In clinical trials especially those events-driven trials, it is often of great interest to predict the timing of pre-specified landmark event accurately in order to prepare for the interim or final analyses. Both parametric and non-parametric approaches have been proposed in the literature to estimate the underlying survival functions which is the key to the prediction of future event times. However, the existing approaches are neither not applicable to double-blind clinical trials or assuming smooth survival functions which might not hold in real clinical settings. In this talk, a hybrid parametric and non-parametric approach is proposed to predict event times in double-blind clinical trials with time-to-event outcomes. A greedy algorithm is first developed to detect change points in survival functions. The survival function before the last change point is estimated non-parametrically and the tail distribution beyond the last change point is estimated parametrically. Numerical results show that the proposed approach provides accurate predictions for future event times.

Objectives

Leiomyomas (uterine fibroids) are benign tumors that are prevalent among women of reproductive age and affect approximately 80% of premenopausal women. Hysterectomy is one of the most common surgical procedures for removal of uterine fibroids. Laparoscopic hysterectomy (LH) has increasingly been used as an alternative to the open surgical hysterectomy (OSH). The purpose of this study was to identify predictors of LH procedures among leiomyoma patients.

Methods

The National Inpatient Sample (NIS) is the largest publicly available inpatient healthcare database in the US, containing a 20% stratified systematic random sample of all US community hospital discharges. Thirteen years of available data, 2002–2014, were used for this analysis. International Classification of Diseases 9th revision (ICD-9) codes were used to identify all records of women with primary discharge diagnosis of uterine leiomyoma who had undergone either LH or OSH. Multivariate logistic regression analysis was conducted to identify significant predictors of LH. Population sampling weights were used to extrapolate results to national estimates.

Results

There were 310,150 leiomyoma LH procedures from 2002 to 2014, corresponding to 22.5% of the total number of leiomyoma hysterectomy procedures. The mean age of patients undergoing LH was significantly higher than those undergoing OSH (45.69 years vs 43.29 years; p<0.001). After adjusting for covariates, significant predictors of LH procedures included age > 50 years (OR=3.18; 95% CI=3.57-4.07), hospital location in the West (OR=1.46; 95% CI=1.38-1.86), and hospital urban non-teaching status (OR=1.20; 95% CI=1.04-1.38).

Conclusions

The study shows that despite the advances in surgical technologies, there are very few predictors of non-conventional laparoscopic approach to leiomyoma hysterectomy. Our findings suggest that there is a differential access to LH procedures among the US population which may be, in part, a result of racial and socioeconomic differences and centralization of laparoscopy services in urban areas.

Objectives

To quantify the approval, denial, and withdrawal rates and identify any predictors of success or failure for all new technology add-on payment (NTAP) applications from FY 2003 to FY 2018 in the United States.

Methods

The Center for Medicare and Medicaid (CMS) releases inpatient payment methodology rulemaking annually in the Federal Register, including details of NTAP submissions. The proposed and final rulemaking documents were analyzed to quantify the approval, denial, and withdrawal rates of all applications and determine primary reasons for denial or withdrawal from FY 2003 to FY 2018. Raw data were coded to further examine any predictors of application success such as product type, therapeutic category, manufacturer type, reapplication status, and proposed rule determination.

Results

There were 95 NTAP applications submitted over the last 15 fiscal years. Approximately 30%, 25%, and 45% of applications were approved, withdrawn prior to final rule, or denied, respectively. Inability to meet the “newness criteria” developed by CMS was the primary reason for denied and withdrawn applications. Product type, therapeutic category, and reapplication status have minor to significant impact on the approval rate of an application. However, manufacturer type and proposed rule determination have little to no impact on application outcome.

Conclusions

While there are a few factors that may positively influence the outcome of a NTAP application, the approval rates for the program are low overall. Without additional reimbursement from the NTAP program, inpatient hospitals may be deterred from adopting innovative therapies because of financial burdens. CMS and manufacturers should strive to find a better consensus for a framework that adequately incentivizes the utilization of new technologies for Medicare beneficiaries.

Abstract 00357: Open surgical repair has historically been the only corrective procedure for mitral regurgitation. There are few contemporary data examining evolving patient characteristics and surgical outcomes of mitral valve surgery. We sought to characterize trends in patient characteristics and outcomes after isolated mitral valve repair surgery over the past decade in the United States

Abstract 00165: Mitral intervention has historically been open surgical and has been limited to a small subset of patients who could tolerate surgery and where there was perceived clinical benefit. With the emergence of transcatheter mitral interventions, we sought to compare the types of patients undergoing surgical and transcatheter mitral interventions. Moreover, we sought to determine if there were any major differences in adverse events, length of stay (LOS), and cost to the healthcare system

Objectives

Percutaneous coronary intervention (PCI) is a common major medical procedure in the United States. It has previously been estimated there are 600,000–1,000,000 PCIs performed annually, although these results are based on older data. The aim of this study was to accurately estimate the number of PCI procedures and describe potential trends in PCI use during 2010 to 2013.

Methods

The National Inpatient Sample (NIS) is the largest publicly available all-payer inpatient healthcare database in the United States, containing a 20% stratified systematic random sample of discharges from all US community hospital discharges. The most recent 4 years of available data, 2010–2013, were used for this analysis. International Classification of Diseases 9th revision (ICD-9) codes were used to identify patients undergoing PCI, as well as to segment high-risk patients based on certain diagnoses. Population sampling weights were used to extrapolate results to national estimates.

Results

There were 559,219 PCI procedures in 2010 decreasing to 519,100 in 2013, corresponding to a change of -7.2% (p<0.0001). The corresponding rate of PCIs per 10,000 population was 18.08 in 2010 and 16.40 in 2013. Despite the overall decrease in PCI volume, procedures among high-risk patients increased during this time period. PCI procedures among patients with cardiogenic shock increased from 19,932 in 2010 (3.56% of all PCIs) to 22,685 in 2013 (4.37%) and procedures among patients with left ventricular heart failure increased from 40,417 (7.23%) in 2010 to 59,110 (11.39%) in 2013. CONCLUSIONS: This study shows that the volume of PCIs in the United States has decreased in recent years, and is significantly lower than previous estimates. Despite the overall decrease in PCI volume, procedures among patients with high-risk characteristics increased during the same time period. This suggests a potential shift in the application of PCI to more severe patients.

Chronic kidney disease (CKD) is progressive condition characterizing the gradual loss of kidney function. Patients with CKD are at high risk for hyperkalemia, which can cause severe cardiac electrophysiology alterations and sudden death. The objective of this study was to describe the prevalence of CKD among US adults and identify predictors of hyperkalemia among CKD patients during the time period 2007 to 2014.

Introduction

Valvular heart disease (VHDs), of which the mitral valve is frequently implicated, are collectively responsible for more than 20,000 annual deaths in the United States and the economic burden on the public health system due to VHDs is estimated in the billions of dollars.

Objective

The aim of this study was to assess the number of mitral valve disease hospitalizations and surgical procedures in the United States during the period 2010-2013.

Materials and Methods

Patient data were extracted from the Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample (NIS). The NIS is the largest publicly available all-payer inpatient healthcare database in the United States, containing a 20% stratified systematic random sample of discharges from all US community hospitals. The most recent 4 years of available data, 2010-2013, were used for this analysis. Patients diagnosed with mitral valve disease were identified using the appropriate International Classification of Diseases 9th revision (ICD-9) diagnosis codes. ICD-9 procedure codes were used to identify surgeries to repair or replace the mitral valve, including transcatheter mitral valve repair (TMVR), open repair, tissue graft, and synthetic replacement. Categorical variables were analyzed using the chi-squared test, while continuous variables were analyzed using a one-way analysis of variance (ANOVA) or independent samples t-test.

Results

There were 174,2587 discharge records in 2013 with a diagnosis of mitral valve disease. Using HCUP’s sampling weights, this extrapolates to 871,290 hospitalizations in the United States. This is a slight decrease from 2010, when there were 883,933 estimated hospitalizations. The mean age of patients hospitalized for mitral valve disease increased from 71.15 (95% CI: 71.08-71.23) years in 2010 to 71.68 (95% CI: 71.61-71.76) years in 2013 (p<0.0001). 61.7% of mitral valve hospitalizations were females in 2010 compared with 59.2% in 2013 (p<0.0001). Mitral valve surgical procedures increased from 35,111 procedures in 2010 to 38,030 procedures in 2013. Concurrently, the proportion of mitral valve disease patients that underwent a surgical procedure to repair or replace the mitral valve increased from 3.5% in 2010 to 3.9% in 2013 (p<0.0001). Among elderly patients ≥80 years of age, mitral valve disease represented an increasing proportion of all hospitalizations, increasing from 6.0% in 2010 to 6.6% in 2013. Patients with a primary diagnosis of mitral valve disease who underwent a mitral valve surgical procedure were younger on average than those with a primary diagnosis of mitral valve disease who did not undergo a surgical procedure: 64.50 (95% CI: 64.31-64.69) years vs. 70.73 (95% CI: 70.51-70.95) years (p<0.0001). Conclusions This study reveals that mitral valve disease is a considerable public health problem in the United States, particularly among older adults. During 2010-2013, an increasing proportion of hospitalized patients with mitral valve disease underwent surgical procedures to repair or replace the valve. Overall, the volume of mitral valve surgical procedures increased approximately 8.3% from 2010 to 2013.

Introduction

Cardiovascular disease (CVD) is the leading cause of death in the United States. Aspirin therapy has been shown to be an effective prevention measure to reduce the risk of new (primary) or recurring (secondary) cardiovascular events.

Objective

The aim of this study was to provide an epidemiological analysis of the use of low-dose aspirin for primary and secondary CVD prevention from 2012–2014.

Materials and Methods

The data source was the National Health Interview Survey (NHIS), a multipurpose survey conducted by the Centers for Disease Control and Prevention (CDC). It is the principal source of information on the health of the civilian, oninstitutionalized population of the United States. Estimates of self-reported low-dose aspirin use for primary and secondary CVD prevention among adults ≥40 years of age were obtained from the NHIS for the years 2012–2014. Demographics and health characteristics data were used to analyze intergroup differences for the combined time period, as well as intragroup differences from year to year. Categorical variables were analyzed using the chi-squared test, while continuous variables were analyzed using a oneway analysis of variance (ANOVA) or independent samples t-test.

Results

Among adults ≥40 years of age during 2012-2014, 18.7% selfreported as taking aspirin for primary CVD prevention and 8.9% self-reported as taking aspirin for secondary CVD prevention. During the full time period of 2012-2014, females represented 55.1% of patients taking aspirin for primary CVD prevention, but 45.3% of patients taking aspirin for secondary CVD prevention (p<0.0001). Adults taking aspirin for secondary CVD prevention were significantly older on average than those taking aspirin for primary CVD prevention (68.5 ± 11.1 vs. 65.8 ± 11.2 years; p<0.0001). The proportion of adults taking aspirin for primary CVD prevention significantly increased from 18.3% in 2012 to 19.4% in 2014 (p=0.003), while there was no statistically significant change in the proportion taking aspirin for secondary CVD prevention. Compared with adults taking aspirin for primary CVD prevention, those taking aspirin for secondary CVD prevention had a significantly higher prevalence of comorbidities such as cancer (22.0% vs. 17.4%; p<0.0001), diabetes (43.5% vs. 33.2%; p<0.0001), hypertension (67.1% vs. 52.9%; p<0.0001), and high cholesterol (56.2% vs. 44.2%; p<0.0001). Conclusions This study shows that over 25% of the population over 40 years of age in the United States self-reports as taking lowdose aspirin for primary or secondary CVD prevention, with primary CVD prevention patients outnumbering secondary CVD prevention patients at a ratio of more than 2:1. Aspirin use for primary CVD prevention increased throughout the study period. While there was no statistically significant change in aspirin use for secondary CVD prevention, the trend appeared to be decreasing during the study period.

Background

Molecular testing is central to the management of many cancers. We measured the extent to which FDA-approved and lab-developed tests (LDTs) are used in stage IV adenocarcinoma non-small cell lung cancer (NSCLC) patients, with a focus on EGFR mutations and ALKrearrangements.

Methods

A retrospective chart review of NSCLC patients was conducted between April and September of 2015. A representative sample of US oncologists was asked to complete an online form containing fields for anonymized patient information, including year of diagnosis, whether molecular testing was performed, and if so, which markers were tested, type of test used, and test location. 8,129 NSCLC-treating oncologists in the US were invited to participate in the study. Oncologists were eligible to participate if they personally managed at least 5 NSCLC patients per month, and diagnosed at least one NSCLC patients in the past 12 months

Results

Information on 765 patient records was obtained from the 153 physicians who met eligibility criteria. Histological subtype split was as follows: 76% of patients had adenocarcinoma, 19% had squamous cell carcinoma and 5% had other types. Among the 579 patients with adenocarcinoma, 95% and 84% were tested for EGFR mutations and ALK-rearrangements, respectively. Sample was further reduced to only include patients for which the type of test was known, and whose diagnostic was performed after FDA approval of the relevant test (May 14, 2013 for EGFR and August 26, 2011 for ALK), resulting in 424 patients for EGFR, and 399 for ALK. 87% of adenocarcinoma patients were tested with non-FDA-approved tests for EGFR, and 49% for ALK.

Conclusions

This study highlights the prevalence of non-FDA-approved testing in the treatment of lung cancer. Further evaluation of the comparability of approved and unapproved tests currently being used interchangeably is needed to mitigate uncharacterized variability between tests.