Early Menarche and Psychopathological Symptoms in Young Chinese Women

Abstract

Objective:

To evaluate the psychopathological symptoms, suicide, and self-harming behaviors among students with early, on-time, and late menarche in high school and college and the association of early menarche with these disorders.

Methods:

The design consisted of a cross-sectional study of 5597 high school students and 2768 college students. Menarche age, suboptimal mental health status, anxiety, depression, suicide, and self-harming behaviors were obtained by self-report questionnaire.

Results:

In high school students, all the disorders occurred at significantly higher frequency in those with early menarche than in those with on-time and late menarche. In college students, only suboptimal mental health status, depression, and suicidal ideation happened at significantly higher frequency in the early menarche group than in the other two groups. The college group had a lower frequency of all the disorders than the high school group for all three groups of girls, that is, with early, on-time, or late menarche. In a multivariate logistic regression model, early menarche persisted as a risk factor for all the disorders after other factors were controlled.

Conclusions:

Psychopathological symptoms, suicide, and self-harming behaviors are more common in early menarche students than in on-time and late menarche students. The effects of early menarche on the disorders might dissipate over time. Early menarche might serve as a predictor for the disorders in Chinese girls.

Introduction

With the improvements in general health, nutrition, and other living conditions, pubertal timing has had a secular trend of declining in developed countries. The prevalence of early menarche has been increasing during the past years.^1

–5 In China, a dramatic and steadily increasing trend of early menarche was witnessed in both urban and rural girls.^6,7 Many mental disorders, such as depression, certain anxiety disorders and substance use disorders, increase in prevalence during adolescence.^8
–10 A total of 15%–25% of adolescents meet the diagnostic criteria of specific mental disorders.¹¹ Early puberty in girls has been identified as an important risk factor for various mental health problems, such as depression, eating disorder, anxiety, psychosomatic symptoms, suicide attempts, substance abuse, and delinquent behavior.^12

–18

In China, the popularization of adolescent sex education is relatively lower than in developed countries,^19
–21 and adolescent girls pay more attention to and are more bewildered by menarche and other pubertal events. Whether or not early menarche has a greater influence than on-time or late menarche on psychological symptoms is unknown, and there are no reports based on large sample size and across districts. This study looks into this question.

Because menarche is a particularly dramatic and shocking signal of physical development compared to other indicators of pubertal timing, we used age at menarche as an indicator of pubertal timing.²² We aimed to examine the links between early menarche and psychopathological symptoms, suicide, and self-harming behaviors in students of different grades and to evaluate to what extent early menarche affects those disorders and if the effects persist in young adulthood.

Materials and Methods

We used baseline data from a large school-based cohort from the National High Technology Research and Development Program (863 Program) in China. Stratified cluster sampling was conducted in eight representative districts of China. In each district, we selected one college, one urban and one rural junior high school, and one urban and one rural senior high school. Thus, eight urban and eight rural junior high schools, eight urban and eight rural senior high schools, and eight colleges participated in the study.

Sample

Two classes from grades 7, 8, 10, and 11 in each junior and senior high school and two classes from grades 1 and 2 (freshman and sophomore) in each college were randomly selected for the study. The study protocol was approved by the ethical review board of Anhui Medical University, and written consent was obtained from the presidents of the schools, the teachers, and the students. A total of 9,619 students participated in the study. Every student agreed to answer the questionnaires before the study; however, 6.2% of them didn't answer all the questions. Because the ratio was relatively low and the sample size was large, recruitment was not done. The students completed the questionnaires in their classrooms 1 hour after the teacher informed them about the consent form. A total of 9023 students aged 9–29 years from 40 schools submitted complete questionnaires. Exclusion criteria were no menarche at ≤14 years and being unsure about menstrual age. All data collection was carried out from March to June in 2008.

Measures

A self-report questionnaire was used to collect information on sociodemographic data, suboptimal mental health status, depression, anxiety, suicide, and self-harming behaviors. Menarche age was evaluated by self-retrospective review. Participants self-reported their birth date, sex, grade, registered residence (rural = 1, urban = 2), household structure (only child = 1, more than one child = 2), family economic status (above average = 1, medium = 2, lower average = 3), self-perceived body shape (low weight = 1, normal = 2, overweight = 3, obesity = 4), mother's education level (unknown = 1, primary school or below = 2, high school = 3, college degree or above = 4), and father's education level (unknown = 1, primary school or below = 2, high school = 3, college degree or above = 4).

Onset of menstruation was elicited by asking: How old were you when you had your first menstruation? Response alternatives were as follows: I have not yet experienced menstruation/before 10 years/between ages 10 and 11/between ages 11 and 12/between ages 12 and 13/between ages 13 and 14/after age 14. A majority of girls had experienced menarche around age 12 or 13.^17,23,24 In this study, 80.2% respondents had menarche between the ages of 11 and 14. We used the phrase, early onset of menstruation, for menarche that occurred earlier than in the majority of peers. Specifically, early onset refers to menarche before 11 years, and late onset refers to menarche after 14 years, including those who had not experienced menstruation after 14 years old.

We developed The Multidimensional Sub-health Questionnaire of Adolescents (MSQA),^25

–29 and we used the female student results on the MSQA as our basic sample. Suboptimal mental health status was measured using the psychological domain (α = 0.958) of the MSQA, which consists of 39 questions on three dimensions: emotional symptoms, behavioral symptoms, and social adaptation problems. Emotional symptoms were measured using the 17 short questions for students (e.g., Do you always feel nervous?). Behavioral symptoms were measured using 9 questions (e.g., Do you always have the impulse to damage something?). Social adaptation problems were measured using 13 questions (e.g., Do you always feel difficulty in adapting to school life?). The answers to all the questions fell into six categories, in accordance with the duration of each symptom (none or last <1 week/last ≥1 week/last ≥2 weeks/last ≥1 month/last ≥2 months/last ≥3 months). In the data analysis, 0 or 1 was assigned to each answer; 0 was assigned if the symptom duration time was <1 month (negative items), and 1 was assigned otherwise (positive items). Then, we calculated the total scores of each participant's positive items, and the 90th percentile of the score was used as the threshold of a corresponding psychopathology status or symptoms. We defined those whose score reached 8 as having a suboptimal mental health status. The 90th percentiles of the scores were 3, 1, and 4, respectively, for emotional symptoms, behavioral symptoms, and social adaptation problems.

The Zung Self-Rating Depression Scale (SDS) and Zung Self-Rating Anxiety Scale (SAS) are commonly used self-evaluation instruments for measuring depression and anxiety symptoms, respectively.^30,31 Both the SAS and the SDS have 20 questions related to the frequency of various symptoms. The standard total score is the sum of the scores of 20 items multiplied by 1.25. An SDS standard score ≥53 indicates the presence of depression, and an SAS standard score ≥50 indicates the presence of anxiety.

Suicidal behaviors were determined using three yes/no questions about suicidal ideation, suicidal plan, and attempted suicide in the previous year. The three questions, respectively, were: Have you seriously thought about suicide? Have you made any plan to implement suicide? Have you ever attempted suicide? All students who answered yes (one or more times) were judged as having suicidal behaviors.

Deliberate self-harm was elicited in a self-report by answering: Have you ever tried to injure yourself during the past year? The response options were no or yes. The criteria for self-harm were an act with a nonfatal outcome in which an individual deliberately did one or more of the following: self-cutting, self-hitting, pulling hair, striking head, self-pinching, self-clawing, self-biting, self-scalding. All students who answered yes (one or more times) to any such action were judged as being self-harming and those with four or more self-harming actions were judged as being multiply self-harming.

Data analyses

SPSS 10.0 software was used in data analyses (SPSS, Chicago, IL). The Pearson chi-square test was used for analysis of sociodemographic data of students with early menarche. Differences in psychopathological symptoms, suicide, and self-harming behaviors between high school and college students and among early, on-time, and late menarche groups were analyzed using the Pearson chi-square or Fisher's exact test. To evaluate the effects of early menarche, a multivariate logistic regression model using the forward stepwise method was applied to psychopathological symptoms, suicide and self-harming behaviors, after controlling for other possible risk factors (grade in school, registered residence, only child in a family, self-reported family economic status, self-perceived body shape, and parental education level). The level of significance was set at p < 0.05.

Results

Four hundred forty-three students (all in junior high school) who had no menarche before age 14 and 215 students who were uncertain about their menstrual age were excluded. The final sample was 8365 (92.7%) participants. Of the respondents, 1032 were in 7th grade, with a mean age of 12.72 years (standard deviation [SD] 0.78); 1392 were in 8th grade, with a mean age of 13.61 years (SD 0.76); 1553 were in 10th grade, with a mean age of 15.74 years (SD 0.71); 1620 were in 11th grade, with a mean age of 16.83 years (SD 0.63); 1353 were college freshmen, with a mean age of 18.94 years (SD 1.03); and 1415 were college sophomore, with a mean age of 19.91 years (SD 0.96).

Among 8365 participants, 561 (6.7%) had early menarche, 6712 (80.2%) had on-time menarche, and 1092 (13.1%) had late menarche. Early menarche occurred at significantly higher frequency in the following groups: junior high school, urban, only child, better family economic status, obesity, and better parental education level. These groups had the highest percentages compared with the other groups in the same category, and there was a gradually falling trend among the groups in each variable (Table 1). In evaluating the father's education level, 240 students did not know this mainly because their fathers died or they did not live with their fathers. In evaluating mother's education, 121 students did not know their mothers' education.

Table 1.

Description of Students with Early Menarche

	Full sample n	Menarche <11 years n (%)	Chi-square value	p value
Grade			101.077	0.000
Junior high school (grades 7 and 8)	2424	262 (10.8)
Senior high school (grades 10 and 11)	3173	189 (6.0)
College	2768	110 (4.0)
Registered residence			52.232	0.000
Urban	4899	410 (8.4)
Rural	3466	151 (4.4)
Only child in family			64.059	0.000
Yes	3960	357 (9.0)
No	4405	204 (4.6)
Self-reported family economic status			16.581	0.001
Above average	1236	109 (8.8)
Medium	5743	385 (6.7)
Below average	1386	67 (4.8)
Self-perceived body shape			68.773	0.000
Obese	219	36 (16.4)
Overweight	2205	196 (8.9)
Normal	4417	265 (6.0)
Low weight	1524	64 (4.2)
Father's education level			15.704	0.001
College degree or above	1190	104 (8.7)
High school	5784	370 (6.4)
Primary school or less	1151	63 (5.5)
Unknown	240	24 (10.0)
Mother's education level			30.719	0.000
College degree or above	901	83 (9.2)
High school	5478	392 (7.2)
Primary school or less	1865	77 (4.1)
Unknown	121	9 (7.4)
Total n	8365	561 (6.7)

Three dimensions of the MSQA are detailed. For high school students, the percentages of psychopathology in emotional symptoms were 20.6%, 15.7%, and 16.9%, respectively, for the early, on-time, and late menarche groups (p = 0.024); the percentages in behavioral symptoms were 27.7%, 19.3%, and 19.3%, respectively (p = 0.000); the percentages in social adaptation problems were 18.2%, 12.0%, and 12.7%, respectively (p = 0.001). For college students, the percentages of psychopathology in emotional symptoms were 17.3%, 9.4%, and 12.5%, respectively, for the early, on-time, and late menarche groups (p = 0.006); the percentages in behavioral symptoms were 18.2%, 9.8%, and 13.3%, respectively (p = 0.003); the percentages in social adaptation problems were 17.3%, 6.8%, and 7.0%, respectively (p = 0.000).

Almost all psychopathological symptoms, suicide, and self-harming behaviors occurred at a higher frequency in the early menarche group than in the other two groups (Table 2). For college students, the early menarche group had 1 student who attempted suicide, and the frequency of occurrence was lower than in the late menarche group, although there was no significant difference among the three groups. For high school students, those with early menarche had a statistically significantly higher frequency of all disorders than those with on-time and late menarche. For college students, the early menarche group had a significantly higher frequency of suboptimal mental health status, depression, and suicidal ideation than the other two groups. There was no difference in disorders between the on-time and late menarche groups.

Table 2.

Association Between Menarche and Psychopathology in High School and College Students

	Early menarche <11 years n (%)	On-time menarche 11–14 years n (%)	Late menarche >14 years n (%)	Chi-square value	p value
Full sample
High school students	451 (8.1)	4541 (81.1)	605 (10.8)
College students	110 (4.0)	2171 (78.4)	487 (17.6)
Suboptimal mental health status
High school	82 (18.2)	600 (13.2)	81 (13.4)	8.637	0.013
College	17 (15.5)	153 (7.0)	45 (9.2)	12.119	0.002
Chi-square value/p value	0.453/0.501	56.057/0.000	4.549/0.033
Depression symptoms
High school	196 (43.5)	1673 (36.8)	225 (37.2)	7.685	0.021
College	36 (32.7)	494 (22.8)	120 (24.6)	6.236	0.044
Chi-square value/p value	4.199/0.0400	133.335/0.000	19.661/0.000
Anxiety symptoms
High school	100 (22.2)	816 (18.0)	126 (20.8)	6.969	0.031
College	20 (18.2)	265 (12.2)	65 (13.3)	3.648	0.161
Chi-square value/p value	0.838/0.360	36.107/0.000	10.459/0.001
Suicidal ideation
High school	115 (25.5)	753 (16.6)	100 (16.5)	23.082	0.000
College	15 (13.6)	123 (5.7)	29 (6.0)	11.739	0.003
Chi-square value/p value	6.990/0.0080	154.249/0.000	28.959/0.000
Suicidal plan
High school	52 (11.5)	307 (6.8)	42 (6.9)	14.081	0.001
College	4 (3.6)	46 (2.1)	10 (2.1)	1.173	0.610
Chi-square value/p value	6.132/0.013	63.514/0.000	14.219/0.000
Attempted suicide
High school	30 (6.7)	154 (3.4)	22 (3.6)	12.307	0.002
College	1 (0.9)	20 (0.9)	7 (1.4)	1.070	0.350
Chi-square value/p value	5.587/0.018	35.489/0.000	5.047/0.025
Self-harm (≥1)
High school	135 (29.9)	929 (20.5)	114 (18.8)	24.149	0.000
College	17 (15.5)	189 (8.7)	45 (9.2)	5.804	0.055
Chi-square value/p value	9.385/0.002	146.134/0.000	19.999/0.000
Multiple self-harm (≥4)
High school	23 (5.1)	110 (2.4)	15 (2.5)	11.496	0.003
College	2 (1.8)	12 (0.6)	2 (0.4)	3.205	0.219
Chi-square value/p value	2.237/0.135	28.769/0.000	7.534/0.006

The college groups had a lower frequency of all psychopathological symptoms, suicide, and self-harming behaviors than the high school groups with either early, on-time, or late menarche (Table 2). In girls with on-time and late menarche, this difference was statistically significant. In girls with early menarche, only the frequencies of depression, suicidal ideation, suicidal plan, attempted suicide, and self-harm differed significantly between high school and college groups.

In binary logistic regression analysis, the time of menarche was set as an independent variable and on-time menarche was an indicator. Suboptimal mental health status, anxiety, depression, suicidal ideation, suicidal plan, attempted suicide, self-harm, and multiple self-harm were dependent variables. In these analyses, early menarche was a risk factor for all the disorders. In multivariate analysis, grade, registered residence, only child in a family, self-reported family economic status, self-perceived body shape, and parental education level were added to the model stepwise to ascertain if early menarche persisted as a risk factor. After those factors were controlled, it was found that early menarche did persist as a risk factor for all the disorders (Table 3).

Table 3.

Effects of Early Menarche on Psychopathology, Suicide, and Self-Harming Behaviors

	Early menarche (<11 years)
Psychopathology	Crude OR (95% CI)	Adjusted OR (95% CI)
Suboptimal mental health	1.696 (1.347-2.134)	1.529 (1.210-1.931)
Depression symptoms	1.479 (1.241-1.763)	1.392 (1.164-1.665)
Anxiety symptoms	1.417 (1.147-1.752)	1.420 (1.146-1.760)
Suicidal ideation	2.009 (1.631-2.475)	1.709 (1.380-2.116)
Suicidal plan	1.998 (1.486-2.686)	1.713 (1.267-2.315)
Attempted suicide	2.198 (1.485-3.253)	1.804 (1.213-2.685)
Self-harm (≥1)	1.860 (1.527-2.264)	1.646 (1.347-2.011)
Multiple self-harm (≥4)	2.519 (1.624-3.909)	2.165 (1.388-3.377)

All analyses adjusted for grade, registered residence, only child in a family, self-reported family economic status, self-perceived body shape, and parental education level.

CI, confidence interval; OR, odds ratio.

Discussion

In this study, high school students with early menarche had a significantly higher frequency of all psychopathological symptoms, suicide, and self-harming behaviors than those with on-time and late menarche. College students in the early menarche group had a significantly higher frequency than the other two groups only in suboptimal mental health status, depression, and suicidal ideation. The college students had a lower frequency than the high school group of all disorders in girls with early, on-time, and late menarche. In the early menarche group, however, the lower frequency was not significant for suboptimal mental health status, anxiety, and multiple self-harm. In multivariate analysis, early menarche persisted as a risk factor for all disorders after other factors were controlled. Thus, in China as in other countries, adolescents with early menarche manifested a high frequency of subsequent psychopathological symptoms, suicide, and self-harming behaviors.

Many studies had examined the relationship between early menarche and increased mental health problems. Kaltiala-Heino et al.,¹⁷ in their study on the relationship between early menarche and emotional and behavioral problems in middle adolescence, found that the earlier the menarche, the more common the studied problems, including depression, bulimia nervosa, psychosomatic symptoms, anxiety, drinking, substance abuse, smoking, bullying, and truancy. Another study also confirmed the association between early puberty and eating disorder and anxiety, but early puberty was not associated with alcohol use, personality, and sensation seeking.³² Striegel-More et al.,³³ nevertheless, found that there were no differences in the relationship between time of menarche and symptoms of eating disorders. However, previous studies seldom focused on the relationship between early menarche and suicide and self-harming behaviors. In our study, we examined the association between menarche and suboptimal mental health status, anxiety, depression, suicide, and self-harming behaviors and found that the early menarche group had a higher frequency of almost all those disorders, whereas there was no difference between the on-time and late menarche groups.

In our study, the number of early menarche girls decreased from junior high school [262 (10.8)], to senior high school [189 (6.0)], to college [110 (4.0)]. Although we excluded some junior high school students ≤14 years without menarche, this exclusion did not influence the results because the absolute number of girls with early menarche among junior high school students was very high. This trend of decreasing may be associated with the declining trend of the age of menarche in China, and college students might have recall bias. At the same time, early maturing girls were more likely to report getting in trouble at school, absenteeism, and truancy.³⁴ They might have had less interest in academic subjects and were less likely to pursue college educations. Thus, the decreasing number might suggest the impact of early menarche on academic learning. Analysis of the father's education level showed that 240 students whose father died or did not live with them had the highest percentage (10%) of early menarche. This suggested that the father's absence might be associated with early puberty, which was also reported by another study.³⁵

In China, the only-child policy (aimed at the urban crowd) may cause such families to have socioeconomic advantages compared with multichildren families (parents in rural areas can have 2 children). Living conditions are better in urban areas than in rural ones. So the only child receives better nutrition from childhood, has a higher household income, and has a better living environment. A report on the physical fitness and health surveillance of Chinese school students every 5 years showed that the age of menarche is about 1 year earlier in economically developed areas than in undeveloped areas. The surveillance also showed that the age of menarche is about 1 year earlier in urban areas than in rural ones. Therefore, the only child might have earlier menarche, which is also the case in our data. Because the only child in a family receives excessive attention, it is easier for them to be self-centered and to lack social skills. Because an only child might have more psychological problems, this was set as a controlling variable.

In our study, girls in families with a higher economic status and girls who had parents with higher education experienced earlier menarche. Recent epidemiological evidence suggests that both lower parental educational attainment and lower household income have been linked to early puberty, but this phenomenon mainly exists in developed countries.^36,37 A comprehensive review published in 2003, however, reported inconsistent associations between socioeconomic circumstances and the onset of puberty: developed vs. developing countries showed particularly marked variations.³⁸ A recent study in the United States showed that parental education was not a significant predictor of early menarche. White girls in the highest quartile of household income were at a significantly lower risk of early menarche, whereas black girls in the highest quartile of household income were at an increased risk of early menarche.³⁹ Therefore, this association between socioeconomic status and menarche might be different for different racial/ethnic groups. Because components of socioeconomic status have different meanings in different racial/ethnic groups, our results may also be associated with the selected socioeconomic status factor (household income vs. parental education).

All psychopathological symptoms, suicide, and self-harming behaviors were more common in high school students than in college students. This may be associated with the removal of study pressure after they entered college, and college students might have more mental health resource services and better psychological adjustment ability. In the early menarche group, however, this decreased frequency from high school to college was not significant in suboptimal mental health status, anxiety, and multiple self-harm. There were significant associations between all the studied symptoms and early menarche in high school students, whereas in college students, only suboptimal mental health status, depression, and suicidal ideation had a significant association with early menarche. The effects of early menarche on some of these disorders may dissipate over time. Several studies determined if the timing of puberty continued to be associated with psychopathology. El-Khouri and Mellner,⁴⁰ for example, studied the relationship between self-reported somatic complaints (headaches, stomach problems, sleeping difficulties, and sadness) and timing of menarche cross-sectionally at ages 15 and 43 and longitudinally between the two ages. They found significant associations between all the included symptoms in adolescence and early menarcheal timing. However, no associations remained between menarcheal timing and symptoms at age 43. Another study reported that young women who had had early onset of menarche had higher rates of lifetime history of disorder, along with current elevation of psychosocial symptoms compared with women who were on-time maturers.⁴¹

Pubertal development involves biological, psychological, and social changes, all of which may contribute to mental health problems. In a study among adolescent girls, the association between early menarche and mental distress was weakened after socioeconomic and behavioral factors were controlled.⁴² However, we found in our study that early menarche was more common in the following groups: junior high school, urban, only child, better family economic status, obesity, and higher parental education level. In logistic regression analyses, early menarche was a risk factor for all these factors. After social and psychological factors were controlled for, early menarche persisted as a predictor for all the factors.

The limitations of this study are that the data were collected using a single self-report method. A single informant may cause inflated correlations or common method bias. Diagnostic interviews would be more reliable for detecting mental disorders. Physical examination could give a more accurate estimate of pubertal timing, and personal interviews might give a better understanding of an adolescent's suicidal behavior. Some steps were taken to increase the validity of these reports. Anonymous answering, multiple indicators of psychological functioning, and assessment of confounding variables may reduce the influence of those limitations. The most significant feature of this study is the large sample size, so we should select simple and specific methods to collect data. Menarche is a specific event in pubertal development, and age of menarche is easy to recall in a retrospective assessment of pubertal timing. Some of the mental health problems, such as suicide ideation and suicide plan, were assessed with single-item measures only. We used the Youth Risk Behavior Survey developed by the U.S. Centers for Disease Control and Prevention. These are consistent with basic epidemiological methods.

The present results are based on a large population sample. The sample is representative in that students from junior high schools, senior high schools, and colleges in eight municipalities of China participated in the study, and almost all the students who were at school on the survey day returned a response of acceptable quality.

Conclusions

Early pubertal timing is associated with psychopathological symptoms, suicide, and self-harming behaviors in adolescence and young adulthood that may lead to detrimental sequelae and even fatal outcomes. Professionals working with adolescents in health and social services and schools should pay attention to the psychological health needs of adolescents who mature earlier than the majority of their peers. Maturing earlier than peers seems to create additional stress in the process of adapting to the changes in one's own body and social role brought about by pubertal maturation. Early maturing adolescents might benefit from tailored health education and counseling.

Footnotes

Acknowledgments

This study was supported by the National High Technology Research and Development Program (863 Program) of China (grant 2006AA02Z427) and Natural Science Foundation of Anhui province (grant KJ2010A163). We thank all the students who kindly agreed to take part in the study and all the health workers who took part in data collection.

Disclosure Statement

The authors have no conflicts of interest to report.

References

Herman-Giddens

, Slora

, Wasserman

et al. Secondary sexual characteristics and menses in young girls seen in office practice: A study from the Pediatric Research in Office Settings Network. Pediatrics, 1997; 99:505–512.

Herman-Giddens

, Kaplowitz

, Wasserman

. Navigating the recent articles on girls' puberty in Pediatrics: What do we know and where do we go from here? Pediatrics, 2004; 113:911–917.

Chumlea

, Schubert

, Roche

et al. Age at menarche and racial comparisons in U.S. girls. Pediatrics, 2003; 111:110–113.

Cabanes

, Ascunce

, Vidal

et al. Decline in age at menarche among Spanish women born from 1925 to 1962. BMC Public Health, 2009; 4:449.

Harris

, Prior

, Koehoorn

. Age at menarche in the Canadian population: Secular trends and relationship to adulthood BMI. J Adolesc Health, 2008; 43:548–554.

Chen

, Shu

, Xiong

et al. Survey on age of menarche in 56, 924 women recruited from Pudong district of Shanghai. Chin J Obstet Gynecol, 2009; 44:500–503.

Chen

, Ji

. Geographic distribution and secular shift of menarcheal age of Chinese girls in provincial capitals. Chin Prev Med, 2003; 4:264–266.

Angold

, Costello

, Worthman

. Puberty and depression: The roles of age, pubertal status and pubertal timing. Psychol Med, 1998; 28:51–61.

Feehan

, McGee

, Raja

, Williams

. DSM-III-R disorders in New Zealand 18-year-olds. Aust NZ J Psychiatry, 1994; 28:87–99.

10.

Perkonigg

, Pfister

, Hofler

et al. Substance use and substance use disorders in a community sample of adolescents and young adults: Incidence, age effects and patterns of use. Eur Addict Res, 2006; 12:187–196.

11.

Newman

, Moffitt

, Caspi

, Magdol

, Silva

, Stanton

. Psychiatric disorder in a birth cohort of young adults: Prevalence, comorbidity, clinical significance, and new case incidence from ages 11 to 21. J Consult Clin Psychol, 1996; 64:552–562.

12.

Lanza

, Collins

. Pubertal timing and the onset of substance use in females during early adolescence. Prev Sci, 2002; 3:69–82.

13.

Costello

, Sung

, Worthman

, Angold

. Pubertal maturation and the development of alcohol use and abuse. Drug Alcohol Depend, 2007; 88:S50–59.

14.

, Kim

, Brody

et al. It's about timing and change: Pubertal transition effects on symptoms of major depression among African American youths. Dev Psychol, 2003; 39:430–439.

15.

Glowinski

, Madden

, Bucholz

, Lynskey

, Heath

. Genetic epidemiology of self-reported lifetime DSM-IV major depressive disorder in a population-based twin sample of female adolescents. J Child Psychol Psychiatry, 2003; 44:988–996.

16.

Stice

, Presnell

, Bearman

. Relation of early menarche to depression, eating disorders, substance abuse, and comorbid psychopathology among adolescent girls. Dev Psychol, 2001; 37:608–619.

17.

Kaltiala-Heino

, Marttunen

, Rantanen

, Rimpelä

. Early puberty is associated with mental health problems in middle adolescence. Soc Sci Med, 2003; 57:1055–1064.

18.

Lien

, Dalgard

, Heyerdahl

, Thoresen

, Bjertness

. The relationship between age of menarche and mental distress in Norwegian adolescent girls and girls from different immigrant groups in Norway: Results from an urban city cross-sectional survey. Soc Sci Med, 2006; 63:285–295.

19.

Wang

. Professional development: The pursuit path of pubertal education. Soc Sci Nanjing, 2009; 4:102–106.

20.

Min

, Wang

. The enlightenment of international adolescent sex education and development trends to China. Educ Sci Res, 2001; 11:56–59.

21.

Luo

, Gao

. Sexual knowledge level studies in primary and secondary school students. Educ Res Exp, 2000; 3:50–54.

22.

, Conger

, Elder

Jr . Pubertal transition, stressful life events, and the emergence of gender differences in adolescent depressive symptoms. Dev Psychol, 2001; 37:404–417.

23.

Kaltiala-Heino

, Kosunen

, Rimpelä

. Pubertal timing, sexual behaviour and self-reported depression in middle adolescence. J Adolesc, 2003; 26:531–545.

24.

Hou

, Li

, Sun

, Xia

, Chen

, Mi

. Update pubertal development in Beijing school-aged girls. Chin J Evidence-based Pediatr, 2006; 1:264–268.

25.

Tao

, Hu

, Sun

, Hao

. The development and application of Multidimensional Sub-health Questionnaire of Adolescents (MSQA) Chin J Dis Control Prev, 2008; 12:309–314.

26.

, Tao

, Hu

et al. Study on of Multidimensional Sub-health Questionnaire of Adolescents. Chin J Public Health, 2008; 24:1025–1028.

27.

Xing

, Tao

, Yuan

et al. Evaluation of reliability and validity of the Multidimensional Sub-health Questionnaire of Adolescents. Chin J Public Health, 2008; 24:1031–1033.

28.

Wan

, Hu

, Tao

, Qi

, Xing

, Yuan

. Research on responsiveness of Multidimensional Sub-health Questionnaire of Adolescents. Chin J Public Health, 2008; 24:1035–1036.

29.

Wan

, Tao

, Hu

, Yuan

, Qi

, Xing

. Predictive value of adolescent sub-health for suicidal psychology and behaviors. Chin J Sch Health, 2009; 30:295–297.

30.

Dugan

, McDonald

, Passik

, Rosenfeld

, Theobald

, Edgerton

. Use of the Zung Self-Rating Depression Scale in cancer patients: Feasibility as a screening tool. Psycho-oncology, 1998; 7:483–493.

31.

Zung

. A rating instrument for anxiety disorders. Psychosomatics, 1971; 12:371–379.

32.

Zehr

, Culbert

, Sisk

, Klump

. An association of early puberty with disordered eating and anxiety in a population of undergraduate women and men. Horm Behav, 2007; 52:427–435.

33.

Striegel-Moore

, McMahon

, Biro

, Schreiber

, Crawford

, Voorhees

. Exploring the relationship between timing of menarche and eating disorder symptoms in black and white adolescent girls. Int J Eat Disord, 2001; 30:421–433.

34.

Graber

, Lewinsohn

, Seeley

, Brooks-Gunn

. Is psychopathology associated with the timing of pubertal development? J Am Acad Child Adolesc Psychiatry, 1997; 36:1768–1776.

35.

Bogaert

. Age at puberty and father absence in a national probability sample. J Adolesc, 2005; 28:541–546.

36.

Lee

, Appugliese

, Kaciroti

, Corwyn

, Bradley

, Lumeng

. Weight status in young girls and the onset of puberty. Pediatrics, 2007; 119:e624–630.

37.

Ellis

, Essex

. Family environments, adrenarche, and sexual maturation: A longitudinal test of a life history model. Child Dev, 2007; 78:1799–1817.

38.

Parent

, Teilmann

, Juul

, Skakkebaek

, Toppari

, Bourguignon

. The timing of normal puberty and the age limits of sexual precocity: Variations around the world, secular trends, and changes after migration. Endocr Rev, 2003; 24:668–693.

39.

Braithwaite

, Moore

, Lustig

et al. Socioeconomic status in relation to early menarche among black and white girls. Cancer Causes Control, 2009; 20:713–720.

40.

El-Khouri

, Mellner

. Symptom development and timing of menarche: A longitudinal study. Int J Methods Psychiatr Res, 2004; 13:40–53.

41.

Graber

, Seeley

, Brooks-Gunn

, Lewinsohn

. Is pubertal timing associated with psychopathology in young adulthood? J Am Acad Child Adolesc Psychiatry, 2004; 43:718–726.

42.

Lien

, Dalgard

, Heyerdahl

, Thoresen

, Bjertness