The prevalence of autoimmunity, characterized as the immune system responding to non-threatening self-antigens, was recently shown in a nationally representative sample of U.S. adolescents to have increased substantially over the last 25 years (1). Drivers of this trend remain unknown (1). Developing immune systems in children and adolescents require education through immune challenges to function properly later in life (2). The hygiene hypothesis posits that improved hygiene and clean environments in childhood lead to declining exposure to infections and other historically common immune challenges and to improper immune responses to non-threatening exposures, such as allergens later in life (3–6). Indeed, asthma and allergy are commonly used as indicators for inadequate immune challenge in childhood (7) and asthma prevalence in adolescents has also steadily increased over time while the prevalence of common infection indicators has decreased (8, 9).

Immune mechanisms underlying the hygiene hypothesis are not fully-established but include the possibility that reduced microbial exposures in early life may prevent the activation of regulatory T-cells, thereby enabling both the development of allergic diseases and activation of autoimmune pathways (4). To elucidate whether autoimmunity, measured by antinuclear antibodies (ANA), may be an additional indicator of inadequate immune system education in childhood and adolescence, we explored age and birth cohort associations of common hygiene hypothesis indicators, including asthma, allergy, and persistent infections, with the presence of ANA in a representative sample of U.S. adolescents from the National Health and Nutrition Examination Survey (NHANES) ( Figure 1 ).

We hypothesized that an increase in allergy and asthma prevalence seen over 25 years in the U.S. would be correlated with the increase in ANA prevalence observed in adolescents. Further, we hypothesized that lower prevalence of common types of infections, would be inversely associated with ANA prevalence.

The study population was all adolescents, aged 12-19 years, from four cross-sectional NHANES cycles for which ANA data were available (overall sample, N=2,709): 1988-1991 (N=676), 1999-2000 (N=609), 2003-2004 (N=581) and 2011-2012 (N=843). The 2001-2002 cycle was not included as no individuals aged 12-19 were tested for ANA. NHANES is a representative survey of the non-institutionalized U.S. population. Provided subsample specific weights were used to adjust for nonresponse and the probability of selection into each ANA subsample (10). The NHANES protocol was approved by the National Center for Health Statistics (NCHS) Ethics Review Board of the Centers for Disease Control and Prevention (CDC). All participants or guardians gave written informed consent. Laboratory assays were conducted previously as part of standardized protocols available at: https://wwwn.cdc.gov/nchs/nhanes/Search/DataPage.aspx?Component=Laboratory.

ANA prevalence has steadily increased in U.S. adolescents over a recent 25-year span (1). Adolescence is a time of immense change in physical growth and development. In particular, immune system development extends from the prenatal period through adolescence (12), and immune competence continually builds as neoantigens are encountered (13). Due to the pace of growth and hormonal shifts, late childhood and adolescence is a sensitive period where adolescents are particularly vulnerable to environmental exposures (13). The observed increase in ANA among adolescents, a 200% increase in the odds of ANA in 2011-12 compared to 1988-1991, is alarming and may represent temporal shifts in environmental exposures for this age range. We investigated whether the increase in adolescent ANA prevalence correlated with other factors associated with improved hygiene and cleaner environments in childhood, consistent with the hygiene hypothesis. Overall, our findings were mixed. Increases in ANA prevalence over time did not appear to be explained by asthma, allergy, or the infections that we studied.

However, we did observe a positive association between asthma diagnosis before 6 years and ANA positivity. It is possible that the airway inflammation (14) characteristic of this asthma phenotype and ANA may have a common underlying cause, including early life exposures. The co-occurrence of ANA and early-onset asthma may become more common in U.S. adolescents, though additional data from more proximal time periods is needed to monitor for a potential trend. Understanding the pathophysiological pathways that may link early-onset asthma and biomarkers of autoimmunity will be an important area of future research.

Though we did not observe consistent inverse relationships between the infections examined and ANA, as would be expected under the hygiene hypothesis, the data did suggest an inverse association between H. pylori and ANA positivity. H. pylori was previously shown to be protective for early-onset (less than 5 years) childhood asthma in NHANES (15). H. pylori reduces host inflammatory responses, even early in life, which in animal models have been shown to provide protective effects against allergic airway inflammation in first and second generation offspring (16–18). As with other persistent pathogens, the early life benefits of H. pylori infection may wane and undermine the mature adult immune system, resulting in detrimental outcomes (19–21). Indeed, a recent study of H. pylori in U.S. adults aged 20+ years found that H. pylori seropositivity was associated with higher odds of ANA positivity, suggesting that chronic inflammation resulting from H. pylori may impact immunological regulation (22). Additional research is needed to understand the relationship of persistent pathogens, such as H. pylori, and immune self-tolerance over the life course.

Elevated odds of ANA were observed for adolescents with atopic asthma in one time period (2003-2004), but these results were not replicated in other periods. Overall, atopic asthma was rare in adolescents (2.8%), so analyses were underpowered. Previous research on atopic asthma and autoimmune disease, in part motivated by the idea of antagonism between allergic and autoimmune responses, has shown mixed results spanning positive, negative, and no associations, with heterogenous definitions of autoimmune disease and atopy (23). Cross-regulatory T-helper Type 1 (TH1) versus Type 2 (TH2) immune responses have been hypothesized to lead to negative relationships or antagonism between allergic and autoimmune diseases (24, 25). According to the contemporary hygiene hypothesis, cleaner environments and less exposure to infections provide inadequate education of the immune system, resulting in higher likelihood of allergic and autoimmune disease. Thus, we would expect the occurrence of atopic asthma and autoimmunity to be correlated (26). Whether the co-occurrence is due to the same underlying dysregulation of immunological processes is an area of active research. Studies investigating atopic dermatitis in children have shown increased risk for autoimmune diseases (27). Studies have found positive associations between IgE autoantibodies and atopic dermatitis in children (26), and that children with atopic dermatitis develop ANA earlier than children without atopic dermatitis (27), suggesting potential for an underlying common cause including autoreactive T-cells and molecular mimicry (26). Further research on atopic asthma and autoimmunity is needed to understand common underlying mechanisms.

Our study has several limitations. We were limited to self-reported allergy, asthma and atopy information because NHANES did not link to participant medical records. Further, we do not have information on family history of autoimmune disorders because this was not asked in the NHANES surveys. Some of the conditions and infections examined had low prevalence, reducing power and precision of statistical analyses. Other infections examined in studies of the hygiene hypothesis, such as Hepatitis A (28), were not included due to vaccine licensure and varying vaccine updates during the study period. Previous studies have shown that ANA are more commonly detected during infections, and though autoimmunity may have a positive role during certain infections (29), in our study, ANA were not associated with self-reported infection (head or chest cold) in the last four weeks. The serological data in NHANES are cross-sectional so we cannot determine the temporal relationships of infections with ANA incidence. Longitudinal studies are warranted to better understand the age at onset and persistence of ANA, particularly in existing pediatric cohorts including biomarkers related to allergy and asthma phenotype (30). Detection of ANA by immunofluorescence identifies a wide range of autoantibodies and staining patterns. While we did not have data on disease-specific autoantibodies, these are uncommon in the general population (<10%) (31). The clinical interpretation of specific patterns is an important area of ongoing research. There is little known about ANA patterns in children and the natural history of ANA over time as they age into adulthood, though studies suggest varying factors, such as genetic variability and immune phenotype may be related to risk of developing autoimmune diseases (32). Investigation of specific cellular patterns with atopy is an important area for future research.

Despite these limitations, our study has strengths. We used nationally representative data spanning 25 years to understand increasing prevalence of ANA in U.S. adolescents. We contrasted ANA and serostatus to several infections whereas previous studies only looked at the association between ANA and infections among ANA positive individuals (33). NHANES is designed to examine national trends in health over time and the infection assays are of high quality and comparable over time. Further, we were able to control for several important variables in examining drivers of hygiene hypothesis exposures, including age, sex, race/ethnicity, BMI, and education. In sensitivity analyses, we additionally controlled for country of birth, smoking status and household size, as these factors may influence exposure to hygiene hypothesis indicators, but the addition of these potential confounders did not meaningfully change effect estimates. Our previous work has shown that vitamin D deficiency was associated with ANA positivity in older US adults (aged 50+) (34). In the current study, however, no statistically significant differences in vitamin D levels by ANA status were observed, therefore it was not considered as a confounder. Our previous work (35) also found that some medications were associated with ANA in the general adult (18+) population, including bronchodilators in those ages 60 and older, but medication use and ANA in adolescents was not examined. A thorough examination of medication use goes beyond the scope of the current study, but is an important area of future research.

Though the results from our study were mixed, ANA may be an additional indicator of inadequate immune system education in childhood and adolescence. The immunology of the hygiene hypothesis is complex with multiple mechanisms potentially driving associations. Proper immune maturation requires antigenically rich environments; the greater frequency of self-tolerance loss in younger age groups observed in our study may be the result of reduced microbiome diversity, poor immunologic education, and molecular mimicry (29, 36). Characterizing more recent birth cohorts by ANA status, as well as allergy, atopic asthma, and infections will be an important next step in determining if trends observed in this study are maintained or diminished. In addition, as the birth cohorts examined in this study enter mid-life, monitoring for increases in autoimmune disease will shed light on the connection between early appearance of autoimmunity markers and autoimmune disease risk. Early ANA occurrence may also be a sentinel for poor immunological regulation and cellular aging is associated with ANA incidence, therefore, future studies should investigate whether children with ANA have more aged immune systems than their chronological age may represent.

Publicly available datasets were analyzed in this study. This data can be found here: https://www.cdc.gov/nchs/nhanes/index.htm.

The studies involving human participants were reviewed and approved by National Center for Health Statistics (NCHS) Ethics Review Board of the Centers for Disease Control and Prevention (CDC). Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

Conceived of and designed the study: HM, CP, and DS. Acquired the data and performed the analysis: JW. Drafted the manuscript: HM and JW. Consulted on analytic strategy and critically revised manuscript: CP, DS, FM, and GD. All authors contributed to the article and approved the submitted version.

This work was supported in part by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (Z01-ES049028) and contract HHSN273201600011C to Social & Scientific Systems, and in part by the Shaw Scientist Award from the Greater Milwaukee Foundation.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.