The impact of the Wenchuan earthquake on early puberty: a natural experiment

Background

Pubertal timing is of great interest to both health providers and the public. There’s evidence that early puberty is associated with a broad range of short- and long-term adverse health consequences, especially for adolescent girls (Stroud & Davila, 2016). The physical and emotional risks of early maturation include not only depression, anxiety, substance abuse and early sexual debut during adolescence (Mensah et al., 2013; Walvoord, 2010; Copeland et al., 2010), but also reproductive cancers and metabolic syndrome in later life (Day et al., 2015; Walvoord, 2010; Copeland et al., 2010). Pubertal timing is a complex trait affected by gene-environment interactions (Karapanou & Papadimitriou, 2010). A deeper understanding of the potential contributors to pubertal timing is essential due to a secular trend toward earlier pubertal onset in many countries (Papadimitriou, 2016; Sørensen et al., 2012).

In addition to chemical exposures and obesity, emotional stress is another culprit that could jump-start puberty earlier (Ellis & Garber, 2000; Rickard, Frankenhuis & Nettle, 2014; Sun et al., 2017). According to psychosocial acceleration theory, an evolutionary-biological perspective on human development, differences in environmental contexts tend to channel children into divergent developmental trajectories in pubertal timing (Belsky, Steinberg & Draper, 1991; Webster et al., 2014). The stresses caused by harsh environments regulate reproductive strategies more susceptible to enhance reproductive fitness and maximize the probability of leaving descents (Pesonen et al., 2008), including early pubertal timing (Belsky, 2012; Webster et al., 2014). There is considerable evidence that childhood abuse, family conflict, father absence, and parental mental disorder are associated with earlier puberty in girls (Ellis & Garber, 2000; Rickard, Frankenhuis & Nettle, 2014). A cohort study also linked household socioeconomic disadvantage to earlier pubertal timing in both boys and girls (Sun et al., 2017).

War is another harsh environmental stressor that could influence pubertal timing (Belsky, 2008). The 1934–1944 Helsinki Birth Cohort Study in Finland revealed that former Helsinki evacuated girls who were sent to live with foster families in Sweden and Denmark during World War II had earlier menarche than the non-separated cohort members who remained at home (Pesonen et al., 2008). Additionally, hurricane and earthquake are two of the most common deadly types of natural disasters that could risk the physical and mental health of the child survivors both short-term and long-term (Marsee, 2008; Zheng et al., 2012). Based on psychosocial acceleration theory, disaster related stressors could affect pubertal timing (Belsky, Steinberg & Draper, 1991; Webster et al., 2014). However, no study has investigated the association between the earthquake and early puberty.

The 8.0-magnitude Wenchuan earthquake that occurred on May 12, 2008, in Sichuan Province was the biggest and deadliest earthquake since the 1950 Tibet earthquake and 1976 Tangshan earthquake in China. This disaster impacted over 15 million people, rendered at least 4.8 million homeless, and left 69,197 people dead, 374,176 injured and 18,222 listed as missing (Fan et al., 2017). The disaster generated a natural experimental setting for measuring the impact of the earthquake for early puberty among the child survivors of the quake. We hypothesized that exposure to the Wenchuan earthquake was associated with an increased risk of early puberty, with the younger the age at exposure, the higher the risk of early puberty.

Methods

Results

A total of 8,883 adolescents from 29 schools in eight provinces were included in the analysis, of which 4,543 were boys (51.14%), and 4,340 were girls (48.86%) (Table 1). The mean age of the final pooled sample was 15.15 years (standard deviation [SD] =1.81). There were eight schools in Sichuan exposed to the Wenchuan earthquake, and 21 schools in other provinces were unexposed. In the final sample, there were 2,658 earthquake survivors and 6,225 adolescents who did not experience the Wenchuan earthquake. Nearly 14% of the participants were preschoolers and the rest, 86%, were schoolchildren when the Wenchuan earthquake struck.

The rate of early puberty was 2.55% for boys and 10.14% for girls (Table 2). As illustrated in Table 3, the rate of early oigarche was 4.9% for male earthquake survivors and 1.54% for boys who did not experience the Wenchuan earthquake. The RR of the association between earthquake exposure and early oigarche was 3.18 [95%CI [2.21–4.57], p < 0.001]. Similarly, Table 3 also presents the rate of early menarche (19.50% for exposure and 6.17% for non-exposure) and the RR between earthquake exposure and early menarche (3.16 [2.65–3.78], p < 0.001).

We conducted subgroup analysis for males and females respectively, given that the children’s age at the time of the earthquake may serve as an effect-measure modifier on the association between earthquake exposure and early puberty (Table 4). The earthquake exposure appeared to double the risk of early puberty (boys:adjusted RR = 1.90 [1.19–3.03], P < 0.050; girls: 2.22 [1.75–2.80], P < 0.050). As shown in Table 4, older age at the time of the earthquake was a protective factor that reduced the risk of early menarche more than 50% compared with younger age at the time of the Wenchuan earthquake (preschool girls: RR = 3.91 [1.31–11.72], p = 0.003; schoolgirls: RR = 2.09 [1.65–2.64], p < 0.001). The association was non-significant among adolescent boys who were exposed to the Wenchuan earthquake before age seven years (RR = 1.73 [0.43–6.91], p = 0.425), but significant among the surviving schoolboys (RR = 1.93 [1.18–3.16], p = 0.007)

Discussion

To our best knowledge, this is the first study to investigate the potential impacts of the Great earthquake on pubertal timing. The 2008 Wenchuan earthquake offers a unique opportunity to explore the effects of the earthquake on the quake survivors’ later lives. In this natural experiment, exposure to the Wenchuan earthquake doubled the risk ratio of early puberty (1.90 for boys, 2.22 for girls). What’s more, there may be effect modification by children’s age at earthquake exposure. Earthquake exposure predicted almost a fourfold increased risk ratio of early puberty in preschool girls, whereas the increase was about twofold in schoolgirls. Similarly, earthquake exposure doubled the risk ratio in schoolboys. However, the association of earthquake exposure with early puberty was not statistically significant in preschool boys. The main reason for the unanticipated result was the sample was too small (Unexposed = 50[Cases = 2]).

Factors that influence pubertal maturation have gained much attention recently because of the secular trend to earlier puberty (Karapanou & Papadimitriou, 2010). There are many determinants of early puberty, but none of these serve as potential confounders in the present study because the Wenchuan earthquake was apparently not affected by any of these determinants. The age at earthquake exposure might mediate the impact of earthquake exposure on early puberty as an intermediary step in the causal pathway. On the one hand, younger age at earthquake exposure could prolong the time of the earthquake shock action. Children who experienced the earthquake earlier, on the other hand, may suffer more from earthquake-related stress and trauma, which put them at higher risk for earlier puberty. We found that preschool students were more vulnerable to early puberty when they were exposed to the earthquake, which made a new contribution to psychosocial acceleration theory that the first five to seven years of life might be especially influential in shaping reproductive strategy.

In the opinion of evolutionary-mind thinkers, accelerated somatic development was adaptive under dangerous or unpredictable environments because this tactic increased the chances of dispersing genes across generations before dying (Belsky, 2012; Belsky, Steinberg & Draper, 1991). The adaptive developmental mechanism that enhances reproductive fitness may have negative consequences on individuals in modern society, including early menarche, early sexual debut, and risk-taking (Belsky, 2012; Ellis & Bjorklund, 2012).

Considerable research on adverse health consequences of disaster events including earthquakes and hurricanes has focused on mental health (Carroll & Frakt, 2017; Fan et al., 2017; Shultz & Galea, 2017). Our findings revealed the profound impacts of the earthquake on early puberty both in boys and girls. A better understanding of these impacts could inform post-disaster interventions, which are crucial to the well-being of survivors, especially children (Carroll & Frakt, 2017). Given that children take their cues from adults, parents/guardians/caretakers should act calmly and confidently when disasters occur (Carroll & Frakt, 2017). Pediatricians should screen the health conditions of the young survivors of disasters in the years to come. Policies that provide children with supportive environments and caring relationships are critical in rebuilding their lives after disasters.

The underlying biological mechanisms that earthquakes might influence the early puberty are still barely understood. The stressors, including early childhood adversity and earthquake exposure, can affect not only the hypothalamic-pituitary-adrenal (HPA) and hypothalamic-pituitary-gonadal (HPG) axes individually, but the cross-talk between the two axes, possibly mediating HPA-HPG coupling (Ruttle et al., 2015; Shirtcliff et al., 2015). The mechanism explains how stressors may trigger pubertal development. Animal studies also suggested that social environment could influence the HPG axis at the molecular level (Maruska & Fernald, 2011). However, the role of genomic changes regulating pubertal development remains unclear (Sun et al., 2017) and is a promising field (Lomniczi & Ojeda, 2016). Studies also suggested that major adversity in early childhood could impair the brain’s reward circuits processing threat and stress (Boyce, Sokolowski & Robinson, 2012; Hanson, Hariri & Williamson, 2015), which might influence the regulation of reproductive hormones, resulting in early-onset puberty (Boyce, Sokolowski & Robinson, 2012).

Conclusions

Exposure to the Wenchuan earthquake elevated the risk of early puberty in both boys and girls, and preschoolers are more vulnerable to earthquakes compared with schoolchildren. The findings suggest that the health of children, especially younger children, must remain a focus in the recovery from disasters.

Supplemental Information