In this large sample of the US population, we quantitatively assessed the exposure-response association of age at menarche with risk of childhood- and adult-onset asthma separately. A novel effect of early menarche was only observed on childhood-onset asthma. Each one-year increase in age at menarche was associated with 17% decrease in risk of childhood asthma. There was about 60% increased risk of childhood-onset asthma for early menarche and 41% decreased risk for late menarche when compared to those with normal menarche. None of race, family income, or family history of asthma could modify the associations.
The link between age at menarche and asthma onset appears to be equivocal. Studies from Germany and Swiss did not observe any effect of menarche on asthma among young populations 21, 25. In contrast, a 60%, 108% and 134% increase in risk of asthma onset were reported in British, US and Canadian girls who had early menarche, respectively.16, 18, 19 Notably, the study from British reported that the girls with early menarche had more asthma at all ages, but the association reached statistical significance only among girls age 11 years at the time of follow-up.18 Similar results were also observed in a Sweden study of 2492 adults aged 18-60 years,20 but inconsistent with the findings from a recently published study which included 0.24 million women aged 40-69 years based on UK biobank data and reported 8% increase in the risk of asthma for early menarche and 8% decrease for late menarche by using mendelian randomization.15 One possible reason for the inconsistent results in these studies might be the different ages at menarche used as the reference category (including age of 12-13,18, 26 12-14,15 >12,16 ≥12,20, 22 and 12.6619 years). A meta-analysis helped solve the problem by combining the findings from of seven cohort studies with consistent definition of early menarche as age at menarche>12 years. Results of the meta-analysis study showed a 37% increase in risk of asthma for early menarche when comparing groups with late menarche in women aged from 7 to 57 years.27 Our findings confirmed and extended the adverse effect of early menarche on asthma onset in US general population. By further distinguishing the childhood- and adult-onset asthma in our study, we found the novel effect of early menarche only on childhood asthma rather than adult-onset asthma. Although there was no guideline for clear distinction between childhood and adult-onset asthma, several obvious differences such as symptoms and sensitivity to allergens were observed.28 The results from our study indicate potential differences in risk factors and pathogenesis of the two types of asthma.
The underlying mechanisms for the early menarche related asthma is still unclear. Menarche is an indicator of pubertal onset for women and early menarche is generally considered to be a sign of precocious puberty, companying with changes of sex hormone levels. A recent nationwide study of US adults reported the association between sex hormone levels and the risk of asthma,29 providing a possible reason for increased risk of asthma onset among participants with early menarche. Besides, other risk factors associated with precocious puberty such as obesity, depression, anxiety, and psychosomatic symptoms30, 31 have been demonstrated to participate in activating inflammatory mediators,32 which have been regarded as important role on asthma development.27, 33, 34
Our results have important public health implication. Asthma is one of the leading chronic diseases, affecting 339.4 million people and contributing to 420,000 deaths globally in 2016.35, 36 It was estimated that there were about 6.2 million children under the age of 18 with asthma.37 The present study confirms and extends previous research by quantifying the effect of age at menarche on the risks of childhood asthma, suggesting early menarche may be a risk factor for childhood-onset asthma in US. Timely and effective management of individuals with early menarche should be a part of the prevention for asthma.
There are several strengths in this study. To our knowledge, this is the first representative study to quantitatively estimate the association of age at menarche with asthma onset in a general population of US girls and women, covering all races in the US and those aged from 12 to 79 years. Furthermore, we identified childhood- and adult-onset asthma according to age at asthma onset, accounting for the effect of asthma phenotypes on the association.38 In addition, most previous studies used logistic regression models to assess the associations,18-20, 22, 39, 40 ignoring the effect of the event time. Due to the potential impact of age at menarche upon the time asthma takes to happen, it seems to be more appropriate to account for time scales in the analyses. Therefore, we performed Cox proportional-hazards regression models with age as the timescale to study the age at menarche that are significantly associated with the timing of the asthma onset.
The main limitation of this study is that the definitions of age at menarche and asthma were based on self-report rather than medical records, which might cause recall bias. Continuous available examination data or prospective study design should be considered to help increase the reliability in the further studies.