Lewis H Kuller

University of Pittsburgh, Pittsburgh, Pennsylvania, United States

Are you Lewis H Kuller?

Claim your profile

Publications (770)5073.83 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Novel approaches to observational studies and clinical trials could improve the cost-effectiveness and speed of translation of research. Hybrid designs that combine elements of clinical trials with observational registries or cohort studies should be considered as part of a long-term strategy to transform clinical trials and epidemiology, adapting to the opportunities of big data and the challenges of constrained budgets. Important considerations include study aims, timing, breadth and depth of the existing infrastructure that can be leveraged, participant burden, likely participation rate and available sample size in the cohort, required sample size for the trial, and investigator expertise. Community engagement and stakeholder (including study participants) support are essential for these efforts to succeed.
    Contemporary clinical trials 11/2015; DOI:10.1016/j.cct.2015.11.017 · 1.94 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: -Increasingly, the diagnostic codes from administrative claims data are being used as clinical outcomes. Methods and results: -Data from the Cardiovascular Health Study (CHS) were used to compare event rates and risk-factor associations between adjudicated hospitalized cardiovascular events and claims-based methods of defining events. The outcomes of myocardial infarction (MI), stroke, and heart failure (HF) were defined in three ways: 1) the CHS adjudicated event (CHS[adj]); 2) selected ICD9 diagnostic codes only in the primary position for Medicare claims data from the Center for Medicare and Medicaid Services (CMS[1(st)]); and 3) the same selected diagnostic codes in any position (CMS[any]). Conventional claims-based methods of defining events had high positive predictive values (PPVs) but low sensitivities. For instance, the PPV of an ICD9 code of 410.x1 for a new acute MI in the first position was 90.6%, but this code identified only 53.8% of incident MIs. The observed event rates were low. For MI, the incidence was 14.9 events per 1000 person years for CHS[adj] MI, 8.6 for CMS[1(st)] and 12.2 for CMS[any]. In general, CVD risk factor associations were similar across the three methods of defining events. Indeed, traditional CVD risk factors were also associated with all first hospitalizations not due to an MI. Conclusions: -The use of diagnostic codes from claims data as clinical events, especially when restricted to primary diagnoses, leads to an underestimation of event rates. Additionally, claims-based events data represent a composite endpoint that includes the outcome of interest and selected (misclassified) non-event hospitalizations.
    Circulation 11/2015; DOI:10.1161/CIRCULATIONAHA.115.018610 · 14.43 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: Increasing life expectancy has resulted in a larger population of older individuals at risk of dementia. Methods: The Cardiovascular Health Study-Cognition Study followed 532 participants from 1998-99 (mean age 79) to 2013 (mean age 93) for death and dementia. Results: Risk of death was determined by extent of coronary artery calcium, high-sensitivity cardiac troponin, brain natriuretic peptide, and white matter grade. Significant predictors of dementia were age, apolipoprotein-E4, vocabulary raw score, hippocampal volume, ventricular size, cognitive performance, and number of blocks walked. By 2013, 160 of 532 were alive, including 19 cognitively normal. Those with normal cognition had higher grade education, better cognition test scores, greater hippocampal volume, faster gait speed, and number of blocks walked as compared with survivors who were demented. Discussion: Few survived free of dementia and disability. Prevention and delay of cognitive decline for this older population is an imperative.
    Alzheimer's & dementia: the journal of the Alzheimer's Association 10/2015; DOI:10.1016/j.jalz.2015.08.165 · 12.41 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: There is controversy regarding whether to report concentrations of high-sensitivity cardiac troponin T (hs-cTnT) to the limit of blank (LOB) (3 ng/L) or the limit of detection (LOD) (5 ng/L) of the assay in community-based cohorts. We hypothesized that hs-cTnT concentrations between the LOB and LOD would be associated with poorer cardiovascular outcomes compared to concentrations below the LOB. Methods: hs-cTnT was analyzed in a total of 10 723 participants from the Cardiovascular Health Study (CHS), Atherosclerosis Risk in Communities (ARIC) study, and Dallas Heart Study (DHS). Participants were divided into 2 groups, those with hs-cTnT concentrations below the limit of blank (LOB) (<3 ng/L) and those with hs-cTnT between the LOB and limit of detection (LOD) (3-4.99 ng/L). Cross-sectional associations with traditional cardiovascular risk factors and cardiac structural measurements, and longitudinal associations with long-term cardiovascular outcomes of incident heart failure and cardiovascular death, were determined. Results: Participants with hs-cTnT between the LOB and LOD for all 3 cohorts were older and more likely to be male and have a higher burden of cardiovascular risk factors and structural pathology. A metaanalysis of the 3 cohorts showed participants with hs-cTnT between the LOB and LOD were at increased risk of new-onset heart failure (hazard ratio, 1.18; 95% CI, 1.02-1.38) and cardiovascular mortality (hazard ratio, 1.29; 95% CI, 1.06-1.57). Conclusions: hs-cTnT concentrations between the LOB and LOD (3-4.99 ng/L) are associated with a higher prevalence of traditional risk factors, more cardiac pathology, and worse outcomes than concentrations below the LOB (<3 ng/L).
    Clinical Chemistry 10/2015; DOI:10.1373/clinchem.2015.244160 · 7.91 Impact Factor

  • Arthritis and Rheumatology 10/2015; DOI:10.1002/art.39471
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: The American Heart Association's "Simple 7" offers a practical public health conceptualization of cardiovascular health (CVH). CVH predicts incident cardiovascular disease (CVD) in younger populations, but has not been studied in a large, diverse population of aging postmenopausal women. The extent to which CVH predicts cancer in postmenopausal women is unknown. Methods: Multivariable Cox regression estimated hazard ratios and 95% CIs for the association between CVH and incident CVD, any cancer, and cancer subtypes (lung, colorectal, and breast) among 161,809 Women's Health Initiative observational study and clinical trial participants followed from 1993 through 2010. Data were analyzed in 2013. CVH score was characterized as the number (0 [worst] to 7 [best]) of the American Heart Association's ideal CVH behaviors and factors at baseline: smoking, BMI, physical activity, diet, total cholesterol, blood pressure, and fasting glucose. Results: Median follow-up was approximately 13 years. Fewer minorities and less educated women achieved ideal CVH, a common benchmark. In adjusted models, compared with women with the highest (best) CVH scores, those with the lowest (worst) CVH scores had nearly seven times the hazard of incident CVD (6.83, 95% CI=5.83, 8.00) and 52% greater risk of incident cancer (1.52, 95% CI=1.35, 1.72). Ideal CVH was most strongly inversely associated with lung cancer, then colorectal cancer, and then breast cancer. Conclusions: Lower ideal CVH is more common among minority and less educated postmenopausal women and predicts increased risk of CVD and cancer in this population, emphasizing the importance of prevention efforts among vulnerable older adults.
    American journal of preventive medicine 10/2015; DOI:10.1016/j.amepre.2015.07.039 · 4.53 Impact Factor

  • [Show abstract] [Hide abstract]
    ABSTRACT: Alzheimer disease is the most common form of dementia in the elderly, and the complex relationships among risk factors produce highly variable natural histories from normal cognition through the prodromal stage of mild cognitive impairment (MCI) to clinical dementia. We used a novel statistical approach, mixed membership trajectory models, to capture the variety of such pathways in 652 participants in the Cardiovascular Health Study Cognition Study over 22 years of follow-up (1992-2014). We identified 3 trajectories: a "healthy" profile with a peak probability of MCI between 95 and 100 years of age and only a 50% probability of dementia by age 100; an "intermediate" profile with a peak probability of MCI between 85 and 90 years of age and progression to dementia between 90 and 95 years; and an "unhealthy" profile with a peak probability of progressing to MCI between ages 75 and 80 years and to dementia between the ages of 80 and 85 years. Hypertension, education, race, and the ϵ4 allele of the apolipoprotein E gene all affected the closeness of an individual to 1 or more of the canonical trajectories. These results provide new insights into the natural history of Alzheimer disease and evidence for a potential difference in the pathophysiology of the development of dementia. © The Author 2015. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
    American journal of epidemiology 07/2015; 182(4). DOI:10.1093/aje/kwv051 · 5.23 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The Seven Countries Study in the 1960s showed very low mortality from coronary heart disease (CHD) in Japan, which was attributed to very low levels of total cholesterol. Studies of migrant Japanese to the USA in the 1970s documented increase in CHD rates, thus CHD mortality in Japan was expected to increase as their lifestyle became Westernized, yet CHD mortality has continued to decline since 1970. This study describes trends in CHD mortality and its risk factors since 1980 in Japan, contrasting those in other selected developed countries. We selected Australia, Canada, France, Japan, Spain, Sweden, the UK and the USA. CHD mortality between 1980 and 2007 was obtained from WHO Statistical Information System. National data on traditional risk factors during the same period were obtained from literature and national surveys. Age-adjusted CHD mortality continuously declined between 1980 and 2007 in all these countries. The decline was accompanied by a constant fall in total cholesterol except Japan where total cholesterol continuously rose. In the birth cohort of individuals currently aged 50-69 years, levels of total cholesterol have been higher in Japan than in the USA, yet CHD mortality in Japan remained the lowest: > 67% lower in men and > 75% lower in women compared with the USA. The direction and magnitude of changes in other risk factors were generally similar between Japan and the other countries. Decline in CHD mortality despite a continuous rise in total cholesterol is unique. The observation may suggest some protective factors unique to Japanese. © The Author 2015; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association.
    International Journal of Epidemiology 07/2015; DOI:10.1093/ije/dyv143 · 9.18 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The literature on environment and obesity is characterized by studies that are often cross-sectional and lack racial diversity. This study examined associations between neighborhood features and BMI development over 6 years in an urban sample of 2,295 girls (56% African American; mean age at baseline, 11.2 years) in 2004. Analyses were conducted in 2011-2015. Girls, caregivers, and study staff completed annual neighborhood questionnaires. Linear mixed-effects modeling examined annual changes in neighborhood features and BMI and assessed whether baseline neighborhood features modified BMI growth over time. At baseline, 40% of participants were overweight/obese. Participants' neighborhoods had few neighborhood problems, moderate levels of safety issues and inconvenient features, low levels of neighborhood disorder, few cases of loitering youth, and substantial traffic volume. Adverse neighborhood features were more common for African American than white participants. Neighborhood features were relatively stable over the follow-up period. African American girls with helpful neighbors had lower annual BMI growth (-0.09 kg/m(2)) than others. For white girls, BMI increased more for girls with helpful neighbors (+0.09 kg/m(2) annually). Regardless of race, living in a U.S. Census tract with low levels of educational achievement was linked with higher BMI growth (an additional 0.07 kg/m(2) annually). Girls living in Census tracts with high (versus low) levels of poverty gained an additional 0.08 kg/m(2) gain annually. Social environment features are associated with BMI change in white and African American urban girls and may be helpful for identifying girls at risk for early adolescent weight gain. Copyright © 2015 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.
    American journal of preventive medicine 07/2015; 49(6). DOI:10.1016/j.amepre.2015.05.021 · 4.53 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The prevalence of cardiometabolic multimorbidity is increasing. To estimate reductions in life expectancy associated with cardiometabolic multimorbidity. Age- and sex-adjusted mortality rates and hazard ratios (HRs) were calculated using individual participant data from the Emerging Risk Factors Collaboration (689 300 participants; 91 cohorts; years of baseline surveys: 1960-2007; latest mortality follow-up: April 2013; 128 843 deaths). The HRs from the Emerging Risk Factors Collaboration were compared with those from the UK Biobank (499 808 participants; years of baseline surveys: 2006-2010; latest mortality follow-up: November 2013; 7995 deaths). Cumulative survival was estimated by applying calculated age-specific HRs for mortality to contemporary US age-specific death rates. A history of 2 or more of the following: diabetes mellitus, stroke, myocardial infarction (MI). All-cause mortality and estimated reductions in life expectancy. In participants in the Emerging Risk Factors Collaboration without a history of diabetes, stroke, or MI at baseline (reference group), the all-cause mortality rate adjusted to the age of 60 years was 6.8 per 1000 person-years. Mortality rates per 1000 person-years were 15.6 in participants with a history of diabetes, 16.1 in those with stroke, 16.8 in those with MI, 32.0 in those with both diabetes and MI, 32.5 in those with both diabetes and stroke, 32.8 in those with both stroke and MI, and 59.5 in those with diabetes, stroke, and MI. Compared with the reference group, the HRs for all-cause mortality were 1.9 (95% CI, 1.8-2.0) in participants with a history of diabetes, 2.1 (95% CI, 2.0-2.2) in those with stroke, 2.0 (95% CI, 1.9-2.2) in those with MI, 3.7 (95% CI, 3.3-4.1) in those with both diabetes and MI, 3.8 (95% CI, 3.5-4.2) in those with both diabetes and stroke, 3.5 (95% CI, 3.1-4.0) in those with both stroke and MI, and 6.9 (95% CI, 5.7-8.3) in those with diabetes, stroke, and MI. The HRs from the Emerging Risk Factors Collaboration were similar to those from the more recently recruited UK Biobank. The HRs were little changed after further adjustment for markers of established intermediate pathways (eg, levels of lipids and blood pressure) and lifestyle factors (eg, smoking, diet). At the age of 60 years, a history of any 2 of these conditions was associated with 12 years of reduced life expectancy and a history of all 3 of these conditions was associated with 15 years of reduced life expectancy. Mortality associated with a history of diabetes, stroke, or MI was similar for each condition. Because any combination of these conditions was associated with multiplicative mortality risk, life expectancy was substantially lower in people with multimorbidity.
    JAMA The Journal of the American Medical Association 07/2015; 314(1):52-60. DOI:10.1001/jama.2015.7008 · 35.29 Impact Factor
  • Rachel H. Mackey · Lewis H. Kuller ·

    Journal of the American College of Cardiology 06/2015; 65(23). DOI:10.1016/j.jacc.2015.03.574 · 16.50 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: To understand the incidence and persistence of severe obesity (≥1.2× 95(th) BMI percentile-for-age) in girls across the transition to adolescence, and map developmental trajectories of adolescent severe obesity in a high-risk sample. Methods: We examined ten years of prospectively collected data from a population sample of urban girls (n=2,226; 53% African American, aged 7-10 in 2003-2004). We determined severe obesity prevalence and incidence by age. Logistic regression evaluated for secular trend in the association between age and severe obesity prevalence. Unconditional latent growth curve models (LGCMs) compared BMI development through the adolescence transition between girls with severe obesity versus healthy BMI. Results: Severe obesity prevalence was 8.3% at age 7-10 and 10.1% at age 16-19 (white: 5.9%; African American: 13.2%; p<0.001). Age-specific prevalence increased more rapidly among the latest-born, versus earliest-born, girls (p=0.034). Incidence was 1.3% to 2.4% annually. When we compared 12-15 year-old girls with severe obesity versus healthy BMI, average body weight was already distinct 5 years earlier (16.5 kg versus 25.7 kg; p<0.001) and the BMI difference between groups increased annually. LCGMs between ages 7-10 and 11-14 indicated an increase of 3.32 kg/m(2) in the healthy-BMI group and 8.50 kg/m(2) in the severe obesity group, a 2.6-fold difference. Conclusions: Youth-onset severe obesity warrants particular concern in urban girls due to high prevalence and an increasing secular prevalence trend. Late childhood and early adolescence may represent a key developmental window for prevention and treatment, but is too late to prevent youth-onset severe obesity entirely.
    Journal of Clinical and Translational Endocrinology 06/2015; 2(4). DOI:10.1016/j.jcte.2015.04.001
  • [Show abstract] [Hide abstract]
    ABSTRACT: This report evaluates incidence of cardiovascular disease (CVD) morbidity and mortality over 10 years among the >160,000 postmenopausal women in the Women's Health Initiative (WHI) in relation to self-reported RA, disease modifying anti-rheumatic drugs (DMARD) use, anti-CCP+, RF+, CVD risk factors, joint pain, and inflammation (white blood cell (WBC) count and IL-6.) Methods: Anti-CCP and RF were measured on a sample (n=9,988) of WHI participants with self-reported RA. RA was classified as self-reported RA plus anti-CCP+ positivity and/or use of DMARDs. Self-reported RA that was both anti-CCP- and DMARD- was classified as "unverified RA." Age-adjusted rates of coronary heart disease (CHD), stroke, CVD, fatal CVD and total mortality were higher for women with RA vs. no RA, with multivariable-adjusted HR(95%CI) of 1.46(1.17, 1.83) for CHD, and 2.55(1.86, 3.51) for fatal CVD. Within RA, anti-CCP+ and RF+ were not significantly associated with higher risk of any outcomes, despite slightly higher risk of fatal CVD and death for anti-CCP+ vs. anti-CCP- RA. Joint pain severity and CVD risk factors were strongly associated with CVD risk, even for women with no RA. CVD incidence was increased for RA vs. no RA at almost all risk factor levels, except low levels of joint pain or inflammation. Within RA, inflammation was more strongly associated with fatal CVD and total mortality than CHD or CVD. Among postmenopausal women, RA was associated with 1.5-2.5 higher CVD risk, strongly associated with CV risk factors, joint pain severity, and inflammation, but similar for anti-CCP+ and RF+. This article is protected by copyright. All rights reserved. © 2015, American College of Rheumatology.
    Arthritis and Rheumatology 05/2015; 67(9). DOI:10.1002/art.39198
  • [Show abstract] [Hide abstract]
    ABSTRACT: Given conflicting data regarding the association of HIV infection and ischemic stroke risk, we sought to determine whether HIV infection conferred an increased ischemic stroke risk among male veterans. The Veterans Aging Cohort Study-Virtual Cohort consists of HIV-infected and uninfected veterans in care matched (1:2) for age, sex, race/ethnicity, and clinical site. We analyzed data on 76,835 male participants in the Veterans Aging Cohort Study-Virtual Cohort who were free of baseline cardiovascular disease. We assessed demographics, ischemic stroke risk factors, comorbid diseases, substance use, HIV biomarkers, and incidence of ischemic stroke from October 1, 2003, to December 31, 2009. During a median follow-up period of 5.9 (interquartile range 3.5-6.6) years, there were 910 stroke events (37.4% HIV-infected). Ischemic stroke rates per 1,000 person-years were higher for HIV-infected (2.79, 95% confidence interval 2.51-3.10) than for uninfected veterans (2.24 [2.06-2.43]) (incidence rate ratio 1.25 [1.09-1.43]; p < 0.01). After adjusting for demographics, ischemic stroke risk factors, comorbid diseases, and substance use, the risk of ischemic stroke was higher among male veterans with HIV infection compared with uninfected veterans (hazard ratio 1.17 [1.01-1.36]; p = 0.04). HIV infection is associated with an increased ischemic stroke risk among HIV-infected compared with demographically and behaviorally similar uninfected male veterans. © 2015 American Academy of Neurology.
    Neurology 04/2015; 84(19). DOI:10.1212/WNL.0000000000001560 · 8.29 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Brachial-ankle pulse wave velocity (baPWV) is a simple and reproducible measure of arterial stiffness and is extensively used to assess cardiovascular disease (CVD) risk in eastern Asia. We examined whether baPWV is associated with coronary atherosclerosis in an international study of healthy middle-aged men. A population-based sample of 1131 men aged 40-49years was recruited - 257 Whites and 75 Blacks in Pittsburgh, US, 228 Japanese-Americans in Honolulu, US, 292 Japanese in Otsu, Japan, and 279 Koreans in Ansan, Korea. baPWV was measured with an automated waveform analyzer (VP2000, Omron) and atherosclerosis was examined as coronary artery calcification (CAC) by computed-tomography (GE-Imatron EBT scanner). Association of the presence of CAC (defined as ≥10Agatston unit) was examined with continuous measure as well as with increasing quartiles of baPWV. As compared to the lowest quartile of baPWV, the multivariable-adjusted odds ratio (95% Confidence Interval [CI]) for the presence of CAC in the combined sample was 1.70 (0.98, 2.94) for 2nd quartile, 1.88 (1.08, 3.28) for 3rd quartile, and 2.16 (1.19, 3.94) for 4th quartile (p-trend=0.01). The odds for CAC increased by 19% per 100cm/s increase (p<0.01), or by 36% per standard-deviation increase (p<0.01) in baPWV. Similar effect-sizes were observed in individual races, and were significant among Whites, Blacks and Koreans. baPWV is cross-sectionally associated with CAC among healthy middle-aged men. The association was significant in Whites and Blacks in the US, and among Koreans. Longitudinal studies are needed to determine its CVD predictive ability. Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.
    International journal of cardiology 04/2015; 189(1):67-72. DOI:10.1016/j.ijcard.2015.04.020 · 4.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We hypothesized that higher concentrations of LDL particles (LDL-P) and leptin, and lower concentrations of HDL particles (HDL-P), and total and high molecular weight (HMW) adiponectin, would predict incident coronary heart disease (CHD) among severely obese postmenopausal women. In a case-cohort study nested in the Women's Health Initiative Observational Study, we sampled 677 of the 1852 white or black women with body mass index (BMI) ≥40 kg/m(2) and no prevalent cardiovascular disease (CVD), including all 124 cases of incident CHD over mean 5.0 year follow-up. Biomarkers were assayed on stored blood samples. In multivariable-adjusted weighted Cox models, higher baseline levels of total and small LDL-P, and lower levels of total and medium HDL-P, and smaller mean HDL-P size were significantly associated with incident CHD. In contrast, large HDL-P levels were inversely associated with CHD only for women without diabetes, and higher total and HMW adiponectin levels and lower leptin levels were associated with CHD only for women with diabetes. Higher total LDL-P and lower HDL-P were associated with CHD risk independently of confounders including CV risk factors and other lipoprotein measures, with adjusted HR (95%CIs) of 1.55(1.28, 1.88) and (0.70 (0.57, 0.85), respectively, and similar results for medium HDL-P. Higher CHD risk among severely obese postmenopausal women is strongly associated with modifiable concentrations of LDL-P and HDL-P, independent of diabetes, smoking, hypertension, physical activity, BMI and waist circumference. Severely obese postmenopausal women should be considered high risk candidates for lipid lowering therapy.
    Biochimica et Biophysica Acta - Clinical 03/2015; 120. DOI:10.1016/j.bbacli.2015.03.005
  • Akira Sekikawa · Margaret F Doyle · Lewis H Kuller ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent long-term randomized clinical trials (RCTs) of long-chain n-3 polyunsaturated fatty acids (LCn-3 PUFAs) on coronary heart disease (CHD) among high-risk patients conducted in Western countries all failed to show their clinical benefits. In striking contrast, an RCT of LCn-3 PUFAs on CHD conducted in Japan, which is a combination of secondary and primary prevention, showed a significant 19% reduction. Potential reasons for this discrepancy are large differences in doses of LCn-3 PUFAs administered (300-900mg/day in Western countries vs. 1800 mg/day in Japan) and background dietary intake of LCn-3 PUFAs (<300mg/day in Western countries vs. >1000mg/day in Japan). These observations suggest that higher doses of LCn-3 PUFAs than examined in RCTs in Western countries may be cardio-protective. Atherosclerosis is the major underlying cause of CHD. Recent observational studies and an RCT of LCn-3 PUFAs on atherosclerosis in Japan show that LCn-3 PUFAs are anti-atherogenic. In this brief review, we focus on recent epidemiological and clinical findings of LCn-3 PUFAs on atherosclerosis and CHD, contrasting studies in Western countries to those in Japan. We also discuss mechanisms of high-dose LCn-3 PUFAs on atherosclerosis. Copyright © 2015 Elsevier Inc. All rights reserved.
    Trends in cardiovascular medicine 03/2015; 25(8). DOI:10.1016/j.tcm.2015.03.001 · 2.91 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study was to identify, at the voxel level, brain regions associated with the time to develop mild cognitive impairment (MCI) or Alzheimer's disease (AD) from normal cognition. We analyzed incident MCI (n = 58) or AD (n = 151) in 292 cognitively normal participants in the Cardiovascular Health Study-Cognition Study (mean age = 79.2 ± 3.6 years). We used segmented, modulated grey matter maps from 3D (spoiled gradient echo) MRI scans obtained in 1998/99 (with clinical follow-up through 2012) that were smoothed with a 3-D 4 mm Gaussian filter. We fit approximately 1.92 million voxel-level Cox proportional hazard models to examine the grey matter volume effect on time to event, adjusting for age, sex, and diabetes. We used the significance threshold of p < 0.005 with contiguity threshold of at least 68 voxels (false detection probability <2.5 × 10-8). Areas within the mesial temporal lobe (MTL), anterior temporal lobe, hippocampus, and posterior cingulate gyrus were associated with time to MCI or AD. The presence of white matter lesions (a marker of small vessel disease in the brain) was associated with the volumes of the MTL and precuneus; MRI-identified infarcts also predicted MTL volume. These findings are important because we identified critical brain regions that predict a person's increased likelihood of developing MCI or AD over a decade prior to the onset of clinical symptoms; these critical brain regions were themselves affected by the presence of vascular disease.
    Journal of Alzheimer's disease: JAD 02/2015; 46(1). DOI:10.3233/JAD-150047 · 4.15 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We study regularized estimation in high-dimensional longitudinal classification problems, using the lasso and fused lasso regularizers. The constructed coefficient estimates are piecewise constant across the time dimension in the longitudinal problem, with adaptively selected change points (break points). We present an efficient algorithm for computing such estimates, based on proximal gradient descent. We apply our proposed technique to a longitudinal data set on Alzheimer's disease from the Cardiovascular Health Study Cognition Study, and use this data set to motivate and demonstrate several practical considerations such as the selection of tuning parameters, and the assessment of model stability.

Publication Stats

41k Citations
5,073.83 Total Impact Points


  • 1975-2015
    • University of Pittsburgh
      • • Graduate School of Public Health
      • • Department of Epidemiology
      • • Department of Medicine
      • • Department of Neurological Surgery
      • • Department of Psychiatry
      Pittsburgh, Pennsylvania, United States
  • 2013
    • Wake Forest University
      • Department of Internal Medicine
      Winston-Salem, North Carolina, United States
  • 2011
    • Case Western Reserve University School of Medicine
      • Department of Epidemiology and Biostatistics
      Cleveland, Ohio, United States
    • Weill Cornell Medical College
      • Department of Medicine
      New York, New York, United States
  • 1994-2011
    • Fred Hutchinson Cancer Research Center
      • Division of Public Health Sciences
      Seattle, Washington, United States
    • University of Benin
      Benim, Edo, Nigeria
    • University of Alabama at Birmingham
      • Department of Medicine
      Birmingham, AL, United States
    • University of Melbourne
      • Department of Ophthalmology
      Melbourne, Victoria, Australia
  • 1987-2010
    • University of Minnesota Duluth
      Duluth, Minnesota, United States
  • 1999-2009
    • University of Vermont
      • Department of Pathology
      Burlington, VT, United States
    • North Carolina Department of Health and Human Services
      Raleigh, North Carolina, United States
  • 2007
    • University of California, Davis
      • Center for Neuroscience
      Davis, CA, United States
    • University of California, Los Angeles
      • Department of Epidemiology
      Los Angeles, CA, United States
    • University of North Carolina at Chapel Hill
      North Carolina, United States
  • 2003-2007
    • University of Helsinki
      • Department of Psychology
      Helsinki, Province of Southern Finland, Finland
    • Beth Israel Deaconess Medical Center
      • Division of General Medicine and Primary Care
      Boston, MA, United States
    • San Diego State University
      San Diego, California, United States
  • 2006
    • University at Buffalo, The State University of New York
      • Department of Social and Preventive Medicine
      Buffalo, NY, United States
    • VU University Amsterdam
      Amsterdamo, North Holland, Netherlands
  • 2004-2006
    • Johns Hopkins University
      • Department of Mental Health
      Baltimore, Maryland, United States
    • Saint Louis University
      Сент-Луис, Michigan, United States
    • New England Baptist Hospital
      Boston, Massachusetts, United States
  • 2005
    • Harvard Medical School
      • Department of Medicine
      Boston, MA, United States
  • 2000-2004
    • University of Washington Seattle
      • Department of Medicine
      Seattle, Washington, United States
    • University of California, San Francisco
      San Francisco, California, United States
  • 1997
    • University of California, Irvine
      Irvine, California, United States
  • 1991-1997
    • Western Psychiatric Institute and Clinic
      Pittsburgh, Pennsylvania, United States
  • 1996
    • University of Benin Teaching Hospital
      Benim, Edo, Nigeria
  • 1995
    • Graduate School USA
      Washington, Washington, D.C., United States
  • 1989-1992
    • Childrens Hospital of Pittsburgh
      • Department of Pediatrics
      Pittsburgh, Pennsylvania, United States
  • 1990
    • National Institutes of Health
      • Office of Disease Prevention
      Bethesda, MD, United States
  • 1988
    • Pennsylvania Department of Health
      Harrisburg, Pennsylvania, United States
  • 1986
    • Georgia Health Sciences University
      • Department of Medicine
      Augusta, Georgia, United States
  • 1984
    • Rutgers New Jersey Medical School
      • Department of Medicine
      Newark, New Jersey, United States