Trends in Apolipoprotein B, Non–High-Density Lipoprotein, and Low-Density Lipoprotein for Adults 60 Years and Older by Use of Lipid-Lowering Medications
United States, 2005 to 2006 Through 2013 to 2014
Over the years, guidelines for cholesterol lowering have focused on total cholesterol (TC) or low-density lipoprotein cholesterol (LDL-C). However, in recent years, apolipoprotein B (apo B) and non–high-density lipoprotein cholesterol (HDL-C) have been proposed as better measures of the atherosclerotic burden of lipids and improved measures for risk prediction.1 In this research letter, we examine trends in mean apo B, non–HDL-C, and LDL-C in adults ≥60 years of age by use of lipid-lowering medication from 2005 to 2006 through 2013 to 2014 using five 2-year cross-sectional NHANES (National Health and Nutrition Examination) surveys.
NHANES uses a stratified, multistage probability design to produce samples representative of the US noninstitutionalized population. Written informed consent was obtained from adult participants. The survey was approved by the National Center for Health Statistics Research Ethics Review Board.
NHANES includes a home interview consisting of health-related questions, including lipid-lowering medication use, and an examination at a mobile examination center that included blood-derived measures. Participants in the examination component were randomly assigned to a morning session (and asked to fast at least 9 hours before examination) or an afternoon/evening session. Each examined participant was eligible for TC and HDL-C; only participants examined in the morning were eligible for apo B and triglycerides.
In 2013 to 2014, 57.8% of adults ≥60 years of age eligible for the survey were examined. Comparable response rates for 2011 to 2012, 2009 to 2010, 2007 to 2008, and 2005 to 2006 were 56.9%, 66.7%, 67.7%, and 65.0%, respectively.
Apo B, TC, HDL-C, and triglycerides were analyzed on venous samples following standardized protocols.2 Across survey periods, there were changes in laboratories, methods, and analyzers used to measure these lipids, including apo B.3 Measurement of TC, HDL-C, and triglycerides has been described elsewhere.2 LDL-C was calculated by the Friedewald equation [LDL-C=TC−(HDL-C+triglycrides/5)]4 and non–HDL-C as TC−HDL-C. All measures were performed according to the criteria of the Centers for Disease Control’s lipid standardization program (http://www.cdc.gov/labstandards/lsp_faq.html), ensuring accuracy and comparability.
Use of lipid-lowering medications was assessed with the question, “Are you currently taking medication to lower your blood cholesterol?” regardless of specific types of lipid-lowering medications.
Trends in apo B, non–HDL-C, and LDL-C were tested at the α=.05 level with a 2-sided t statistic and orthogonal contrast matrices.5
For adults on lipid-lowering medications, the effect of sex on trends in these lipids was tested with multiple linear regression.
Examination sample weights for non–HDL-C and morning fasting sample weights for apo B and LDL-C adjusted for differential sampling, nonresponse, and noncoverage were used to estimate all population parameters. Standard errors were estimated and statistical hypotheses were tested, accounting for stratification and clustering in addition to weighting. Analysis was conducted with SAS 9.4 (SAS Institute Inc) and SUDAAN 11.0 (RTI International).
There were 3693 adults ≥60 years of age with measured apo B, 8217 with calculated non–HDL-C, and 3642 with calculated LDL-C during 2005 to 2014.
Forty-two percent of men and 39% of women ≥60 years of age reported using lipid-lowering medications in 2005 to 2014.
From 2005 to 2006 through 2013 to 2014, mean apo B declined in men on lipid-lowering medications by 12 mg/dL (P<0.01; Figure). Declines were also seen in mean non–HDL-C (15 mg/dL; P<0.01) and mean LDL-C (17 mg/dL; P<0.01; Figure). For women, significant declines were seen only for mean non–HDL-C (10 mg/dL; P<0.05). The observed declines seen in mean apo B (6 mg/dL; P=0.19) and LDL-C (7 mg/dL; P=0.36) were not statistically significant. There were no significant interactions between survey year and sex for adults on lipid-lowering medications for apo B (P=0.07), non–HDL-C (P=0.17), or LDL-C (P=0.08).
From 2005 to 2006 through 2013 to 2014, there were no significant changes in these lipids in either sex not on lipid-lowering medications. Mean apo B, non–HDL-C, and LDL-C were consistently lower in adults on lipid-lowering medications than in those not on lipid-lowering medications for both men and women.
Treated men showed significant declines in these lipids over time. The observed declines in women were of smaller magnitude but not statistically significant for apo B or LDL-C. Similar to all surveys, NHANES is subject to measurement and sampling variability, which could explain in part these differential sex trends. The sample sizes for apo B and LDL-C, based on adults examined in the morning, may be too small to detect a statistically significant trend of a smaller magnitude seen in women when a statistically significant trend exists. Limitations of the data include the use of lipid-lowering medications based on self-report and lack of information on medication type, dose, and duration of use. The next steps for future research include an analysis to understand the reasons for the declines among users of lipid-lowering medications but not among nonusers.
Analyses incorporating changes in the class or intensity of medications used and shifts in nonpharmacological correlates of serum lipid concentrations (including diet and physical activity) may provide insight into the patterns observed and inform clinical and public health interventions.
The author thank David Lacher, MD, M.Ed, for ensuring the quality of the data. The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the National Center for Health Statistics, the Centers for Disease Control and Prevention, or the National Heart, Lung, and Blood Institute.
Sources of Funding
The National Health and Nutrition Examination Survey is funded by a collaboration of US government agencies, including the Centers for Disease Control and the National Heart, Lung, and Blood Institute.
Data sharing: The sources of the data used to produce the results and of the measurement of apo B during 2005 to 2006 through 2013 to 2014 are available on the Centers for Disease Control website (https://www.cdc.gov/nchs/nhanes/index.htm). The statistical methods used in the article, including the software packages (SAS and SUDAAN), are outlined in the research letter.
- © 2018 American Heart Association, Inc.
- Ballantyne CM,
- Pitt B,
- Loscalzo J,
- Cain VA,
- Raichlen JS
- Rosinger A,
- Carroll MD,
- Lacher D,
- Ogden C
- 3.↵Centers for Disease Control and Prevention/National Center for Health Statistics/National Health and Nutrition Examination Survey. 2007–2008 Data documentation, codebook, and frequencies, apolipoprotein B (ApoB_E), analytic notes. 2010. https://wwwn.cdc.gov/Nchs/Nhanes/2007–2008/APOB_E.htm. Accessed March 27, 2018.
- Friedewald WT,
- Levy RI,
- Fredrickson DS
- Fisher RA,
- Yates F