Risk of Thyroid Cancer Associated with Use of Liraglutide and Other Antidiabetic Drugs in a US Commercially Insured Population.

BACKGROUND: Quantify association between the glucagon-like peptide-1 receptor agonist liraglutide and risk of thyroid cancer (TC) compared to other antidiabetics. PATIENTS AND METHODS: Initiators of liraglutide, exenatide, metformin, pioglitazone or groups of dipeptidyl peptidase-4 inhibitors or sulfonylureas were identified in a US health plan (2010-2014) and followed for a median of 17 months. Thyroid cancer cases during follow-up were identified via a validated algorithm. Incidence rates of TC among liraglutide and comparators were assessed using relative risks estimated within propensity score-matched cohorts using intention to treat (ITT) and time on drug analyses. Latency effects and potential surveillance bias were evaluated. RESULTS: Relative risks from ITT analyses ranged from 1.00 (95% confidence interval (CI) 0.56-1.79) versus metformin to 1.70 (95% CI 1.03-2.81) versus all comparators excluding exenatide. Effect estimates from latency analyses were slightly attenuated. Time on drug analyses suggested no increased risk for either longer duration or higher cumulative dose of liraglutide. Medical record review found 85% were papillary or a follicular variant of papillary or both; 46% were microcarcinomas (≤10 millimeters), which were more prevalent in the liraglutide cohort (67% versus 43% in all comparators). CONCLUSION: Relative risks were elevated for several comparisons, which should be interpreted cautiously because of potential residual confounding and surveillance bias. Liraglutide cases had smaller thyroid nodules and shorter time-to-diagnosis, suggesting increased surveillance for TC among liraglutide initiators, especially shortly after the drug´s approval. After adjusting the primary analyses (ITT) for latency, no significant elevated risk of TC was observed among liraglutide initiators.

Development and Validation of a Risk Prediction Model for Cesarean Delivery After Labor Induction.

Objective: The goal of the study was to develop and validate a prediction model for cesarean delivery after labor induction that included factors known before the start of induction, unlike prior studies that focused on characteristics at the time of induction. Materials and Methods: Using 17,370 term labor inductions without documented medical indications occurring at 14 U.S. hospitals, 2007-2012, we created and evaluated a model predicting cesarean delivery. We assessed model calibration and discrimination, and we used bootstrapping for internal validation. We externally validated the model by using 2122 labor inductions from a hospital not included in the development cohort. Results: The model contained eight variables-gestational age, maternal race, parity, maternal age, obesity, fibroids, excessive fetal growth, and history of herpes-and was well calibrated with good risk stratification at the extremes of predicted probability. The model had an area under the curve (AUC) for the receiver operating characteristic curve of 0.82 (95% confidence interval 0.81-0.83), and it performed well on internal validation. The AUC in the external validation cohort was 0.82. Conclusion: This prediction model can help providers estimate a woman's risk of cesarean delivery when planning a labor induction.

Liraglutide use and evaluation of pancreatic outcomes in a US commercially insured population.

AIMS: Both acute pancreatitis (AP) and pancreatic cancer (PC) have been areas of focus for studies of incretin drugs. This 5-year prospective cohort study aimed to quantify possible associations between liraglutide and risk of AP and PC as compared to other antidiabetic drugs (ADs). MATERIALS AND METHODS: Patients initiating liraglutide or other ADs who were enrolled in a US health plan (2010-2014) were included. Comparisons of AP and PC incidence rates were made between matched cohorts of liraglutide initiators and initiators of other ADs. Adjudicated AP cases and algorithm-based PC cases were identified. Propensity score-matched intention-to-treat (ITT) and time-on-drug (TOD) analyses were completed using Poisson regression. A latency analysis was performed for PC. RESULTS: Median follow-up was 405 days for AP cohorts (9995 liraglutide, 1:1 matched to all comparators) and 503 days for PC cohorts (35 163 liraglutide, 1:1 matched to all comparators). In the primary AP analysis, "current" use of liraglutide was not significantly associated with elevated risk across comparators (all comparators relative risk [RR] = 1.2; 95% confidence interval [CI], 0.6-2.3). ITT results were similar where, in the primary analysis, no RRs were significantly associated with PC (all comparators RR = 0.7; 95% CI, 0.3-1.4); latency and TOD analyses did not alter findings. There was no evidence of a dose-response effect. CONCLUSIONS: Liraglutide was not associated with an increased risk of AP or PC, although risk estimates were more variable for AP, and numbers of cases for both outcomes were limited because of the rarity of outcomes.

Exenatide use and incidence of pancreatic and thyroid cancer: A retrospective cohort study.

A retrospective cohort study, supplemented with a nested case-control study, was performed using two administrative databases from commercial health plans in the United States to compare the incidence of pancreatic and thyroid cancer among users of exenatide versus other antidiabetic drugs (OADs). Patients with type 2 diabetes who initiated exenatide or OADs between 1 June 2005 and 30 June 2015 were included. Pancreatic and thyroid cancers were identified using chart-validated algorithms in the cohort study. Cases in the nested case-control study were chart-confirmed pancreatic or thyroid cancers, and controls were sampled using risk-set sampling. The time-fixed analyses comparing 33 629 exenatide initiators with 49 317 propensity-score-matched OAD initiators yielded hazard ratios of 0.76 (95% confidence interval [CI] 0.47-1.21) for pancreatic cancer and 1.46 (95% CI 0.98-2.19) for thyroid cancer. Results in the time-dependent analyses by cumulative duration or dose were similar. Nested case-control analyses yielded rate ratios of 0.61 (95%CI, 0.37-1.00) for pancreatic cancer and 0.89 (95% CI, 0.64-1.24) for thyroid cancer. This observational study suggested exenatide use was not associated with an increased risk of pancreatic or thyroid cancer.

Is there an association between liraglutide use and female breast cancer in a real-world setting?

BACKGROUND: Liraglutide is a human glucagon-like peptide-1 receptor agonist approved for treatment of adults with type 2 diabetes mellitus at a maximum dose of 1.8 mg/day (Victoza®) and more recently at 3.0 mg/day for weight management (Saxenda®). During the evaluation of liraglutide for approval in weight management, a minor imbalance in the numbers of reported breast neoplasms was observed, motivating the present study. Our objective was to quantify the association between liraglutide and incidence of breast cancer (BC) among women in a real-world setting. PATIENTS AND METHODS: Women initiating liraglutide or other antidiabetic therapies and who were enrolled in a large US health plan (2010-2014) were included. Comparisons of BC incidence rates were made between matched cohorts of initiators of liraglutide and cohorts of initiators of exenatide, metformin, pioglitazone, sulfonylureas, and dipeptidyl peptidase-4 inhibitors separately and as two "all comparators" groupings: with or without exenatide. Women with two or more claims with BC diagnosis codes within 61days of each other were identified as possible cases, with additional confirmation by clinician review of comprehensive claims listings. Propensity score matched intention-to-treat (ITT) and time-on-drug (TOD) analyses were completed via Poisson regression. A latency analysis was performed. RESULTS: Relative risks for BC for liraglutide vs comparators from the ITT analyses ranged from 0.90 (95% CI: 0.67-1.22) for both the "all comparator" and "all comparator except exenatide" cohorts to 1.46 (95% CI: 0.96-2.22) relative to exenatide. Latency analyses excluding the first year of follow-up yielded slightly attenuated point estimates. The TOD analyses of cumulative use of liraglutide suggested no increased risk of BC. CONCLUSION: Neither the ITT (overall or latency analysis) nor cumulative TOD analyses suggested an elevated risk of BC among liraglutide initiators. Short length of follow-up and the potential for confounding by unmeasured factors limit the full assessment of long-term risk.

Performance of claims-based algorithms for identifying incident thyroid cancer in commercial health plan enrollees receiving antidiabetic drug therapies.

BACKGROUND: Thyroid cancer incidence is increasing in the United States (US) and many other countries. The objective of this study was to develop and evaluate algorithms using administrative medical claims data for identification of incident thyroid cancer. METHODS: This effort was part of a prospective cohort study of adults initiating therapy on antidiabetic drugs and used administrative data from a large commercial health insurer in the US. Patients had at least 6 months of continuous enrollment prior to initiation during 2009-2013, with follow-up through March, 2014 or until disenrollment. Potential incident thyroid cancers were identified using International Classification of Diseases, 9th Revision (ICD-9) diagnosis code 193 (malignant neoplasm of the thyroid gland). Medical records were adjudicated by a thyroid cancer specialist. Several clinical variables (e.g., hospitalization, treatments) were considered as predictors of case status. Positive predictive values (PPVs) and 95% confidence intervals (CIs) were calculated to evaluate the performance of two primary algorithms. RESULTS: Charts were requested for 170 patients, 150 (88%) were received and 141 (80%) had sufficient information to adjudicate. Of the 141 potential cases identified using ≥1 ICD-9 diagnosis code 193, 72 were confirmed as incident thyroid cancer (PPV of 51% (95% CI 43-60%)). Adding the requirement for thyroid surgery increased the PPV to 68% (95% CI 58-77%); including the presence of other therapies (chemotherapy, radio-iodine therapy) had no impact. When cases were required to have thyroid surgery during follow-up and ≥2 ICD-9 193 codes within 90 days of this surgery, the PPV was 91% (95% CI 81-96%); 62 (82%) of the true cases were identified and 63 (91%) of the non-cases were removed from consideration by the algorithm as potential cases. CONCLUSIONS: These findings suggest a significant degree of misclassification results from relying only on ICD-9 diagnosis codes to detect thyroid cancer. An administrative claims-based algorithm was developed that performed well to identify true incident thyroid cancer cases.

Inappropriate Fentanyl Prescribing Among Nursing Home Residents in the United States.

OBJECTIVE: We quantified transdermal fentanyl prescribing in elderly nursing home residents without prior opioid use or persistent pain, and the association of individual and facility traits with opioid-naïve prescribing. DESIGN: Cross-sectional study. SETTING: Linked Minimum Data Set (MDS) assessments; Online Survey, Certification and Reporting (OSCAR) records; and Medicare Part D claims. PARTICIPANTS: From a cross-section of all long-stay US nursing home residents in 2008 with an MDS assessment and Medicare Part D enrollment, we identified individuals (≥65 years old) who initiated transdermal fentanyl, excluding those with Alzheimer disease, severe cognitive impairment, cancer, or receipt of hospice care. MEASUREMENTS: We used Medicare Part D to select beneficiaries initiating transdermal fentanyl in 2008 and determined whether they were "opioid-naïve," defined as no opioid dispensing during the previous 60 days. We obtained resident and facility characteristics from MDS and OSCAR records and defined persistent pain as moderate-to-severe, daily pain on consecutive MDS assessments at least 90 days apart. We estimated associations of patient and facility attributes and opioid-naïve fentanyl initiation using multilevel mixed effects logistic regression modeling. RESULTS: Among 17,052 residents initiating transdermal fentanyl, 6190 (36.3%) were opioid-naïve and 15,659 (91.8%) did not have persistent pain. In the regression analysis with adjustments, residents who were older (ages ≥95 odds ratio [OR] 1.69, 95% confidence interval [CI] 1.46-1.95) or more cognitively impaired (moderate-to-severe cognitive impairment, OR 1.99, 95% CI 1.73-2.29) were more likely to initiate transdermal fentanyl without prior opioid use. CONCLUSION: Most nursing home residents initiating transdermal fentanyl did not have persistent pain and many were opioid-naïve. Changes in prescribing practices may be necessary to ensure Food and Drug Administration warnings are followed, particularly for vulnerable subgroups, such as the cognitively impaired.

Implications of immortal person-time when outcomes are nonfatal.

PURPOSE: The amount of immortal time bias in studies with nonfatal outcomes is unclear. To quantify the magnitude of bias from mishandling of immortal person-time in studies of nonfatal outcomes. METHODS: We derived formulas for quantifying bias from misclassified or excluded immortal person-time in settings with nonfatal outcomes, assuming a constant rate of outcome. In the situation of misclassified or excluded immortal person-time, the quantification includes the immortal time and corresponding events mistakenly attributed to the exposed group (misclassified) or excluded from study (excluded) that must be attributed to the comparison group. RESULTS: With misclassified immortal person-time, the magnitude of bias varies according to the incidence rate ratio of immortal time and comparison group as well as the rate ratio of immortal time and exposed group: toward null for both ratios less than 1, no bias for both ratios equal to 1, away from null for both ratios greater than 1. For one ratio less than 1 and the other greater than 1, the direction and magnitude of bias can be obtained from the formula provided. With excluded immortal person-time, the magnitude of bias is associated with the incidence rate ratio of immortal time and comparison group: toward null for the ratio less than 1, no bias for the ratio equal to 1, and away from null for the ratio greater than 1. CONCLUSIONS: Bias due to immortal person-time in studies with nonfatal outcomes can vary widely and can be quantified under assumptions that apply to many studies.

Patterns of Lipid Lowering Therapy among Children Ages 8-20 Years.

OBJECTIVES: Pediatric guidelines in 2008 and 2011 recommended lipid lowering therapy in children ≥ 8 years of age with high-risk cardiovascular conditions, such as familial hypercholesterolemia (FH). Our objective was to describe the patterns and predictors of lipid lowering therapy initiation in commercially insured children between 2005 and 2010. STUDY DESIGN: Using commercial health plan data on children ages 8-20 years from 2004-2010, we estimated rates of lipid lowering therapy initiation overall and stratified by age. Using a nested case-control design, we used multivariable logistic regression to identify temporal, demographic, clinical, and health utilization characteristics associated with lipid lowering therapy initiation. RESULTS: Among >13 million children, 665 initiated lipid lowering therapy for an incidence rate 2.6/100,000 person-years (PY). Incidence rates were highest in 2005 (4.1/100,000 PY) and 2008 (3.9/100,000 PY), with no discernable secular trend. Rates of lipid lowering therapy initiation were significantly greater in children ≥ 15 years of age (OR 2.9 [95% CI 5.2-13.0]), males (2.1 [1.7-2.4]), and those with a diagnosis of FH (165.2 [129.0-211.6]), other dyslipidemia (175.5 [143.2-215.3]), diabetes type I (7.7 [4.7-12.4]), diabetes type II (13.6 [8.5-21.7]), hypertension (8.1 [4.9-13.3]), obesity (7.8 [4.7-12.7]), and ≥ 5 outpatient visits (1.5 [1.2-1.7]), and children with dispensing of ≥ 2 nonlipid lowering therapy prescriptions were less likely to initiate lipid lowering therapy (0.2 [0.2-0.3]). CONCLUSIONS: Despite new guidelines, lipid lowering therapy initiation in children is low and has not increased through 2010. Although diagnosis of FH and other dyslipidemias was associated with higher probability of lipid lowering therapy initiation, our findings suggest lipid lowering therapy is underutilized in this population given the prevalence of these disorders.

A pooled analysis of exenatide use and risk of acute pancreatitis.

OBJECTIVE: To estimate the association between exenatide BID use and acute pancreatitis across two claims-based studies. RESEARCH DESIGN AND METHODS: We pooled two cohort studies within separate commercial health insurance databases. We included initiators of exenatide BID and all other antihyperglycemic drugs without prior pancreatitis from 2005-2008. Poisson regression models provided rate ratios (RRs) and 95% confidence intervals (CIs) of the association of exenatide BID with acute pancreatitis adjusted for quintiles of propensity scores. MAIN OUTCOME MEASURES: Primary inpatient diagnoses of acute pancreatitis with correction for misclassification via a validation sub-study. RESULTS: There were 49,956 initiators of exenatide BID and 692,333 initiators of other antihyperglycemic drugs. Patients in the two studies were similar on many demographic and clinical characteristics. Exenatide BID initiators had a higher prevalence of diagnoses consistent with diabetes complications (e.g. peripheral neuropathy) and cardiovascular risk factors (e.g. hypertension). In both studies, current exenatide BID use was not associated with uncorrected outcomes of acute pancreatitis (pooled RR 1.0; CI 0.8-1.3). PPV correction resulted in a slightly higher point estimate for current use (pooled RR 1.3; CI 1.0-1.7) and past use (pooled RR 1.6; 95% CI 1.2-2.1). CONCLUSIONS: These data are consistent with little or no higher risk of acute pancreatitis associated with current exenatide BID use relative to nonuse. Although previous work identified non-causal mechanisms, an increased incidence of acute pancreatitis following cessation of treatment remains a possibility. Bias due to residual confounding or outcome misclassification may remain, and should be considered a potential explanation for these findings.

Different analyses estimate different parameters of the effect of erythropoietin stimulating agents on survival in end stage renal disease: a comparison of payment policy analysis, instrumental variables, and multiple imputation of potential outcomes.

OBJECTIVE: To compare the assumptions and estimands across three approaches to estimate the effect of erythropoietin-stimulating agents (ESAs) on mortality. STUDY DESIGN AND SETTING: Using data from the Renal Management Information System, we conducted two analyses using a change to bundled payment that, we hypothesized, mimicked random assignment to ESA (pre-post, difference-in-difference, and instrumental variable analyses). A third analysis was based on multiply imputing potential outcomes using propensity scores. RESULTS: There were 311,087 recipients of ESAs and 13,095 non-recipients. In the pre-post comparison, we identified no clear relationship between bundled payment (measured by calendar time) and the incidence of death within 6 months (risk difference -1.5%; 95% confidence interval [CI] -7.0%, 4.0%). In the instrumental variable analysis, the risk of mortality was similar among ESA recipients (risk difference -0.9%; 95% CI -2.1, 0.3). In the multiple imputation analysis, we observed a 4.2% (95% CI 3.4%, 4.9%) absolute reduction in mortality risk with the use of ESAs, but closer to the null for patients with baseline hematocrit level >36%. CONCLUSION: Methods emanating from different disciplines often rely on different assumptions but can be informative about a similar causal contrast. The implications of these distinct approaches are discussed.

Linkage of routinely collected oncology clinical data with health insurance claims data--an example with aromatase inhibitors, tamoxifen, and all-cause mortality.

PURPOSE: Studies of cancer based solely on health insurance claims data typically lack information on cancer clinical characteristics that are strong predictors of treatment and prognosis. Our objective was to evaluate routinely collected cancer clinical data for adjustment of confounding using an example evaluation of mortality associated with aromatase inhibitors and tamoxifen. METHODS: This cohort study identified women with breast cancer from 2008 through 2010 using health insurance claims data linked to clinical information on stage at diagnosis, current clinical status, histology, and other clinical markers. Estimated mortality rates (MRs) and 95% confidence intervals (CI) were compared between users of aromatase inhibitors or tamoxifen, adjusted for claims-identified covariates and additionally for the clinical variables using propensity scores and proportional hazards regression models. RESULTS: The overall (n = 7974) estimated MR was 69/1000 person-years (95%CI = 62-76 person-years), 308/1000 person-years (95% CI = 273-345 person-years) for women with metastasis, and 12/1000 person-years (95%CI = 8-16 person-years) for women without active cancer. Propensity score matching of aromatase inhibitor users (n = 777) with tamoxifen users (n = 535) removed many, but not all, covariate imbalances. The hazard ratios (HRs) of all-cause mortality comparing users of aromatase inhibitors with tamoxifen users ranged from 1.0 to 1.6, with the HR most similar to previous clinical trials (0.87) coming from the claims-only analysis. CONCLUSIONS: We were able to address potential unmeasured confounders by linking clinical information to the claims data; however, there was no apparent improvement in confounding control in the chosen example. Conditioning eligibility on the clinical data restricted the sample size substantially.

Incidence of health insurance claims for thyroid neoplasm and pancreatic malignancy in association with exenatide: signal refinement using active safety surveillance.

OBJECTIVES: As part of a regulatory postmarketing commitment, we assessed the risk of claims for thyroid and pancreatic cancer among users of exenatide using an active drug safety surveillance system. METHODS: This active surveillance assessment used cohort methodology and commercial health insurance claims data to identify initiators of exenatide and propensity score-matched initiators of metformin or glyburide between June 2005 and September 2009, with up to 1 year of follow up through December 2009. The primary analysis estimated absolute and relative risk (RR) of inpatient or outpatient claims with diagnosis codes for thyroid neoplasm (benign or malignant) or pancreatic malignancies after exclusion of patients with a history of the same diagnosis at baseline. RESULTS: Among the matched comparison cohorts (N ≈ 32,800 each), there were 37 claims-suggested thyroid malignancies among exenatide initiators and 26 among metformin or glyburide initiators [RR 1.4; 95% confidence interval (CI) 0.8-2.4]. This association was attenuated when limited to inpatient thyroid cancer claims (RR 0.9; CI 0.3-2.6). Exenatide use was not associated with an increased risk of benign thyroid neoplasm (RR 0.7; CI 0.3-1.7), or pancreatic cancer (RR 0.8; CI 0.5-1.6). CONCLUSIONS: Use of exenatide was associated with a modestly higher incidence of inpatient and outpatient claims, but not inpatient claims for thyroid malignancies. Exenatide was not associated with higher risk of benign thyroid neoplasm or pancreatic cancer. Misclassification of outcomes and exposure, and residual confounding remain limitations of this analysis to be considered when interpreting the results. We have initiated a formal epidemiologic investigation to explore these relationships.

Atypical antipsychotic use and risk of fracture in persons with Parkinsonism.

Our objective was to estimate the effect of atypical antipsychotics (AAs) on the rate of fractures in a parkinsonism population. We conducted an age- and state-matched nested case-control study in five states (CA, FL, NY, OH, IL) using the Medicaid analytic extract from 2001 to 2002. Eligible participants had a diagnosis of parkinsonism, excluding persons with secondary parkinsonism, bone cancer, bone infections, schizophrenia, schizoaffective disorder, and those who used conventional antipsychotics. The primary outcome was the occurrence of a fracture of the femur, ankle, fibula, tibia, humerus, radius, or ulna (N = 851). Risk-set sampling defined controls (N = 4220). We used conditional-logistic regression to derive adjusted odds ratios (AOR) and 95% confidence intervals of the association between fracture and use of quetiapine, risperidone, or olanzapine in the 60 days before the index date compared to nonuse. After adjustment for confounding, use of quetiapine (AOR 2.4; 95% CI 1.5-3.8), risperidone (AOR 1.2; 95% CI 0.9-1.7), or olanzapine (AOR 1.7; 95% CI 1.2-2.4) was associated with a higher rate of fracture. Use of an AA was associated with a higher rate of fracture in persons with parkinsonism. Prescribers must be cautious when using these agents in elderly persons with parkinsonism.

Use of a claims-based active drug safety surveillance system to assess the risk of acute pancreatitis with exenatide or sitagliptin compared to metformin or glyburide.

OBJECTIVE: To estimate risk and relative risk (RR) of acute pancreatitis among patients using incretin-based diabetes therapies (exenatide or sitagliptin) compared to patients treated with agents with established safety profiles (metformin or glyburide). RESEARCH DESIGN AND METHODS: The study population was derived from a large US commercial health insurance transaction database using an active drug safety surveillance system (i3 Aperio). This analysis is based on data from June 2005 through June 2008. Cohorts of exenatide and sitagliptin initiators were each matched to an equal number of metformin or glyburide (met/gly) initiators using propensity scores to reduce confounding in the comparison of outcomes during follow-up. Patients with claims suggesting pancreatic disease in the 6 months prior to cohort entry were excluded. MAIN OUTCOME MEASURE: Claims for hospitalizations associated with a primary diagnosis of acute pancreatitis (ICD-9 577.0). RESULTS: There were 27 996 exenatide initiators and 16 276 sitagliptin initiators and approximately equal numbers of matched comparators. During follow-up of up to 1 year, acute pancreatitis occurred among 0.13% of patients treated with exenatide and 0.12% of patients treated with sitagliptin. The risk of acute pancreatitis was comparable for initiators of exenatide (RR 1.0; 95% confidence interval (CI) 0.6-1.7) and sitagliptin (RR 1.0; 95% CI 0.5-2.0) relative to the comparison cohorts. CONCLUSIONS: These data do not provide evidence for an association of acute pancreatitis among initiators of exenatide or sitagliptin compared to met/gly initiators. These results are limited by the data available in an administrative, healthcare database.

Association between statin use and risk for keratinocyte carcinoma in the veterans affairs topical tretinoin chemoprevention trial.

BACKGROUND: Recent evidence suggests that statins may prevent cancer. OBJECTIVE: To quantify the association between statin use and the occurrence of keratinocyte carcinoma in high-risk veterans. DESIGN: Cohort study. SETTING: 6 Veterans Affairs medical centers. PARTICIPANTS: 1037 participants of the Veterans Affairs Topical Tretinoin Chemoprevention Trial, a randomized, multicenter, double-blind, vehicle-controlled trial of topical tretinoin, 0.1%, for prevention of keratinocyte carcinoma conducted from November 1998 to November 2004. MEASUREMENTS: Time to first occurrence of keratinocyte carcinoma on the face or ears. Participants using a statin at randomization, according to the Veterans Affairs Pharmacy Benefits Management database, were considered exposed. Study dermatologists conducted physical examinations at baseline and every 6 months during follow-up. The association between statin use at randomization and the outcome was evaluated by using propensity score matching (n = 608) and Cox proportional hazards regression (n = 1037). RESULTS: Among the 1037 participants, 37% used a statin at randomization (n = 397) for a median duration of at least 900 days over a median follow-up of 3.5 years. In the propensity score-matched analysis, statin use at randomization was not associated with keratinocyte carcinoma (rate ratio, 0.92 [95% CI, 0.73 to 1.16]), a finding that was consistent with the estimates derived from the Cox proportional hazards regression (rate ratio, 0.84 [CI, 0.70 to 1.02]). LIMITATIONS: The extent of residual confounding is unknown, and the confidence bounds around the measures of association were wide. These data may not be generalizable to lower-risk populations. CONCLUSION: These data show no conclusive or consistent relationship between long-term statin use and risk for keratinocyte carcinoma.