Faculty Bibliography

The Centers for Medicare & Medicaid Services is committed to moving 50% of its fee-for-service care to value-based alternative payment models by 2018. The Comprehensive Care for Joint Replacement model is a mandatory agency program that bundles lower extremity joint arthroplasties into episodes of care that extend from the index admission to 90 days after discharge. This program, which began on April 1, 2016, includes many of the hospitals that perform total joint arthroplasties. As with other bundled payment arrangements, this model is built around seven principles that orthopaedic surgeons should be familiar with to maximize participation.

BACKGROUND:Health disparities exist among many patient populations, with race, payer status, hospital size and access to teaching versus non-teaching hospitals potentially affecting whether certain patients have access to the benefits of total wrist arthroplasty (TWA). METHODS:The National Inpatient Sample Database (NIS) was queried from 2001 to 2013 for TWA using the ICD-9 code 81.73. Patient-level data included age, sex, race, payer status, and year of discharge. Hospital-level data included hospital bed size, location, teaching status, and region. RESULTS:There were 1,213 patients identified who underwent TWA between 2001 and 2013. Total number of procedures decreased from 88 TWAs in 2001 to 65 in 2013. The yearly volume ranged from 33 in 2005 to 128 in 2007. The male-female ratio was 2.5 to 1. The majority of TWA procedures were performed at urban teaching hospitals (60.8%). CONCLUSIONS:The NIS database shows a downward trend of total wrist arthroplasty utilization. The majority of total wrist arthroplasties were performed at urban teaching hospitals indicating treatment occurs most often at academic centers of excellence.

BACKGROUND:We sought to investigate how patient demographics and baseline comorbidities varied between hip fracture and total joint arthroplasty patients across New York State and to determine implications of differences within the contexts of the bundled payment system. METHODS:All Medicare hip and knee arthroplasty and hip fracture cases in the New York State SPARCS database between 2004 and 2014 were identified. Hospitals were categorized geographically into Metropolitan Statistical Areas (MSAs) to determine case distribution. Baseline comorbidities and patient characteristics were stratified. RESULTS:A total of 218,300 cases were identified; 187,720 arthroplasties and 30,580 hip fractures. The distribution of total cases was significantly skewed toward large MSAs and there was wide variability in the arthroplasty/fracture ratio. Despite similar baseline patient age and gender distributions, there were significant inconsistencies observed in comorbidity burden and length of stay between MSAs. While every MSA had higher average comorbidity scores and longer average lengths of stay in the hip fracture cases than the arthroplasty cases, there was also additional variability observed in the hip fracture cases between each MSA. CONCLUSION/CONCLUSIONS:The unpredictability observed in hip fracture cases compared to elective arthroplasty can make cost containment in a bundled payment system difficult, particularly for hospitals serving sicker patient populations.

Accuracy of component and limb alignment are critical parameters for the long-term success of unicompartmental knee implants. In this study, we performed a laboratory evaluation of an instrumentation system which was designed for an early intervention (EI) type of unicompartmental knee. The accuracy of fit was evaluated by implanting in 20 sawbones full leg models. The overall alignment of the limb was compared pre- and postoperatively. The accuracy of placement of each component on its bone was measured. The mean overall alignment angle in the frontal plane was within 1° of target with less than 1° standard deviation. The components were positioned in frontal and sagittal planes with maximum errors of 2°. The angular accuracy was better than in studies reported in the literature for manual instruments, and almost approached the accuracy of computer-assisted systems. The position of the femoral component in the recess was within 1 mm in most cases but the sagittal flexion angle was variable with a standard deviation of 6°. Evaluation of a surgical technique in this way was a valuable method for determining accuracy and for highlighting any deficiencies in the system which could then be corrected.

In recent years, narcotics have been subject to increased regulation and monitoring because of their side effects and potential for misuse. Currently, variation in prescribing patterns of narcotics among orthopedic surgeons is unknown. The Medicare Part D claims database was used to identify orthopedic surgeons who prescribed at least one schedule II or III narcotic during 2014. The median duration of a narcotic prescription was 8.2 days. The median prescription duration was shortest for hand surgeons (5.6 days) and longest for spine surgeons (12.6 days). Orthopedic surgeons in New York (10.1 days) provided the most narcotics per prescription, with physicians in Vermont (6.2 days) providing the least. Substantial variation exists in narcotic prescribing patterns for orthopedic surgeons at the individual, subspecialty, and statewide levels. With public health focus on reducing narcotics abuse, physician stewardship of these medications will become increasingly relevant.

BACKGROUND:The purpose of this study was to examine the geographic and demographic variations and time trends of different types of meniscal procedures in New York State to determine whether disparities exist in access to treatment. METHODS:The New York Statewide Planning and Research Cooperative System (SPARCS) outpatient database was reviewed to identify patients who underwent elective, primary knee arthroscopy between January 1, 2003, and December 31, 2015, for 1 of the following diagnosis-related categories: Group 1, meniscectomy; Group 2, meniscal repair; and Group 3, meniscal allograft transplantation, with or without anterior cruciate ligament reconstruction (ACLR). The 3 groups of meniscal procedures were compared on geographic distribution, patient age, insurance, concomitant ACLR, and surgeon and hospital volume over the study period. RESULTS:A total of 649,470 patients who underwent knee arthroscopy between 2003 and 2015 were identified for analysis. Both meniscectomies and meniscal repairs had a scattered distribution throughout New York State, with allograft volume concentrated at urban academic hospitals. The majority of patients who underwent any meniscal procedure had private insurance, with Medicaid patients having the lowest rates of meniscal surgery. At high-volume hospitals, meniscal repairs and allografts are being performed with increasing frequency, especially in patients <25 years of age. Meniscal repairs are being performed concomitantly with ACLR with increasing frequency. CONCLUSIONS:Meniscal repairs and allografts are being performed at high-volume hospitals for privately insured patients with increasing frequency. Geographic access to these treatments, particularly allografts, is limited. CLINICAL RELEVANCE/CONCLUSIONS:Disparities in the availability of advanced meniscal treatment require further investigation and understanding to improve access to care.

Introduction/objectives: In this study, we evaluate the application of preoperative PRO scores, such as the Hip dysfunction and Osteoarthritis Outcomes Score (HOOS) and EuroQol-5Dimension (EQ-5D), as potential predictive modelling tools to anticipate adverse in-hospital outcomes. Methods: Patients between the ages of 18 to 95 undergoing a primary THA between January 2015 and January 2017 at this institution were chart reviewed for inclusion in this study. 40% of our patient population completed preoperative PRO scores within 1 year of surgery and were included in this study. Nursing documentation was reviewed for patient demographics and in-hospital course metrics, such as visual analogue scale (VAS) for pain and morphine equivalence usages. EQ-5D was noted to be binomially distributed and subsequently transformed into a categorical variable with patients scoring >50% placed into a "high EQ-5D" group, and those scoring below into a "low EQ-5D" group. Results: In total, 349 patients including 157 males and 192 females were recruited for this study. The average age and body mass index (BMI) was 62.4+/-11.0 years and 28.6+/-5.61 kg/m2, respectively. The median American Society of Anesthesiology (ASA) Score within our patient cohort was 2. Of the pre-operative scores, age, BMI, EQ-5D, and HOOS section scores were compared with average daily pain, all but age were significantly correlated. However, these values had low r2 values <0.1, indicating poor predictive strength. Conclusion: Our study demonstrates that baseline PRO scores, such as the HOOS and EQ-5D, contain a small predictive component for in-hospital pain scores and average daily morphine. Furthermore, PRO tools can potentially be used to develop systematic, predictive risk stratification models

Introduction/objectives: As surgeons, we have a moral obligation to address potentially preventable complications in an effort to improve total hip arthroplasty (THA) outcomes. The goal of this study is to identify and report potentially preventable causes for revision THA (rTHA). Methods: A retrospective review of 352 consecutive patients that underwent rTHA or re-revision THA from August 2015 to August 2017 was conducted. 138 of these were identified as primary to rTHA within a 5 year interval. Two adult reconstruction fellowship trained surgeons reviewed perioperative parameters and classified rTHA recipients into two categories: preventable rTHA and nonpreventable rTHA. Basic demographics, surgical characteristics for the primary THA (pTHA), and pre- and post-rTHA variables were analyzed. Results: Sixty (43.5%) rTHAs were deemed preventable. Of these rTHA recipients, 20 were male and 40 were female. Mean age at time of rTHA was 61.5 years and mean body mass index was 27.8 kg/m2. The following were identified as preventable reasons for rTHA: cup malpositioning (70%), instability (53%), intra-operative fracture (40%), history of spinal surgery/deformity (22%), aseptic loosening (20%), femoral component subsidence (15%), and other (18%). The most common bearing surface during pTHA was cobalt chrome on highly crosslinked polyethylene (40%). The most common femoral head size was 36mm (38%). Technology was used for assistance in 8.3% of pTHAs. Four patients (6.7%) underwent re-revision THA, three for instability and one for aseptic loosening. Conclusion: A high proportion (43.4%) of rTHA is potentially preventable. Furthermore, surgeons are responsible for carefully evaluating causes for rTHA and identifying new methods to address these issues

BACKGROUND:The Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) is a nationally reported survey of patients' perspectives of hospital care that is used for hospital comparison and reimbursement. Although the survey attempts to correct for many factors that may affect scoring, socioeconomic factors are not considered in score weighting. The purpose of this study was to analyze the effect of socioeconomic status on HCAHPS scores. PATIENTS AND METHODS/METHODS:HCAHPS scores from 15,789 patients were collected. All patients were seen at a single academic medical center from 2010 to 2014, thus controlling for quality of care. HCAHPS Top Box scores were then compared to patient socioeconomic status based on the median income of the ZIP Code for each patient. RESULTS:Median income was negatively associated with patients' overall hospital rating (p < 0.001) and willingness to recommend hospital (p < 0.001). When controlling for the current adjustment factors (age, education, primary language, health status, and emergency admission), living in a ZIP Code with a median household income above $100,000 per year was independently associated with worse Top Box Scores for the categories of "Overall Hospital Rating" (p = 0.042), "Recommend Hospital" (p = 0.007), "Pain Management" (0.048), "Communication about Medicine" (p = 0.007), "Cleanliness of Hospital Environment" (p = 0.002), and "Quietness of Hospital Environment" (p < 0.001). CONCLUSION/CONCLUSIONS:Socioeconomic status independently affects HCAHPS scores. Patients living in ZIP Codes with lower median incomes generally rated hospitals better than patients with higher incomes. Therefore, treatment of a disproportionate number of low income patients cannot be cited as a pretext for poor HCAHPS scores.