Causal Selection of Covariates in Regression Calibration for Mismeasured Continuous Exposure.

Regression calibration as developed by Rosner, Spiegelman, and Willett is used to adjust the bias in effect estimates due to measurement error in continuous exposures. The method involves two models: a measurement error model relating the mismeasured exposure to the true (or gold-standard) exposure and an outcome model relating the mismeasured exposure to the outcome. However, no comprehensive guidance exists for determining which covariates should be included in each model. In this article, we investigate the selection of the minimal and most efficient covariate adjustment sets under a causal inference framework. We show that to address the measurement error, researchers must adjust for, in both measurement error and outcome models, any common causes (1) of true exposure and the outcome and (2) of measurement error and the outcome. We also show that adjusting for so-called prognostic variables that are independent of true exposure and measurement error in the outcome model, may increase efficiency, while adjusting for any covariates that are associated only with true exposure generally results in efficiency loss in realistic settings. We apply the proposed covariate selection approach to the Health Professional Follow-up Study dataset to study the effect of fiber intake on cardiovascular disease. Finally, we extend the originally proposed estimators to a nonparametric setting where effect modification by covariates is allowed.

Remote cognitive behavioral therapy for older adults with anxiety symptoms: A systematic review and meta-analysis.

INTRODUCTION: In-person cognitive behavioral therapy (CBT) can reduce self-reported anxiety in older adults. However, studies are limited for remote CBT. We assessed the effectiveness of remote CBT in mitigating self-reported anxiety in older adults. METHODS: We conducted a systematic review and meta-analysis based on a literature search of PubMed, Embase, PsycInfo, and Cochrane databases up to March 31, 2021, for randomized controlled clinical trials comparing the effectiveness of remote CBT versus non-CBT controls on mitigating self-reported anxiety in older adults. We calculated within-group pre-to-post-treatment standardized mean difference using Cohen's d, obtained the difference between a remote CBT group and a non-CBT control group as our effect size for cross-study comparison, and conducted a random-effects meta-analysis. Changes in scores on self-reported anxiety symptoms (Generalized Anxiety Disorder-7 item Scale, Penn State Worry Questionnaire, or Penn State Worry Questionnaire - Abbreviated), and self-reported depressive symptoms (Patient Health Questionnaire-9 item Scale or Beck Depression Inventory) were primary and secondary outcomes, respectively. RESULTS: Six eligible studies, containing 633 participants with a pooled mean age of 66.6 years, were included in the systematic review and meta-analysis. There was a significant mitigating effect of intervention on self-reported anxiety, favoring remote CBT over non-CBT controls (between-group effect size: -0.63; 95% CI: -0.99 to -0.28). We also found a significant mitigating effect of intervention on self-reported depressive symptoms (between-group effect size: -0.74; 95% CI: -1.24 to -0.25). DISCUSSION: Remote CBT is more effective in reducing self-reported anxiety and depressive symptoms than non-CBT control in older adults.

Response Patterns to Weekly Short Message Service Health Surveys Among Diverse Youth at High Risk for Acquiring HIV.

HIV researchers use short messaging service (SMS)-based surveys to monitor health behaviors more closely than what would be possible with in-person assessment. Benefits are tempered by nonresponse to completing surveys. Understanding response patterns and their associated study participant characteristics would guide more tailored use of SMS-based surveys for HIV studies. We examined response to weekly 7-item SMS surveys administered as part of an HIV prevention trial. Using Mixture hidden Markov models (MHMM), we identified the underlying response patterns shared by subgroups of participants over time and quantified the association between these response patterns and participant characteristics. Three underlying response patterns were identified; responders, responders with phone-related errors, and non-responders. Non-responders versus responders were more likely to be younger, male, cis-gender, Black and Latinx participants with histories of homelessness, incarceration, and social support service utilization. Responders with phone-related errors compared to non-responders were more likely to be Black, Latinx, female, students, and have a history of incarceration and social support service utilization. More nuanced results from MHMM analyses better inform what strategies to use for increasing SMS response rates, including assisting in securing phone ownership/service for responders with phone-related errors and identifying alternative strategies for non-responders. Actively collecting and monitoring non-delivery notification data available from SMS gateway service companies offers another opportunity to identify and connect with participants when they are willing but unable to respond during follow-up.

Differential impact of mitigation policies and socioeconomic status on COVID-19 prevalence and social distancing in the United States.

BACKGROUND: The spread of COVID-19 has highlighted the long-standing health inequalities across the U.S. as neighborhoods with fewer resources were associated with higher rates of COVID-19 transmission. Although the stay-at-home order was one of the most effective methods to contain its spread, residents in lower-income neighborhoods faced barriers to practicing social distancing. We aimed to quantify the differential impact of stay-at-home policy on COVID-19 transmission and residents' mobility across neighborhoods of different levels of socioeconomic disadvantage. METHODS: This was a comparative interrupted time-series analysis at the county level. We included 2087 counties from 38 states which both implemented and lifted the state-wide stay-at-home order. Every county was assigned to one of four equally-sized groups based on its levels of disadvantage, represented by the Area Deprivation Index. Prevalence of COVID-19 was calculated by dividing the daily number of cumulative confirmed COVID-19 cases by the number of residents from the 2010 Census. We used the Social Distancing Index (SDI), derived from the COVID-19 Impact Analysis Platform, to measure the mobility. For the evaluation of implementation, the observation started from Mar 1st 2020 to 1 day before lifting; and, for lifting, it ranged from 1 day after implementation to Jul 5th 2020. We calculated a comparative change of daily trends in COVID-19 prevalence and Social Distancing Index between counties with three highest disadvantage levels and those with the least level before and after the implementation and lifting of the stay-at-home order, separately. RESULTS: On both stay-at-home implementation and lifting dates, COVID-19 prevalence was much higher among counties with the highest or lowest disadvantage level, while mobility decreased as the disadvantage level increased. Mobility of the most disadvantaged counties was least impacted by stay-at-home implementation and relaxation compared to counties with the most resources; however, disadvantaged counties experienced the largest relative increase in COVID-19 infection after both stay-at-home implementation and relaxation. CONCLUSIONS: Neighborhoods with varying levels of socioeconomic disadvantage reacted differently to the implementation and relaxation of COVID-19 mitigation policies. Policymakers should consider investing more resources in disadvantaged counties as the pandemic may not stop until most neighborhoods have it under control.

Association of depression symptom severity with short-term risk of an initial hospital encounter in adults with major depressive disorder.

BACKGROUND: Despite the availability of pharmacologic and nonpharmacologic treatment options, depression continues to be one of the leading causes of disability worldwide. This study evaluated whether depression symptom severity, as measured by PHQ-9 score, of patients diagnosed with MDD is associated with short-term risk of a hospital encounter (ER visit or inpatient stay). METHODS: Adults with ≥1 PHQ-9 assessment in an outpatient setting (index date) and ≥ 1 MDD diagnosis within 6 months prior were included from the de-identified Optum Electronic Health Record database (April 2016-June 2019). Patients were categorized by depression symptom severity based on PHQ-9 scores obtained by natural language processing. Crude rates, adjusted absolute risks, and adjusted relative risks of all-cause and MDD-related hospital encounters within 30 days following assessment of depression severity were determined. RESULTS: The study population consisted of 280,145 patients with MDD and ≥ 1 PHQ-9 assessment in an outpatient setting. Based on PHQ-9 scores, 26.9% of patients were categorized as having none/minimal depression symptom severity, 16.4% as mild, 24.7% as moderate, 19.6% as moderately severe, and 12.5% as severe. Among patients with none/minimal, mild, moderate, moderately severe, and severe depression, the adjusted absolute short-term risks of an initial all-cause hospital encounter were 4.1, 4.4, 4.8, 5.6, and 6.5%, respectively; MDD-related hospital encounter adjusted absolute risks were 0.8, 1.0, 1.3, 1.6, and 2.1%, respectively. Compared to patients with none/minimal depression symptom severity, the adjusted relative risks of an all-cause hospital encounter were 1.60 (95% CI 1.50-1.70) for those with severe, 1.36 (1.29-1.44) for those with moderately severe, 1.18 (1.12-1.25) for those with moderate, and 1.07 (1.00-1.13) for those with mild depression symptom severity. CONCLUSIONS: These study findings indicate that depression symptom severity is a key driver of short-term risk of hospital encounters, emphasizing the need for timely interventions that can ameliorate depression symptom severity.

Association of antidiabetic therapies with lower extremity amputation, mortality and healthcare cost from a nationwide retrospective cohort study in Taiwan.

We compared risks of clinical outcomes, mortality and healthcare costs among new users of different classes of anti-diabetic medications. This is a population-based, retrospective, new-user design cohort study using the Taiwan National Health Insurance Database between May 2, 2015 and September 30, 2017. An individual was assigned to a medication group based on the first anti-diabetic prescription on or after May 1, 2016: SGLT-2 inhibitors, DPP-4 inhibitors, GLP-1 agonists or older agents (metformin, etc.). Clinical outcomes included lower extremity amputation, peripheral vascular disease, critical limb ischemia, osteomyelitis, and ulcer. We built three Cox proportional hazards models for clinical outcomes and mortality, and three regression models with a log-link function and gamma distribution for healthcare costs, all with propensity-score weighting and covariates. We identified 1,222,436 eligible individuals. After adjustment, new users of SGLT-2 inhibitors were associated with 73% lower mortality compared to those of DPP-4 inhibitors or users of older agents, while 36% lower total costs against those of GLP-1 agonists. However, there was no statistically significant difference in the risk of lower extremity amputation across medication groups. Our study suggested that SGLT-2 inhibitors is associated with lower mortality compared to DPP 4 inhibitors and lower costs compared to GLP-1 agonists.

Early relaxation of community mitigation policies and risk of COVID-19 resurgence in the United States.

This study aimed to assess the impact of coronavirus disease (COVID-19) prevalence in the United States in the week leading to the relaxation of the stay-at-home orders (SAH) on future prevalence across states that implemented different SAH policies. We used data on the number of confirmed COVID-19 cases as of August 21, 2020 on county level. We classified states into four groups based on the 7-day change in prevalence and the state's approach to SAH policy. The groups included: (1) High Change (19 states; 7-day prevalence change ≥50th percentile), (2) Low Change (19 states; 7-day prevalence change <50th percentile), (3) No SAH (11 states: did not adopt SAH order), and (4) No SAH End (2 states: did not relax SAH order). We performed regression modeling assessing the association between change in prevalence at the time of SAH order relaxation and COVID-19 prevalence days after the relaxation of SAH order for four selected groups. After adjusting for other factors, compared to the High Change group, counties in the Low Change group had 33.8 (per 100,000 population) fewer cases (standard error (SE): 19.8, p < 0.001) 7 days after the relaxation of SAH order and the difference was larger by time passing. On August 21, 2020, the No SAH End group had 383.1 fewer cases (per 100,000 population) than the High Change group (SE: 143.6, p < 0.01). A measured, evidence-based approach is required to safely relax the community mitigation strategies and practice phased-reopening of the country.

Development of an Electronic Data Collection System to Support a Large-Scale HIV Behavioral Intervention Trial: Protocol for an Electronic Data Collection System.

BACKGROUND: Advancing technology has increased functionality and permitted more complex study designs for behavioral interventions. Investigators need to keep pace with these technological advances for electronic data capture (EDC) systems to be appropriately executed and utilized at full capacity in research settings. Mobile technology allows EDC systems to collect near real-time data from study participants, deliver intervention directly to participants' mobile devices, monitor staff activity, and facilitate near real-time decision making during study implementation. OBJECTIVE: This paper presents the infrastructure of an EDC system designed to support a multisite HIV biobehavioral intervention trial in Los Angeles and New Orleans: the Adolescent Medicine Trials Network "Comprehensive Adolescent Research & Engagement Studies" (ATN CARES). We provide an overview of how multiple EDC functions can be integrated into a single EDC system to support large-scale intervention trials. METHODS: The CARES EDC system is designed to monitor and document multiple study functions, including, screening, recruitment, retention, intervention delivery, and outcome assessment. Text messaging (short message service, SMS) and nearly all data collection are supported by the EDC system. The system functions on mobile phones, tablets, and Web browsers. RESULTS: ATN CARES is enrolling study participants and collecting baseline and follow-up data through the EDC system. Besides data collection, the EDC system is being used to generate multiple reports that inform recruitment planning, budgeting, intervention quality, and field staff supervision. The system is supporting both incoming and outgoing text messages (SMS) and offers high-level data security. Intervention design details are also influenced by EDC system platform capabilities and constraints. Challenges of using EDC systems are addressed through programming updates and training on how to improve data quality. CONCLUSIONS: There are three key considerations in the development of an EDC system for an intervention trial. First, it needs to be decided whether the flexibility provided by the development of a study-specific, in-house EDC system is needed relative to the utilization of an existing commercial platform that requires less in-house programming expertise. Second, a single EDC system may not provide all functionality. ATN CARES is using a main EDC system for data collection, text messaging (SMS) interventions, and case management and a separate Web-based platform to support an online peer support intervention. Decisions need to be made regarding the functionality that is crucial for the EDC system to handle and what functionality can be handled by other systems. Third, data security is a priority but needs to be balanced with the need for flexible intervention delivery. For example, ATN CARES is delivering text messages (SMS) to study participants' mobile phones. EDC data security protocols should be developed under guidance from security experts and with formative consulting with the target study population as to their perceptions and needs. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/10777.

Impact of serum SP-A and SP-D levels on comparison and prognosis of idiopathic pulmonary fibrosis: A systematic review and meta-analysis.

BACKGROUND AND OBJECTIVE: Idiopathic pulmonary fibrosis (IPF) has a poor prognosis in general; however, it is heterogeneous to detect relative biomarkers for predicting the disease progression. Serum biomarkers can be conveniently collected to detect and help to differentially diagnose IPF and predict IPF prognosis. This meta-analysis aimed to evaluate the use of serum surfactant proteins A and D (SP-A and SP-D) for differential diagnosis and prognosis of IPF. METHODS: Relevant articles were searched in PubMed, Embase, and Chinese National Knowledge Infrastructure databases and reviewed by 2 independent readers. Standard mean difference (SMD) and 95% confidence interval (CI) were calculated to assess the difference in serum levels of SP-A/D among patients with IPF, when compared to patients with non-IPF interstitial lung disease (ILD), pulmonary infection, and healthy control. Hazard ratio (HR) and 95% CI were used to compare the relative risk of mortality. RESULTS: Twenty-one articles (totalling 1289 IPF patients) were included in final meta-analysis. Serum SP-A levels were significantly higher in patients with IPF than in patients with non-IPF ILD (SMD: 1.108 [0.584, 1.632], P < .001), or pulmonary infection (SMD: 1.320 [0.999, 1.640], P < .001) and healthy controls (SMD: 2.802 [1.901, 3.702], P < .001). There was no significant difference in serum SP-D levels between patients with IPF and those with non-IPF ILD patients (SMD: 0.459 [-0.000, 0.919], P = .050). Serum SP-D levels were significantly higher in patients with IPF than in patients with pulmonary infection (SMD: 1.308 [0.813, 1.803], P < .001) and healthy controls (SMD: 2.235 [1.739, 2.731], P < .001). Risk of death in patients with IPF and elevated serum SP-A was increased 39% compared to patients with low SP-A groups. Elevated SP-D increased risk by 111% when compared to low SP-D. In acute exacerbation of IPF, serum SP-A/D were higher than those in stable stage. The comparisons and prognosis might be different in Asian and Caucasian patients. CONCLUSIONS: Serum SP-A/D detection might be useful for differential diagnosis and prediction of survival in patients with IPF.

Determinants analysis of outpatient service utilisation in Georgia: can the approach help inform benefit package design?

BACKGROUND: The healthcare financing reforms initiated by the Government of Georgia in 2007 have positively affected inpatient service utilisation and enhanced financial protection, especially for the poor, but they have failed to facilitate outpatient service use among chronic patients. Non-communicable diseases significantly affect Georgia's ageing population. Consequently, in this paper, we look at the evidence emerging from determinants analysis of outpatient service utilisation and if the finding can help identify possible policy choices in Georgia, especially regarding benefit package design for individuals with chronic conditions. METHODS: We used Andersen's behavioural model of health service utilisation to identify the critical determinants that affect outpatient service use. A multinomial logistic regression was carried out with complex survey design using the data from two nationally representative cross-sectional population-based health utilisation and expenditure surveys conducted in Georgia in 2007 and 2010, which allowed us to assess the relationship between the determinants and outpatient service use. RESULTS: The study revealed the determinants that significantly impede outpatient service use. Low income, 45- to 64-year-old Georgian males with low educational attainment and suffering from a chronic health problem have the lowest odds for service use compared to the rest of the population. CONCLUSIONS: Using Andersen's behavioural model and assessing the determinants of outpatient service use has the potential to inform possible policy responses, especially those driving services use among chronic patients. The possible policy responses include reducing financial access barriers with the help of public subsidies for sub-groups of the population with the lowest access to care; focusing/expanding state-funded benefits for the most prevalent chronic conditions, which are responsible for the greatest disease burden; or supporting chronic disease management programs for the most prevalent chronic diseases and for special age groups aimed at the timely detection, education and management of chronic patients.

Risk of Gastrointestinal Bleeding Among Dabigatran Users - A Self Controlled Case Series Analysis.

This article aims to evaluate the real world risk of gastrointestinal bleeding among users naïve to dabigatran. We adopted a self-controlled case series design. We sampled 1215 eligible adult participants who were continuous insured users between July 1, 2010 and March 31, 2012 with use of dabigatran and at least one gastrointestinal bleeding episode. We used a conditional Poisson regression to estimate incidence rate ratios. The population consisted of 64.69% of male and 60.25% patients equal to or greater than age 65 at start of observation. After adjustment for time-variant confounders, the incidence rate of gastrointestinal bleeding was similar during dabigatran risk period and non-exposed period (incidence rate ratio [IRR] = 1.01, 95% confidence interval [CI] 0.90, 1.15). There was no significant difference in GI incidence rate between periods of dabigatran and warfarin (IRR = 0.99, 95% CI 0.75-1.31). Among this database of young and healthy participants, dabigatran was not associated with increased incidence rate of GI bleeding compared with non-exposed period among naïve dabigatran users. We did not detect an increased risk of GI bleeding over dabigatran vs warfarin risk period. Along with other studies on safety and effectiveness, this study should help clinicians choose the appropriate anticoagulant for their patients.

Understanding HIV Care Provider Attitudes Regarding Intentions to Prescribe PrEP.

INTRODUCTION: Preexposure prophylaxis (PrEP) is a promising approach to reducing HIV incidence. So, garnering the support of HIV providers, who are most familiar with antiretrovirals and likely to encounter patients in HIV serodiscordant relationships, to scale-up PrEP implementation is essential. We sought to determine whether certain subgroups of HIV providers were more likely to intend to prescribe PrEP. METHODS: Surveys were administered to HIV providers in Miami, Florida and Washington, District of Columbia. Composite scores were developed to measure PrEP knowledge, experience, and likelihood of prescribing. Latent class analysis was used to stratify provider attitudes toward PrEP. RESULTS: Among 142 HIV providers, 73.2% had cared for more than 20 HIV-infected patients in the previous 3 months; 17% had ever prescribed PrEP. Latent class analysis identified 2 classes of providers (entropy, 0.904); class 1 (n = 95) found PrEP less effective and perceived barriers to prescribing it; class 2 (n = 47) perceived PrEP as moderately effective and perceived fewer barriers to prescribing it. Compared with class 2, class 1 had significantly less experience with PrEP delivery (t(22.7) = 2.88, P = 0.009) and was significantly less likely to intend to prescribe to patients with multiple sex partners (20% vs. 43%, P = 0.04) and those with a drug use history (7% vs. 24%, P = 0.001). CONCLUSIONS: Although most HIV providers found PrEP to be effective, those considering it less effective had limited knowledge and experience with PrEP and had lesser intentions to prescribe. Provider training regarding whom should receive PrEP and addressing potential barriers to PrEP provision are needed if this HIV prevention method is to be optimized.

Risk of gastrointestinal bleeding associated with oral anticoagulants: population based retrospective cohort study.

OBJECTIVES: To determine the real world safety of dabigatran or rivaroxaban compared with warfarin in terms of gastrointestinal bleeding. DESIGN: Retrospective cohort study. SETTING: Large administrative database of commercially insured people in United States from 1 October 2010 through 31 March 2012. PARTICIPANTS: Enrollees with a prescription of warfarin, dabigatran, or rivaroxaban between 1 October 2010 and 31 March 2012, who were aged 18 years or older, had continuous enrollment and no oral anticoagulant use during the six months before the entry date, with known age and sex, and with no gastrointestinal bleeding for at least six months before the cohort entry date. The final study sample of 46,163 patients included 4907 using dabigatran, 1649 using rivaroxaban, and 39,607 using warfarin. MAIN OUTCOME MEASURE: Time to gastrointestinal bleeding. Hazard ratios were derived from Cox proportional hazard models with propensity score weighting and robust estimates of errors. RESULTS: Dabigatran users tended to be older (dabigatran v rivaroxaban v warfarin: 62.0 v 57.6 v 57.4 years) and more likely to be male (69% v 49% v 53%). The rate of gastrointestinal bleeding was highest among dabigatran users and lowest among rivaroxaban users (dabigatran v rivaroxaban v warfarin: 9.01 v 3.41 v 7.02 cases per 100 person years). After adjustment for potentially confounding covariates, there was no evidence of a statistically significant difference in the risk of gastrointestinal bleeding between dabigatran and warfarin users (adjusted hazard ratio 1.21, 95% confidence interval 0.96 to 1.53) or between rivaroxaban and warfarin users (0.98, 0.36 to 2.69). CONCLUSIONS: Although rates of gastrointestinal bleeding seem to be similar in this commercially insured sample of adults in the United States, we cannot rule out as much as a 50% increase in the risk of gastrointestinal bleeding with dabigatran compared with warfarin or a more than twofold higher risk of bleeding with rivaroxaban compared with warfarin.

Sorting through the lost and found: are patient perceptions of engagement in care consistent with standard continuum of care measures?

BACKGROUND: Indicators for determining one's status on the HIV care continuum are often measured using clinical and surveillance data but do not typically assess patient perspectives. We assessed patient-reported care status along the care continuum and whether it differed from medical records and surveillance data. METHODS: Between June 2013 and October 2014, a convenience sample of clinic-attending HIV-infected persons was surveyed regarding care-seeking behaviors and self-perceived status along the care continuum. Participant responses were matched to DC Department of Health surveillance data and clinic records. Participants' care patterns were classified using Health Resources Services Administration-defined care status: in care (IC), sporadic care (SC), or out of care (OOC). Semistructured qualitative interviews were analyzed using an open coding process to elucidate relevant themes regarding participants' perceptions of engagement in care. RESULTS: Of 169 participants, most were male participants (64%) and black (72%), with a mean age of 50.7 years. Using self-reported visit patterns, 115 participants (68%) were consistent with being IC, 33 (20%) SC, and 21 (12%) OOC. Among OOC participants, 52% perceived themselves to be fully engaged in HIV care. In the previous year, among OOC participants, 71% reported having a non-HIV-related medical visit and 90% reported current antiretroviral use. Qualitatively, most SC and OOC persons did not see their HIV providers regularly because they felt healthy. CONCLUSIONS: Participants' perceptions of HIV care engagement differed from actual care receipt as measured by surveillance and clinical records. Measures of care engagement may need to be reconsidered as persons not receiving regular HIV care maybe accessing other health care and HIV medications elsewhere.

Anti-diabetic therapies and the risk of acute pancreatitis: a nationwide retrospective cohort study from Taiwan.

AIMS: To examine the relationship between different anti-diabetic therapies (dipeptidyl peptidase-4 (DPP-4), metformin and sulfonylureas) and risk of acute pancreatitis among type 2 diabetic patients in Taiwan, and explore each drug's dose-response relationship. MATERIALS AND METHODS: We derived a nationwide retrospective cohort of patients with type 2 diabetes in Taiwan. The inclusion criteria are adult diabetic patients with continuous baseline enrollment, new users of the studied drugs, and without missing demographics. There were 4113/101 498/44 772 DPP-4/Metformin/Sulfonylurea users. Adjusted hazards ratios for pancreatitis associated with DPP-4, derived from Cox proportional hazard models with propensity score weighting, were estimated; dose-response analyses were also conducted. RESULTS: Dipeptidyl peptidase-4 was statistically significantly associated with a decreased risk of acute pancreatitis compared with sulfonylureas (adjusted HR: 0.36, 95%CI [0.17, 0.75]) but not metformin (adjusted HR: 0.67, 95%CI [0.32, 1.41]); metformin was statistically significantly associated with a lower risk of pancreatitis than sulfonylurea (adjusted HR: 0. 53; 95%CI [0.37, 0.76]). In addition, low-dose metformin was statistically significantly associated with a lower risk of pancreatitis compared with high-dose metformin (HR: 0.65; 95%CI [0.44, 0.97]). CONCLUSIONS: Our findings suggest that sulfonylureas may potentially be associated with an increased risk of pancreatitis compared with DPP-4 or metformin. Studies with longer follow up, larger sample sizes, and more precise capture of confounders may be needed to determine the risk of pancreatitis associated with incretin based therapies.