Surgeons Consider Initial Nonoperative Treatment With Potential for Future Conversion to Reverse Arthroplasty a Reasonable Option for Older, Relatively Infirm, and Less-Active Patients

Abstract Statistical models support medical research by facilitating individualized outcome prognostication conditional on independent variables or by estimating effects of risk factors adjusted for covariates. Theory of statistical models is well‐established if the set of independent variables to consider is fixed and small. Hence, we can assume that effect estimates are unbiased and the usual methods for confidence interval estimation are valid. In routine work, however, it is not known a priori which covariates should be included in a model, and often we are confronted with the number of candidate variables in the range 10–30. This number is often too large to be considered in a statistical model. We provide an overview of various available variable selection methods that are based on significance or information criteria, penalized likelihood, the change‐in‐estimate criterion, background knowledge, or combinations thereof. These methods were usually developed in the context of a linear regression model and then transferred to more generalized linear models or models for censored survival data. Variable selection, in particular if used in explanatory modeling where effect estimates are of central interest, can compromise stability of a final model, unbiasedness of regression coefficients, and validity of p‐values or confidence intervals. Therefore, we give pragmatic recommendations for the practicing statistician on application of variable selection methods in general (low‐dimensional) modeling problems and on performing stability investigations and inference. We also propose some quantities based on resampling the entire variable selection process to be routinely reported by software packages offering automated variable selection algorithms.

The need for surgery for the majority of patients with displaced proximal humeral fractures is unclear, but its use is increasing.

the treatment of proximal humeral fractures in the elderly remains controversial. Options include nonoperative treatment, open reduction with internal fixation (ORIF), and hemiarthroplasty. Locking plate technology has expanded the indications for ORIF for certain fracture types in osteoporotic bone. This study was performed to characterize the incidence, treatment, and revision surgery of proximal humeral fractures according to geographic region both before (1999 to 2000) and after (2004 to 2005) the introduction of locking plates. we used a 20% sample of Medicare Part-B data and the Medicare denominator file for the years 1998 to 2006. Proximal humeral fractures were identified by Common Procedural Terminology codes for treatment, categorized as nonoperative, ORIF, or hemiarthroplasty. Geographic variation in treatment type was determined with use of 306 hospital referral regions. Odds ratios for revision surgery were calculated by the need for repeat surgery within one year of the index procedure. Rates were adjusted for age, sex, race, and comorbidities. there were 14,774 proximal humeral fractures in the 20% sample from 1999 to 2000 (an estimated total of 73,870 fractures) and 16,138 fractures in the sample from 2004 to 2005 (an estimated total of 80,690 fractures). The overall age, sex, and race-adjusted incidence of proximal humeral fractures was unchanged from 1999 to 2005 (2.47 vs. 2.48 per 1000 Medicare beneficiaries; p = 0.992). However, the absolute rate of surgically managed proximal humeral fractures rose 3.2 percentage points from 12.5% to 15.7%, a relative increase of 25.6% (p < 0.0001). The relative increase in the percentage of fractures treated with ORIF was 28.5% (p < 0.0001), while the percentage of fractures treated with hemiarthroplasty increased 19.6% (p < 0.0001). There were large regional variations in the proportion treated surgically (range, 0% to 68.18%). The rates of repeat surgery were significantly higher in 2004 to 2005 compared with 1999 to 2000 (odds ratio = 1.47, p = 0.043). although the incidence of proximal humeral fractures in the elderly did not change from 1999 to 2005, the rate of surgical treatment increased significantly. The marked regional variation in the rates of surgical treatment highlights the need for better consensus regarding optimal treatment of proximal humeral fractures. Additional research is needed to help to determine which fractures are best treated operatively in order to maximize outcome and minimize the need for revision surgery. therapeutic Level II. See Instructions to Authors for a complete description of levels of evidence.