RMRobyn L. McClelland

Robyn L. McClelland

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Robyn L. McClelland has a total of 332 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Study design and rationale for the Dog Aging Project Pack: a prospective, long-term, longitudinal, observational study of biological and environmental determinants of healthy aging in US companion dogs

    by Amber J. Keyser, Audrey Ruple, Matthew D. Dunbar, Paul E. Litwin, Mandy Kauffman, Robyn McClelland, Brooke Benton, Crystal R. Arkenberg, Weihsueh A. Chiu, Joshua M. Akey, et al.

    Companion dogs are a valuable source of translational discovery about the basic biology of aging because of their well-characterized genome, phenotypic variation in lifespan, aging, and disease profiles, and diverse environmental exposures shared with their human owners. The Dog Aging Project (DAP)’s prospective, longitudinal open cohort study was begun in 2019 to investigate the biological and environmental determinants of healthy aging in companion dogs. As of December 31, 2024, the study has enrolled 50,188 companion dogs across all 50 US states and the District of Columbia, including over half with follow-up data, spanning up to five years, and 14,978 documented deaths. The DAP is a community science research program in which owners volunteer to provide detailed longitudinal information about their dogs. This manuscript describes the recruitment and retention of this study population, known as the DAP Pack, including the data collection processes, and the means by which DAP Open Data are shared with the research community. The goals of this manuscript are to present the detailed methodology of the DAP as a resource to support authors and readers of DAP publications, to provide the scientific community with a framework to support the design of similar future studies, and to facilitate use of DAP data by investigators with Open Data access.

  • Development and Validation of a Coronary Artery Calcium Volume and Density Score

    by Harpreet S. Bhatia, Alexander C. Razavi, Zeina Dardari, Robyn L. McClelland, David R. Jacobs, James G. Terry, J. Jeffrey Carr, Daniel Duprez, Parag H. Joshi, Anand Rohatgi, et al.

    BACKGROUND: Coronary artery calcium (CAC) scoring is a useful tool for coronary heart disease (CHD) risk prediction. The current standard method, the Agatston score, weights plaque area upward by density. However, higher density is inversely associated with CHD. We aimed to develop and validate a CAC score accounting for the inverse association of density with CHD, and the nonlinear association between CAC volume and density in a prospective cohort study. METHODS: A CAC volume and density score was developed, which adjusts CAC volume for density based on whether the volume level was 0 and follow-up for CHD events. Score performance for CHD risk prediction was compared with the pooled cohort equations and the Agatston score using Cox proportional hazards models and Harrell C -index and continuous and categorical net reclassification improvement. Analyses were also stratified by Agatston score, age, sex, race/ethnicity, and statin use. Finally, performance was evaluated in 4 longitudinal cohorts for external validation. RESULTS: The MESA cohort included 3360 individuals. CAC volume and density score improved the C -index for CHD to a greater degree over the pooled cohort equation (0.699 [95% CI, 0.675–0.723]) compared with the Agatston score (0.687 [95% CI, 0.663–0.711]) and resulted in greater continuous net reclassification improvement (0.54 [95% CI, 0.43–0.64]) compared with the Agatston score (0.45 [95% CI, 0.34–0.56]) when added to the pooled cohort equation. The greatest improvement in risk prediction was in those with lower burden of CAC (Agatston <300). Consistent results were seen in stratified analyses and external validation cohorts. CONCLUSIONS: The CAC volume and density score, which adjusts CAC volume downward for higher density, and can be calculated from existing CAC reports, significantly improves risk discrimination and reclassification compared with the Agatston score.


  • Open Access

    Association between depressive symptoms and thoracic aortic calcification, and their independent and joint prognostic value for incident atherosclerotic cardiovascular disease in the MESA cohort

    by Daiki Arai, Quinn White, Keishi Ichikawa, Robyn L. McClelland, W. Craig Johnson, Mio Yabuki, Spencer Hansen, Matthew J. Budoff

    Depression is highly prevalent; however, its association with atherosclerotic cardiovascular disease (ASCVD) events remains incompletely understood. This study aimed to evaluate the association between depressive symptoms and thoracic aortic calcification (TAC) and assess their joint prognostic value for future ASCVD events. We analyzed 2,600 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) without baseline ASCVD. TAC was quantified using non-contrast chest CT as a marker of subclinical atherosclerosis. Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D ≥ 16). Linear regression assessed the association between depressive symptoms and log-transformed TAC. Cox proportional hazards models evaluated independent and joint associations of depressive symptoms and TAC tertiles with incident ASCVD, adjusting for covariates. Overall, 389 participants (15%) exhibited depressive symptoms. In fully adjusted models, depressive symptoms were not associated with TAC (β = 0.08; 95% CI -0.13–0.29), without sex-specific differences. Both depressive symptoms (HR 1.63; 95% CI 1.16–2.29) and TAC (HR 1.22; 95% CI 1.13–1.33) were independently associated with ASCVD events. Individuals with depressive symptoms had significantly higher ASCVD risk in the highest tertile TAC group, whereas no difference was observed in lower TAC tertiles. Although depressive symptoms were not directly associated with TAC, their presence was significantly associated with increased ASCVD risk, particularly among participants with elevated TAC. Coexisting depressive symptoms and substantial atherosclerotic burden confer a higher ASCVD risk. Thus, evaluating depressive symptoms may provide additional context for cardiovascular risk stratification, particularly among individuals with advanced subclinical atherosclerosis.

  • Predictive Value of Coronary Artery Calcium Score From Nongated Chest CT Scans Compared With Gated Scans and Incidence of Cardiovascular Events

    by Dhiran Verghese, Robert Trujillo, Divya Abraham, Robyn L. McClelland, April Kinninger, Sanjay Manubolu, Jairo Aldana, Robert J. Cubeddu, Douglas Drachman, Ignacio Inglessis-Azuaje, et al.

    BACKGROUND: Data on the prognostic value of incidental coronary artery calcium (CAC) from nongated chest computed tomography scans are limited. Using paired computed tomography scans from the MESA (Multi-Ethnic Study of Atherosclerosis) exam 5, we evaluated the relationship between nongated CAC, gated CAC, and future cardiovascular events. METHODS: Between April 2010 and December 2011, a total of 2601 participants underwent same-day gated and nongated chest computed tomography scans. After excluding 106 with previous coronary heart disease or cardiovascular disease and 23 for missing covariates, 2472 participants formed the study population. Gated and nongated scans were interpreted at a core laboratory, blinded to acquisition methodology. Cox regression examined associations between CAC (gated and nongated) and incident coronary heart disease/cardiovascular disease events. Correlation was assessed with Pearson coefficients, model performance with receiver operating characteristic curves and C-statistic, and overall accuracy with Brier scores. RESULTS: Among the 2472 participants, 53% were female, 38% white, 13% Chinese, 26% Black, and 23% Hispanic/Latino. Compared with participants with zero CAC, those with moderate (101–299) and severe (≥300) nongated CAC had higher coronary heart disease risk (hazard ratio, 2.67 [95% CI, 1.14–6.27]; hazard ratio, 5.22 [95% CI, 2.37–11.5]) and cardiovascular disease risk (hazard ratio, 1.32 [95% CI, 1.32–4.04]; hazard ratio, 2.89 [95% CI, 1.68–4.96]). Gated and nongated log-standardized CAC highly correlated ( r =0.961; P <0.001). The area under the receiver operating characteristic curves, C-statistic, and Brier scores were statistically similar for gated and nongated CAC. CONCLUSIONS: Nongated CAC predicts cardiovascular events with performance comparable to gated CAC. Given the large number of nongated scans performed annually, incorporating their quantification into clinical practice offers a scalable approach to personalized preventive care. Our findings are particularly relevant in light of the recent 2026 American College of Cardiology/American Heart Association dyslipidemia guidelines, which endorse the use of incidental CAC from nongated computed tomography scans for atherosclerotic cardiovascular disease risk stratification and guiding lipid-lowering therapy.


  • Reevaluating the Use of Race/Ethnicity in the MESA Risk Score

    by Quinn White, Spencer Hansen, Brittany Saldivar Murphy, Craig Johnson, Andrew DeFilippis, Wendy Post, Robyn McClelland

    Background: Cardiovascular disease risk scores are widespread in research and clinical settings. The MESA (Multi‐Ethnic Study of Atherosclerosis) risk score estimates the 10‐year risk of coronary heart disease and was novel in its use of coronary artery calcium alongside traditional risk factors, including race/ethnicity. However, including race/ethnicity in a risk score model requires careful consideration. We developed a race‐free MESA risk score and compared it with the original. Methods: We used data from the MESA cohort, a community‐based cohort of individuals aged 45 to 84 years who identified as White, Hispanic/Latino, Black, or Chinese and were free of prevalent cardiovascular disease upon study entry, to develop a race‐free risk score estimating 10‐year coronary heart disease risk. We used the same 2‐step regularization procedure as that for the original MESA risk score: (1) least absolute shrinkage and selection operator for variable selection, and (2) ridge regression to prevent overfitting. Neither race/ethnicity nor any interaction term with race/ethnicity was included as a candidate predictor. Results: When excluding race/ethnicity, several interaction terms were retained that were not retained in the original MESA risk score. The area under the receiver operating characteristic curve for the race‐free risk score was 0.820 (95% CI, 0.801–0.840), and the difference in area under the receiver operating characteristic curve was not significant (estimate, 0.0032 [95% CI, −0.0008 to 0.0073]). The discrimination slope in the race‐free model was 0.101, while that of the original MESA risk score was 0.0937. Conclusions: We developed a race‐free MESA risk score that performs comparably to the MESA risk score in both calibration and discrimination.

  • Open Access

    Association of Subclinical Cardiac Remodeling With Incident Cancer in the MESA Cohort

    by Xinjiang Cai, Quinn White, Jordan Juarez, Omar Chehab, Bharath Ambale‐Venkatesh, W. Craig Johnson, Spencer L. Hansen, Kosuke Inoue, Joao A.C. Lima, Karol E. Watson, et al.

    Background: Patients with established cardiovascular disease have an increased risk of cancer; however, the association between subclinical cardiac remodeling and cancer development remains unclear. We aimed to examine the association between cardiac magnetic resonance imaging–derived measures of subclinical cardiac remodeling and incident cancer in the MESA (Multi‐Ethnic Study of Atherosclerosis) cohort. Methods: Multivariable Cox proportional hazards models were used to estimate the association between each cardiac magnetic resonance imaging measure of interest and the incidence of cancer, separately, for an all‐inclusive cancer end point as well as prostate, breast, female‐specific, colorectal, and lung cancer subtypes. Results: Of 6214 participants with cancer data available (median age, 61.0 years; 52.7% women), 4595 had at least 1 cardiac magnetic resonance imaging–derived measure of interest, with 563 cancer events over a median follow‐up of 18.7 years. Overall cancer incidence was higher with greater left ventricular mass index or left ventricular mass/volume ratio, and lower with higher left atrial emptying fraction or peak left atrial longitudinal strain. In adjusted Cox proportional hazards models, left ventricular mass index was significantly associated with the hazard of breast cancer (hazard ratio [ HR] , 1.88 [95% CI , 1.19–2.97]; P =0.007), and peak left atrial strain was significantly associated with the hazard of colorectal cancer ( HR , 0.52 [95% CI , 0.33–0.82]; P =0.005). Conclusions: In the MESA cohort, elevated left ventricular mass index and worse (lower) peak left atrial strain were associated with an increased risk of breast cancer and colorectal cancer, respectively. Subclinical cardiac remodeling may serve as an early marker to identify individuals at risk for both cardiovascular disease and cancer.

  • Open Access

    Impact of Renal Disease Markers on Cardiovascular Disease Risk Prediction When Coronary Calcium Score Is Known

    by Joshua Barzilay, Robert Trujillo, Spencer Hansen, Michael Blaha, Matthew Budoff, Robyn L. McClelland

    Background: Markers of renal function have been added to the Predicting Risk of Cardiovascular Disease Events and Systematic Coronary Risk Evaluation cardiovascular risk calculators to enhance risk prediction. Here we examine the role of estimated glomerular filtration (eGFR) and urine albumin creatinine ratio (UACR)—and related dichotomous cut points (eGFR 2 and albuminuria ≥30 mg albumin/gram creatinine)—for prediction of coronary heart disease, cardiovascular disease (CVD) outcomes (CVD mortality, heart failure, stroke, total CVD), and total mortality, using the MESA (Multi‐Ethnic Study of Athersclerosis) calculator, which includes coronary artery calcium scores as a predictor in addition to traditional CVD risk factors. Methods: The study included 6707 participants without clinical CVD with coronary artery calcium scoring. Cox proportional hazards models, adjusted for covariates including coronary artery calcium, were used to gauge the association of UACR, eGFR, albuminuria, and eGFRP 65 years of age, and for all outcomes in participants with diabetes. Conclusions: When coronary artery calcium is known, albuminuria improves heart failure prediction. Albuminuria and UACR each improve total mortality prediction as well. UACR and albuminuria improve prediction for all outcomes in people with diabetes.

  • Association between absolute blood eosinophil count and carotid plaque and coronary artery calcification in the Multi-Ethnic Study of Atherosclerosis (MESA)

    by Nyla E Mathis, Mats W Johansson, Spencer L Hansen, Alison S Dasiewicz, Stephane Esnault, Sameer K Mathur, Loren C Denlinger, Robyn L McClelland, Nizar N Jarjour, James H Stein, et al.

    Prior studies have demonstrated associations between eosinophil activation products and incident stroke. We investigated whether circulating eosinophil concentration is associated with imaging markers of atherosclerotic cardiovascular disease (ASCVD), analyzing 2 arterial beds (carotid artery plaque [CAP] and coronary artery calcium [CAC]) in the Multi-Ethnic Study of Atherosclerosis (MESA). The MESA enrolled individuals aged 45 to 84, without ASCVD at baseline. ASCVD risk factors, blood eosinophils, CAP presence/score (0 to 12), and CAC presence/Agatston score were measured at MESA Exam No. 5. Using logistic and linear regression analyses, the relationships between blood eosinophils, CAP or CAC presence, or scores (log[score + 1]), adjusted for confounders, were evaluated. The 2,166 participants were a mean (standard deviation [SD]) 69.6 (9.3) years old, 53% female, 29% Hispanic, 28% Black, 1% Chinese American. The eosinophil count was median (interquartile range) 0.10 (0.10, 0.20)×103/μl, CAP score 2 (0, 4), and CAC score 45 (0, 292) Agatston units. In risk-factor adjusted models, increased blood eosinophil count (per 1 SD [0.15 × 103/μl]) was associated with log scores of CAP (β: 0.05 [95% confidence interval 0.02 to 0.08, P = 0.001) or CAC (β: 0.11, [0.01 to 0.21], P = 0.03). Similar associations were observed for eosinophils and CAP (odds ratio [OR]: 1.12, [1.01 to 1.25], P = 0.03) and CAC (OR: 1.15, [1.02 to 1.30], P = 0.02) presence. In conclusion, in a large, multiethnic US cohort, blood eosinophils were independently linked to imaging atherosclerosis measures in 2 arterial distributions. These data suggest potential roles of eosinophilic inflammation in atherosclerosis.


  • Open Access

    Validation of normal reference ranges in cardiac magnetic resonance imaging: The Multi-Ethnic Study of Atherosclerosis

    by Nadine Kawel-Boehm, Spencer L. Hansen, Bharath Ambale-Venkatesh, J. Jeffrey Carr, J. Paul Finn, Michael Jerosch-Herold, Steven M. Kawut, Robyn L. McClelland, Wendy Post, Martin R. Prince, et al.


  • Open Access

    Air pollution and cognitive performance in late life: the role of lipid profile – results from the Multi-Ethnic Study of Atherosclerosis (MESA)

    by Xi Pan, Robyn L. McClelland, Lilah M. Besser, Jana A. Hirsch, Joel D. Kaufman, Jing Cao, Timothy M. Hughes