Hospital-acquired pressure injuries (HAPI) are a societal burden and considered potentially preventable. Data on risk factors and HAPI burden are important for effective prevention initiatives. This study of the 2009-2014 US Premier Healthcare Database identified HAPI risk factors and compared outcomes after matching HAPI to non-HAPI patients. The cumulative incidence of HAPI was 0.28% (47 365 HAPI among 16 967 687 total adult inpatients). Among the matched sample of 110 808 patients (27 702 HAPI), the strongest risk factors for HAPI were prior PI (odds ratio [OR] = 12.52, 95% confidence interval [CI] = 11.93-13.15), prior diabetic foot ulcer (OR = 3.43, 95% CI = 3.20-3.68), and malnutrition (OR = 3.11, 95% CI = 3.02-3.20). HAPI patients had longer adjusted length of stay (3.7 days, P < .0001), higher total hospitalization cost ($8014, P < .0001), and greater odds of readmissions through 180 days (OR = 1.60, 95% CI = 1.55-1.65). This study demonstrates how big data may help quantify HAPI burden and improve internal hospital processes by identifying high-risk patients and informing best practices for prevention.
e16073 Background: North America has one of the highest rates of RCC, and in 2018, 65,340 new cases were diagnosed in the US alone. Tyrosine kinase inhibitors have been a mainstay treatment for RCC. However, the first CPI, nivolumab, was approved by the FDA as a second-line treatment for RCC in late 2015 and as a first-line therapy in combination with ipilimumab in 2018. In this study, real-world outcomes in patients with RCC receiving CPIs were examined. Methods: An observational study was conducted from March 1, 2015, to December 31, 2017, using the Premier Healthcare Database, a hospital discharge database. Patients with RCC aged ≥18 years were identified by ICD-9/10 codes and were included if they received a CPI during the study period, irrespective of treatment line. Comorbidities were assessed in a 6-month look-back period commencing at the time of first CPI treatment and RCC diagnosis. 34 distinct immune-related adverse events (irAEs) were identified in a 90-day look-back period from time of the irAE to determine if patients had received a CPI during that time. Patients were followed for 90 days after the irAE to determine if CPI treatment was reinitiated. Results: During the study period, 1228 patients with RCC received a CPI, of whom 719 (59%) had ≥1 irAE. Approximately 95% of patients received nivolumab and the remaining 5% received an off-label CPI. At the time of initial CPI treatment, patients who experienced any irAE during the study period had a higher Charlson Comorbidity Index (5.1±3.2 vs 4.2±3.0; p < 0.001) and were more likely to be hospitalized on an emergent basis than those without an irAE (9.0% vs 5.1%; p = 0.005). Patients with an irAE also had a higher average number of previous immune-related or immunocompromised comorbid conditions (0.33±0.57 vs 0.12±0.36; p < 0.001). After experiencing an irAE, 65% of patients reinitiated a CPI within 90 days. Conclusions: The majority of RCC patients receiving a CPI, mostly anti-PD-1 agents, experienced an irAE, and patients with irAEs had a higher comorbidity burden than those without irAEs. Most patients with irAEs reinitiated CPI treatment within 90 days.
Abstract Background A 6-valent vaccine (VLA15) is being tested in clinical trials for the prevention of Lyme disease caused by Borrelia burgdorferi sensu lato strains expressing OspA serotypes 1-6. Background incidence rates (IRs) of health outcomes in Lyme disease endemic and non-endemic regions of the US may help to contextualize whether the frequencies of events reported during vaccine clinical trials or post-marketing are consistent with expected population level rates. The objective of this study was to estimate and compare IRs of health outcomes in Lyme disease endemic vs. non-endemic regions using US administrative claims data. Methods IQVIA PharMetrics® Plus commercial claims database was used to estimate IRs of 63 outcomes relevant to vaccine safety monitoring in the US during 01/01/2017-12/31/2019. Endemic regions were classified using 3-digit zip codes that overlapped with Lyme disease high incidence counties (10 cases/100,000 persons) according to the CDC. Analyses included all individuals aged ≥ 2 years with ≥ 1 year of enrollment. Outcomes were defined by ICD-10-CM diagnosis codes according to the literature or expert input and required ≥ 1 inpatient or ≥ 2 outpatient claims/codes. Crude IRs and 95% confidence intervals (CIs) were calculated for each outcome and compared between endemic vs. non-endemic regions using IR ratios (IRR). Results The study population included 8.7M in endemic and 27.8M in non-endemic regions. Mean age was slightly higher in endemic (37.7 yrs [SD=18.9]) vs. non-endemic (36.8 yrs [SD=19.5]) cohorts, and 51% in both cohorts were female. Table 1 provides a summary of the IRs and IRRs for the 10 highest and 10 lowest ranking IRRs for health conditions by endemic region status. IRRs (95% CI) ranged from a low of 0.74 (0.71, 0.78) for systemic lupus erythematosus to a high of 2.14 (1.93, 2.37) for meningoencephalitis. Conclusion This study identified potential differences between Lyme endemic and non-endemic regions of the US in background IRs of health conditions in vaccine safety monitoring. Differences in background IRs between endemic and non-endemic regions should be considered when contextualizing possible safety signals in clinical trials and post-marketing. Disclosures Jill Dreyfus, PhD, MPH, Pfizer, Inc.: Employment|Pfizer, Inc.: Stocks/Bonds Swapna Munnangi, PhD, IQVIA: Employment Camilla Bengtsson, PhD, IQVIA: Employment|Pfizer, Inc: Advisor/Consultant Barbara Correia, PhD, IQVIA: Employee|Pfizer, Inc.: Advisor/Consultant Rejane Figueiredo, PhD, IQVIA: Biostatistician Sarah Galvin, BS, Pfizer, Inc.: Employment James H. Stark, PhD, Pfizer: Employee|Pfizer: Stocks/Bonds Michele Zawora, MD, FAAFP, Pfizer: Employment|Pfizer: Stocks/Bonds Mark Riddle, MD, DrPH, Pfizer: Employee salary Jason Maguire, MD, Pfizer, Inc.: Employee|Pfizer, Inc.: Stocks/Bonds Qin Jiang, PhD, Pfizer: Employee|Pfizer: Employee|Pfizer: Stocks/Bonds|Pfizer: Stocks/Bonds Juan Naredo Turrado, MS, IQVIA: Employee Steven Bailey, MD, MPH, MBA, Pfizer, Inc.: Employment
Introduction: Age at menarche is associated with cardiometabolic disease risk later in life; whether hepatic steatosis (fatty liver) or regional depots of abdominal fat play a role in this association is unclear. In a community-based sample of African American and White women free of diagnosed liver disease, we tested the hypothesis that age at menarche is inversely associated with hepatic steatosis and abdominal fat depots in midlife, after control for demographic and lifestyle factors and early adult BMI. Methods: Women in the Coronary Artery Risk Development in Young Adults (CARDIA) study with complete measures of recalled age at menarche (at exam years 0 (1985-86) and 2), multiple slice abdominal CT (at year 25) of hepatic steatosis (HS), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and intermuscular adipose tissue (IAT) were included. We defined nonalcoholic fatty liver disease as liver attenuation ≤ 48 Hounsfield units (HU) after excluding other causes of liver disease. We used multivariable linear regression (continuous outcomes) and Poisson regression (binary outcomes). Results: The mean (±sd) menarcheal age for the 1215 women eligible for this analysis was 12.6±1.5 y. A 1-year increment in age at menarche was associated with lower liver steatosis (0.8±0.2 HU; greater HU = less steatosis), VAT (-6.6±1.2 cc), IAT (-7.3±1.4 cc), SAT (-19.3±3.3 cc), and NAFLD prevalence (prevalence ratio=0.86; 95% CI: 0.78, 0.94) after confounder adjustment ( Table ). Associations remained after further adjustment for year 0 BMI (ages 18-30 y; all p for trend <0.05), but were attenuated (all p for trend >0.05) after adjustment for year 25 BMI (a mediator) when participants were 43-55 y. Conclusion: These findings indicate age at menarche is inversely associated with midlife liver steatosis and visceral and subcutaneous abdominal fat, independent of early, but not late, adult BMI. Whether these associations mediate the higher cardiometabolic disease risk observed in early maturing females needs to be determined.
Abstract Background Although Staphylococcus aureus is a leading cause of postsurgical infections, national estimates of these infections after elective surgeries based on microbiology data are limited. This study assessed cumulative 180-day postsurgical S. aureus incidence in real-world hospital settings. Methods A retrospective study of adults (≥18 years) undergoing inpatient or hospital-based outpatient elective surgeries from 1/7/2010–30/6/2015 at hospitals (N = 181) reporting microbiology results in the Premier Healthcare Database (PHD). 86 surgical categories were identified from the National Healthcare Safety Network procedures. We classified positive S. aureus cultures using a hierarchy (bloodstream [BSI], surgical site [SSI], and all other types [urinary tract, respiratory, other/unknown site]) and calculated incidence (number of infections divided by the number of elective surgery discharges). We estimated national infection case volumes by multiplying incidence by national inpatient elective surgical discharge estimates using the entire PHD and weights based on hospital characteristics. Results Following 884 803 inpatient elective surgical discharges, 180-day S. aureus infection incidence was 1.35% (0.30% BSI, 0.74% SSI no BSI, 0.32% all other types only). Among 1 116 994 hospital-based outpatient elective surgical discharges, 180-day S. aureus incidence was 1.19% (0.25% BSI, 0.75% SSI no BSI, 0.19% all other types only). Methicillin resistance was observed in ~45% of the S. aureus infections. We estimated 55 764 S. aureus postsurgical infections occurred annually in the US following 4.2 million elective inpatient surgical discharges. Conclusions The high burden of S. aureus infections after both inpatient and outpatient elective surgeries highlights the continued need for surveillance and novel infection prevention efforts.
Abstract Introduction A younger age at menarche and an older age at menopause are well established risk factors for breast cancer. Recent genome-wide association studies have identified several novel genetic loci associated with these two traits. However, the association between these loci and breast cancer risk is unknown. Methods In this study, we investigated 19 and 17 newly identified single nucleotide polymorphisms (SNPs) from the ReproGen Consortium that have been associated with age at menarche and age at natural menopause, respectively, and assessed their associations with breast cancer risk in 6 population-based studies among up to 3,683 breast cancer cases and 34,174 controls in white women of European ancestry. In addition, we used these SNPs to calculate genetic risk scores (GRSs) based on their associations with each trait. Results After adjusting for age and potential population stratification, two age at menarche associated SNPs (rs1079866 and rs7821178) and one age at natural menopause associated SNP (rs2517388) were associated with breast cancer risk (p values, 0.003, 0.009 and 0.023, respectively). The odds ratios for breast cancer corresponding to per-risk-allele were 1.14 (95% CI, 1.05 to 1.24), 1.08 (95% CI, 1.02 to 1.15) and 1.10 (95% CI, 1.01 to 1.20), respectively, and were in the direction predicted by their associations with age at menarche or age at natural menopause. These associations did not appear to be attenuated by further controlling for self-reported age at menarche, age at natural menopause, or known breast cancer susceptibility loci. Although we did not observe a statistically significant association between any GRS for reproductive aging and breast cancer risk, the 4 th and 5 th highest quintiles of the younger age at menarche GRS had odds ratios of 1.14 (95% CI, 1.01 to 1.28) and 1.13 (95% CI, 1.00 to 1.27), respectively, compared to the lowest quintile. Conclusions Our study suggests that three genetic variants, independent of their associations with age at menarche or age at natural menopause, were associated with breast cancer risk and may contribute modestly to breast cancer risk prediction; however, the combination of the 19 age at menarche or the 17 age at natural menopause associated SNPs did not appear to be useful for identifying a high risk subgroup for breast cancer.
Hospital-acquired bloodstream infections have a definite impact on patient encounters and cause increased length of stay, costs, and mortality. However, methods for estimating these effects are potentially biased, especially if the time of infection is not incorporated into the estimation strategy. We focused on matching patient encounters in which a hospital-acquired infection occurred to comparable encounters in which an infection did not occur. This matching strategy is susceptible to a selection bias because inpatients that stay longer in the hospital are more likely to acquire an infection and thus also are more likely to have longer and more costly stays. Instead, we have proposed risk-set matching, which matches infected encounters to similar encounters still at risk for infection at the corresponding time of infection. Matching on the one-dimensional propensity score can create comparable pairs for a large number of characteristics; an analogous propensity score is described for risk-set matching. We have presented dramatically different estimates using these 2 approaches with data from a pediatric cohort from the Premier Healthcare Database, United States, 2009-2016. The results suggest that estimates that did not incorporate time of infection exaggerated the impact of hospital-acquired infections with regard to attributed length of stay and costs.
