Search

One way to communicate skin cancer risk is through ultraviolet (UV) photographs, which can depict the target person (tailored visual) or someone else (stock visual). There is a need for more longitudinal research examining the relative impact of tailored UV photographs compared with other message interventions that could increase sun safe behaviors.Students 14-18 years of age (Compared with the implementation intervention, participants in the tailored UV condition reported increased fear and freedom threat and decreased appearance norms and benefits of tanning immediately following exposure to the intervention and decreased outdoor tanning 1 month after the intervention. Indirect effects also emerged with tailored UV exposure decreasing outdoor tanning via appearance benefits and increasing outdoor tanning when immediate fear triggered psychological reactance.The results contribute to research on lay reactions to tailored visuals, implementation interventions, and theorizing the indirect effects of affect and cognition across time. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Long-term mental health outcomes were characterized in patients who were diagnosed with Hodgkin lymphoma (HL), and risk factors for the development of mental health disorders were identified.Patients who were diagnosed with HL between 1997 and 2014 were identified in the Utah Cancer Registry. Each patient was matched with up to five individuals from a general population cohort identified within the Utah Population Database, a unique source of linked records that includes patient and demographic data.In total, 795 patients who had HL were matched with 3575 individuals from the general population. Compared with the general population, patients who had HL had a higher risk of any mental health diagnosis (hazard ratio, 1.77; 95% confidence interval, 1.57-2.00). Patients with HL had higher risks of anxiety, depression, substance-related disorders, and suicide and intentional self-inflicted injuries compared with the general population. The main risk factor associated with an increased risk of being diagnosed with mental health disorders was undergoing hematopoietic stem cell transplantation, with a hazard ratio of 2.06 (95% confidence interval, 1.53-2.76). The diagnosis of any mental health disorder among patients with HL was associated with a detrimental impact on overall survival; the 10-year overall survival rate was 70% in patients who had a mental health diagnosis compared with 86% in those patients without a mental health diagnosis (p .0001).Patients who had HL had an increased risk of various mental health disorders compared with a matched general population. The current data illustrate the importance of attention to mental health in HL survivorship, particularly for patients who undergo therapy with hematopoietic stem cell transplantation.

The development of an advanced robotic platform in 2014 led to increased adoption of minimally invasive (MI) approaches in thoracic surgery. Due to dataset reporting lag, a comprehensive assessment of trends in thoracic approaches has not been analyzed to date.We queried the National Cancer Database (NCDB) for patients with Stage I-III who underwent lung resection from 2010 to 2018. Most published NCDB analyses on lung cancer using pre-2015 data. Overall treatment trends were analyzed, with geographic, institutional, and socioeconomic characteristics evaluated for approach.There were 162 335 lung resections, and 131 958 were anatomic. Robotic resection saw a steady increase through 2012 but plateaued in 2013-2014. From 2015 to 2018, another increase correlated with the release of a new platform. Video-assisted thoracoscopic surgery lung resection plateaued in 2014 and decreased in 2018. Open resection steadily decreased. Tumors requiring neoadjuvant radiation had an increase in MI approach with corresponding decreases in the open. On multivariable analysis, African-American race, low volume, Medicaid insurance, and nonacademic setting were associated with a lower likelihood of MI surgery.The open approach has decreased since 2010. More than 65% of anatomic resections are now performed in MI. As this trend will continue, it is important that all patients are afforded the opportunity of the least invasive approach.

Objectives:The Core Elements of Outpatient Antibiotic Stewardship provides a framework to improve antibiotic use, but cost-effectiveness data on implementation of outpatient antibiotic stewardship interventions are limited. We evaluated the cost-effectiveness of Core Element implementation in the outpatient setting.Methods:An economic simulation model from the health-system perspective was developed for patients presenting to outpatient settings with uncomplicated acute respiratory tract infections (ARI). Effectiveness was measured as quality-adjusted life years (QALYs). Cost and utility parameters for antibiotic treatment, adverse drug events (ADEs), and healthcare utilization were obtained from the literature. Probabilities for antibiotic treatment and appropriateness, ADEs, hospitalization, and return ARI visits were estimated from 16,712 and 51,275 patient visits in intervention and control sites during the pre- and post-implementation periods, respectively. Data for materials and labor to perform the stewardship activities were used to estimate intervention cost. We performed a one-way and probabilistic sensitivity analysis (PSA) using 1,000,000 second-order Monte Carlo simulations on input parameters.Results:The proportion of ARI patient-visits with antibiotics prescribed in intervention sites was lower (62% vs 74%) and appropriate treatment higher (51% vs 41%) after implementation, compared to control sites. The estimated intervention cost over a 2-year period was $133,604 (2018 US dollars). The intervention had lower mean costs ($528 vs $565) and similar mean QALYs (0.869 vs 0.868) per patient compared to usual care. In the PSA, the intervention was dominant in 63% of iterations.Conclusions:Implementation of the CDC Core Elements in the outpatient setting was a cost-effective strategy.

ImportanceInappropriate use of antibiotics for diarrheal illness can result in adverse effects and increase in antimicrobial resistance.ObjectiveTo determine whether the diarrheal etiology prediction (DEP) algorithm, which uses patient-specific and location-specific features to estimate the probability that diarrhea etiology is exclusively viral, impacts antibiotic prescriptions in patients with acute diarrhea.Design, Setting, and ParticipantsA randomized crossover study was conducted to evaluate the DEP incorporated into a smartphone-based electronic clinical decision-support (eCDS) tool. The DEP calculated the probability of viral etiology of diarrhea, based on dynamic patient-specific and location-specific features. Physicians were randomized in the first 4-week study period to the intervention arm (eCDS with the DEP) or control arm (eCDS without the DEP), followed by a 1-week washout period before a subsequent 4-week crossover period. The study was conducted at 3 sites in Bangladesh from November 17, 2021, to January 21, 2021, and at 4 sites in Mali from January 6, 2021, to March 5, 2021. Eligible physicians were those who treated children with diarrhea. Eligible patients were children between ages 2 and 59 months with acute diarrhea and household access to a cell phone for follow-up.InterventionsUse of the eCDS with the DEP (intervention arm) vs use of the eCDS without the DEP (control arm).Main Outcomes and MeasuresThe primary outcome was the proportion of children prescribed an antibiotic.ResultsA total of 30 physician participants and 941 patient participants (57.1% male; median [IQR] age, 12 [8-18] months) were enrolled. There was no evidence of a difference in the proportion of children prescribed antibiotics by physicians using the DEP (risk difference [RD], −4.2%; 95% CI, −10.7% to 1.0%). In a post hoc analysis that accounted for the predicted probability of a viral-only etiology, there was a statistically significant difference in risk of antibiotic prescription between the DEP and control arms (RD, −0.056; 95% CI, −0.128 to −0.01). No known adverse effects of the DEP were detected at 10-day postdischarge.Conclusions and RelevanceUse of a tool that provides an estimate of etiological likelihood did not result in a significant change in overall antibiotic prescriptions. Post hoc analysis suggests that a higher predicted probability of viral etiology was linked to reductions in antibiotic use.Trial RegistrationClinicaltrials.gov Identifier: NCT04602676

Advances in immunotherapy including immune checkpoint inhibitors (ICIs) have transformed the standard of care of many cancers including melanoma. However, the benefit of ICIs is hampered by immune related adverse events (irAEs). The frequency of irAEs varies between ICI treatments and cancer types. These irAEs represent autoimmune reactions that can affect diverse organs, with distinct biology, onset, and severity from their de novo autoimmune disease counterparts. Some irAEs could persist and develop into chronic comorbidities, and a subset of severe irAEs could led to death. Yet, little is known about the underlying molecular mechanisms and no biomarkers are currently available to predict irAEs after ICI treatment. We aimed to identify soluble predictive biomarkers of irAE by non-invasive means using high-throughput Olink Proteomics assay. We collected plasma from 31 melanoma patients before ICI treatment. 25 patients developed severe irAE (grade 3 or above) at one or more organ system including hepatobiliary (n=9), endocrine (n=8), gastrointestinal (n=6), skin and subcutaneous tissue (n=6), musculoskeletal and connective tissue (n=5), blood and lymphatic (n=4) and respiratory (n=1) after receiving ICI, whereas 6 patients have no severe irAEs. The relative expression of 1,472 probes corresponding to 1,463 proteins as part of the Olink Cardiometabolic, Neurology, Inflammation and Oncology panels were quantified as Normalized Protein eXpresion (NPX) on a log-2 scale. Two-tailed T-test was performed in comparing the severe irAEs vs no irAE groups to determine the difference proteins and p < 0.05 and FDR < 0.1 was deemed to be significant. Similarly, we performed organ-specific irAEs using the same approach. Proteins were assessed by enrichment analysis using EnrichR. Exploratory analysis revealed elevated expression of CD4 and CD8 cell markers such as IL7R, S100A4, GZMA and GZMB in patients that developed severe irAEs. More broadly, pathway enrichment analysis suggest chemokines and cytokine receptor related pathways were enriched in these patients implicating the role of immune system and inflammatory mediators as potential biomarkers of irAE. In particular, the expression of some interleukin family members appears to be a good predictor of irAE toxicity. Interestingly, our analysis revealed irAE at various organ tissues were associated with different set of proteins suggesting an organ specific protein signature. In conclusion, we demonstrated the potential utility of soluble protein in the peripheral blood measurement at baseline prior to ICI treatment for prediction of irAE risk. This finding will be evaluated in a larger independent cohort. Citation Format: Muhammad Zaki Fadlullah Wilmot, Magdalena Kovacsovics, Xuechen Wang, Samuel Coleman, John Marsiglio, Berit Gibson, Yoko DeRose, Qin Zhou, Annaleah Larson, John Hyngstrom, Ben Haaland, Siwen Hu-Lieskovan, Aik Choon Tan. Proteomic study reveals predictive biomarkers of immune-related adverse events in melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2144.

237 Background: We performed plasma-based high-plex proteomic profiling for identifying classifiers of clinical outcomes in metastatic prostate cancer (PC). Olink Explore NGS-based proteome profiling platform was used for high-precision analysis of 736 cancer associated plasma proteins in plasma samples from non-metastatic stage prostate cancer (PC), metastatic hormone-sensitive PC (mHSPC) and metastatic castrate resistant PC (mCRPC) states. Methods: Plasma was collected prospectively in a cohort of 108 PC patients (24 with non-metastatic PC; 28 mHSPC; 56 mCRPC of which 37 patients were collected before starting any mCRPC treatments). Proteomic data were generated with Proximity Extension Assay (PEA) on the Olink platform from 100 µL plasma per sample. Levels of 736 cancer-associated protein assays were denoted as normalized protein expression (NPX) units through a QC and normalization process developed and provided by Olink. Data generation of NPX consists of normalization to the extension control, log2 -transformation, and level adjustment using the plate control (plasma sample). Temporal trends of differentially expressed assays in non-metastatic PC, mHSPC and mCRPC states were identified using linear mixed effects model (FDR with Benjamini-Hochberg (BH) adjustment; q-value

Genomic and morphologic heterogeneity in clear cell renal cell carcinoma (ccRCC) presents a barrier to prognostication and treatment decisions. Data from pathology are used with clinical markers to predict disease progression after nephrectomy. However, determining the risk of cancer recurrence, and survival with metastatic cancer remains challenging. Recently, analysis of histologic growth patterns (HGP) in ccRCC revealed promising associations with survival outcomes.To investigate whether HGPs can be used to predict overall survival (OS) after nephrectomy, we examined 24 HGPs in primary tumors of 147 patients that included 107 patients with metastatic disease.The median number of HGPs per case was 5 and was greater in metastatic and larger tumors. After adjustment for 6 pathologic and demographic variables, HGPs were significantly associated with OS post nephrectomy. Small nests, expansile nests and nests with high nuclear to cytoplasmic ratio were associated with favorable outcomes; while spindled low grade, fused nests/solid sheets, rhabdoid, and sarcomatoid patterns were associated with unfavorable outcomes. A 3-tiered and a 2-tiered risk model were developed based on combinations of HGPs. The models performed equally well as WHO/ISUP nucleolar plus necrosis grade (necrosis grade), and better than WHO/ISUP nucleolar grade alone in predicting OS at the median OS of 6 years. Pairwise correlations between HGPs revealed 2 tumor evolutionary branches that differed in risk of metastatic disease: one with mesenchymal differentiation, and other with epithelial tubulopapillary differentiation. While 44 of 107 (41%) patients with metastatic ccRCC displayed evidence of mesenchymal differentiation, mesenchymal features were only observed in 1 of 40 (3%) patients without evidence of metastatic disease.These findings suggest that HGPs may provide a novel path to refine the estimation of OS after nephrectomy and to determine the molecular basis of tumor evolution.

218 Background: Adolescents and young adults (AYAs; ages 18-39) with cancer report worries about costs and feeling uninformed about their insurance, which may impair their ability to navigate the health care system, use their insurance, and manage costs. We conducted a pilot randomized controlled trial (RCT) to assess a virtually delivered health insurance navigation intervention (“CHAT”) to improve health insurance literacy (HIL) and knowledge, increase awareness of insurance protections in the Affordable Care Act (ACA), and reduce financial toxicity and stress. Methods: CHAT is a 4-session program delivered by a patient navigator. Session content included learning about insurance, navigating your plan, insurance-related laws, & managing care costs. Eligible participants had access to a wireless device, were within their first year of cancer diagnosis, and were treated at oncology sites from the University of Utah Healthcare and Intermountain Healthcare systems. We assessed feasibility and preliminary efficacy of CHAT compared to usual navigation care, including HIL (9 items on confidence using insurance), insurance knowledge (13 items on terms such as deductible & copay), ACA protections (8 items), Perceived Stress Scale (PSS; 4 items), and COmprehensive Score for financial Toxicity (COST; 11 items) using t-tests. Results: From November 2020 to January 2022, 86 AYAs enrolled (46.2% approached consented); 68.6% were female, 72.1% white, 18.6% Hispanic, 34.9% age 18-25, 65.1% age 26-39, 83.7% had private insurance, and 10.5% public. Mean baseline HIL scores were 22.6 (range = 9-33, SD = 5.5). At baseline, 65.1% of participants were unaware of the ACA provision allowing appeals for coverage denials. 64.4% of intervention participants completed all 4 sessions; 86.0% completed the 5-month follow-up survey with data collection ongoing (74/86). At follow-up, compared to usual navigation care, CHAT participants had improvements in HIL, knowledge of insurance, and knowledge of ACA provisions (Table). While PSS improved more for CHAT than usual care, COST did not differ. Conclusions: Results support the feasibility and preliminary efficacy of CHAT with related improvements in HIL and knowledge of the ACA. Clinical trial information: NCT04448678. [Table: see text]

15 Background: We previously demonstrated that Huntsman at Home (HH), a novel oncology hospital at home program, was associated with reduced healthcare utilization and costs. HH was also linked to shorter hospital stays and fewer emergency room (ER) visits. In this study, we sought to understand the impact of HH in specific patient subgroups. Methods: We compared outcomes among 169 patients consecutively admitted to HH against 198 usual care patients. Five dichotomous subgroups were created based upon patient a) sex b) age c) area level median income d) Charlson Comorbidity Index (CCI), and e) current use of systemic therapy (ST). Outcomes included 30-day costs, unplanned hospitalizations (UH), length of hospital stays, and ER visits. HH and usual care were compared via inverse propensity weighted regression models. Treatment propensities were estimated via random forests based on age, race, stage, cancer site, presence of metastases, CCI, and area level median income.Results: The between group difference favoring HH achieved statistical significance (p < 0.05) for at least two out of the four outcomes in each subgroup except for patients with higher comorbid illnesses. While HH participants did not always experience statistically better outcomes than usual care, none of the outcomes examined favored usual care for any subgroup. Sex. Female and male HH patients experienced fewer UH and lower costs than usual care. Male HH patients also had shorter hospital stays and fewer ED visits. Age. When stratifying age at 65 years, older HH patients experienced fewer days in the hospital and fewer UH. Younger HH patients had lower costs, and fewer UH and ED visits. Area level Income. All outcomes were better for high and low income HH patients compared to usual care except for ED visits among those with low income. CCI. Among those with a low CCI score, all four outcomes were better among HH patients. In contrast, differences between groups with higher comorbid illness did not achieve statistical significance for any outcome. Systemic Therapy. HH participants on ST experienced shorter hospital stays and fewer UH compared to usual care. Among those not on ST, HH patients experienced lower costs, and fewer UH and ED admissions. Conclusions: In this exploratory analysis, we found that the utilization and cost benefit associated with HH was robust, favoring better outcomes in each subgroup including lower 30-day costs, shorter hospital length of stay and fewer unplanned hospitalizations or ER visits. While medically complex patients may not receive similar benefit from HH as other subgroups, no outcomes favored patients managed by usual care. Taken together, this suggests that health care utilization and cost reductions associated with HH occur across multiple subgroups, but patients with high comorbidity may require additional intervention to realize lower utilization and costs.

377 Background: Among patients diagnosed with mNSCLC, significant declines in fat and muscle mass are common, yet there is a paucity of data related to body composition, QoL, TD, and survival. Accordingly, we evaluated the relationship between body composition via routine CT scans, patient-reported outcomes (PROs), and overall survival (OS) among a convenience sample of mNSCLC patients. Methods: Data from 80 mNSCLC patients with initial CT scans and NCI PROMIS questionnaires within the first three months of diagnosis were analyzed. Body composition from CT scans (sliceOmatic software) extracted Skeletal muscle (SM), intermuscular adipose tissue (IAT), visceral VAT, and subcutaneous (SAT) in area (cm2) and a discovery set was expressed as HU. PROMIS PROs (pain, fatigue, anxiety, depression, and physical function) were collected in clinic using an iPad as an institutional quality initiative. Results: Median time to OS was 16 months. When stratified by sex, females had longer median survival time (female 25 months, male 14 months). When all body composition variables were considered together, greater amounts of skeletal muscle were linked with a 63% reduction in mortality risk 0.37 (95% CI 0.16, 0.87) in the adjusted model. Per one-unit increase in VAT (area cm2) was linked with a 0.04 unit decrease in pain (95% CI -0.08, -0.01). Associations between HU and clinical were also observed. Greater amount of VAT (mean HU) was linked with 2.57(95% CI 1.10, 6.01) times the odds of death in the crude model. Also, in crude model, one unit of increase in skeletal muscle (mean HU) was associated with 0.3 (95% CI 0.03, 0.57) unit increase in physical functionality, and -0.34(95% CI -0.63, -0.05)) unit decrease in fatigue. Conclusions: Body composition data can be collected from retrospective scans and appears to be prognostic of OS, PRO and TD.

Objective: High-throughput metabolomics assays can generate thousands of biomarker measurements and provide novel opportunities for prognostic modeling in colorectal cancer (CRC) research. The high dimensionality of metabolic data brings unforeseen statistical challenges and traditional variable selection methods may not perform well due to simultaneous challenges of computational expediency, statistical accuracy, and algorithmic stability. The Iterative Sure Independence Screening (ISIS) has demonstrated superior theoretical properties in handling such situations, and may be a viable alternative. Methods: In a prospective study of 77 newly diagnosed CRC patients (stage I-IV), pre-surgical urinary samples were analyzed on a gas chromatography-mass spectrometry platform. After exclusion of metabolites with >30% coefficient of variation, 168 metabolites remained for statistical analysis. Raw measures were processed following a standard normalization pipeline. The primary outcome was overall survival (OS) as measured from date of cancer diagnosis. In addition to metabolomics data, the predictor set included baseline clinical characteristics, such as age, sex, body mass index, tumor site, tumor stage, and receipt of neo-adjuvant and/or adjuvant treatment. We applied the ISIS method with Lasso penalty on a Cox regression model (ISIS-Lasso) to identify features associated with OS. Cox models with either Lasso regularization or backward selection were also considered as competing methods. The performance of the models was assessed through two standard performance matrices: Uno's time-dependent Area Under the Curve (tAUC) and Brier's score, both with resample-based validation. Results: Based on bootstrapped tAUC curves, we demonstrated that the screening step in ISIS can significantly improve model performance, since its predicted mean (and median) AUC are larger across clinically relevant follow-up time points (2 and 5 years after diagnosis) relative to the corresponding measures from the other two models. The prediction error based on Brier's score with 0.632+ bootstrap from ISIS-Lasso was also noticeably lower compared to the model from backward selections. Based on the ISIS-Lasso model, we identified two features, that were predictive of OS in CRC patients: tumor stage and cystine. When fixing all other clinical measures, patients with early stage (I-III) had 52% lower risk of death, compared to late stage (IV); 1 standard deviation of increase of cystine level was associated with 62% increased risk of death. Conclusion: We have demonstrated the feasibility and effectiveness of an ISIS-based method to improve selection of prognostic models derived from metabolomics data. This may be especially useful for studies with moderate sample sizes. We have identified cystine as a potentially important prognostic biomarker. Citation Format: Tengda Lin, Biljana Gigic, Kenneth Boucher, Ben Haaland, David Liesenfeld, Robert Owen, Petra Schrotz-King, Jourgen Boehm, Anita Peoples, Augustin Scalbert, Martin Schneider, Jane Figueiredo, William Grady, Christopher Li, David Shibata, Erin Siegel, Adetunji Toriola, Alexis Ulrich, Neli Ulrich, Jincheng Shen, Jennifer Ose. Application of iterative sureindependence screening to improve urinary metabolomics-based prediction of survival in colorectal cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 892.

Background We report VA-wide trends over time in acute respiratory infection (ARI) antibiotic prescribing, and early assessment of an intervention to improve ARI management. Methods We created a retrospective cohort of ARI (sinusitis, pharyngitis, bronchitis, and URI-NOS) visits between 2009 and April 2018. Patients with complicating conditions were excluded. Antibiotic prescribing rates were calculated. A provider-directed VA-wide ARI campaign was initiated in October 2017. The Campaign was implemented locally by antibiotic stewards or regional personnel trained in academic detailing (AD). Campaign components: dashboards for tracking provider and facility prescribing metrics, printable feedback reports, and AD educational materials. Metrics include: ARI antibiotic prescribing rates, bronchitis/URI-NOS antibiotic prescribing rates, guideline-concordant antibiotic selection for sinusitis or pharyngitis, and proportion of ARI visits with a sinusitis diagnosis. A Logistic generalized estimating equation model assessed metrics over time pre-/postintervention and χ2 tests compared guideline concordant antibiotic proportions pre-/postintervention. Results There were 1,580,612 and 137,421 ARI visits pre-/postintervention, respectively. Antibiotic prescribing decreased from 2009, annual odds ratio (OR) 0.94 [95% CI 0.93, 0.96; P < 0.001]. An additional effect was observed postintervention [OR 0.88, (0.84, 0.88), P < 0.001]. Bronchitis/URI-NOS prescribing rates decreased from 2009 [annual OR 0.94 (CI 0.93, 0.95), P < 0.001]. Additional effect was observed postintervention [OR 0.86, (0.81, 0.91), P < 0.001]. Overall, the proportion of ARI visits diagnosed with sinusitis increased [annual OR 1.09 (1.08, 1.10), P < 0.01], but the proportion of sinusitis diagnoses decreased [OR 0.72 (0.69, 0.75), P < 0.001] postintervention. Guideline-concordant antibiotic selection was 61.5% vs. 71.2% for sinusitis and 63.3% vs. 67.8% for pharyngitis pre-/postintervention, respectively (both P < 0.001). Conclusion Antibiotic prescribing rates for ARIs within the VA have steadily declined since 2010. Additional decline in antibiotic prescribing was associated with the launch of a national campaign to improve ARI management. Disclosures All authors: No reported disclosures.

Background Though more often associated with persistent diarrhea, protozoal gastroenteritis may also present as an acute illness. Suspicion for gastrointestinal protozoal infection is historically based on travel or recreational water exposure and duration of illness, which can delay diagnosis and effective treatment. The use of multiplex PCR panels has the potential to improve diagnosis of protozoal illness. To better understand when testing for protozoa is warranted, we examined predictors of protozoal gastroenteritis within a prospective multicenter study of U.S. children presenting to emergency departments with acute gastroenteritis (AGE). Methods The analysis utilized data from the IMPACT study, a prospective trial of the clinical impact of a multiplex PCR panel for detecting GI pathogens in AGE in five pediatric hospitals across the United States. From these data, we evaluated 72 potential predictors, including patient demographics, medical histories, exposures, symptoms, and vital signs. Using Random Forest Algorithm, 10 variables were selected based on variable importance measure. These were then entered into a Stepwise logistic regression model. Results Out of 962 patients there were 41 (4.3%) patients with protozoal detections, including 18 for Giardia and 24 for Cryptosporidium. Of these, 21 (51%) were male and the median age was 4.3 years old. Detection rates varied by site with 76% (31) of cases at two Midwestern sites. In 23 (56%) of cases protozoa was the sole pathogen detected. Logistic regression modeling of the top 10 variables identified by Random Forest showed the strongest predictor was living in a household with a child 5 years old or younger, followed by study site and younger age (table). Notably, travel, recreational water exposure, and duration of diarrhea were not in the top 10 variables identified. Conclusion Our analysis suggests that age and household exposure predict higher likelihood of protozoal infection in children with AGE. Classic epidemiologic exposures including travel and recreational water exposure were not predictive. These data could improve appropriate test selection. Future studies are still needed for external validation of this model. Disclosures A. Pavia, BioFire Dx: Investigator, Research support.

Background Infectious diarrheal illness is a significant contributor to healthcare costs in the US pediatric population. New multi-pathogen PCR-based panels have shown increased sensitivity over previous methods; however, they are costly and clinical utility may be limited in many cases. Clinical Prediction Rules (CPRs) may help optimize the appropriate use of these tests. Furthermore, Natural Language Processing (NLP) is an emerging tool to extract clinical history for decision support. Here, we examine NLP for the validation of a CPR for pediatric diarrhea. Methods Using data from a prospective clinical trial at 5 US pediatric hospitals, 961 diarrheal cases were assessed for etiology and relevant clinical variables. Of 65 variables collected in that study, 42 were excluded in our models based on a scarcity of documentation in reviewed clinical charts. The remaining 23 variables were ranked by random forest (RF) variable importance and utilized in both an RF and stepwise logistic regression (LR) model for viral-only etiology. We investigated whether NLP could accurately extract data from clinical notes comparable to study questionnaires. We used the eHOST abstraction software to abstract 6 clinical variables from patient charts that were useful in our CPR. These data will be used to train an NLP algorithm to extract the same variables from additional charts, and be combined with data from 2 other variables coded in the EMR to externally validate our model. Results Both RF and LR models achieved cross-validated area under the receiver operating characteristic curves of 0.74 using the top 5 variables (season, age, bloody diarrhea, vomiting/nausea, and fever), which did not improve significantly with the addition of more variables. Of 270 charts abstracted for NLP training, there were 41 occurrences of bloody diarrhea annotated, 339 occurrences of vomiting, and 145 occurrences of fever. Inter-annotator agreement over 9 training sets ranged between 0.63 and 0.83. Conclusion We have constructed a parsimonious CPR involving only 5 inputs for the prediction of a viral-only etiology for pediatric diarrheal illness using prospectively collected data. With the training of an NLP algorithm for automated chart abstraction we will validate the CPR. NLP could allow a CPR to run without manual data entry to improve care. Disclosures All authors: No reported disclosures.

Sudoku is played by millions of people across the globe. It has simple rules and is very addictive. The game board is a nine-by-nine grid of numbers from one to nine. Several entries within the grid are provided and the remaining entries must be filled in subject to no row, column, or three-by-three subsquare containing duplicate numbers. By exploiting these three types of uniformity, we propose an approach to constructing a new type of design, called a Sudoku-based space-filling design. Such a design can be divided into groups of subdesigns so that the complete design and each subdesign achieve maximum uniformity in univariate and bivariate margins. Examples are given illustrating the proposed construction method. Applications of such designs include computer experiments with qualitative and quantitative factors, linking parameters in engineering and crossvalidation. Copyright 2011, Oxford University Press.

Background Audit-feedback of antibiotic prescribing rates for acute respiratory infections (ARIs) is a promising approach to reduce antibiotic use; however, the generalizability and sustainability are unknown. We describe an audit-feedback intervention and outcomes across multiple seasons in different clinic settings. Methods Two VA Medical Centers distributed audit-feedback reports targeting providers with frequent ARI visits in emergency department (ED) and primary care (PC) during 2015–2016 and 2016–2017. An academic detailing visit delivered by local peers accompanied the initial audit-feedback report. The intervention was expanded to ED and PC clinics (n = 10) in three other VA facilities in 2017–2018. Outcomes included rates of antibiotics prescribed, recurrent visits for ARIs within 30 days, and adverse events. We assessed intervention sustainability in initiating VAs, and intervention generalizability in expansion VAs. Mixed-effect logistic regression models were used to assess intervention effect on antibiotic prescribing and outcomes. Results Antibiotic prescribing for uncomplicated ARI visits (n = 7,814) declined from 53.8% to 27.9% post intervention. The intervention was associated with a reduction in odds of prescribing antibiotics in initiating facilities (odds ratio [OR] 0.6 (95% CI 0.3, 0.9), which declined further with an annual OR 0.8 [95% CI 0.7, 1.1] per year. Preliminary 6-month postintervention results were available from pilot clinics (n = 3) within two of the expansion VAs, which indicated similar effectiveness (OR 0.5 [0.4, 0.7]). Recurrent visits for ARIs (8.2% vs. 8.6%, P = 0.14) and adverse events (2.3% vs. 2.1%, P = 0.90) were not different pre-/post-intervention. Receipt of an antibiotic was not associated with recurrent visits for ARI (8.6% vs. 8.0%, P = 0.45) or adverse events (1.9% vs. 1.7%, P = 0.11). Conclusion An audit-feedback intervention sustained a reduction in antibiotic prescribing for ARIs over 3 years, and resulted in similar reductions in antibiotic use in varied ED and PC settings without affecting ARI-related return visit rates. Disclosures All authors: No reported disclosures.

Temperature control for a large data center is both important and expensive. On the one hand, many of the components produce a great deal of heat, and on the other hand, many of the components require temperatures below a fairly low threshold for reliable operation. A statistical framework is proposed within which the behavior of a large cooling system can be modeled and forecast under both steady state and perturbations. This framework is based upon an extension of multivariate Gaussian autoregressive hidden Markov models (HMMs). The estimated parameters of the fitted model provide useful summaries of the overall behavior of and relationships within the cooling system. Predictions under system perturbations are useful for assessing potential changes and improvements to be made to the system. Many data centers have far more cooling capacity than necessary under sensible circumstances, thus resulting in energy inefficiencies. Using this model, predictions for system behavior after a particular component of the cooling system is shut down or reduced in cooling power can be generated. Steady-state predictions are also useful for facility monitors. System traces outside control boundaries flag a change in behavior to examine. The proposed model is fit to data from a group of air conditioners within an enterprise data center from the IT industry. The fitted model is examined, and a particular unit is found to be underutilized. Predictions generated for the system under the removal of that unit appear very reasonable. Steady-state system behavior also is predicted well.

Background Despite the high prevalence and global impact of knee osteoarthritis (KOA), current treatments are palliative. No disease modifying anti-osteoarthritic drug (DMOAD) has been approved. We recently demonstrated significant involvement of uric acid and activation of the innate immune response in osteoarthritis (OA) pathology and progression, suggesting that traditional gout therapy may be beneficial for OA. We therefore assess colchicine, an existing commercially available agent for gout, for a new therapeutic application in KOA. Methods/Design COLKOA is a double-blind, placebo-controlled, randomized trial comparing a 16-week treatment with standard daily dose oral colchicine to placebo for KOA. A total of 120 participants with symptomatic KOA will be recruited from a single center in Singapore. The primary end point is 30% improvement in total Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score at week 16. Secondary end points include improvement in pain, physical function, and quality of life and change in serum, urine and synovial fluid biomarkers of cartilage metabolism and inflammation. A magnetic resonance imaging (MRI) substudy will be conducted in 20 participants to evaluate change in synovitis. Logistic regression will be used to compare changes between groups in an intention-to-treat analysis. Discussion The COLKOA trial is designed to evaluate whether commercially available colchicine is effective for improving signs and symptoms of KOA, and reducing synovial fluid, serum and urine inflammatory and biochemical joint degradation biomarkers. These biomarkers should provide insights into the underlying mechanism of therapeutic response. This trial will potentially provide data to support a new treatment option for KOA. Trial registration The trial has been registered at clinicaltrials.gov as NCT02176460. Date of registration: 26 June 2014. Electronic supplementary material The online version of this article (doi:10.1186/s13063-015-0726-x) contains supplementary material, which is available to authorized users.

A framework is proposed for making quality predictions in situations for which only systematically inaccurate data are available. The predictions are based on the systematically inaccurate data, complete data from similar situations, and expert knowledge. The proposed predictive model is well suited to functional data and is computationally simple, fast, and stable. We focus primarily on a particular problem presenting itself in the pharmaceutical industry. Predicting both side effect and endpoint dose responses before the initiation of a clinical trial has enormous ethical and financial importance in the pharmaceutical industry. The proposed Bayesian semiparametric predictive model is used to predict unobserved clinical dose-response curves conditional on preclinical data, data from similar compounds, and prior knowledge. The model allows for nonlinear dose-response curves and the incorporation of relevant prior information. Posterior sampling is achieved through a simple and computationally efficient Gibbs sampler. The predictions from the model are drawn from the posterior distribution of the average dose-response curve for the candidate compound, allowing straightforward incorporation into a risk assessment model unlike the deterministic predictions often used currently. The model is used on actual data from the pharmaceutical industry, showing that the model is capable of predicting lack or presence of trend with appropriate uncertainty.

Large-scale computer experiments are becoming increasingly important in science. A multi-step procedure is introduced to statisticians for modeling such experiments, which builds an accurate interpolator in multiple steps. In practice, the procedure shows substantial improvements in overall accuracy, but its theoretical properties are not well established. We introduce the terms nominal and numeric error and decompose the overall error of an interpolator into nominal and numeric portions. Bounds on the numeric and nominal error are developed to show theoretically that substantial gains in overall accuracy can be attained with the multi-step approach., Published in at http://dx.doi.org/10.1214/11-AOS929 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Greater trochanter pain syndrome due to tendinopathy or bursitis is a common cause of hip pain. The previously reported magnetic resonance (MR) findings of trochanteric tendinopathy and bursitis are peritrochanteric fluid and abductor tendon abnormality. We have often noted peritrochanteric high T2 signal in patients without trochanteric symptoms. The purpose of this study was to determine whether the MR findings of peritrochanteric fluid or hip abductor tendon pathology correlate with trochanteric pain. We retrospectively reviewed 131 consecutive MR examinations of the pelvis (256 hips) for T2 peritrochanteric signal and abductor tendon abnormalities without knowledge of the clinical symptoms. Any T2 peritrochanteric abnormality was characterized by size as tiny, small, medium, or large; by morphology as feathery, crescentic, or round; and by location as bursal or intratendinous. The clinical symptoms of hip pain and trochanteric pain were compared to the MR findings on coronal, sagittal, and axial T2 sequences using chi-square or Fisher’s exact test with significance assigned as p