Dr. Steven Mattheos reported sharp increases in the volume of imaging services and subsequent expenditures attributing them to the advent of entrepreneurial activity by physicians, the practice of defensive medicine to thwart malpractice suits, and the patients who demand more tests. He commented on the potential conflicts of interests and the ability of surgeons, radiologist and third parties to bill to different aspects of the same service. To determine the appropriateness of self-referral and ordering habits, Dr. Mattheos reviewed 119 medical records of 10 orthopedic surgeons in five separate offices in a community, non-academic setting in eastern Massachusetts. The compilation of the orthopedic surgeons included General orthopedic practitioners, as well as specialists in Sports, Shoulder, Spine, and Hand. MRI’s ordered included the knee (34), shoulder (34), spine (29), ankle/foot (9), hip (4), elbow (3), wrist (3), and leg (3). These charts were reviewed independently by three reviewers and were evaluated according to the Appropriateness Score Scale, a measure developed for this study. Dr. Mattheos reported that the No Stake group had an average Appropriateness Score of 2.98 (standard deviation of 1.00) and that the Stake group had an average Appropriateness Score of 3.68 (standard deviation of 1.05, p <0.001). Although it is commonly believed that self-referral leads to an increase in the ordering of “inappropriate” studies, Dr. Mattheos’ findings suggest that this may not be the case in Massachusetts. Dr. Mattheos, along with the audience, speculated that this may be due to the surgeons in the stake group who take time to meticulously document indications for imaging believing that they have to justify their self-referral.

Dr. Craig A. Rineer commented on current classification systems failure to incorporate the essence of the entire injury pattern and suggested that lack of radial head cortical contact as a surrogate of instability. He believed that among partial fractures displaced greater than 2 millimeters (Broberg and Morrey Modified Mason type 2), complete loss of cortical contact of at least one radial head fracture fragment is predictive of a complex injury pattern. Dr. Rineer used billing records to identify 298 consecutive skeletally mature patients with radial head fractures treated over a 6-year period. He reviewed medical records for gender, age, ipsilateral upper extremity fractures or dislocations, overall fracture pattern, and Mason type. He classified fractures based on whether or not there was complete lack of cortical contact between a fracture fragment and the rest of the proximal radius. Predictors of isolated vs. complex injury patterns among the Broberg and Morrey Modified Mason type 2 fractures were determined in univariate and multivariable analyses. Dr. Rineer found that 133 fractures (45%) were Mason type 1, 65 (22%) were Mason type 2, and 100 (34%) were Mason type 3. He found that 163 fractures (55%) were classified as having cortical contact and 135 (45%) were classified as not having cortical contact. Nineteen of 163 fractures (12%) with cortical contact were part of a complex elbow injury while 127 of 135 fractures (94%) without cortical contact were part of a complex elbow injury (p <0.001). In his multivariable analysis among the Broberg and Morrey Modified Mason type 2 fractures, Dr. Rineer found that loss of cortical contact was the only significant predictor of a complex injury pattern. Dr. Rineer concluded that among Broberg and Morrey Modified Mason type 2 fractures, complete loss of cortical contact of at least one fracture fragment is predictive of a complex injury pattern.

Focusing on the advent of pay-for-performance medicine, Dr. Nina Shervin set out to validate a new version of the most widely used outcome assessment tool in total hip arthroplasty, the Harris Hip Score (HHS). Currently, the HHS has largely remains a cumbersome physician-administered paper-based tool and as practice demands increase, physicians will rely on self-administered touch screen and web-based computer formats for efficiency. Dr. Shervin used an IRB-approved prospective protocol to recruit patients during their office visit and pre-operative educational class visits. Dr. Shervin enrolled 69 patients and 8 were excluded due to incomplete data. Dr. Shervin found no significant differences in the remaining 61 patients who completed the doctor administered, self-administered touch screen and web-based computer forms of the questionnaire (p>0.13). Analysis revealed a moderately strong association between all three methods. She found the web-based modality to have the least variability. Dr. Shervin’s study drew several comments from the audience undoubtedly in part due to the fact that validation of new outcome measures today has the potential to become the basis of tomorrow’s reimbursement.