Association of endoscopic biopsy sampling methods with detection of precursor lesions of gastric cancer.

BACKGROUND AND AIMS: The yield of various endoscopic biopsy sampling methods for detection of precursor lesions of noncardia gastric cancer in a real-world setting remains unclear. Our objective was to evaluate the association of endoscopic biopsy sampling methods with detection of gastric intestinal metaplasia (GIM) and gastric dysplasia (GD). METHODS: We conducted a case-control study of adult patients who underwent EGD with biopsy sampling between 2010 and 2021 in a racially and ethnically diverse U.S. healthcare system. Cases were patients with histopathologic findings of GIM and/or GD. Control subjects were matched 1:1 by age, procedure date, and medical center. We compared the detection of GIM and GD using 4 different biopsy sampling methods: unspecified, specified stomach location, 2+2, and the Sydney protocol. Additionally, we assessed trends in use of sampling methods (Cochrane-Armitage) and identified patient and endoscopist factors associated with their use (logistic regression). RESULTS: We identified 20,938 GIM and 455 GD matched pairs. A greater proportion of GIM cases were detected using 2+2 (31.3% vs 25.3%, P < .0001) and the Sydney protocol (9.1% vs 1.0%, P < .0001) compared with control subjects. Similarly, a greater proportion of GD cases were detected using the Sydney protocol (15.6% vs .4%, P < .0001). We observed an increasing trend in the use of the Sydney protocol during the study period (3.8%-16.1% in cases, P < .0001; 1%-1.1% in control subjects, P = .005). Male and Asian American patients were more likely to undergo 2+2 or the Sydney protocol, whereas female and Hispanic endoscopists were more likely to perform sampling using these protocols. CONCLUSIONS: The application of the Sydney protocol is associated with an increased detection of precursor lesions of gastric cancer in routine clinical practice.

Derivation and External Validation of Machine Learning-Based Model for Detection of Pancreatic Cancer.

INTRODUCTION: There is currently no widely accepted approach to screening for pancreatic cancer (PC). We aimed to develop and validate a risk prediction model for pancreatic ductal adenocarcinoma (PDAC), the most common form of PC, across 2 health systems using electronic health records. METHODS: This retrospective cohort study consisted of patients aged 50-84 years having at least 1 clinic-based visit over a 10-year study period at Kaiser Permanente Southern California (model training, internal validation) and the Veterans Affairs (VA, external testing). Random survival forests models were built to identify the most relevant predictors from >500 variables and to predict risk of PDAC within 18 months of cohort entry. RESULTS: The Kaiser Permanente Southern California cohort consisted of 1.8 million patients (mean age 61.6) with 1,792 PDAC cases. The 18-month incidence rate of PDAC was 0.77 (95% confidence interval 0.73-0.80)/1,000 person-years. The final main model contained age, abdominal pain, weight change, HbA1c, and alanine transaminase change (c-index: mean = 0.77, SD = 0.02; calibration test: P value 0.4, SD 0.3). The final early detection model comprised the same features as those selected by the main model except for abdominal pain (c-index: 0.77 and SD 0.4; calibration test: P value 0.3 and SD 0.3). The VA testing cohort consisted of 2.7 million patients (mean age 66.1) with an 18-month incidence rate of 1.27 (1.23-1.30)/1,000 person-years. The recalibrated main and early detection models based on VA testing data sets achieved a mean c-index of 0.71 (SD 0.002) and 0.68 (SD 0.003), respectively. DISCUSSION: Using widely available parameters in electronic health records, we developed and externally validated parsimonious machine learning-based models for detection of PC. These models may be suitable for real-time clinical application.

Development of Parkinson Disease and Its Relationship with Incidentally Discovered White Matter Disease and Covert Brain Infarction in a Real-World Cohort.

OBJECTIVE: This study aimed to examine the relationship between covert cerebrovascular disease, comprised of covert brain infarction and white matter disease, discovered incidentally in routine care, and subsequent Parkinson disease. METHODS: Patients were ≥50 years and received neuroimaging for non-stroke indications in the Kaiser Permanente Southern California system from 2009 to 2019. Natural language processing identified incidentally discovered covert brain infarction and white matter disease and classified white matter disease severity. The Parkinson disease outcome was defined as 2 ICD diagnosis codes. RESULTS: 230,062 patients were included (median follow-up 3.72 years). A total of 1,941 Parkinson disease cases were identified (median time-to-event 2.35 years). Natural language processing identified covert cerebrovascular disease in 70,592 (30.7%) patients, 10,622 (4.6%) with covert brain infarction and 65,814 (28.6%) with white matter disease. After adjustment for known risk factors, white matter disease was associated with Parkinson disease (hazard ratio 1.67 [95%CI, 1.44, 1.93] for patients <70 years and 1.33 [1.18, 1.50] for those ≥70 years). Greater severity of white matter disease was associated with increased incidence of Parkinson disease(/1,000 person-years), from 1.52 (1.43, 1.61) in patients without white matter disease to 4.90 (3.86, 6.13) in those with severe disease. Findings were robust when more specific definitions of Parkinson disease were used. Covert brain infarction was not associated with Parkinson disease (adjusted hazard ratio = 1.05 [0.88, 1.24]). INTERPRETATION: Incidentally discovered white matter disease was associated with subsequent Parkinson disease, an association strengthened with younger age and increased white matter disease severity. Incidentally discovered covert brain infarction did not appear to be associated with subsequent Parkinson disease. ANN NEUROL 2022;92:620-630.

Machine learning versus regression for prediction of sporadic pancreatic cancer.

BACKGROUND/OBJECTIVES: There is currently no widely accepted approach to identify patients at increased risk for sporadic pancreatic cancer (PC). We aimed to compare the performance of two machine-learning models with a regression-based model in predicting pancreatic ductal adenocarcinoma (PDAC), the most common form of PC. METHODS: This retrospective cohort study consisted of patients 50-84 years of age enrolled in either Kaiser Permanente Southern California (KPSC, model training, internal validation) or the Veterans Affairs (VA, external testing) between 2008 and 2017. The performance of random survival forests (RSF) and eXtreme gradient boosting (XGB) models were compared to that of COX proportional hazards regression (COX). Heterogeneity of the three models were assessed. RESULTS: The KPSC and the VA cohorts consisted of 1.8 and 2.7 million patients with 1792 and 4582 incident PDAC cases within 18 months, respectively. Predictors selected into all three models included age, abdominal pain, weight change, and glycated hemoglobin (A1c). Additionally, RSF selected change in alanine transaminase (ALT), whereas the XGB and COX selected the rate of change in ALT. The COX model appeared to have lower AUC (KPSC: 0.737, 95% CI 0.710-0.764; VA: 0.706, 0.699-0.714), compared to those of RSF (KPSC: 0.767, 0.744-0.791; VA: 0.731, 0.724-0.739) and XGB (KPSC: 0.779, 0.755-0.802; VA: 0.742, 0.735-0.750). Among patients with top 5% predicted risk from all three models (N = 29,663), 117 developed PDAC, of which RSF, XGB and COX captured 84 (9 unique), 87 (4 unique), 87 (19 unique) cases, respectively. CONCLUSIONS: The three models complement each other, but each has unique contributions.

Stratifying Future Stroke Risk with Incidentally Discovered White Matter Disease Severity and Covert Brain Infarct Site.

BACKGROUND: Covert cerebrovascular disease (CCD) includes white matter disease (WMD) and covert brain infarction (CBI). Incidentally discovered CCD is associated with increased risk of subsequent symptomatic stroke. However, it is unknown whether the severity of WMD or the location of CBI predicts risk. OBJECTIVES: The aim of this study was to examine the association of incidentally discovered WMD severity and CBI location with risk of subsequent symptomatic stroke. METHOD: This retrospective cohort study includes patients aged ≥50 years old in the Kaiser Permanente Southern California health system who received neuroimaging for a nonstroke indication between 2009 and 2019. Incidental CBI and WMD were identified via natural language processing of the neuroimage report, and WMD severity was classified into grades. RESULTS: A total of 261,960 patients received neuroimaging; 78,555 patients (30.0%) were identified to have incidental WMD and 12,857 patients (4.9%) to have incidental CBI. Increasing WMD severity is associated with an increased incidence rate of future stroke. However, the stroke incidence rate in CT-identified WMD is higher at each level of severity compared to rates in MRI-identified WMD. Patients with mild WMD via CT have a stroke incidence rate of 24.9 per 1,000 person-years, similar to that of patients with severe WMD via MRI. Among incidentally discovered CBI patients with a determined CBI location, 97.9% are subcortical rather than cortical infarcts. CBI confers a similar risk of future stroke, whether cortical or subcortical or whether MRI- or CT-detected. CONCLUSIONS: Increasing severity of incidental WMD is associated with an increased risk of future symptomatic stroke, dependent on the imaging modality. Subcortical and cortical CBI conferred similar risks.

Stratifying future stroke risk with incidentally-discovered white matter disease severity and covert brain infarct site.

BACKGROUND: Covert cerebrovascular disease (CCD) includes white matter disease (WMD) and covert brain infarction (CBI). Incidentally-discovered CCD is associated with increased risk of subsequent symptomatic stroke. However, it is unknown whether the severity of WMD or the location of CBI predicts risk. OBJECTIVES: To examine the association of incidentally-discovered WMD severity and CBI location with risk of subsequent symptomatic stroke. METHOD: This retrospective cohort study includes patients 50 years old in the Kaiser Permanente Southern California health system who received neuroimaging for a non-stroke indication between 2009-2019. Incidental CBI and WMD were identified via natural language processing of the neuroimage report, and WMD severity was classified into grades. RESULTS: 261,960 patients received neuroimaging; 78,555 (30.0%) were identified to have incidental WMD, and 12,857 (4.9%) to have incidental CBI. Increasing WMD severity is associated with increased incidence rate of future stroke. However, the stroke incidence rate in CT-identified WMD is higher at each level of severity compared to rates in MRI-identified WMD. Patients with mild WMD via CT have a stroke incidence rate of 24.9 per 1,000 person-years, similar to that of patients with severe WMD via MRI. Among incidentally-discovered CBI patients with a determined CBI location, 97.9% are subcortical rather than cortical infarcts. CBI confers a similar risk of future stroke, whether cortical or subcortical, or whether MRI- or CT-detected. CONCLUSIONS: Increasing severity of incidental WMD is associated with an increased risk of future symptomatic stroke, dependent on the imaging modality. Subcortical and cortical CBI conferred similar risks.

Risk Prediction of Pancreatic Cancer in Patients With Recent-onset Hyperglycemia: A Machine-learning Approach.

BACKGROUND: New-onset diabetes (NOD) has been suggested as an early indicator of pancreatic cancer. However, the definition of NOD by the American Diabetes Association requires 2 simultaneous or consecutive elevated glycemic measures. We aimed to apply a machine-learning approach using electronic health records to predict the risk in patients with recent-onset hyperglycemia. MATERIALS AND METHODS: In this retrospective cohort study, health plan enrollees 50 to 84 years of age who had an elevated (6.5%+) glycated hemoglobin (HbA1c) tested in January 2010 to September 2018 with recent-onset hyperglycemia were identified. A total of 102 potential predictors were extracted. Ten imputation datasets were generated to handle missing data. The random survival forests approach was used to develop and validate risk models. Performance was evaluated by c -index, calibration plot, sensitivity, specificity, and positive predictive value. RESULTS: The cohort consisted of 109,266 patients (mean age: 63.6 y). The 3-year incidence rate was 1.4 (95% confidence interval: 1.3-1.6)/1000 person-years of follow-up. The 3 models containing age, weight change in 1 year, HbA1c, and 1 of the 3 variables (HbA1c change in 1 y, HbA1c in the prior 6 mo, or HbA1c in the prior 18 mo) appeared most often out of the 50 training samples. The c -indexes were in the range of 0.81 to 0.82. The sensitivity, specificity, and positive predictive value in patients who had the top 20% of the predicted risks were 56% to 60%, 80%, and 2.5% to 2.6%, respectively. CONCLUSION: Targeting evaluation at the point of recent hyperglycemia based on elevated HbA1c could offer an opportunity to identify pancreatic cancer early and possibly impact survival in cancer patients.

Risk for Shoulder Conditions After Vaccination: A Population-Based Study Using Real-World Data.

BACKGROUND: Although shoulder conditions have been reported as an adverse event after intramuscular vaccination in the deltoid muscle, epidemiologic data on shoulder conditions after vaccination are limited. OBJECTIVE: To estimate the risk for shoulder conditions after vaccination and assess possible risk factors. DESIGN: Retrospective cohort study. SETTING: Kaiser Permanente Southern California, a large integrated health care organization. PARTICIPANTS: Kaiser Permanente Southern California members aged 3 years or older who had an intramuscular vaccination administered in the deltoid muscle between 1 April 2016 and 31 December 2017. MEASUREMENTS: A natural language processing (NLP) algorithm was used to identify potential shoulder conditions among vaccinated persons with shoulder disorder diagnosis codes. All NLP-identified cases were manually chart confirmed on the basis of our case definition. The characteristics of vaccinated persons with and without shoulder conditions were compared. RESULTS: Among 3 758 764 administered vaccinations, 371 cases of shoulder condition were identified, with an estimated incidence of 0.99 (95% CI, 0.89 to 1.09) per 10 000 vaccinations. The incidence was 1.22 (CI, 1.10 to 1.35) for the adult (aged ≥18 years) and 0.05 (CI, 0.02 to 0.14) for the pediatric (aged 3 to 17 years) vaccinated populations. In the adult vaccinated population, advanced age, female sex, an increased number of outpatient visits in the 6 months before vaccination, lower Charlson Comorbidity Index, and pneumococcal conjugate vaccine were associated with a higher risk for shoulder conditions. Among influenza vaccines, quadrivalent vaccines were associated with an increased risk for shoulder conditions. Simultaneous administration of vaccines was associated with a higher risk for shoulder conditions among elderly persons. LIMITATION: Generalizability to other health care settings, use of administrative data, and residual confounding. CONCLUSION: These population-based data suggest a small absolute risk for shoulder conditions after vaccination. Given the high burden of shoulder conditions, clinicians should pay attention to any factors that may further increase risks. PRIMARY FUNDING SOURCE: Centers for Disease Control and Prevention.

Genetic polymorphisms in DNA repair genes and their association with risk of cervical cancer: A systematic review and meta-analysis.

BACKGROUND: There have been a large number of epidemiologic studies regarding the association between genetic polymorphisms in DNA repair genes and onset of cervical cancer. However, results are inconsistent. METHODS: Articles published before June 2021 and regarding genetic polymorphisms in DNA repair genes and cervical cancer were searched in following databases: PubMed, Web of Science, Google Scholar, and CNKI. With at least three articles for each polymorphism, we made meta-analysis to compute multivariate odds ratios (ORs) and their 95% confidence intervals (CIs). RESULTS: The present study showed significant associations between XRCC1 Arg399Gln polymorphisms and risk of cervical cancer in Asian, whereas no significant association between them were showed in Caucasian (Asian: GA vs. GG: OR = 1.27, 95%CI 1.06-1.52; AA vs. GG: OR = 1.91, 95%CI 1.29-2.83; GA + AA vs. GG: OR = 1.36, 95%CI 1.12-1.65; AA vs. GG + GA: OR = 1.66, 95%CI 1.17-2.37; Caucasian: GA vs. GG: OR = 1.08, 95%CI 0.83-1.41; AA vs. GG: OR = 2.18, 95%CI 0.75-6.31; GA + AA vs. GG: OR = 1.23, 95%CI 0.85-1.78; AA vs. GG + GA: OR = 1.70, 95%CI 0.69-4.18). In addition, there were significant associations between ERCC2 rs13181 polymorphisms and risk of cervical cancer in Asian (AC vs AA: OR = 0.53, 95%CI 0.37-0.75, I2 = 0.0%, p value of Q test = 0.847; AC + CC vs AA: OR = 0.50, 95%CI 0.36-0.70, I2 = 0.0%, p value of Q test = 0.856). CONCLUSIONS: The meta-analysis showed that there were significant associations between XRCC1 Arg399Gln and ERCC2 rs13181 polymorphisms and risk of cervical cancer.

Severity of Diabetic Retinopathy and the Risk of Future Cerebrovascular Disease, Cardiovascular Disease, and All-Cause Mortality.

PURPOSE: To determine the relationship between the severity of diabetic retinopathy and the future risk of cerebrovascular accident (CVA), myocardial infarction (MI), congestive heart failure (CHF), and all-cause mortality in patients with type 2 diabetes mellitus. DESIGN: Retrospective cohort study. PARTICIPANTS: Patients with type 2 diabetes who underwent diabetic retinopathy screening via fundus photography. METHODS: The relationship between retinopathy status and the 5-year risk of first-time CVA, MI, CHF, and all-cause mortality was investigated using multivariate Cox proportional hazards regressions that controlled for age, gender, race or ethnicity, hemoglobin A1c, duration of diabetes, high-density lipoprotein level, low-density lipoprotein level, history of hypertension, systolic blood pressure, diastolic blood pressure, tobacco use, statin use, body mass index, urine microalbumin-to-creatinine ratio, and estimated glomerular filtration rate. MAIN OUTCOME MEASURES: Five-year risk of first-time CVA, MI, CHF, and all-cause mortality. RESULTS: Seventy-seven thousand three hundred seventy-six patients were included in this study. The average age was 59.8 years with 53.6% male, 31.2% non-Hispanic White, and 41.4% Hispanic patients. Diabetic retinopathy was significantly associated with all outcomes on multivariate analysis. Compared with patients with no retinopathy, those with minimal nonproliferative diabetic retinopathy (NPDR) had a higher risk of CVA (hazard ratio [HR], 1.31; 95% confidence interval [CI], 1.18-1.46), MI (HR, 1.30; 95% CI, 1.15-1.46), CHF (HR, 1.29; 95% CI, 1.19-1.40), and death (HR, 1.15; 95% CI, 1.05-1.25). Similarly, patients with moderate to severe NPDR had a higher risk of each outcome (CVA: HR, 1.56; 95% CI, 1.29-1.89; MI: HR, 1.92; 95% CI, 1.57-2.34; CHF: HR, 1.90; 95% CI, 1.66-2.18, and death: HR, 1.55; 95% CI, 1.32-1.82), as did patients with proliferative diabetic retinopathy (CVA: HR, 2.53; 95% CI, 1.84-3.48; MI: HR, 1.89; 95% CI, 1.26-2.83; CHF: HR, 1.96; 95% CI, 1.47-2.59; and death: HR, 1.87; 95% CI, 1.36-2.56). CONCLUSIONS: Diabetic retinopathy is significantly associated with future risk of CVA, MI, CHF, and death, with higher degrees of retinopathy appearing to carry a heightened risk for each outcome. Retinal information may provide valuable insights into patients' risk of future vascular disease and death.

Temporal trends in heart failure mortality in an integrated healthcare delivery system, California, and the US, 2001-2017.

BACKGROUND: In recent years, decreases in mortality rates attributable to cardiovascular diseases have slowed but mortality attributable to heart failure (HF) has increased. METHODS: Between 2001-2017, trends in age-adjusted mortality with HF as an underlying cause for Kaiser Permanente Southern California (KPSC) members were derived through linkage with state death files and compared with trends among California residents and the US. Average annual percent change (AAPC) and 95% confidence intervals (CI) were calculated using Joinpoint regression. Analyses were repeated examining HF as a contributing cause of death. RESULTS: In KPSC, the age-adjusted HF mortality rates were comparable to California but lower than the US, increasing from 23.9 per 100,000 person-years (PY) in 2001 to 44.7 per 100,000 PY in 2017, representing an AAPC of 1.3% (95% CI 0.0%, 2.6%). HF mortality also increased in California from 33.9 to 46.5 per 100,000 PY (AAPC 1.5%, 95% CI 0.3%, 2.7%), while remaining unchanged in the US at 57.9 per 100,000 PY in 2001 and 2017 (AAPC 0.0%, 95% CI - 0.5%, 0.5%). Trends among KPSC members ≥ 65 years old were similar to the overall population, while trends among members 45-64 years old were flat between 2001-2017. Small changes in mortality with HF as a contributing cause were observed in KPSC members between 2001 and 2017, which differed from California and the US. CONCLUSION: Lower rates of HF mortality were observed in KPSC compared to the US. Given the aging of the US population and increasing prevalence of HF, it will be important to examine individual and care-related factors driving susceptibility to HF mortality.

Validation of the Enriching New-Onset Diabetes for Pancreatic Cancer Model in a Diverse and Integrated Healthcare Setting.

BACKGROUND: The risk of pancreatic cancer is elevated among people with new-onset diabetes (NOD). Based on Rochester Epidemiology Project Data, the Enriching New-Onset Diabetes for Pancreatic Cancer (END-PAC) model was developed and validated. AIMS: We validated the END-PAC model in a cohort of patients with NOD using retrospectively collected data from a large integrated health maintenance organization. METHODS: A retrospective cohort of patients between 50 and 84 years of age meeting the criteria for NOD in 2010-2014 was identified. Each patient was assigned a risk score (< 1: low risk; 1-2: intermediate risk; ≥ 3: high risk) based on the values of the predictors specified in the END-PAC model. Patients who developed pancreatic ductal adenocarcinoma (PDAC) within 3 years were identified using the Cancer Registry and California State Death files. Area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were estimated. RESULTS: Out of the 13,947 NOD patients who were assigned a risk score, 99 developed PDAC in 3 years (0.7%). Of the 3038 patients who had a high risk, 62 (2.0%) developed PDAC in 3 years. The risk increased to 3.0% in white patients with a high risk. The AUC was 0.75. At the 3+ threshold, the sensitivity, specificity, PPV, and NPV were 62.6%, 78.5%, 2.0%, and 99.7%, respectively. CONCLUSIONS: It is critical that prediction models are validated before they are implemented in various populations and clinical settings. More efforts are needed to develop screening strategies most appropriate for patients with NOD in real-world settings.

The use of natural language processing to identify vaccine-related anaphylaxis at five health care systems in the Vaccine Safety Datalink.

PURPOSE: The objective was to develop a natural language processing (NLP) algorithm to identify vaccine-related anaphylaxis from plain-text clinical notes, and to implement the algorithm at five health care systems in the Vaccine Safety Datalink. METHODS: The NLP algorithm was developed using an internal NLP tool and training dataset of 311 potential anaphylaxis cases from Kaiser Permanente Southern California (KPSC). We applied the algorithm to the notes of another 731 potential cases (423 from KPSC; 308 from other sites) with relevant codes (ICD-9-CM diagnosis codes for anaphylaxis, vaccine adverse reactions, and allergic reactions; Healthcare Common Procedure Coding System codes for epinephrine administration). NLP results were compared against a reference standard of chart reviewed and adjudicated cases. The algorithm was then separately applied to the notes of 6 427 359 KPSC vaccination visits (9 402 194 vaccine doses) without relevant codes. RESULTS: At KPSC, NLP identified 12 of 16 true vaccine-related cases and achieved a sensitivity of 75.0%, specificity of 98.5%, positive predictive value (PPV) of 66.7%, and negative predictive value of 99.0% when applied to notes of patients with relevant diagnosis codes. NLP did not identify the five true cases at other sites. When NLP was applied to the notes of KPSC patients without relevant codes, it captured eight additional true cases confirmed by chart review and adjudication. CONCLUSIONS: The current study demonstrated the potential to apply rule-based NLP algorithms to clinical notes to identify anaphylaxis cases. Increasing the size of training data, including clinical notes from all participating study sites in the training data, and preprocessing the clinical notes to handle special characters could improve the performance of the NLP algorithms. We recommend adding an NLP process followed by manual chart review in future vaccine safety studies to improve sensitivity and efficiency.

Influence of Ambulatory Triglyceride Levels on Risk of Recurrence in Patients with Hypertriglyceridemic Pancreatitis.

BACKGROUND AND AIMS: To evaluate impact of ambulatory triglyceride levels on risk of recurrent pancreatitis in patients with hypertriglyceridemic pancreatitis. METHODS: We conducted a longitudinal retrospective cohort study of patients with serum triglyceride level ≥ 500 mg/dL during index hospitalization for acute pancreatitis within a regional integrated healthcare system between 2006 and 2013 (follow-up through 2015). Cases were identified based on combination of diagnosis codes and serum amylase/lipase. We used multivariable robust Poisson regression to determine independent effect of baseline (first outpatient) triglyceride measurement on risk of recurrent pancreatitis. Ambulatory triglyceride levels were categorized as normal (0-200 mg/dL), moderately elevated (201-500 mg/dL), and highly elevated (> 500 mg/dL). We further assessed factors related to likelihood of normalization of serum triglycerides (< 200 mg/dL) in the outpatient setting. RESULTS: One hundred and fifty-one patients met study inclusion criteria with median follow-up of 3 years. Overall, 45 (29.8%) patients experienced at least 1 recurrent attack with 25 (16.6%) experiencing multiple episodes. In multivariable analysis, patients that continued to have moderately elevated ((adjusted rate ratio RR 5.47 (95% CL 1.80, 16.65)) as well as highly elevated (RR 8.45 (2.55, 27.96)) triglycerides were at increased risk of disease recurrence compared to patients that achieved normalization. Patients with triglyceride measurement performed within 30 days from discharge were more likely to achieve normalization, 40 versus 26%, p = 0.03. CONCLUSIONS: For patients with hypertriglyceridemic pancreatitis, even modest elevation in subsequent triglyceride levels was associated with increased risk of recurrence. Future efforts should focus on ensuring timely care in the outpatient setting with a goal of normalizing triglycerides.

Controller therapy in Asians and whites with persistent Asthma.

BACKGROUND: Little is known concerning the relative effectiveness of LTRAs compared to ICSs as monotherapy or LABA as add-on therapy in the Asian population. OBJECTIVES: In this retrospective cohort study, we examined the comparative effectiveness of montelukast to ICS as a first-line monotherapy and as an add-on in comparison with LABA on asthma exacerbations among Asian and non -Hispanic white persistent asthma patients in a large managed care organization. METHODS: The three add-on comparisons were montelukast plus low-dose ICS versus LABA plus low-dose ICS, montelukast plus low-dose ICS versus medium-dose ICS, and montelukast plus medium-dose ICS versus LABA plus medium-dose ICS. Patients were identified based on ICD-9 diagnosis codes and administrative pharmacy dispensing. Exacerbations were defined as asthma emergency department visit or hospitalization, or asthma outpatient visits requiring systemic corticosteroid dispensing. Patient demographic and clinical characteristics were balanced by using inverse probability treatment weighting. Multivariable robust Poisson and Cox-proportional hazards regression models were applied to estimate rate ratios and hazard ratios. RESULTS: Compared with low-dose ICS monotherapy, montelukast monotherapy evidenced a lower incidence rate (RR 0.89, CI 0.79-0.99, p = 0.03) but similar hazard rate (HR 0.96, CI 0.86-1.06, p = 0.43) of asthma exacerbation in white patients 12 years of age or older. No difference was observed in Asian patients or in white children 4-11 years of age. All other comparisons did not reveal a statistically significant difference in incidence or hazard rate. CONCLUSION: In a real-world comparative effectiveness study, asthma exacerbation rates were similar among guideline alternative controller regimens in Asians and whites.

Comparing performance between log-binomial and robust Poisson regression models for estimating risk ratios under model misspecification.

BACKGROUND: Log-binomial and robust (modified) Poisson regression models are popular approaches to estimate risk ratios for binary response variables. Previous studies have shown that comparatively they produce similar point estimates and standard errors. However, their performance under model misspecification is poorly understood. METHODS: In this simulation study, the statistical performance of the two models was compared when the log link function was misspecified or the response depended on predictors through a non-linear relationship (i.e. truncated response). RESULTS: Point estimates from log-binomial models were biased when the link function was misspecified or when the probability distribution of the response variable was truncated at the right tail. The percentage of truncated observations was positively associated with the presence of bias, and the bias was larger if the observations came from a population with a lower response rate given that the other parameters being examined were fixed. In contrast, point estimates from the robust Poisson models were unbiased. CONCLUSION: Under model misspecification, the robust Poisson model was generally preferable because it provided unbiased estimates of risk ratios.

Dynamic Measurement of Disease Activity in Acute Pancreatitis: The Pancreatitis Activity Scoring System.

OBJECTIVES: Acute pancreatitis has a highly variable course. Currently there is no widely accepted method to measure disease activity in patients hospitalized for acute pancreatitis. We aimed to develop a clinical activity index that incorporates routine clinical parameters to assist in the measurement, study, and management of acute pancreatitis. METHODS: We used the UCLA/RAND appropriateness method to identify items for inclusion in the disease activity instrument. We conducted a systematic literature review followed by two sets of iterative modified Delphi meetings including a panel of international experts between November 2014 and November 2015. The final instrument was then applied to patient data obtained from five separate study cohorts across Southern California to assess profiles of disease activity. RESULTS: From a list of 35 items comprising 6 domains, we identified 5 parameters for inclusion in the final weighted clinical activity scoring system: organ failure, systemic inflammatory response syndrome, abdominal pain, requirement for opiates and ability to tolerate oral intake. We applied the weighted scoring system across the 5 study cohorts comprising 3,123 patients. We identified several distinct patterns of disease activity: (i) overall there was an elevated score at baseline relative to discharge across all study cohorts, (ii) there were distinct patterns of disease activity related to duration of illness as well as (iii) early and persistent elevation of disease activity among patients with severe acute pancreatitis defined as persistent organ failure. CONCLUSIONS: We present the development and initial validation of a clinical activity score for real-time assessment of disease activity in patients with acute pancreatitis.

Hospitalizations, outcomes, and management costs of febrile neutropenia in patients from a managed care population.

PURPOSE: The study objective was to evaluate chemotherapy treatment patterns and incidence, cost, and resource utilization of febrile neutropenia-related hospitalization (FNH) in patients with breast cancer, lung cancer, and non-Hodgkin's lymphoma (NHL) from Kaiser Permanente Southern California (KPSC), a large integrated delivery system. METHODS: Adults ≥18 years with any stage breast cancer, lung cancer, or NHL who initiated myelosuppressive chemotherapy from 01/01/2006 to 12/31/2009 were included. Chemotherapy dose delays ≥7 days, relative dose intensity (RDI), regimen switching, FNH and all-cause mortality, granulocyte colony-stimulating factor (G-CSF) and antibiotic use, and healthcare utilization/cost were evaluated by cancer type, regimen, and/or cycle. RESULTS: Among 3314 breast cancer patients, 25.3% received an RDI ≤85%, 13.9% experienced FNH with an all-cause mortality rate of 2.0%, and 20.2% received primary prophylaxis with G-CSF. Among those with FNH, mean hospital length of stay (LOS) was 4.1 days, and mean total costs were $20,462. Among 1443 lung cancer patients, 17.9% had an RDI ≤85%, 8.0% experienced FNH with an all-cause mortality rate of 25.2%, and 4.5% received primary prophylaxis with G-CSF. Among those with FNH, mean LOS was 6.8 days, and mean total costs were $32,964. Among 581 NHL patients, 27.9% had an RDI ≤85% and 22.4% experienced FNH with an all-cause mortality rate of 13%. Among those with FNH, mean LOS was 7.9 days, and mean total costs were $37,555. CONCLUSIONS: Marked variability was observed among different cancer types and chemotherapy regimens. Given the variability, detailed insight into incidence, management, and burden of FN can help inform clinical decision making.

Development and validation of algorithms to identify acute diverticulitis.

PURPOSE: The objectives of this study were to develop and validate algorithms to accurately identify patients with diverticulitis using electronic medical records (EMRs). METHODS: Using Kaiser Permanente Southern California's EMRs of adults (≥18 years) with International Classification of Diseases, Clinical Modifications, Ninth Revision diagnosis codes of diverticulitis (562.11, 562.13) between 1 January 2008 and 31 August 2009, we generated random samples for pilot (N = 692) and validation (N = 1502) respectively. Both samples were stratified by inpatient (IP), emergency department (ED), and outpatient (OP) care settings. We developed and validated several algorithms using EMR data on diverticulitis diagnosis code, antibiotics, computed tomography, diverticulosis history, pain medication and/or pain diagnosis, and excluding patients with infections and/or conditions that could mimic diverticulitis. Evidence of diverticulitis was confirmed through manual chart review. Agreement between EMR algorithm and manual chart confirmation was evaluated using sensitivity and positive predictive value (PPV). RESULTS: Both samples were similar in socio-demographics and clinical symptoms. An algorithm based on diverticulitis diagnosis code with antibiotic prescription dispensed within 7 days of diagnosis date, performed well overall. In the validation sample, sensitivity and PPV were (84.6, 98.2%), (95.8, 98.1%), and (91.8, 82.6%) for OP, ED, and IP, respectively. CONCLUSION: Using antibiotic prescriptions to supplement diagnostic codes improved the accuracy of case identification for diverticulitis, but results varied by care setting.

Phenotypes determined by cluster analysis in severe or difficult-to-treat asthma.

BACKGROUND: Asthma phenotyping can facilitate understanding of disease pathogenesis and potential targeted therapies. OBJECTIVE: To further characterize the distinguishing features of phenotypic groups in difficult-to-treat asthma. METHODS: Children ages 6-11 years (n = 518) and adolescents and adults ages ≥12 years (n = 3612) with severe or difficult-to-treat asthma from The Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens (TENOR) study were evaluated in this post hoc cluster analysis. Analyzed variables included sex, race, atopy, age of asthma onset, smoking (adolescents and adults), passive smoke exposure (children), obesity, and aspirin sensitivity. Cluster analysis used the hierarchical clustering algorithm with the Ward minimum variance method. The results were compared among clusters by χ(2) analysis; variables with significant (P < .05) differences among clusters were considered as distinguishing feature candidates. Associations among clusters and asthma-related health outcomes were assessed in multivariable analyses by adjusting for socioeconomic status, environmental exposures, and intensity of therapy. RESULTS: Five clusters were identified in each age stratum. Sex, atopic status, and nonwhite race were distinguishing variables in both strata; passive smoke exposure was distinguishing in children and aspirin sensitivity in adolescents and adults. Clusters were not related to outcomes in children, but 2 adult and adolescent clusters distinguished by nonwhite race and aspirin sensitivity manifested poorer quality of life (P < .0001), and the aspirin-sensitive cluster experienced more frequent asthma exacerbations (P < .0001). CONCLUSION: Distinct phenotypes appear to exist in patients with severe or difficult-to-treat asthma, which is related to outcomes in adolescents and adults but not in children. The study of the therapeutic implications of these phenotypes is warranted.