HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 60:1409-1413 (2005)
© 2005 The Gerontological Society of America

Changes in Biological Markers of Health: Older Americans in the 1990s

Eileen M. Crimmins1,, Dawn Alley1, Sandra L. Reynolds2, Melanie Johnston3, Arun Karlamangla4 and Teresa Seeman4

1 Andrus Gerontology Center, University of Southern California, Los Angeles.
2 School of Aging Studies, University of South Florida, Tampa.
3 Department of Preventive Medicine and Biometrics, University of Colorado Health Sciences Center, Denver.
4 Division of Geriatrics, University of California, Los Angeles, School of Medicine.

Address correspondence to Eileen M. Crimmins, PhD, Andrus Gerontology Center, University of Southern California, 3715 McClintock Avenue, MC 0191, Los Angeles, CA 90089-0191. E-mail: crimmin{at}usc.edu


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Background. Many studies that show improved health in older adults have relied on subjective measures of health. This article assesses changes in the physiological status of older Americans during the 1990s using biological measures of high-risk for morbidity and mortality.

Methods. Changes in the prevalence of clinically-defined, high risk for 10 biological markers were assessed in respondents age 65 years and older from National Health and Nutrition Examination Surveys (NHANES) III (1988–1994) and IV (1999–2000).

Results. Some changes in prevalence of high-risk values of biological markers indicate improved health among older adults in the 1990s: a 6% reduction in the prevalence of high-risk total cholesterol (p <.001) and a 7% reduction in the prevalence of high-risk homocysteine (p <.001). Other changes indicate worsening health: a 9% increase in the prevalence of high-risk systolic blood pressure (p <.01), a 10% increase in obesity (p <.001), and an 8% increase in the prevalence of high-risk C-reactive protein (p <.001). These changes remained significant after adjusting for age, sex, and education. Results of logistic regressions indicate that changes in the frequency of medication usage, medication efficacy, prevalence of chronic disease, and diet explained some of these changes.

Conclusions. Changes in the prevalence of high-risk values of biological markers in the 1990s are mixed. Greater use and effectiveness of lipid-lowering medication has contributed to the reduction in percentage of the population with high-risk lipid levels, and folate supplementation accounted for a decline in the percentage with high-risk homocysteine. However, increases in the percentage with high-risk systolic blood pressure occurred despite an increase in the use of antihypertensive medications, in part because of the limited ability of antihypertensive medications to bring blood pressure below high-risk levels.

HEALTH in the older population, as measured by many dimensions, appears to have improved during the last two decades (1–4). Mortality has continued to decline, and disability and loss of functioning are less common than in the past. Not all indicators point toward improving health, however. The prevalence of most diseases has increased in the older population (1,5,6).

Most studies examining trends in the health of older adults have focused on self-reports of health that can be affected by circumstances other than the intrinsic health of the individual (7,8). For instance, self-reports of conditions and diseases are subject to variations in knowledge, access to health care, and disease diagnosis. Disability measures can be confounded by role expectations and living environments (9,10).

In contrast, the physiological status that underlies disability, loss of functioning, disease, and mortality can be more directly measured through biological indicators of risk. Trends in these biological markers should be a good indicator of trends in the innate health status of the older population as all of these markers have been linked to higher mortality, poorer physical and cognitive functioning, and more heart disease (11–16).

Past research on trends in biological risk markers has focused on cardiovascular risk factors, namely blood pressure, serum cholesterol levels, and relative weight (17–19). From the 1960s through the early 1990s, decreases in hypertension were fueled by increased diagnosis and treatment of blood pressure problems (17,20). However, there is some evidence that improvement in blood pressure may have stopped and that blood pressure may have actually increased through the 1990s (21). Cholesterol appears to have decreased steadily from the 1960s through 2000 (22). One of the most adverse health trends in recent decades has been the increase in weight and obesity (23). Several other markers, such as triglycerides, glycated hemoglobin, C-reactive protein (CRP), and homocysteine level, are increasingly considered clinically relevant and change over time in these markers also requires attention. The intent of this article is to examine changes in the prevalence of high-risk values for indicators of physiological functioning in older Americans through the 1990s and to explore reasons for these observed changes.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Data
The National Health and Nutrition Examination Surveys (NHANES) are cross-sectional studies of the noninstitutionalized U.S. population including interview, clinical examination, and laboratory tests (24). In this analysis, data from NHANES III (1988–1994) and NHANES IV (1999–2000) were used to examine change in biological risk in the population aged 65 years and older. Although data collection for these studies took place over a number of years, each person was interviewed only once. NHANES III centered on 1991, and NHANES IV centered on 1999; differences between the two surveys represent change for an average of 8 years. There were 4495 respondents 65 and older in NHANES III and 1196 in NHANES IV who completed both the interview and the examination and laboratory components. All analyses were conducted using STATA (25) to account for the design effect, and the sample completing the laboratory components was weighted to reflect the noninstitutionalized population of the United States.

Measures
We examined change across the two studies in the percentage of the population with measured values above (or below) clinically defined "high risk" levels for 10 biological markers: diastolic blood pressure, systolic blood pressure, high-density lipoprotein (HDL) cholesterol, fasting low-density lipoprotein (LDL) cholesterol, total cholesterol, glycated hemoglobin, body mass index (BMI), fasting triglycerides, CRP, and homocysteine. Clinical cut points used to define high-risk levels are shown in Table 1. For all variables, at least 87% of those eligible to provide data did so at both dates. Measures of LDL cholesterol and triglycerides were available for the participants in the morning sample fasting at least 6 hours (about 40% of the total laboratory sample), and measures of homocysteine were only available for the second half of NHANES III (1991–1994). For these reasons, these three indicators have data for a smaller number of respondents.


View this table:
[in this window]
[in a new window]
  		Table 1. Percentage of Population Aged 65+ Years at High Risk for Biological Markers in NHANES III and IV.

 
Because assays used to measure CRP differed somewhat at the two dates, an adjustment of NHANES IV assay values was performed using information derived from comparisons made by the laboratory doing both sets of assays. The correlation between the two assays at levels greater than 3.0 mg/l was 0.993, with a slope indicating that NHANES IV values need to be reduced by 9.9% for comparability. Because the lower level of sensitivity changed, we used 4.0 mg/l as the cut point for risk across the surveys (26). Persons with levels of CRP >10 mg/l were assumed to have acute infections and were eliminated from CRP analyses.

Measured values of biological markers were used to identify individuals with high-risk levels of each indicator, regardless of medication use status. Thus, persons on medication whose measured levels were outside the risk range were not considered to be at risk. Subsequent analyses examined the changing prevalence of medication usage relative to measured risk. Self-report data were used to examine age, gender, education, disease diagnosis, and medication prescription.

Analysis
First, we examined change across the surveys in the percentage with high-risk values. For each biomarker, we then used logistic regression to examine the relative odds of being at high risk in the NHANES IV sample versus the NHANES III sample. Regression models examined the relative odds of being high-risk after controlling for changes in age and sex of the population (Model 1) and educational level (Model 2). Because weight increased markedly between the two surveys, Model 3 also controlled for BMI. A final set of analyses explored changes in factors likely to affect change in individual biomarkers, including diagnosis and medication use, prevalence of chronic disease, and the level of serum folate.


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
Table 1 presents the percentage of the U.S. population aged 65 years and older with high-risk levels of each biological measure in the two studies. By NHANES IV, a majority of older Americans had high-risk levels of systolic blood pressure, almost one third had weight levels considered obese, and one quarter had high-risk levels of total cholesterol. The percentage of older Americans with high-risk systolic blood pressure increased by 9% from NHANES III to IV, but there was no significant change in the prevalence of high-risk diastolic blood pressure (Table 1). The prevalence of high-risk levels of total and LDL cholesterol declined significantly (6.1% and 8.9%, respectively); however, there was no significant change in the percentage with high-risk levels of HDL cholesterol or glycated hemoglobin. The prevalence of obesity among the elderly population increased from 21.4% to 30.8%. The prevalence of high-risk levels of triglycerides appears to have increased marginally, from 18.4% to 21.8%, and the prevalence of CRP greater than 4.0 mg/l increased significantly (from 21.3% to 29.6%). However, the prevalence of high-risk levels of homocysteine declined by 7%, from 17.6% to 10.3%.

Explanations for Changes in Percentage of Population With High-Risk Values of Biological Markers
Table 2 presents results from regression models testing the robustness of the observed differences in prevalence of high risk to controls for change in population composition. Model 1 indicates the odds ratios for being in the high-risk group in NHANES IV versus NHANES III adjusted for age and sex. Generally, the results are the same as those in Table 1, indicating that changes in age and in sex composition, or both, do not explain observed changes over time. The odds for high-risk levels of systolic blood pressure were 26% higher, the odds for high-risk BMI were 43% higher, and the odds for high-risk CRP were 68% higher for the NHANES IV sample than for the NHANES III sample. However, the odds of having high-risk levels of LDL and total cholesterol were 28% and 35% lower, respectively, and the odds of having a high-risk level of homocysteine were 40% lower. The odds of having high-risk diastolic blood pressure, HDL cholesterol, glycated hemoglobin, and fasting triglycerides did not change significantly.


View this table:
[in this window]
[in a new window]
  		Table 2. Likelihood of Being in a High-Risk Group in NHANES IV Relative to NHANES III.

 
Because educational attainment has been linked to better health and the educational level of the older population increased markedly in the last decade (the percentage of Americans aged 65 and older who have more than a high school education increased from 26% in NHANES III to 34% in NHANES IV), Model 2 adjusted for years of education. This did not significantly change any of the results. Because body weight is related to many other biological indicators of risk, Model 3 controlled for BMI. Controlling for BMI did not change the significance or direction of the results, although results indicate that the increase in prevalence of high-risk CRP would have been somewhat smaller without the concurrent increase in obesity.

Explanations for Changes in Specific Markers
To further investigate the causes for these changes, we assessed changes in related factors. For blood pressure and cholesterol, we examined changes in diagnosis, medication prescription, and use and effectiveness of medication. To do this, we divided the sample by diagnosis, use of medication, and measured risk group. For CRP, we examined changes in the prevalence of concurrent disease. For homocysteine, we examined changes in level of serum folate.

Systolic blood pressure.-- Self-report of physician diagnosis of high blood pressure increased during the 1990s, with 41% of older adults reporting that they had been diagnosed with high blood pressure in NHANES III (46% in NHANES IV). However, the use of medication to control hypertension in the older population also increased during the 1990s, from 35% in NHANES III to 43% in NHANES IV. If we used an alternative approach to defining high-risk blood pressure that included those taking medication as well as those with measured high-risk systolic blood pressure, the increase in the presence of high-risk blood pressure between NHANES III and NHANES IV would have been even greater, from 59% to 70%.

Among those persons who reported having been diagnosed with high blood pressure, almost all (90%) had been told to take medication in each survey. Taking blood pressure medication does not necessarily reduce an individual's blood pressure below the risk level; among those taking medication, 56% in NHANES III and 60% in NHANES IV (19.4% and 25.1% of the total samples) had high measured systolic blood pressure. Another reason for the rise in measured systolic blood pressure is the increase in undiagnosed high blood pressure, in which participants measured at-risk did not report prior diagnosis (17% in NHANES III to 22% in NHANES IV). Thus, despite the increase in treatment, an increase in undiagnosed high blood pressure and diagnosed, but inadequately controlled, hypertension are factors in the growing prevalence of high-risk systolic blood pressure.

Cholesterol.-- Although we observed a decline in measured high-risk LDL and total cholesterol, the percentage reporting that they had been diagnosed with high cholesterol increased (31% in NHANES III vs 45% in NHANES IV). There was also an increase in the use of medication to control cholesterol, from 20% to 35%. Medication prescription among those diagnosed with high cholesterol increased markedly during the 1990s (from 33% in NHANES III to 62% in NHANES IV). In addition, the effectiveness of medications available to treat high cholesterol increased (33). In NHANES III, 40% of those taking medication still had high-risk levels of cholesterol; this figure fell to only 21% in NHANES IV. Thus, increases in cholesterol testing and diagnosis, as well as increased use and efficacy of medications, may have contributed to the observed decrease in high-risk total cholesterol.

C-reactive protein.-- The increase in CRP observed here was paralleled by an increase in BMI (Table 2, Models 2 and 3). Controlling for BMI suggests that, without the increase in weight, the odds of high-risk CRP in NHANES IV versus NHANES III would have been 59% higher, instead of 70%. Higher CRP levels in NHANES IV may also be associated with the increased prevalence of chronic diseases in the later survey period. Controlling for the change in prevalence of seven chronic conditions related to inflammation (hypertension, diabetes, chronic heart failure, heart attack, asthma, chronic bronchitis, and arthritis) reduced the odds ratio (OR) for the indicator of change in high-risk CRP by 0.12 (results not shown), suggesting that some of the observed increase in CRP was related to the concurrent increasing prevalence of chronic conditions.

Homocysteine.-- We considered whether the decreasing prevalence of high-risk homocysteine might be the result of changes in folate supplementation. Higher levels of serum folate led to lower levels of homocysteine (34), and folate consumption in the population increased after 1998 because of regulated folate supplementation of cereals and flours to prevent neural tube defects in fetuses (35). Although this policy was oriented toward improving pregnancy outcomes, it also resulted in improved dietary intake of folate among elderly persons. The mean level of serum folate in the older population increased from 9.8 ng/ml in NHANES III to 20.9 ng/ml in NHANES IV. These increasing levels of serum folate appear to completely explain the decrease in high-risk homocysteine levels between the two surveys, as addition of serum folate to Model 3 in Table 2 resulted in an insignificant OR for the indicator of change in homocysteine (OR = 1.14, results not shown).


   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
Comparisons of the prevalence of high-risk levels of a range of clinically relevant biological markers in older adults at the beginning versus the end of the 1990s revealed a mixed pattern of changes in health. There was an increase in the prevalence of high-risk systolic blood pressure reflecting increases in both diagnosed and undiagnosed cases and in the fact that blood pressure may not be reduced below risk levels even among those using medication. There were also significant increases in the percentage of the older population with high-risk levels of CRP during the 1990s. Concurrent changes in factors related to CRP, including weight and the prevalence of diseases, may explain part of the increase.

In contrast to these patterns of increasing risk, prevalence of high-risk cholesterol showed considerable improvement during the 1990s. This appears to partially result from the dramatic increase in the use of more effective lipid-lowering medications. Another notable health improvement during the 1990s was the reduction in high-risk homocysteine—a change that appears to have been accomplished through public health efforts that led to elevations in serum folate levels in the older U.S. population.

Conclusion
We found mixed patterns of change in the prevalence of high-risk levels of various clinically relevant biomarkers in older Americans through the 1990s: increasing prevalence of high-risk levels for some measures, but decreasing levels for others. Overall, the data suggest that the prevalence of high-risk levels would have risen more without the increased use of medications.


   Acknowledgments
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Support for this project was provided by the National Institutes of Health (grants P30 AG17265, R01 AG023347, K12 AG01004, and T32 AG00037) and by fellowship support from the University of Southern California.

Preliminary results were presented at the Annual Meeting of the Population Association of America, Boston, Massachusetts, April 2, 2004.


   Footnotes
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Decision Editor: John E. Morley, MB, BCh

Received September 24, 2004

Accepted December 20, 2004


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Crimmins EM. Trends in the health of the elderly. Annu Rev Public Health. 2004; 25:17.1-17.20/10.1146/annurev.publhealth.25.102802.124401.
  2. Freedman VA, Crimmins EM, Schoeni RF, et al. Resolving inconsistencies in old-age disability trends: report from a technical working group. Demography. 2004;41:417-441.[Medline]
  3. Freedman VA, Martin LG, Schoeni RF. Recent trends in disability and functioning among older Americans: a critical review of the evidence. JAMA. 2002;288:3137-3146.[Abstract/Free Full Text]
  4. Manton KG, Gu X. Changes in the prevalence of chronic disability in the U.S. Black and Non-black population among 65 from 1982–1999. Proc Natl Acad Sci U S A. 2001;98:6354-6359.[Abstract/Free Full Text]
  5. Crimmins EM, Saito Y. Change in the prevalence of diseases among older Americans: 1984-1994. Demographic Research. 2000, Available at: http://www.demographic-research.org/Volumes/Vol3/9/.
  6. Cutler DM, Richardson E. Measuring the health of the U.S. Population. In: Bailey MN, Reiss PC, Winston C, ed. Brookings Papers on Economic Activity: Microeconomics. Washington, DC: The Brookings Institution; 1997:217–271.
  7. Crimmins EM. Is disability declining among the elderly? Defining disability and examining trends. Crit Issues Aging. 1998;2:10-11.
  8. Jette AM. How measurement techniques influence estimates of disability in older populations. Soc Sci Med. 1994;38:937-942.
  9. Freedman VA, Martin LG. Understanding trends in functional limitations among older Americans. Am J Public Health. 1998;88:1457-1462.[Abstract/Free Full Text]
  10. Simonsick EM, Kasper JD, Guralnik JM, et al. Severity of upper and lower extremity functional limitation: scale development and validation with self-report and performance-based measures of physical function. J Gerontol B Psychol Sci Soc Sci. 2001;56:S10-S19.[Abstract/Free Full Text]
  11. Danesh J, Whincup P, Walker M, et al. Low grade inflammation and coronary heart disease: prospective study and updated meta-analysis. BMJ. 2000;321:199-204.[Abstract/Free Full Text]
  12. Karlamangla AS, Singer BH, McEwen BS, Rowe J, Seeman TE. Allostatic load as a predictor of functional decline: MacArthur studies of successful aging. J Clin Epidemiol. 2002;55:696-710.[Medline]
  13. Ozbey N, Sencer E, Molvalilar S, Orhan Y. Body fat distribution and cardiovascular disease risk factors in pre- and post-menopausal obese women with similar BMI. Endocrine J. 2002;49:503-509.
  14. Seeman TE, Crimmins EM, Huang M, et al. Cumulative biological risk and socio-economic differences in mortality: MacArthur Studies of Successful Aging. Soc Sci Med. 2004;58:1985-1997.
  15. Seeman TE, Singer BH, Rowe JW, Horwitz RI, McEwen BS. Price of adaptation – allostatic load and its health consequences. Arch Intern Med. 1997;157:2259-2268.[Abstract/Free Full Text]
  16. Zemel PC, Sowers JR. Relation between lipids and atherosclerosis: epidemiologic evidence and clinical implications. Am J Cardiol. 1990;66:71-121.
  17. Barker WH, Mullooly JP, Linton KL. Trends in hypertension prevalence, treatment, and control in a well-defined elderly population. Hypertension. 1998;31;:552-559.[Abstract/Free Full Text]
  18. Burke GL, Sprafka JM, Folsom AR, Hahn LP, Luepker RV, Blackburn H. Trends in serum cholesterol levels from 1980 to 1987: The Minnesota Heart Survey. N Engl J Med. 1991;324:941-946.[Abstract]
  19. Freedman DS, Khan LK, Serdula MK, Galuska DA, Dietz WH. Trends and correlates of Class 3 obesity in the United States from 1990 through 2000. JAMA. 2000;288:1758-1761.
  20. Burt VL, Cutler JA, Higgins M, et al. Trends in the prevalence, awareness, treatment, and control of hypertension in the adult US population: data from the Health Examination Surveys, 1960 to 1991. Hypertension. 1995;26:60-69.[Abstract/Free Full Text]
  21. Ayala C, Greenlund KJ, Croft JB, et al. State-specific trends in self-reported blood pressure screening and high blood pressure – United States, 1991–1999. MMWR. 2002;5:456-460.
  22. National Center for Health Statistics. Health, United States, 2002: With chartbook on trends in the health of Americans. 2002. Hyattsville, Md: National Center for Health Statistics.
  23. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA. 2002;288:1723-1727.[Abstract/Free Full Text]
  24. Centers for Disease Control (2004). NHANES III and IV Documentation. Available at: http://www.cdc.gov/nchs/nhanes.htm. Accessed May 12, 2004.
  25. StataCorp., Stata Statistical Software: Release 8, 2001; Stata Corporation; College Station, TX.
  26. Ridker PM. C-reactive protein: a simple test to help predict risk of heart attack and stroke. Circulation. 2003;108:e81-e85.[Free Full Text]
  27. Chobanian AV, Bakris GL, Black HR, et al. Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003;42:1206-1252.[Abstract/Free Full Text]
  28. National Cholesterol Education Program. Detection, evaluation, and treatment of high blood cholesterol in adults. Available at: http://www.nhlbi.nih.gov/guidelines/cholesterol/atp3xsum.pdf. Accessed March 4, 2004.
  29. United States Preventive Services Task Force. Use of glycated hemoglobin and microalbuminuria in the monitoring of diabetes mellitus. Available at: http://www.ahrq.gov/clinic/epcsums/glycasum.pdf. Accessed March 4, 2004.
  30. World Health Organization expert consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157-163.[Medline]
  31. Clarke R, Smith D, Jobst KA, Refsum H, Sutton L, Uleland PM. Folate, vitamin B12, and serum total homocysteine levels in confirmed Alzheimer disease. Arch Neurol. 1998;55:1449-1455.[Abstract/Free Full Text]
  32. Figlin E, Chetrit A, Shahar A, et al. High prevalences of vitamin B12 and folic acid deficiency in elderly subjects in Israel. Br J Haematol. 2003;123:696-701.[Medline]
  33. Stein EA. New statins and new doses of older statins. Curr Atheroscler Rep. 2001;3:14-18.[Medline]
  34. Selhub J, Jacques PE, Wilson PW, Rush D, Rosenberg IH. Vitamin status and intake as primary determinants of homocysteinemia in an elderly population. JAMA. 1993;270:2693-2698.[Abstract/Free Full Text]
  35. Choumenkovitch SF, Selhub J, Wilson PWF, Rader JI, Rosenberg IH, Jacques PF. Folic acid intake from fortification in United States exceeds predictions. J Nutr. 2002;132:2792-279.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 GerontologistHome page
M. G. Parker and M. Thorslund
Health Trends in the Elderly Population: Getting Better and Getting Worse
Gerontologist, April 1, 2007; 47(2): 150 - 158.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS