| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
From the Department of Neurology (Dr. Knopman), Mayo Clinic, Rochester, MN; Division of Epidemiology (L.L. Boland and Drs. McGovern and Folsom), School of Public Health, University of Minnesota, Minneapolis; Department of Medicine (Geriatrics) (Dr. Mosley), Medical Center, University of Mississippi, Jackson; Department of Biostatistics (Dr. Howard), University of Alabama, Birmingham; Health Evaluation Sciences (Dr. Liao), Penn State College of Medicine, Hershey, PA; and Department of Epidemiology (Dr. Szklo), The Johns Hopkins Medical Institutions, Baltimore, MD.
Address correspondence and reprint requests to Dr. David Knopman, Department of Neurology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905.
 |
Article Abstract |
|---|
 Top. Article Abstract IntroductionMethods.Results.Discussion.References |
|---|
METHODS: The authors administered cognitive assessments to 10,963 individuals (8,729 white individuals and 2,234 black individuals) on two occasions separated by 6 years. Subjects ranged in age at the first assessment from 47 to 70 years. The cognitive assessments included the delayed word recall (DWR) test, a 10-word delayed free recall task in which the learning phase included sentence generation with the study words, the digit symbol subtest (DSS) of the Wechsler Adult Intelligence Scale–Revised and the first-letter word fluency (WF) test using letters F, A, and S.
RESULTS: In multivariate analyses (controlling for demographic factors), the presence of diabetes at baseline was associated with greater decline in scores on both the DSS and WF (p < 0.05), and the presence of hypertension at baseline was associated with greater decline on the DSS alone (p < 0.05). The association of diabetes with cognitive decline persisted when analysis was restricted to the 47- to 57-year-old subgroup. Smoking status, carotid intima–media wall thickness, and hyperlipidemia at baseline were not associated with change in cognitive test scores.
CONCLUSIONS: Hypertension and diabetes mellitus were positively associated with cognitive decline over 6 years in this late middle-aged population. Interventions aimed at hypertension or diabetes that begin before age 60 might lessen the burden of cognitive impairment in later life.
|
Introduction |
|---|
|
Top.Article AbstractIntroductionMethods.Results.Discussion.References |
|---|
Studies have also shown clear relationships between diabetes mellitus, cognitive decline, and dementia (usually vascular dementia). The largest associations have been seen in studies with case-control designs9,10 of clinically referred patients. Population-based studies have usually,2,11,12 but not always,13 shown an association of diabetes with cognitive decline in the elderly. In studies of prevalent dementia, there is a consistent link between AD14,15 and diabetes, as well as vascular dementia14,16,17 and diabetes.
The Atherosclerosis Risk in Communities (ARIC) study was initiated in 1987 as a multiethnic, multicenter investigation of vascular disease and its consequences.18 At the second visit, a cognitive battery was administered to participants. Cross-sectional results of the first (visit 2) cognitive battery have been published previously.19 At the fourth ARIC visit 6 years later, the cognitive battery was repeated. The purpose of this report is to describe the relationships between baseline cardiovascular risk factors and change in cognitive function over the 6-year interval.
|
Methods. |
|---|
|
Top.Article AbstractIntroductionMethods.Results.Discussion.References |
|---|
At the second visit (in 1990 to 1992), conducted 3 years after the initial enrollment visit, 14,348 individuals in the ARIC cohort underwent a cognitive evaluation. For most of the current analysis, 300 of these subjects were excluded because they gave a history of stroke or TIA. This visit ("visit 2" in the overall ARIC study) constitutes the baseline visit for the current report. At visit 4, the follow-up visit for this report 6 years later, 10,963 (76%) of the individuals in the ARIC cohort were retested.
Subjects ranged in age at the first assessment from 47 to 70 years. Among the retested individuals, there were 8,729 white subjects and 2,234 black subjects. Women (n = 6,126) were more numerous than men (n = 4,837). Educational attainment varied widely, with 6% of subjects with <9th-grade education, 45% with 9 to 12 years of education, 9% with postsecondary vocational education, and 39% with at least some college.
Cognitive testing. The cognitive assessments and the baseline performance have been presented in detail.19 The battery included the delayed word recall (DWR) test, the digit symbol subtest (DSS) of the Wechsler Adult Intelligence Scale–Revised (WAIS-R) and the first-letter word fluency (WF) test. The tests were administered by trained interviewers in a standardized order during one session in a quiet room. Interviewer performance was monitored by tape recording, and a sample of the testing sessions were reviewed to confirm that there were no systematic differences between the different interviewers who administered the tests. The same test versions were used at both visits.
The DWR20 is a test of verbal learning and recent memory that requires the participant to recall 10 common nouns following a 5-minute interval. In order to produce elaborative processing during the encoding stage of learning, individuals were required to construct sentences incorporating the words presented. During the delay interval, the digit symbol substitution test was given. After 5 minutes, free recall of the words was sought.
The DSS of the WAIS-R21 is a paper-and-pencil task requiring timed translation of numbers to symbols using a key given at the top of the test page. The test was scored as the number of correct translations completed within 90 seconds.
The WF test22 required the participant to generate as many words as possible, but not proper names or places, beginning with a particular letter of the alphabet within 60 seconds. Three separate 1-minute trial periods were used for the each of the letters F, A, and S.
Risk factor assessments. The pertinent patient status and cardiovascular risk states were defined as follows, and are described in greater detail elsewhere.18 All of these health status measures were assessed at ARIC visit 2, and interim events between visits 2 and 4 were identified.
Hypertension was considered to be present at visit 2 if systolic blood pressure (BP) was 
140 mm Hg, if diastolic BP was 90 mm Hg, or if antihypertensive medications were being used. Systolic and diastolic BP were measured three times using a random zero sphygmomanometer in the right arm of seated participants. The mean of the last two measurements was used in all analyses.
Prevalent diabetes mellitus was defined as a fasting glucose of 126 mg/dL, nonfasting glucose of 200mg/dL, a self-reported history of diabetes, or treatment for diabetes. Participants were asked to fast for 12 hours before the clinic visit. Blood was drawn from the antecubital vein of seated participants. Serum glucose was assessed by the hexokinase method.
Hyperlipidemia was defined as a low-density lipoprotein (LDL) cholesterol of 140 mg/dL or the use of cholesterol-lowering agents. Plasma lipids and lipoproteins were determined by enzymatic methods in a laboratory standardized by the Centers for Disease Control. LDL cholesterol was estimated by the Friedewald equation.
All participants were asked to bring to each ARIC clinic visit all prescription and nonprescription medications taken in the previous 2 weeks. Non-steroidal anti-inflammatory drug (NSAID) use was determined by querying subjects for all prescription and nonprescription medication use at visit 2. Although it was possible to code aspirin separately, aspirin is included among the NSAID in these analyses. We coded a subject as taking a CNS-relevant medication if they were receiving antipsychotics, antidepressants, anxiolytics, narcotic analgesics, anticonvulsants, or antineoplastic agents.
Carotid wall intima–media thickness (IMT) was determined by high resolution B-mode ultrasound, as previously described.23 IMT was measured at three sites bilaterally, and an overall average of the mean IMT of the six arterial segments was calculated. The mean IMT used in analysis was further adjusted for reader differences and time trends. For the purposes of the analyses, carotid IMT was divided into tertiles: <0.65 mm, 0.65 to 0.76 mm, and >0.76 mm.
Prevalent stroke was defined as those strokes that were self-reported by subjects at the second ARIC visit.
Incident stroke was defined as a stroke meeting ARIC criteria, as verified by an ARIC clinician through review of medical records.24,25 To date, the identification and validation of incident strokes is complete only from baseline through December 31, 1996. Because the testing cycle for the follow-up cognitive assessments occurred over the 3-year period of January 1996 to January 1999, some incident strokes have not been enumerated. Because adjustment for incident stroke would be only partial, subjects with incident stroke were excluded from multivariate analyses. We chose to include in this report data on subjects with incident stroke for descriptive purposes and because their data illuminates other aspects of the analysis.
Statistical methods. Change scores for cognitive tests were calculated as visit 4 minus visit 2 values for each individual. Mean change scores were tabulated by age level. Change scores were computed for strata of each putative vascular risk factor19 using linear regression to control for age, gender, race-center, educational level, site, and CNS-relevant medications. We performed the analyses both with and without adjusting for the visit 2 performance. Because the adjustment for the visit 2 scores had virtually no effect on the outcomes, we report only the analyses without the adjustment. Additional analyses computed change scores for strata of each putative risk factor broken down by race, but still controlling for age, gender, center, educational level, site, and CNS-relevant medications. Medication use was determined from visit 2 only. Age was treated as a continuous variable in risk-factor analyses. The SAS statistical package (SAS Institute, Cary, NC) was used for these analyses.
|
Results. |
|---|
|
Top.Article AbstractIntroductionMethods.Results.Discussion.References |
|---|
|
|
We also performed analyses using the systolic blood pressure cut points used by Glynn et al.,5 and did not find a U-shaped association with cognitive function on any of the three tests in our cohort. In fact, for the DSS and WF, the group with the lowest systolic BP (<130 mm Hg) had the least cognitive decline.
Interaction terms between age and diabetes and hypertension in the basic multivariate analyses showed that the hypertension association with DSS test change score differed by age (p < 0.05). In order to better define the relationship between cognitive decline and age, we stratified the cohort into those <58 years of age and those 58 years of age ( table 3). Diabetes mellitus was a risk factor for cognitive decline even in the younger age group on both DSS and WF. Hypertension proved to be a risk factor only in older participants.
|
58 years) and race, the associations between diabetes and DSS and WF change scores were significant (p < 0.05) in the older black subjects, as well as in both the younger and older white subgroups, and similar in magnitude to the values shown in table 3 for the entire group. In the younger black subgroup, the diabetes–cognitive change score associations were in the same directions and nearly the same magnitude as observed for the entire cohort, but did not reach significance. For hypertension and cognitive test scores, there were no significant associations among black participants, regardless of age. However, in the older black subjects, the association between hypertension and change scores on the DSS was of the same magnitude as shown in table 3 for the entire cohort. The association between change in scores on the DSS and hypertension was significant (p < 0.05) in the older white subjects. Among older black subjects, current smokers (compared to former smokers but not never smokers) and those in tertile III of carotid intimal thickness (compared to tertile II but not tertile I) had greater decline on the DSS. There were no other race-specific associations with other risk factors. We compared subjects who returned to visit 4 to those who did not ( table 4). There were substantial differences demographically and, in terms of risk factors and baseline cognitive status, the latter was worse in those who did not return.
|
|
Discussion. |
|---|
|
Top.Article AbstractIntroductionMethods.Results.Discussion.References |
|---|
In the baseline assessment of the individuals in the ARIC cohort, cardiovascular risk factors were also associated with lower cognition.19,28 Longitudinal studies such as the current one provide stronger evidence than cross-sectional studies for the existence of linkage between cardiovascular risk factors and cognitive decline.
Of our three cognitive tests, change on the DSS was most strongly associated with risk factors. Because it is a timed test, it may be more sensitive to subcortical vascular lesions in the striatum and in cortico-cortical and cortico-subcortical pathways. The association of hypertension and diabetes mellitus with the decline on the DSS is notably similar to that observed in the Cardiovascular Health Study.2 Our data extend the observations in the Cardiovascular Health Study of older patients to a younger age group. The association between diabetes mellitus and change on the WF task in our study also expands upon the generality of the relationship between diabetes and cognition by showing the association between diabetes and another mental agility task.
Almost all studies that have examined the association of hypertension with declining cognition have found a positive association between the two. The negative study by Prince et al.7,29 may be explained by the absence of nonhypertensive subjects in the study sample. Several studies with older subjects5,30 found a U-shaped association of blood pressure with subsequent cognitive change scores in population-based cohorts. We failed to find a negative association of low blood pressure and cognition. Two studies5,30 used different cognitive screening tools (neither study used timed tests) than we did. In the older age range of the other two studies,5,30 lower blood pressure might be more likely to be due to a pathologic process, whereas in our young patients, low blood pressure may be part of the normal range of variation.
Three studies in older subjects3,4,6 have found that mid-life hypertension was associated with diminished cognitive performance in later life. Our study measured blood pressure at nearly as young an age as that performed in the Honolulu Heart Study4 or the Framingham Heart Study,3 but not as young as a study of twins by Carmelli et al.6 Our study design enabled us to detect an association with cognitive decline at a younger age than these three other studies.
Studies of diabetes mellitus and cognition have shown some inconsistency. Case-control studies9,10 from clinical samples have shown cognitive impairment in diabetic subjects compared to controls. Some groups found no associations either with cross-sectionally measured cognition13 or with a diagnosis of dementia,16 whereas other population studies reported diabetes to be an independent risk factor for either cognitive impairment11,12 or dementia.14,15,17 Only two other studies have measured cognitive function longitudinally in diabetics. One showed an association similar to ours,2 while the other, based on volunteers in the Baltimore Longitudinal Study,31 showed no evidence for greater cognitive decline in diabetics.
The effect of incident stroke on cognitive decline was substantial and reflected in the cognitive tests. The number of incident strokes in our cohort was small and not complete, however. Several other longitudinal studies have shown that incident stroke is a risk factor for cognitive decline and dementia.32-37 Clinicopathologic studies have shown that patients with AD and infarcts are more demented than patients with only Alzheimer pathology.38,39
The lack of an association of carotid artery IMT with change in cognition is notable because carotid artery IMT was correlated with baseline DSS performance in the ARIC cohort19 and is a risk factor for stroke.40,41 A longitudinal study such as this one helps to identify conditions—cognitive impairment and carotid artery disease—that may have common mediators, but may not be causally related themselves.
Our memory test, the DWR, showed negligible changes over 6 years in the study population as a whole. It was surprising to us that change in the DWR was not associated with diabetes or hypertension. Glynn et al.5 also failed to show a consistent association of hypertension with longitudinal cognitive change on their six-item memory test. Perhaps the DWR and the memory test used by Glynn et al.5 lack sufficient sensitivity. Conversely, ARIC subjects who had strokes in the interval between visits 2 and 4 experienced decline on the DWR in addition to the other two tests. Perhaps the effects of diabetes and hypertension are somewhat distinct from that of infarction, per se, in terms of the anatomic locus of the brain injury.
Among ARIC participants, race had a relatively minor impact on associations between risk factors and cognition. Because sample sizes were lower in race-specific analyses, we were not able to detect associations of small magnitude, but the associations with diabetes and hypertension that were seen in the combined cohort were quite similar across racial groups.
The mechanism by which hypertension and diabetes mellitus cause impaired cognition was not addressed in the current study, but one candidate other than overt cerebral infarction might be demyelination or microinfarction in the cerebral white matter. In both the ARIC study42 and Cardiovascular Health Study cross-sectional data,43 white matter lesions by MRI and hypertension were correlated. Furthermore, white matter changes have been correlated with poorer cognitive performance in twin studies44 and population-based studies.37,45 A report of the relationships between brain MRI findings and cognition assessed at visit 3 in ARIC is currently being prepared. Silent cerebral infarctions are another mechanism by which hypertension produces brain injury. In a subset of the ARIC cohort, hypertension was associated with an increased risk for silent cerebral infarcts.46 The associations of risk factors, brain lesions, and cognition is complex, however. In one study,46 cigarette smoking had the most powerful association with silent cerebral infarcts; in the current study, cigarette smoking had no associations with cognitive change.
Strengths of this study include its population-based, ethnically diverse cohort, its very large sample size, and its longitudinal nature. The limitations of this study are few. Longitudinal studies have much attrition, and as we have shown (table 4), ARIC participants who dropped out or died were more impaired than those who returned for follow-up. The attrition of more impaired individuals may have lessened our ability to show associations between cognitive decline and risk factors. Our cognitive battery was limited. It is possible that other measures of abstract reasoning, executive function, visuospatial function, or language could have revealed a different pattern of associations with cardiovascular risk factors. In addition, our analyses involved testing of associations of multiple risk factors, raising the possibility of type I errors.
Our longitudinal observations show that diabetes and hypertension were associated with erosion of cognitive function, as early as late middle age. We hypothesize that the interaction of vascular risk factor–induced cognitive decline and the rising probability of Alzheimer pathology in the 7th and 8th decades of life increases the probability of dementia. Interventions aimed at hypertension or diabetes that begin before age 60 years might lessen the burden of cognitive impairment in later life. Although not all studies that have examined the effects of treatment of hypertension on dementia have been positive,47 some encouraging findings have been reported.48
|
Acknowledgments |
|---|
The authors thank the staff and participants in the ARIC study for their important contributions.
|
Footnotes |
|---|
|
References |
|---|
|
Top.Article AbstractIntroductionMethods.Results.Discussion.References |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Scazufca, P. R Menezes, R. Araya, V. D Di Rienzo, O. P Almeida, D. Gunnell, and D. A Lawlor Risk factors across the life course and dementia in a Brazilian population: results from the Sao Paulo Ageing & Health Study (SPAH) Int. J. Epidemiol., August 1, 2008; 37(4): 879 - 890. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sabia, M. Marmot, C. Dufouil, and A. Singh-Manoux Smoking History and Cognitive Function in Middle Age From the Whitehall II Study Arch Intern Med, June 9, 2008; 168(11): 1165 - 1173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wiederkehr, M. Simard, C. Fortin, and R. van Reekum Comparability of the Clinical Diagnostic Criteria for Vascular Dementia: A Critical Review. Part I J Neuropsychiatry Clin Neurosci, May 1, 2008; 20(2): 150 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Dickinson, J. M. Gold, F. B. Dickerson, D. Medoff, and L. B. Dixon Evidence of Exacerbated Cognitive Deficits in Schizophrenia Patients With Comorbid Diabetes Psychosomatics, April 1, 2008; 49(2): 123 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Knecht, H. Wersching, H. Lohmann, M. Bruchmann, T. Duning, R. Dziewas, K. Berger, and E. B. Ringelstein High-Normal Blood Pressure Is Associated With Poor Cognitive Performance Hypertension, March 1, 2008; 51(3): 663 - 668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. van Dijk, K. G.M. Moons, H. M. Nathoe, E. H.L. van Aarnhem, C. Borst, A. M.A. Keizer, C. J. Kalkman, R. Hijman, and Octopus Study Group Cognitive Outcomes Five Years After Not Undergoing Coronary Artery Bypass Graft Surgery Ann. Thorac. Surg., January 1, 2008; 85(1): 60 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. A. Selnes, M. A. Grega, M. M. Bailey, L. Pham, S. Zeger, W. A. Baumgartner, and G. M. McKhann Neurocognitive Outcomes 3 Years After Coronary Artery Bypass Graft Surgery: A Controlled Study Ann. Thorac. Surg., December 1, 2007; 84(6): 1885 - 1896. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Anstey, C. von Sanden, A. Salim, and R. O'Kearney Smoking as a Risk Factor for Dementia and Cognitive Decline: A Meta-Analysis of Prospective Studies Am. J. Epidemiol., August 15, 2007; 166(4): 367 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Miyasaka, M. E. Barnes, R. C. Petersen, S. S. Cha, K. R. Bailey, B. J. Gersh, G. Casaclang-Verzosa, W. P. Abhayaratna, J. B. Seward, T. Iwasaka, et al. Risk of dementia in stroke-free patients diagnosed with atrial fibrillation: data from a community-based cohort Eur. Heart J., August 2, 2007; 28(16): 1962 - 1967. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Smith, J. A. Blumenthal, M. A. Babyak, B. M. Hoffman, P. M. Doraiswamy, R. Waugh, A. Hinderliter, and A. Sherwood Cerebrovascular Risk Factors, Vascular Disease, and Neuropsychological Outcomes in Adults With Major Depression Psychosom Med, July 1, 2007; 69(6): 578 - 586. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Hailpern, M. L. Melamed, H. W. Cohen, and T. H. Hostetter Moderate Chronic Kidney Disease and Cognitive Function in Adults 20 to 59 Years of Age: Third National Health and Nutrition Examination Survey (NHANES III) J. Am. Soc. Nephrol., July 1, 2007; 18(7): 2205 - 2213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Hogue Jr, K. Freedland, T. Hershey, R. Fucetola, A. Nassief, B. Barzilai, B. Thomas, S. Birge, D. Dixon, K. B. Schechtman, et al. Neurocognitive Outcomes Are Not Improved by 17{beta}-Estradiol in Postmenopausal Women Undergoing Cardiac Surgery Stroke, July 1, 2007; 38(7): 2048 - 2054. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Baker, E. K. Marino Larsen, L. H. Kuller, R. Klein, B. E.K. Klein, D. S. Siscovick, C. Bernick, T. A. Manolio, and T. Y. Wong Retinal Microvascular Signs, Cognitive Function, and Dementia in Older Persons: The Cardiovascular Health Study Stroke, July 1, 2007; 38(7): 2041 - 2047. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Hogue, O. A. Selnes, and G. McKhann Should All Patients Undergoing Cardiac Surgery Have Preoperative Psychometric Testing: A Brain Stress Test? Anesth. Analg., May 1, 2007; 104(5): 1012 - 1014. [Full Text] [PDF]
|
|
|
|
 |
|
 J Ellul, N Archer, C M L Foy, M Poppe, H Boothby, H Nicholas, R G Brown, and S Lovestone The effects of commonly prescribed drugs in patients with Alzheimer's disease on the rate of deterioration J. Neurol. Neurosurg. Psychiatry, March 1, 2007; 78(3): 233 - 239. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. F. Elias, L. M. Sullivan, P. K. Elias, R. B. D'Agostino Sr, P. A. Wolf, S. Seshadri, R. Au, E. J. Benjamin, and R. S. Vasan Left Ventricular Mass, Blood Pressure, and Lowered Cognitive Performance in the Framingham Offspring Hypertension, March 1, 2007; 49(3): 439 - 445. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Staessen, T. Richart, and W. H. Birkenhager Less Atherosclerosis and Lower Blood Pressure for a Meaningful Life Perspective With More Brain Hypertension, March 1, 2007; 49(3): 389 - 400. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Monastero, K. Palmer, C. Qiu, B. Winblad, and L. Fratiglioni Heterogeneity in Risk Factors for Cognitive Impairment, No Dementia: Population-Based Longitudinal Study From the Kungsholmen Project Am J Geriatr Psychiatry, January 1, 2007; 15(1): 60 - 69. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Young, A. G. Mainous III, and M. Carnemolla Hyperinsulinemia and Cognitive Decline in a Middle-Aged Cohort Diabetes Care, December 1, 2006; 29(12): 2688 - 2693. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Akomolafe, A. Beiser, J. B. Meigs, R. Au, R. C. Green, L. A. Farrer, P. A. Wolf, and S. Seshadri Diabetes Mellitus and Risk of Developing Alzheimer Disease: Results From the Framingham Study Arch Neurol, November 1, 2006; 63(11): 1551 - 1555. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S.C. Korf, L. R. White, P. Scheltens, and L. J. Launer Brain Aging in Very Old Men With Type 2 Diabetes: The Honolulu-Asia Aging Study Diabetes Care, October 1, 2006; 29(10): 2268 - 2274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. A. Ajani and E. S. Ford Has the Risk for Coronary Heart Disease Changed Among U.S. Adults? J. Am. Coll. Cardiol., September 19, 2006; 48(6): 1177 - 1182. [Abstract] [Full Text] [PDF]
|
|
|
|
