Author Affiliations: Department of Epidemiology and the Welch Center for Prevention, Epidemiology, and Clinical Research, Bloomberg School of Public Health (Drs Bower, Lazo, and Selvin and Mr Juraschek), School of Medicine (Mr Juraschek), and Division of General Internal Medicine, Department of Medicine (Drs Lazo and Selvin), Johns Hopkins University, Baltimore, Maryland.
C-reactive protein (CRP) is a marker of systemic inflammation and cardiovascular disease.1- 3 Based on findings from recent clinical trials, CRP has been recommended as an adjunct screening tool to stratify cardiovascular risk in the general population.4 However, evidence regarding within-person variability of CRP in the general population is limited. Short-term variability in CRP has important implications for its use and interpretation in clinical practice and research studies. Thus, the objective of this study was to evaluate the short-term, within-person variability in CRP measurements and to quantify the impact of repeated testing on CRP-based cardiovascular risk classification.
We included 541 participants aged 16 to 69 years who completed repeated examinations of the 2001-2002 National Health and Nutrition Examination Survey (NHANES). Briefly, a 5% nonrandom sample of 2001-2002 NHANES participants was recruited for the second examination, occurring approximately 2.5 weeks after the original examination. Participants represented a uniform distribution of individuals by age, sex, and race/ethnicity. The study design and methods for NHANES are detailed elsewhere.5
High-sensitivity serum CRP was measured using latex-enhanced nephelometry.5 We used a cut point of a minimum level of 10.0 mg/L to define an elevated CRP level, based on the NHANES laboratory reference values and American Heart Association/Centers for Disease Control and Prevention recommendations.1 We also conducted sensitivity analyses using a higher cut point (CRP level ≥ 20.0 mg/L). (To convert CRP to nanomoles per liter, multiply by 9.524.)
The Spearman rank correlation and intraclass correlation (ICC) coefficients, and the within-person coefficient of variation (CVw) were used to characterize short-term, within-person variability.6 A persistently elevated CRP level was defined as CRP level of at least 10.0 mg/L at both examinations. We used scatterplots and Bland-Altman plots to visually display measurement variability.6 Finally, we calculated the percentage of participants whose risk category was reclassified owing to repeated testing.
The mean (SD) age of participants was 38.0 (16.5) years. Fifty percent of the study population were female, and 48% were of non-Hispanic white race/ethnicity. The mean time between examinations was 18.9 days. The mean CRP level was 4.5 mg/L (95% CI, 3.9-5.1) at the first examination and 4.3 mg/L (95% CI, 3.8-4.9) at the second examination (P value for the difference, .45). The Spearman rank correlation between visits was 0.65, the ICC was 0.77 (95% CI, 0.69-0.84), and the CVw was 46.2% (95% CI, 42.9%-49.3%). The high variability in CRP can be seen visually on the scatterplot (Figure), although the Bland-Altman plot shows that most of the discordance between examinations occurred at higher values (>10.0 mg/L) (eFigure). The variability was particularly high among persons with CRP levels greater than 20.0 mg/L.
Figure. Scatterplot of high-sensitivity C-reactive protein (hs-CRP) measurements in a subsample of the 2001-2002 National Health and Nutrition Examination Survey.5 Repeated measurements occurred approximately 19 days apart. A, Measurements are plotted on the arithmetic scale. B, Measurements are plotted on the natural logarithm scale. Solid lines are placed on the x-axis and y-axis where the CRP level = 10.0 mg/L. (To convert CRP to nanomoles per liter, multiply by 9.524.)
The prevalence of an elevated CRP level of at least 10.0 mg/L was 10.5% at the first examination and 10.4% at the second; 7.2% of participants had persistently elevated CRP levels (eTable). Of those with a normal CRP level at the first examination, only 3.5% had CRP levels of at least 10.0 mg/L at the second. Of those with CRP levels of at least 10.0 mg/L at the first examination, 32% were reclassified as having CRP levels lower than 10.0 mg/L at the second. The prevalence of a CRP level of at least 20.0 mg/L at the first examination was 4.3% and was 2.8% at the second; 1.5% of participants had CRP levels of at least 20.0 mg/L at both examinations, representing an approximately 65% decrease in prevalence.
In this sample of the general population, we observed significant short-term (approximately 2.5 weeks) within-person variability in CRP levels, particularly at high values. Approximately one-third of persons with elevated CRP levels were reclassified as having normal CRP levels after repeated testing. Our results are consistent with those of previous studies conducted in small selected populations (eg, patients with ischemic heart disease) or in which measurements were months or years apart.7,8 Of note, we observed greater variation at higher values in cases in which clinicians are most likely to intervene.
The 2010 American College of Cardiology Foundation/American Heart Association guidelines for the assessment of cardiovascular risk in asymptomatic adults includes recommendations for CRP level to select patients for statin therapy when the low-density lipoprotein cholesterol concentration is level than 130.0 mg/L.9 Our results suggest that use of a single CRP measure for riskstratification may lead to substantial misclassification. Recommendations for repeated testing to confirm elevations in CRP level prior to altering medical decision-making may be warranted, particularly among those with CRP values near the risk cut points.
Correspondence: Dr Selvin, Welch Center for Prevention, Epidemiology and Clinical Research and the Bloomberg School of Public Health, Johns Hopkins University, 2024 E Monument St, Ste 2-600, Baltimore, MD 21287 (email@example.com).
Published Online: September 3, 2012. doi:10.1001/archinternmed.2012.3712
Author Contributions: Dr Bower had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Bower, Lazo, and Selvin. Analysis and interpretation of data: Bower, Lazo, Juraschek, and Selvin. Drafting of the manuscript: Bower. Critical revision of the manuscript for important intellectual content: Bower, Lazo, Juraschek, and Selvin. Statistical analysis: Bower and Selvin. Obtained funding: Selvin. Study supervision: Lazo and Selvin.
Financial Disclosure: None reported.
Funding/Support: Drs Bower and Juraschek were supported by National Institutes of Health/National Heart, Lung, and Blood Institute grant T32 HL007024.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Internal Medicine editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Bower JK, Lazo M, Juraschek SP, Selvin E. Within-Person Variability in High-Sensitivity C-Reactive Protein. Arch Intern Med. 10.1001/archinternmed.2012.3712.
eTable. Percentage (95% CI) of Participants With Elevated C-Reactive Protein Measured at 2 Time Points (18.9 Days Apart) in a Subsample of NHANES 2001-2002 (n=541)
eFigure. Bland–Altman Plots of the Difference in CRP (mg/dL) Between Measure 1 and Measure 2.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 4
Customize your page view by dragging & repositioning the boxes below.
The Rational Clinical Examination
An abnormal peripheral WBC count, ESR, and CRP were found to have limited diagnostic power for...
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.