs. Fluoroscopic imaging under anesthesia is beneficial for intraoperative decision-making. When radiographs cannot be done after the injury in select patient populations, performing an examination under anesthesia while taking the elbow through gentle ROM is useful. 

Nonstandard views may need to be obtained for joints that are stiff and the position cannot be changed to further assess the integrity and alignment of articulating surfaces. 

There are several classifications available that can assist in further diagnostics. These are the Mason, Regan and Morrey, and O’Driscoll Classifications [18]. 

The Mason Classification for Radial Head Fractures:

• Type I radial head fractures are either nondisplaced or minimally displaced (less than 2 mm), with no mechanical block to rotation.

• Type II are displaced (more than 2 mm) or angulated fractures, with possible mechanical block to forearm rotation.

• Type III have fracture comminution and displacement with confirmed mechanical block to motion.

• Type IV radial head fractures are associated with elbow dislocation[11].

The Regan and Morrey Classification system identifies three types of coronoid fractures: 

• Type I involves the coronoid tip

• Type II describes a fracture involving 50% or less of coronoid height

• Type III is determined by a fracture of greater than 50% of coronoid height [6,11].  

The O’Driscoll Classification system subdivides coronoid injuries based on location and the number of coronoid fragments. It recognizes that anteromedial facet fractures are caused by varus posteromedial rotatory forces [19].

Treatment 

The aim of treatment is to reestablish enough stability to permit early movements of the elbow [9,20,21]. Anatomic alignment of osseous structures is re-established. This is followed by restoration of the radial head and radiocapitellar contact. Ligaments are repaired if necessary.

If the elbow is sufficiently stable to allow early mobilization, non-operative management with immobilization in 90 degrees of flexion for 7 to 10 days is indicated. Non-operative treatment is also indicated if the coronoid fracture is small, the radial head fracture does not need surgery, and the humeroulnar and radiocapitellar joints have been anatomically reduced. A progressive range of movement exercises is routinely instituted following one week of immobilization. Strengthening protocols are begun after six weeks. 

An unstable radial head fracture and type III coronoid fracture, with associated elbow dislocation, is an indication for operative intervention. Open reduction internal fixation (ORIF) of the radial head, LCL reconstruction, and coronoid ORIF, with possible MCL reconstruction is carried out. In some situations, a radial head replacement may be necessary. If the instability persists after addressing the radial head and LCL complex, the next step is to proceed with operative MCL reconstruction. 

Isolated dislocations of the elbow are treated by immediate closed reduction. The reduction can be done in the emergency department under sedation. Successful reduction is accompanied by a clunking sound. The elbow should then be tested for stability by moving the elbow through a range of movements. Postreduction X-rays are done to confirm reduction. The elbow is then splinted in 90 degrees of flexion. Splinting should not proceed beyond three weeks. 

When surgery is indicated for radial head fractures, open reduction internal fixation (ORIF) is usually done. Radial head resection in fracture-dislocations may lead to Essex-Lopresti instability and arthrosis. Every effort should be made to maintain radial head integrity [10].

Open reduction and internal fixation is ideally carried out for radial head fractures when the fracture is non-comminuted and involves more than 40% of the articular surface and demonstrates bony continuity between the radial head and neck. Intra-osseous screws, compression screws, retrograde pinning, or anatomic plates can be utilized for fixation. When a plate is used it must be positioned posterolaterally in the safe zone, with the forearm in neutral, to minimize the risk of injuring the posterior interosseous nerve. Radial head arthroplasty is done for patients with a comminuted/displaced fracture of more than three fragments.

For radial head arthroplasty appropriately-sized implants must be used. A prosthetic head that is too small provides a very narrow area of contact which causes LCL laxity. If the head is too large it leads to poor congruence and excessive LCL tensioning, leading to postoperative stiffness. Patients undergoing arthroplasty with radial head replacement have demonstrated fewer postoperative complications, with significantly better ROM, than radial head repair [20]. When there is no coronoid fracture, a radial head fracture with elbow dislocation can be treated non-operatively. 

Type 3 coronoid fractures should be treated operatively [10]. In type 1 and type 2 fractures, there should be radial head conservation, elbow stability, and bony column congruence following soft tissue reconstruction. The coronoid can be repaired with sutures, anchors, screws, or the “suture lasso” technique. Open reduction internal fixation is the most common treatment for terrible triad injuries [6,10]. Coronoid fractures can be fixed via ORIF through the radial head defect laterally.

When a medial approach is used, the median antebrachial cutaneous nerve should be preserved. The coronoid fracture can be exposed between the two heads of the flexor carpi ulnaris. The ulnar nerve should always be visualized and protected. The lateral approach provides better access to the coronoid process. Postoperatively, active and active-assist ROM therapy is begun after 10-14 days. 

The LCL is repaired with the forearm in pronation if the MCL is intact. If MCL is injured, LCL is repaired with the forearm in supination. Postoperatively, it is important to avoid excessive shoulder abduction because that places undue stress on the LCL repair. Stability and adequate elbow function could be operatively restored without repairing the MCL. However, in cases where the elbow remains unstable after fracture fixation and lateral soft tissue (LCL) repair, especially in extension beyond 30 degrees, the MCL should be repaired [10,11,21].

Terrible triad injuries following high-energy insults are often accompanied by severe soft tissue injuries. This further prolongs the time to operative treatment, as soft tissue requirements for successful surgical outcomes are met in the interim. Several studies have documented that longer delays to surgery from injury produces postoperative elbow stiffness. Zhou et al [3] found that prognostication is optimized when surgical treatment is done between 24 hours and 14 days after injury. 

Lindenhovius et al [4] demonstrated a better range of motion in patients who underwent surgery within two weeks after the injury [4]. Wiigger et al. found that every 24-hour delay in surgery following initial injury more than doubles the risk of postoperative elbow stiffness [4].

Prognosis

Terrible triad injury patterns have historically poor outcomes due to persistent instability, stiffness, and arthrosis. A high index of suspicion is needed to expeditiously proceed through a detailed extremity examination, and appropriate imaging studies to make a correct diagnosis and proceed with early proper treatment [13].

Complications

Complications following elbow fracture-dislocation include synostosis, arthrofibrosis, heterotopic ossification (HO), infection, recurrent instability, post-traumatic arthritis, stiffness, nonunion, ulnar neuropathy, loosening of implant, and symptomatic hardware. Surgery to treat terrible triad injuries is associated with a high risk of complications, with up to a 54.5% reoperation rate, averaging between 22% to 30% [6,21,22].

Anatomic reduction of intraarticular fractures is necessary to prevent arthritic changes. A slight loss of extension can be expected. About 5% to 15% of patients with elbow fractures will experience stiffness following surgery [4]. Arthritis is common after high-energy trauma. It is likely a sequela of initial chondral impact and the degree of recurrent elbow instability. 

Some loss of motion after elbow fracture-dislocation can be expected. Patients usually lose more extension than flexion. The amount of stiffness increases with the energy of the initial injury. Heterotopic bone formation and delay of motion after repair also increase the amount of stiffness. 

Post-traumatic calcium deposition in the collateral ligaments and capsule is relatively common. Some reports document just under a 20% occurrence rate. Heterotopic ossification (HO) has occurred in up to 43% of operatively treated fracture dislocations [23,24].

Heterotopic ossification can cause near-complete ankylosis of the elbow. This can be seen on radiographic imaging 3 to 4 weeks after injury. The frequency and severity of HO are associated with the severity of the injury, the extent of soft tissue damage, length of immobilization, neurological injury, infection, delay to surgery, and the presence of associated burns [24]. Heterotopic ossification most often occurs either anteriorly, between the capsule and brachialis, or posteriorly, between the capsule and triceps. 

Distraction forces across the fracture secondary to flexion or active extension can lead to nonunion. Internal fixation failure is most common following radial neck fracture repairs secondary to its inherently poor vascularity. Recurrent instability rates are low. The most common cause is failure to recognize or treat fracture(s) or ligamentous injury. Recurrent instability is more common following type I or II coronoid fractures. 

Postoperative and Rehabilitation Care

Postoperative splinting is done for up to 10 days, depending on the stability achieved and concurrent injuries. The splint may be placed in flexion with the forearm in pronation to provide stability against posterior subluxation. When both MCL and LCL are repaired, the splint is positioned in flexion and neutral rotation. Some patients can start ROM exercises on the first postoperative day, with a majority beginning active ROM within 48 hours. Forearm rotation is usually allowed. Shoulder and wrist exercises are performed without restrictions. Extension within the terminal 30 degrees of motion is avoided for four weeks. 

The terrible triad of elbow injuries is difficult to treat. Despite optimal treatment and compliance with postoperative rehabilitation, rarely is it possible to achieve a full range of motion. Gomide et al [10] showed a mean flexion-extension range of 113 degrees and average flexion contracture of 24 degrees following surgery for terrible triad injuries of the elbow. 

Conclusion

The terrible triad injury of the elbow is the most complex pattern of all dislocations. It combines ligament damage with radial head and coronoid process fractures. Complete dislocations of the elbow joint should be considered as a terrible triad injury unless proven otherwise. The lack of knowledge of this clinical pattern of injury might be detrimental to elbow function. CT scan assessment should be carried out after the dislocation has been reduced for proper investigation of bony lesions. 

The principle objective of surgical management is to restore the bony stabilizing structures i.e. the radial head and coronoid process and radial collateral ligament reconstruction. Isolated radial head resection should be avoided since it appears as a bad prognosis factor for short and long-term outcome. Arthroplasty is advised if radial head fracture cannot be managed with osteosynthesis.

A medial surgical approach is recommended in patients with persistent posterolateral instability following radial collateral ligament reconstruction or when fixation of a large coronoid process fragment is necessary. External fixation is advocated when there is persistent instability following the reconstruction of bony and ligamentous structures. It provides joint stability and protects the reconstruction. 

References

1. Galbiatti JA, Cardoso FL, Ferro JAS, Godoy RCG, Belluci SOB, Palacio EP. Terrible triad of the elbow: evaluation of surgical treatment. Rev Bras Ortop. 2018 Jul-Aug;53(4):460-466.

2. Ikemoto RY, Murachovsky J, Bueno RS, Nascimento LGP, Carmargo AB, Corrêa VE. TERRIBLE TRIAD OF THE ELBOW: FUNCTIONAL RESULTS OF SURGICAL TREATMENT. Acta Ortop Bras. 2017 Nov-Dec;25(6):283-286. 

3. Zhou C, Lin J, Xu J, Lin R, Chen K, Sun S, Kong J, Shui X. Does Timing of Surgery Affect Treatment of the Terrible Triad of the Elbow? Med Sci Monit. 2018 Jul 09;24:4745-4752.

4. He X, Fen Q, Yang J, Lei Y, Heng L, Zhang K. Risk Factors of Elbow Stiffness After Open Reduction and Internal Fixation of the Terrible Triad of the Elbow Joint. Orthop Surg. 2021 Apr;13(2):530-536. 

5. Ohl X, Siboni R. Surgical treatment of terrible triad of the elbow. Orthop Traumatol Surg Res. 2021 Feb;107(1S):102784.

6. Jones ADR, Jordan RW. Complex Elbow Dislocations and the "Terrible Triad" Injury. Open Orthop J. 2017;11:1394-1404.

7. Doornberg J.N., Ring D.C. Fracture of the anteromedial facet of the coronoid process. J. Bone Joint Surg. Am. 2006;88(10):2216–2224.

8. Kovar FM, Jaindl M, Thalhammer G, Rupert S, Platzer P, Endler G, Vielgut I, Kutscha-Lissberg F. Incidence and analysis of radial head and neck fractures. World J Orthop. 2013 Apr 18;4(2):80-4.

9. Papatheodorou LK, Rubright JH, Heim KA, Weiser RW, Sotereanos DG. Terrible triad injuries of the elbow: does the coronoid always need to be fixed? Clin Orthop Relat Res. 2014 Jul;472(7):2084-91. 

10. Gomide LC, Campos Dde O, Ribeiro de Sá JM, Pamfílio de Sousa MR, do Carmo TC, Brandão Andrada F. TERRIBLE TRIAD OF THE ELBOW: EVALUATION OF SURGICAL TREATMENT. Rev Bras Ortop. 2011 Jul-Aug;46(4):374-9.

11. Xiao K, Zhang J, Li T, Dong YL, Weng XS. Anatomy, definition, and treatment of the "terrible triad of the elbow" and contemplation of the rationality of this designation. Orthop Surg. 2015 Feb;7(1):13-8. 

12. Liman MNP, Avva U, Ashurst JV, Butarbutar JC. StatPearls [Internet]. StatPearls Publishing; Treasure Island (FL): Aug 8, 2022. Elbow Trauma.

13. Ramzi Z, Juanos Cabans J, Jennart H. Terrible triad of the elbow with an ipsilateral Essex-Lopresti injury: case report. J Surg Case Rep. 2020 Jun;2020(6):rjaa103.

14. Wake H., Hashizume H., Nishida K., Inoue H., Nagayama N. Biomechanical analysis of the mechanism of elbow fracture-dislocations by compression force. J. Orthop. Sci. 2004;9(1):44–50.

15. de Klerk HH, Oosterhoff JHF, Schoolmeesters B, Nieboer P, Eygendaal D, Jaarsma RL, IJpma FFA, van den Bekerom MPJ, Doornberg JN., Traumaplatform 3D Consortium. Recognition of the pattern of complex fractures of the elbow using 3D-printed models. Bone Joint J. 2023 Jan;105-B(1):56-63.

16. Gonzalez LJ, Shields CN, Leucht P, Konda SR, Egol KA. Fracture-Dislocations of the Elbow: A Comparison of Monteggia and Terrible Triad Fracture Patterns. Orthopedics. 2022 Dec 02;:1-6.

17. Ozdag Y, Luciani AM, Delma S, Baylor JL, Foster BK, Grandizio LC. Learning Curve Associated With Operative Treatment of Terrible Triad Elbow Fracture Dislocations. Cureus. 2022 Jul;14(7):e27156.

18. Lampaert S, Herregodts J, De Wilde L, Van Tongel A. Radial head fractures: a quantitative analysis. Acta Orthop Belg. 2022 Jun;88(2): 380-386.

19. Shukla DR, Fitzsimmons JS, An KN, O'Driscoll SW. Effect of radial head malunion on radiocapitellar stability. J Shoulder Elbow Surg. 2012 Jun;21(6):789-94.

20. Chen H, Shao Y, Li S. Replacement or repair of terrible triad of the elbow: A systematic review and meta-analysis. Medicine (Baltimore). 2019 Feb;98(6):e13054. 

21. Kim BS, Kim DH, Byun SH, Cho CH. Does the Coronoid Always Need to Be Fixed in Terrible Triad Injuries of the Elbow? Mid-Term Postoperative Outcomes Following a Standardized Protocol. J Clin Med. 2020 Oct 29;9(11).

22. Chen HW, Liu GD, Wu LJ. Complications of treating terrible triad injury of the elbow: a systematic review. PLoS One. 2014;9(5): e97476. 

23. Shukla DR, Pillai G, McAnany S, Hausman M, Parsons BO. Heterotopic ossification formation after fracture-dislocations of the elbow. J Shoulder Elbow Surg. 2015 Mar;24(3):333-8.

24. Tangtiphaiboontana J, Agel J, Beingessner D, Hébert-Davies J. Prolonged dislocation and delay to surgery are associated with higher rates of heterotopic ossification in operatively treated terrible triad injuries. JSES Int. 2020 Jun;4(2):238-241.

 

Autonomy and Accountability in Orthopaedic Surgery

                              Dr. KS Dhillon

Introduction

The orthopaedic surgical practice is becoming increasingly complex. There has been a rapid pace associated with the development of new implants and technologies. This has had several unintended consequences. Hospital clinical staff and doctors have been challenged to be proficient in all aspects of these new technologies. Dependencies have developed on suppliers of these implants and technologies by the physicians and hospitals. Conflict has developed between physicians and hospitals about the cost. The physician-supplier relationship surrounding these products and technologies has led to concern about the industry's influence on the physician's choice of products. External forces have challenged the physician's clinical decision-making and autonomy. The physician accountability has grown. A professional environment that was traditionally characterized by the leadership and autonomy of the surgeon is now strained. It has been redefined to meet the demands of contemporary practice where doctors and staff, suppliers, and hospital managers all have a hand in the coproduction of implant surgery.

The professionalism, autonomy, and accountability of the orthopaedic surgeon need to be reexplored and redefined due to the influence of parties external to the physician i.e. suppliers, payors, hospitals, and patients.  Orthopaedic surgeons need to take the lead and engage in a much more strategic adaptation of their professional environment.

Core to professionalism for orthopaedic surgery is the focus on optimal care for the patient (1). Nonphysicians are increasingly interjecting issues of cost-effectiveness, reimbursement, value, comparative performance, and holding physicians to clinical standards based on “evidence-based” information. Important considerations relative to physician autonomy, accountability, and professionalism include the changing transparency of information about physician performance and judgment, the role of supplier service and relationships, as well as the push for cost-effectiveness and value from hospitals and payors, and the voice of the patient. There are synergies or in some cases lack of synergy between hospitals, surgeons, patients, and the device industry.

Autonomy and Accountability

In the United States, the idea of medicine as a unique form of work has been associated with the concept of professionalism. It is viewed as a compact of autonomy and accountability of the physician and of the whole profession. Eliot Freidson in Profession of Medicine declared that medicine was worthy of the label “profession”. This was because it regulated itself and was not subject to evaluation by others (2). Physicians caring for patients in a medically appropriate and ethical way, based on their own best judgment, represents the hallmark of the physician’s work as a professional. Autonomy is a characteristic that physicians traditionally carry out in their everyday work. Accountability has traditionally been attributed to the patient, self, society, the profession, and medical/ethical standards while carrying the duty. Donald Light (3) suggested that the autonomy of Freidson’s depiction reflected a “constructed reality” of how physicians practiced at a point in time. Light in assessing contemporary medical practice from an overall perspective, explains that with increased observation of practice and access to information, accompanied by the setting up of standards and/or intervention by nonphysicians, the autonomy of the individual physician has been supplanted by accountability as the foundation of medical professionalism (3).

Light has been correct in stating that a myriad of forces has challenged the ability of the individual physician to control the content and conditions of practice, and the traditional operational features of autonomy.  Included in this are hospital protocols, formularies, product standardization, payor formularies, frequent substitution of pharmaceuticals, physician report cards that may be tied to reimbursement/bonuses, requirements of preauthorization for consultations, hospitalization, procedures, denial of certain services, patient response to information on the Internet, direct-to-consumer advertising, declining reimbursements, rising overhead, difficult medical malpractice climate, and frequent decision making in all of these areas by nonphysicians. Review decisions made by entities other than the physician based on large amounts of practice data or “evidence-based medicine” are frequently believed by the physician to challenge his/her autonomy. All of these add further dimensions of accountability to the work of the physician. As Rappolt has stated, clinical guidelines “have a paradoxical relation to professional autonomy, since despite being the quintessence of medical knowledge at the collective level, they diminish the technical autonomy of the individual practitioner” (4). The tension between individual physician autonomy and accountability has to be regarded within the context of specialty practice which, in the modern world of medicine, is characterized by very diverse work processes, procedures, relationships with product manufacturers and their representatives, as well as the work environments. This tension is not merely diametrically opposite ends of a continuum reflecting medical work. It is a very important aspect of the successful strategic management of the profession (5).

For all issues in clinical medicine, accountability to a higher standard bolstered by strong evidence from well-designed clinical trials, does not exist. In some areas, evidence exists but the level of evidence is variable. Some are at a higher level such as from randomized clinical trials while others are at lower levels such as from case series or trials of lesser rigor. In orthopaedic surgery, Level IV evidence such as case series, case-control study, poor reference standard, or no sensitivity analysis is the most common level of evidence (6). This variability in clinical evidence from research highlights the importance of physician anecdotal experience in conjunction with high-quality evidence-based medicine. This also highlights the challenge of forming and imposing clinical guidelines, and achieving physician acceptance, based on evidence-based medicine. It also highlights why physician autonomy and decision making is challenged when hospitals and payors intervene in this arena.

Anybody who participates in or observes medical practice is aware of the fact that autonomy, as it relates to the individual physician, is not dead. Daily physicians engage in encounters with patients where they exercise decision-making grounded in their training and experience to assess the patient’s symptoms, order and evaluate tests, and make clinical recommendations. This is particularly so in patients requiring surgery. Pope (7) has described the contingent nature of surgical work which includes factors specific to the patient such as previous surgeries, unique anatomy, comorbidities, or preferences, surgeon contingencies such as tactile ability, ease of product use, and good patient outcomes and external contingencies such as operating theater environment and level of support. This contingent nature of surgical practice highlights the following:

(1) The extent to which surgeon decision-making may go beyond protocols and standardization as the physician focuses on clinical accountability.

(2) That a clash may exist between clinical and cost accountability surrounding the patient care.

(3) The idea of accountability via standardization may be an enduring aspiration.

This further suggests that characterization of autonomy versus accountability, as dichotomous constructs depicting the profession’s grounding, requires further scrutiny. 

The Impetus for Challenge of Physician Autonomy

The current reality is that healthcare costs form a substantial and increasing percentage of the GDP. In the present environment, reimbursements frequently outpace costs in a highly managed reimbursement system. The practice is inside a technology boom with many high-cost items continually entering the marketplace in a highly fragmented medical system. There has been an erosion of relationships between physicians and hospitals as a consequence of this change. As payors and hospitals have faced these financial realities, they have questioned the wide-ranging variation in the cost and quantity of clinical care. Over time, the length of stay in the hospital has decreased and care has moved to the outpatient clinics. Diagnosis-related groups (DRGs), inpatient chart reviews by insurance companies, capitated payments, and relative ease of outpatient care were all forces in this change. The recent challenge to the overutilization of imaging is only the tip of the iceberg in the assessment of the appropriateness of the application of technology in practice (8). Variation studies in small areas show that it is difficult to assess the extent to which utilization data reflects appropriate, under, or overutilization (9,10). 

Supply costs have increased due to supply-intensive admissions accounting for up to 60% of a hospital’s supply cost. Now strategic management of the healthcare supply chain has emerged as a central characteristic of the best-managed hospitals and systems. The focus on cost reduction has now shifted from commodity items to physician preference items (PPIs). Begun and Lippincott have pointed out that while physicians have been successful in influencing the hospital’s choice of products used in clinical care with rising cost-containment efforts, healthcare delivery organizations are becoming more rigorous in monitoring and influencing the purchase of equipment and supplies (5). In the absence of clear guidelines regarding processes and products, suppliers are engaging both hospitals and physicians in a highly fragmented marketplace (11). A new reimbursement model incorporates the costs of PPIs into a highly bundled episode of care. This includes the hospital, physician, products, and even posthospitalization costs (12).

The hospital’s materials management or supply chain management department has the responsibility of managing the product side of supply intensive admissions. Until more recently, these departments have been transactional with the availability of products at the point of patient care being the measure of success. The supply function has become increasingly strategic as costs attributable to PPIs have escalated. Due to financial concerns, payors and hospitals have started to push for accountability from the orthopaedic surgeon for their choice of product on a clinical and cost basis. Different surgeons utilize different implants, which can vary considerably in cost without clear demonstrable gains in outcomes. There is a question of the individual surgeon’s failure to recognize the financial aspects of supply-intensive admissions and the inability to develop a level of accountability to those who pay for and support the surgeon’s delivery of care. This divide between management and clinical medicine surrounding clinical care is a powerful force that continues to lead hospitals and payors to focus on financial solutions.  This is often to the detriment of physician autonomy and long-term collaborative solutions. The physicians in the US have characteristically lacked organization, even when joined together within groups or “a legal umbrella.” Hence bringing physicians together to agree upon uniform approaches is difficult (13). 

An Apparent Lack of Surgeon Concern?

Although there has been guidance relating to the selection of expensive physician preference items, there has however been little attention to these recommendations by the physicians in both the United States and other nations (14). When it is suggested to the surgeons that the products are “equivalent,” based on available evidence-based information, or that surgeons should be supply indifferent ie, use of any one of similar items will lead to similar clinical outcomes, surgeons often disagree. They cite the importance of their judgment and tacit knowledge. They frequently state that those who manage and pay for care do not understand the several factors that enter into clinical decision making, including their choice of implants. A recent study of surgeons by Kitto et al provides important insight into surgeon views of evidence-based medicine (EBM) (15). The results from the surgeons surveyed suggest the following:

(i) They believe that EBM marginalizes patient involvement in decision making.

(ii) They believe that EBM-generated knowledge is useful and is commonly used in daily clinical decision-making, however, not using EBM does not adversely affect their daily clinical decision-making.

(iii) They have high confidence in their own judgement compared with low confidence in clinical practice guidelines and other sources of evidence.

(iv) Journal summaries of the latest research related to a subject are the most useful resources in clinical practice above clinical practice guidelines. 

To obtain both good clinical and cost evidence on specific products utilized in surgery is often difficult. This results in surgeons deferring to existing clinical evidence, whether it is from research studies, anecdotal, or an accepted community standard. The recent funding and push for comparative effectiveness research has the potential to improve this area. An opportunity to delineate what constitutes “evidence-based” information exists. Also, there is an opportunity to obtain more clarity in clinical effectiveness that could drive cost-effectiveness. Most surgeons choose products based on familiarity and good clinical outcomes. Regardless of this, some see physician commitment to a product as resistance that reflects the political edge of “free choice” associated with autonomy. With supply-intensive surgeries threatening profitability, there are an increasing number of managers who seem prepared to carry out a program that channels physicians toward less expensive but seemingly equivalent products. In the pharmaceutical industry, drugs are frequently “tiered” by payors and/or employers. This is to achieve cost control with three tiers: (1) generics, (2) “preferred brand name drugs”, and (3) “non-preferred brand name drugs”. The cost of the medication to the patient differs. It depends on the tier. At least one U.S. payor has considered “tiering” implant pricing. 

Tiering implants to achieve cost control is quite different from tiering pharmaceuticals. A pharmaceutical can be discontinued and replaced with a different pharmaceutical if found to be suboptimal. An implant, however, once placed, does not have the same relative ease of replacement. There are some who are looking toward pay-for-performance programs where one can reward the surgeon for making choices that consistently lead to superior outcomes. There is a growing consensus that pay-for-performance systems are likely to influence clinician behaviors in unpredictable ways. Attempts to increase reliability may be prohibitively costly or currently beyond the system eg, information management and monitoring capabilities (16). An inquiry by the HSRC-ASU on formal gainsharing suggests that physicians respond to economic incentives around product standardization (17).

It appears that the real issue being confronted is the role of the individual physician as a separate locus of economic power. Physicians are directly involved in deciding what resources are used. They prescribe drugs and diagnostic procedures. They refer to other practitioners, and they direct invasive procedures (18). A full understanding of how physicians generally approach their work and make clinical decisions remains unclear to those outside the profession (19).

Resource Dependency on Suppliers

Work by the HSRC-ASU done previously indicates that the supplier control of resources poses very important challenges to hospitals and the surgeons (20). The supplier, or sometimes the distributor of the supplier’s products who actually interfaces with the surgeon, appears to act on one hand as the agent of the physician by promoting the best possible care and support of the surgeons’ practice and, on the other hand, as the agent of the hospital by informally managing inventory and instrumentation. By creating a resource dependency, the suppliers are working both to maximize their income while assuring continued physician loyalty/hospital loyalty through services and resource management. There is little rigorous research on how supplier dependency affects quality, cost, and outcome.

There are several responses to supplier dependency. Research by the HSRC has shown that there are few hospitals that have managed well in keeping the supplier out of the surgical suite with surgeons, operating room technicians, and those who manage supplies. There are some hospitals that believe that supplier-managed inventory actually reduces hospital costs and that supplier-managed instrumentation is more efficient. Others however decry their inability to manage the flow of goods in and out of the hospital.

The financial support provided by the industry in the form of continuing medical education, research funding, and collaboration with orthopaedic surgeons is yet another area of resource dependency. Many academic medical centers have begun addressing this issue to varying degrees. A consensus statement on this area is still lacking.

Autonomy and Accountability in Orthopaedic Practice

Now it is clear how orthopaedic practice, with a focus on implant surgeries, provides a very special backdrop for exploring the idea of accountability. In orthopaedic practice, there is the downstream result of a very complex value chain that includes organizations, their products, and their people. Accountability is specific to patient characteristics and choices, surgeon ability and outcomes, and external factors that include products, the supplier role, and hospital-imposed choices. 

The Surgeon as the Patient’s Agent

Orthopaedic practice has generally incorporated a “professional-as-agent” model in approaching the care of the patient. Here the professional-as-agent assumes responsibility for directing the healthcare utilization of the patient, as an agent trying to choose what the patient would have chosen, had he/she been as well informed as the professional (21). This is a model where surgeon biases are subverted to the preferences that matter to the patient (22). An orthopaedic surgeon may actively engage a patient to solicit such patient preferences, but such solicitation is not always present or perhaps critical in enacting of this model. The patient could be in no mental state to express preference. In emergency situations, or in choices made during surgery, there is little time for the patient to choose among preferences. A patient might actually express a preference for a passive role in the choices that the physician has to make (22). Surgery continues as a curious mix of active and somewhat passive patient participation, but with the patient’s benefit at the forefront.

Surgeons, Suppliers, and Their Products

The surgeon is one part of the value chain that brings products, processes, and services to the patients. Surgeons hone skills and they become certified within the highly orchestrated system of graduate medical education, specialty certification, and practice. During the professional training and certification, the surgeon is exposed to several products that become central to everyday practice (16).

It is unclear how surgeon relationships with suppliers affect their ability to be “reflective practitioners”. They need to continually question the products (eg, implants) and associated techniques within the context of achieving excellence in care for patients. Familiarity is important. Decisions made in the course of practice frequently become part of what the surgeon defines as the most appropriate materials and processes for achieving excellence in dealing with the patient. This is expected as habitus (23). It results in typifications (patterned responses) (24) that streamline decision-making and rationalize uncertainties that may remain. Research on surgery has not shown how such typifications improve performance. 

The importance of the materials to the act of implant surgery is obvious. What is not obvious is how strong supplier-physician relationships surrounding materials affect professionalism. Instead of directly addressing professionalism, hospital management is attempting to reshape the interface between the physician and supplier through supplier credentialing strategies. These policies mainly address risk issues for the hospital and attempt to disintermediate the physician-supplier relationship so as to affect physician product choice and cost to the hospital. The hospital pressures on surgeons to consider alternatives are usually driven by economic considerations. The hospital may want the physician to consider alternatives, not because they are superior, but because they are deemed “equivalent” and are more cost-effective.

Recently the physician-supplier relationship in orthopaedics was investigated by the U.S. Department of Justice leading to a challenge of physician-supplier financial arrangements (25). Academic medical centers have been questioning the appropriateness of several supplier-provided ancillary benefits (26). Some of these medical centers have banned industry-sponsored gifts, free lunches, and pharmaceutical samples and have also strengthened institutional regulations surrounding industry funding (27,28). The impact of supplier influence on physician professionalism is being questioned in all these cases. Suppliers continue to financially support a great deal of the continuing education and training for implant surgeons. This is frequently done in collaboration with the profession itself.

Surgeons and Hospitals

The site of care is an important part of the value chain. It should not be seen as “neutral” to achieving a high level of professionalism. The hospital as a “doctors workshop” that economist Mark Pauly depicted as an institution designed around the physician’s needs, ready and willing to carry out the physician’s orders. Today, it is subject to forces that can support or detract from professionalism (29). 

The hospital often finds it difficult to understand how surgeons make decisions and resents surgeon resistance to their efforts to bring parties together to address clinical and cost-effectiveness of surgical techniques and products.

Only in the most progressive hospitals do physicians come together to build consensus regarding products and provide the hospital with an understanding of their preferences in a way that the hospital can act in their mutual best interests (30). Hospitals are engaging in much more aggressive programs to control supplier influence and make surgeons aware of the financial ramifications of their clinical choices. It is however unclear which incentives or mix of incentives, ranging from physician/hospital gainsharing to colleague-guided discussions to highly data-driven meetings hold the most promise (17).

The New Professionalism of Shared Decision-Making in Surgery

The idea of shared decision-making has generally been applied to the relationships between physicians and patients as one of the defining features of the new professionalism in medicine (22). The four main characteristics of shared decision-making include (31):

• Both the patient and the doctor are involved

• Both parties share information

• Both parties take steps to build a consensus about the preferred treatment

• An agreement is reached on the treatment to implement. 

Extending this idea, shared decision-making for implant surgery represents surgeons, suppliers, and hospitals being involved in similar ways and sharing in a productive and transparent way their particular areas of expertise.  Currently, these stakeholders are involved in a myriad of relationships and dependencies that affect physician accountability and professionalism.

Pope (7) identified the quality of assistance and the type of equipment as key variables pertaining to surgical quality and outcomes. The strong relationships between individual orthopaedic surgeons and these kinds of variables, however, have not been rigorously studied. No research has reported data on how hospitals and surgeons define, recognize, and authorize appropriate supplier competencies or supplier input. Supplier representatives support high-volume implant surgeons as well as those who do relatively few surgeries. Their role may be highly undervalued in some settings and overvalued in others. These supplier representatives have a variety of incentives associated with their behavior including sustained relationships by way of trust and reliability in the sharing of information. Supplier representatives frequently provide assistance to multiple physicians. They bring knowledge from a much wider range of applications to the operating room. They do not receive high levels of training nor are their conflicts of interest properly managed. Through their intensive relationships with surgeons, supplier representatives reduce the likelihood that surgeons will consider alternative supplies. They work to keep product “switching costs” to a maximum, to sustain surgeon brand commitment (5). Supplier representatives reduce the likelihood of the profession engaging in a key part of its environment on its own. Some are of the opinion that the problem will only be solved by the hospital taking over the nonbusiness aspects of the supplier’s involvement in surgery. The supplier relationship with the physician has been disintermediated with much of the role undertaken by hospital personnel in only a very few instances. In a focus group of over 20 major U.S. health systems and four academic health centers, no participant had undertaken a program to reduce supplier role in the operating room and subsequent dependency on the part of physicians and staff. Smaller hospitals find themselves unable to provide the training and resources necessary to fully replace the supplier’s input.

Research by HSRC-ASU identified a number of hospital supply selection strategies for physician preference items (30). Some hospitals attempt to narrow the number of suppliers while others focus on sustaining a larger number of suppliers but utilize the leverage of “equivalent” designation to assure comparable pricing across suppliers. This is a process where nonphysician managers frequently take the lead. They organize value analysis efforts, build the knowledge necessary to attempt to engage surgeons in a discussion regarding equivalencies, and then manage the process of contracting and channeling the flow of product into and out of the hospital. These engagements between the hospital and surgeons are generally about price and control and not about supporting surgeons who may believe they have made the best choices from a clinical point of view. There are other hospitals and systems that are more critically involved in managing, with physicians, the broader episode of surgical care. They see the surgeon, the surgeon’s staff, the surgeon’s experience, and postsurgical care issues as critical to building a comprehensive commitment with the physician. Today, these efforts may or may not be led by the physician. When these efforts are led by the physician, it results in a broader integration of the physician and hospital’s accountability for clinical care. This is one part of the common interest and accountability of the hospital and physician.

Toward a Professional Culture of Coproduction

This is a period of enhanced efforts by several forces to shape the behavior of orthopaedic surgeons. The changing transparency of information surrounding physician performance and judgment, the role of the supplier, the push for cost-effectiveness, and the stronger patient voice are all playing an influential role. 

Neither the orthopaedic surgeons nor the profession as a whole has heeded the warning that “professions must act boldly to prevent their members from forming relationships with suppliers that exploit the client/professional relationship—thereby risking government intervention or further erosion of the profession’s legitimacy” (5). Much of the contested terrain between hospitals and surgeons has been centered on the belief by surgeons that hospitals only care about costs and have little interest in supporting physician practices in ways that meet their goals. Hospitals have employed a variety of tactics to influence physician choice. This includes prioritization of their relationships with physicians, presentation of good product data, and incentives for physicians (30). They are however hampered in their efforts by the lack of evidence.

There are two opposing strategies about suppliers that are worthy of further discussion:

(1) bringing suppliers into a much more meaningful role to support physician accountability.

(2) substantially reducing physician dependency.

The former is an area where physicians could redefine the supplier’s role in patient care and how their presence impacts the accountability of both physicians as well as the hospital. The ways to integrate suppliers into the supply chain to improve quality and processes have not been sufficiently explored. To accomplish this will require a substantial change on the part of both the hospital and physician, characterized by the creation of a culture of co-production.

Suppliers bring value to the surgeon and the hospital despite questions of undue influence. It is logical to utilize the surgeon to develop new products as the surgeon is most well-versed in product limitations. The first step in transforming this complex relationship is disaggregating the service, sales components, and advisory.

The supplier-surgeon relationship is not well understood by those outside the profession. It is unstudied and is generally viewed in a derogatory light. There is some evidence that dependence on the supplier representative in the operating room is variable. 

The supply chain managers today are in a position to answer to the hospital in terms of risk reduction, cost-effectiveness, and cost savings.  They can also answer to physicians in terms of product availability for clinical care. They often face difficulty due to a lack of infrastructure that supports achieving all these desired outcomes. This is an important opportunity for a physician-led effort to ensure that the physician's voice lends credibility. 

Most of the discussion here has been about the potential for improved strategic management of orthopaedic surgery as an organized entity to redefine professionalism for 21st-century practice. The areas that require attention by the profession include the establishment and monitoring of standards for physician-supplier relationships and surgeon assumption of leadership roles in hospitals pertaining to product selection, supplier choice, and the orchestration of choice based on cost and clinical usefulness. A key ingredient that is usually lacking for physicians to enter these leadership roles is the required knowledge to guide the business aspects and complexity of the relationships. Lacking in most graduate medical education (GME) residency programs is an introduction to basic business concepts and terms, the different priorities of management, and experience in being a part of hospital committees while in clinical training. Also lacking in most hospital management programs is an understanding of physician priorities and the essence of clinical medicine that usually does not fit well into an economic cost model. Physicians who do enter leadership roles may have additional business training or may just have an interest in cost and management issues along with their clinical care. The inclusion of basic business concepts, some level of administrative experience, and focus on potential conflicts of interest with industry during GME is an important consideration to assist physicians. It helps to understand and be influential with issues surrounding their autonomy, cost, accountability, and industry relationships. According to Horton (32), it is important for surgeons to continue taking a leadership role in seeking funding for higher-level outcomes-based studies, comparative analysis, and advocating for large population-based databases.

The terms “shared decision-making” and “coproduction” have been used to represent the myriad of roles, relationships, and factors associated with contemporary implant surgical practice. Orthopaedic surgeons, hospital personnel, supplier representatives, and others come together to orchestrate the surgical episode of care. These synergies, and how each contributes to outcomes and affects accountability, are not properly understood. Due to increased quality and transparency of information, there is a mounting demand for accountability relating to cost and clinical outcomes. This demand which is frequently made by outside agencies dominated by nonphysicians, challenges the idea of the surgeon as an autonomous practitioner and extends his/her accountability. This is an area of opportunity for physicians to lend their clinical expertise and voice and be leaders in comparative effectiveness and in helping to bridge the gap between clinical and cost-effectiveness. Neither the term autonomy or accountability fully reflect the reality of modern surgical practice or the reality of physician professionalism. Increasingly, the invocation of autonomy fails to legitimize surgeon behavior and choice. The challenge for orthopaedic surgeons, as a profession, is to redefine the idea of autonomy, professionalism, and accountability for the 21st-century practice of orthopaedic surgery.

References

1. Accreditation Council for Graduate Medical Education. Advancing Education in Medical Professionalism. Available at: http://www.acgme.org/outcome/implement/profm_resource.pdf.

2. Friedson E. Profession of Medicine: A Study in the Sociology of Applied Knowledge. New York, NY: Harper and Row; 1970.

3. Light D. Towards a new professionalism in medicine: Quality, value, and trust. Tidsskr Nor Laegeforen. 2003;123:1870–1873.

4. Rapport SG. Clinical guidelines and the fate of medical autonomy in Ontario. Soc Sci Med. 1997;44:977–987.

5. Begun J, Lippincott RC. Strategic Adaptation in the Health Professions: Meeting the Challenges of Change. San Francisco, CA: Jossey-Bass; 1993.

6. Obremskey WT, Pappas N, Attallah-Wasif E, Tornetta P, Bhandari M. Level of Evidence in Orthopaedic Journals. J Bone Joint Surg Am. 2005;87:2632–2638.

7. Pope C. Contingency in everyday surgical Work. Sociol Health Illn. 2002;24:369–384.

8. Matthews A. Insurers Hire Radiology Police to Vet Scanning. Wall Street Journal. November 6, 2008. Available at: http://online.wsj.com/article/SB122591900516802409.html.

9. Mullan F. Wrestling with variation: An inteview with Jack Wennberg. Health Aff (Millwood). 2004; Variations Revisited: Web-Exclusive Collection: V73–V80.

10. Weinstein JN, Bronner KK, Morgan TS, Wennberg JE. Trends and geographic variations in major surgery for degenerative diseases of the hip, knee, and spine. Health Aff (Millwood). 2004;Suppl Web Exclusives:VAR81-9.

11. Porter ME. Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York, NY: Free Press; 1980.

12. Prometheus Payment Inc. Available at: http://www.prometheuspayment.org/index.html. Accessed November 22, 2008.

13. Mannion R, Davies HTO, Marshall MN. Cultures for Performance in Health Care. Berkshire, England: McGraw-Hill Education; 2005.

14. National Library of Guidelines Specialist Library. Guidelines Implementation. Available at: http://www.library.nhs.uk/guidelinesfinder/Page.aspx?pagename=IMPL.

15. Kitto S, Villanueva EV, Chesters J, Petrovic A, Waxman BP, Smith JA. Surgeons’ attitudes towards and usage of evidence-based medicine in surgical practice: a pilot study. ANZ J Surg. 2007;77:231–236.

16. Sloan FA, Kasper H, eds. Incentives and Choice in Health Care. Cambridge, MA: MIT Press; 2008.

17. Ketcham J, Furukawa M. Hospital-physician gainsharing in cardiology. Health Aff (Millwood). 2008;27:803–812.

18. Kemp R. Medical dominance and institutional change in the delivery of health care services. Forum Soc Econ. 2007;36:43–51.

19. Kralewski JE, Wingert TD, Barbouche MH. Assessing the Culture of Medical Group Practices. Med Care. 1996;34:377–388.

20. Wilson NA, Schneller ES, Montgomery K, Bozic KJ. Hip and knee implants current trends and policy considerations. Health Aff (Millwood). 2008;27:1587–1598.

21. Evans RG. Strained Mercy: The Economics of Canadian Health Care. Toronto, Ontario, Canada: Butterworth and Co; 1984.

22. Charles C, Gafni A, Whelan T. Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango). Soc Sci Med. 1997;44:681–692.

23. Bordieu P. The Logic of Practice. Cambridge, UK: Polity Press; 1990.

24. Loseke, DR. Thinking about Social Problems: An Introduction to Constructionist Perspectives, 2nd ed. New York, NY: Aldine De Gruyter; 2003.

25. U.S. Department of Justice, U.S. Attorney, District of New Jersey. Five Companies in Hip and Knee Replacement Industry Avoid Prosecution by Agreeing to Compliance Rules and Monitoring. Available at: www.usdoj.gov/usao/nj/press/files/pdffiles/hips0927.rel.pdf.

26. Brennan TA, Rothman DJ, Blank L, Blumenthal D, Chimonas SC Cohen JJ, Golden J, Kassirer JP, Kimball H, Naughton J, Smelser N. Health industry practices that create conflicts of industry: a policy proposal for academic medical centers. JAMA. 2006;295:429–433.

27. Coleman DL, Kazdin AE, Miller LA, Morrow JS, Udelsman R. Guidelines for interactions between clinical faculty and the pharmaceutical industry: one medical school’s approach. Acad Med. 2006;81:154–160.

28. Rothman DJ. Academic medical centers and financial conflicts of interest. JAMA. 2008;299:695–697.

29. Pauly M, Redisch M. The not-for-profit hospital as a physicians’ cooperative. Am Econ Rev. 1973;63:87–99.

30. Montgomery K, Schneller ES. Hospitals’ strategies for orchestrating selection of physician preference items. Milbank Q. 2007;85:307–335.

31. Stevenson FA, Barry CA, Britten N, Barber N, Bradley CP. Doctor-patient communication about drugs: the evidence for shared decision making. Soc Sci Med. 2000;50:829–840.

32. Horton R. Surgical research or comic opera: questions but few answers. Lancet. 1996;347:984–985.