Trends and Racial/Ethnic Disparities in Gluten-Sensitivity
Trends and Racial/Ethnic Disparities in Gluten-Sensitivity
The NHANES is conducted by the National Center for Health Statistics of the Center for Disease Control and Prevention. It collates nationally representative data on the health and nutritional status of the noninstitutionalized, civilian population of the United States. It utilizes a complex, stratified, and multistage probability sampling design and collects information from ~5,000 persons per year using standardized household interviews, physical examinations, and testing of biologic samples. More detailed information on the survey design for the NHANES, including approval from the institutional review board for data collection and analysis, is available from the survey documentation.
NHANES III was conducted from 1988 to 1994. Beginning in 1999, NHANES was redesigned to become a continuous survey; usually data from 2 or more years are combined to achieve adequate sample sizes for statistical analyses. A questionnaire covering only nonsensitive topics was used to interview participants at home to collect demographic data on age, sex, ethnicity, education, poverty index, and so on. Physical examinations and blood collection were also conducted.
For this analysis, data on 14,701 participants who underwent CD testing in the NHANES 2009–2012 period were used. There was no difference in participation in CD testing by sex (P=0.10), but participation was lower among non-Hispanic blacks (83%) compared with non-Hispanic whites (89%) and Hispanics (88%; P<0.001). Participation was also lower in the 6–8-year age group (81%) than in the >20-year age group (89%; P<0.001). To estimate the prevalence of CD in participants aged 6 years or older in NHANES 2009–2012, CD was defined by serological diagnosis or a reported clinical diagnosis of CD. A reported clinical diagnosis of CD was defined by positive responses to two CD-related questions: (i) Has a doctor or other health professional ever told you that you have CD, also called sprue? And (ii) are you on a gluten-free diet? To estimate the prevalence of PWAG in participants aged 6 years or older in NHANES 2009–2012, PWAG was defined by two CD-related questions and a CD serology test; the test was considered "positive" if the individual adhered to a GFD with a concomitant absence of a physician-given diagnosis of CD and/or negative on serologic testing.
As CD serology testing in NHANES III (1988–1994) and NHANES 1999–2004 was limited to adults aged 50 years and older because of the NHANES regulations, data on adults aged 50 years or older with CD serology testing from NHANES III or continuous NHANES (1999–2004, 2009–2012) were used for describing the secular trends in the prevalence of CD between 1998 and 2012. Of 19,864 adults aged ≥50 years enrolled in NHANES III, NHANES 1999–2004, or NHANES 2009–2012, serum samples of 17,010 individuals (9,122 non-Hispanic whites, 3,276 non-Hispanic blacks, and 3,928 Hispanics) were tested for CD.
Serum specimens of NHANES 2009–2012 and stored serum samples of NHANES III and NHANES 1999–2004 were shipped to the Celiac Disease Research Laboratory at Mayo Clinic, Rochester, MN, USA, which performed serological testing using laboratory techniques that have been described previously. Briefly, serum samples were tested for IgA tissue transglutaminase (tTG IgA) with an enzyme-linked immunosorbent assay that uses human recombinant tTG manufactured by Inova Diagnostics, San Diego, CA. If tTG IgA titers were ≥ 4.0 U/ml, serum samples were tested for IgA Endomysial antibodies by indirect immunofluorescence using the reticulin component of the endomysium of the smooth muscle in monkey esophagus tissue (Inova Diagnostics). The result was considered positive if fluorescence was observed at a dilution of at least 1:5 as previously reported by our group and many others (specificity, 99–100%) (refs.). Serological diagnosis was based on positive tTG IgA with confirmatory IgA Endomysial antibodies analysis.
Data were analyzed using Stata version 11 software (College Station, TX) according to the National Center for Health Statistics analytic guidelines. We used appropriate study design and published weights for all analyses. All estimates of prevalence of CD were weighted so as to represent the total US population and to account for oversampling and nonparticipation in the household interview and physical examination. The prevalence of CD or PWAG was estimated in the total population, as well as by race (whites, blacks, and Hispanics), age, gender, and selected demographic factors. Because testing of NHANES III and NHANES 1999–2004 samples included only participants aged 50 years and older, trend analyses of seroprevalence of CD were limited to that age group. The seroprevalence of CD in the three study periods—i.e., 1988–1994, 1999–2004, and 2009–2012—were compared using the χ test. P values<0.05 were considered statistically significant.
Methods
NHANES Survey Background
The NHANES is conducted by the National Center for Health Statistics of the Center for Disease Control and Prevention. It collates nationally representative data on the health and nutritional status of the noninstitutionalized, civilian population of the United States. It utilizes a complex, stratified, and multistage probability sampling design and collects information from ~5,000 persons per year using standardized household interviews, physical examinations, and testing of biologic samples. More detailed information on the survey design for the NHANES, including approval from the institutional review board for data collection and analysis, is available from the survey documentation.
NHANES III was conducted from 1988 to 1994. Beginning in 1999, NHANES was redesigned to become a continuous survey; usually data from 2 or more years are combined to achieve adequate sample sizes for statistical analyses. A questionnaire covering only nonsensitive topics was used to interview participants at home to collect demographic data on age, sex, ethnicity, education, poverty index, and so on. Physical examinations and blood collection were also conducted.
The Prevalence of CD and PWAG in NHANES 2009–2012
For this analysis, data on 14,701 participants who underwent CD testing in the NHANES 2009–2012 period were used. There was no difference in participation in CD testing by sex (P=0.10), but participation was lower among non-Hispanic blacks (83%) compared with non-Hispanic whites (89%) and Hispanics (88%; P<0.001). Participation was also lower in the 6–8-year age group (81%) than in the >20-year age group (89%; P<0.001). To estimate the prevalence of CD in participants aged 6 years or older in NHANES 2009–2012, CD was defined by serological diagnosis or a reported clinical diagnosis of CD. A reported clinical diagnosis of CD was defined by positive responses to two CD-related questions: (i) Has a doctor or other health professional ever told you that you have CD, also called sprue? And (ii) are you on a gluten-free diet? To estimate the prevalence of PWAG in participants aged 6 years or older in NHANES 2009–2012, PWAG was defined by two CD-related questions and a CD serology test; the test was considered "positive" if the individual adhered to a GFD with a concomitant absence of a physician-given diagnosis of CD and/or negative on serologic testing.
The Trends in the Prevalence of CD Between 1988 and 2012
As CD serology testing in NHANES III (1988–1994) and NHANES 1999–2004 was limited to adults aged 50 years and older because of the NHANES regulations, data on adults aged 50 years or older with CD serology testing from NHANES III or continuous NHANES (1999–2004, 2009–2012) were used for describing the secular trends in the prevalence of CD between 1998 and 2012. Of 19,864 adults aged ≥50 years enrolled in NHANES III, NHANES 1999–2004, or NHANES 2009–2012, serum samples of 17,010 individuals (9,122 non-Hispanic whites, 3,276 non-Hispanic blacks, and 3,928 Hispanics) were tested for CD.
Laboratory Testing
Serum specimens of NHANES 2009–2012 and stored serum samples of NHANES III and NHANES 1999–2004 were shipped to the Celiac Disease Research Laboratory at Mayo Clinic, Rochester, MN, USA, which performed serological testing using laboratory techniques that have been described previously. Briefly, serum samples were tested for IgA tissue transglutaminase (tTG IgA) with an enzyme-linked immunosorbent assay that uses human recombinant tTG manufactured by Inova Diagnostics, San Diego, CA. If tTG IgA titers were ≥ 4.0 U/ml, serum samples were tested for IgA Endomysial antibodies by indirect immunofluorescence using the reticulin component of the endomysium of the smooth muscle in monkey esophagus tissue (Inova Diagnostics). The result was considered positive if fluorescence was observed at a dilution of at least 1:5 as previously reported by our group and many others (specificity, 99–100%) (refs.). Serological diagnosis was based on positive tTG IgA with confirmatory IgA Endomysial antibodies analysis.
Statistical Analyses
Data were analyzed using Stata version 11 software (College Station, TX) according to the National Center for Health Statistics analytic guidelines. We used appropriate study design and published weights for all analyses. All estimates of prevalence of CD were weighted so as to represent the total US population and to account for oversampling and nonparticipation in the household interview and physical examination. The prevalence of CD or PWAG was estimated in the total population, as well as by race (whites, blacks, and Hispanics), age, gender, and selected demographic factors. Because testing of NHANES III and NHANES 1999–2004 samples included only participants aged 50 years and older, trend analyses of seroprevalence of CD were limited to that age group. The seroprevalence of CD in the three study periods—i.e., 1988–1994, 1999–2004, and 2009–2012—were compared using the χ test. P values<0.05 were considered statistically significant.
