1. Background
The onset of menstruation and the age which the first menstrual period occurs is considered as the age at menarche, which is an important sign of the puberty in females and a health and socioeconomic status index of societies (1-3). Heredity is the most important factor for determining puberty and age at menarche, which is confirmed by studies that examine the relationship between age of puberty in mother and daughter or sisters (4). However, in addition to inheritance, other factors which have been different and sometimes contradictory in studies also affect the onset time and the progress of puberty. These factors include economic and social status, public health, location and geographical location, nutrition, exposure to light, mental status, chronic diseases, and environmental stimuli (5-7).
The results of studies have shown that during the last decades, the age of puberty and the occurrence of menarche has been reduced in various developed and developing societies, due to improved nutrition and hygiene and improved economic and social status (8, 9). Early age at menarche among females puts them at risk of accelerating skeletal maturity and shortening height in adulthood. Due to the rapid closure of epiphyseal plates in bones, early onset of sexual activity and social and psychological problems are also among other important complications of early puberty (6, 10, 11). According to the studies, early menstruation at an early age is a risk factor for cardiovascular problems, metabolic syndrome, and breast and ovarian cancer, and is associated with an increase in the prevalence of painful menstruation and adolescence pregnancy (1, 6, 10).
The results of some studies have indicated that obesity is one of the causes of early puberty and early menstruation in a person with no specific disease (12). However, in some studies, the effects of overweight and obesity in accelerating the onset of menarche and signs of puberty have been refused (1, 13). In the Sharifi et al.’s study (11) a significant inverse correlation was observed between age at menarche and body mass index (BMI), however, there wasn’t any significant correlation in the Demerath et al.’s study (14).
2. Objectives
Due to the importance of evidence about age at menarche, as an important health indicator for future planning, also, the increasing in weight and decreasing in age at menarche of adolescents in societies and different results of studies about the relationship between BMI and age at menarche, the present study was conducted to determine age at menarche and its relationship with BMI and demographic factors among 10 - 16-year-old students in the city of Qaen, Eastern Iran.
3. Methods
3.1. Study Design and Population
The present study was a cross-sectional study and study population was all girl students studying at elementary and secondary schools in the city of Qaen in East of Iran in 2016. The sample size was estimated based on the mean estimating formula (n = z2s2/d2) and results of the Sharifi et al.’s study (15); 400 girls were selected by multistage sampling. First, the city of Qaen was divided into five geographical regions (North, South, East, West, and Center), and then in each stratum from the list of girls’ schools, an elementary school, and a secondary school were selected randomly. Then in the selected schools, for each desired grade, one class was randomly chosen. All students in the class who remembered the time of menarche and their menstruation had started 6 months ago, were included in the study. The presence of any underlying and systemic diseases that affect the growth of the individual (such as chronic heart disease, diabetes, thyroid, and hormonal disorders), use of hormonal weight-reducing or increasing drugs, and the presence of diseases such as polycystic syndrome, because of the possibility of relation with obesity or effect on incidence of puberty symptoms, were among the exclusion criteria (4, 10).
3.2. Data Collection
Before gathering the data, the objectives of the study were described to students and their parents. It was also assured that all information would remain secret to the researchers and consent was obtained from the parents of students. The present study was approved by the Ethics Committee of Birjand University of Medical Sciences (ethics code: IR.BUMS.REC.1394.86). Data collection were done by interviewing, measuring height and weight, and completing a checklist including demographic questions (including age, studying grade, maternal age, birth order, family size and duration of sleep); questions about the age at menarche was conducted by two midwives in schools. A part of the checklist was due to measuring height and weight for calculating BMI using standard and identical tools including the German Seca portable digital scale along with a graded vertical bar for measuring weight and height, where the weight was measured in light clothing, without shoes, and accuracy of 100 g and height according to an upright position and bare-footed and accuracy of 0.5 cm.
3.3. Statistical Analysis
Finally, after gathering the data, they were transferred to SPSS (PASW Statistics for Windows, version 18.0, Chicago: SPSS Inc., USA), and in addition to representing descriptive statistics, they were analyzed by Pearson’s correlation coefficient, independent t-test, and multiple linear regression, at the level of α = 0.05.
4. Results
From 400 female students participating in this study, 34 (8.5%) were elementary school students and 366 (91.5%) were secondary school students, with the mean age of 13.86 ± 1.10 years old, and with the minimum age of 10 and the maximum age of 16 years. In this study, most of the students (29%) were the first child of the family and most of the families (31.8%) were five persons. On average, the duration of each student's sleep was 7.78 ± 1.82 hours per day and the mean maternal age was 42.63 ± 5.92 years.
The onset of menstruation and menarche for 22.8% of the students was in the spring, 43.3% in the summer, 22% in the autumn and 12% in the winter, and the mean age at menarche was 12.78 ± 1.23 years (95% confidence interval: 12.66 - 12.90). The mean BMI was 20.38 ± 2.72 kg/m2 (Table 1).
| Variable | Mean | Standard Deviation | Minimum | Maximum |
|---|---|---|---|---|
| Age at menarche | 12.78 | 1.23 | 9 | 16 |
| Body mass index (kg/m2) | 20.38 | 2.72 | 14.69 | 36.16 |
| Height (cm) | 157.84 | 6.75 | 132.5 | 176 |
| Weight (kg) | 50.75 | 7.13 | 35.1 | 75.5 |
Description of Age at Menarche, BMI, Height and Weight in Students
According to the results of Pearson’s correlation, there was an inverse and significant relationship between age at menarche and BMI (r = -0.123, P = 0.014). Thus, with the increase in BMI, the age at menarche decreased. Also, there was a positive significant relationship between the age at menarche and height, birth order, maternal age, family size, and duration of sleep; and with the increase of each of these variables, the age at menarche also increased, however, there was no significant relationship between age at menarche and weight (P = 0.86) (Table 2).
| Variable | Body mass index | Height | Weight | Birth Order | Maternal Age | Family Size | Sleep Duration |
|---|---|---|---|---|---|---|---|
| Correlation with age at menarche (r) | -0.123 | 0.195 | 0.009 | 0.132 | 0.232 | 0.177 | 0.156 |
| P value | 0.014 | 0.0001 | 0.864 | 0.008 | 0.0001 | 0.0001 | 0.002 |
The Correlation Between Age at Menarche and BMI, Height, Weight, Birth Order, Maternal Age, Family Size, and Sleep Duration in Students
The mean age at menarche was 11.29 ± 0.76 years in the elementary school students and 12.92 ± 1.17 years in the secondary school students; this difference was significant with independent t-test (P = 0.0001). To predict the age at menarche, using the variables studied in this study, multiple linear regression was performed. After entering the variables by forwarding method, regression coefficients were significant for the variables of height (P = 0.001), sleep duration (P = 0.012), and maternal age, (P = 0.0001), and these three variables could explain 9.6% of age at menarche variance (R2 = 0.096) (Table 3).
| Variable | Regression Coefficient (B) | t | P Value | Model |
|---|---|---|---|---|
| Constant (a) | 5.634 | 3.998 | 0.0001 | Dependent variable: Age at menarche; R2 = 0.096; F = 14.088; P = 0.0001 |
| Height (cm) | 0.03 | 3.453 | 0.001 | |
| Sleep duration (h) | 0.083 | 2.533 | 0.012 | |
| Maternal age (y) | 0.04 | 3.975 | 0.0001 |
Multiple Linear Regression to Estimate the Age at Menarche in Terms of Height, Sleep Duration, and Maternal Age in Students
5. Discussion
Based on the results of the present study, there was a significant inverse correlation between age at menarche and BMI, so that by increase of BMI, the age at menarche decreased, which was consistent with the results of studies by Farahmand et al., in Tehran (16), and Mohamad et al., in Kish Island (17). In the studies of other countries, Wronka in Poland (7), Cho et al., in South Korea (3), and Harris et al., in Canada (8) confirmed this inverse relationship. However, this relationship was not significant in the study by Keshavarzi and Azad-bar study in Kermanshah (2) and the Demerath et al., study in the United States (14).
The reason of this relationship is not completely clear, however, some studies have found it necessary to have a certain percentage of body fat or specific weight and height for menarche occurrence, and believe that thin people catch these necessary values later and have later menarche (18, 19). Other studies consider this as a result of hormonal changes in obese people (20, 21), and some have mentioned the role of childhood obesity in early menarche and obesity in adulthood (22, 23). Due to the importance of obesity and overweight as a major risk factor for non-communicable diseases and complications of early menarche as a risk factor for some diseases such as breast cancer and cardiovascular disease (24-26), planning for suitable interventions for prevention and treatment of obesity and overweight in childhood and adolescence is more necessary.
According to the results, the mean age at menarche in students of Qaen in eastern Iran was 12.78 years, which was similar to the menarche age of girls in Mashhad (27). Various studies in Iran have reported the mean age at menarche in Isfahan as 12.65 (28) and 12.6 in Tehran (16), which has been less than the present study. In the study of girls in Rasht (29) and Kish Island (17), the mean age at menarche was 12.99 and 12.85 years respectively, which is more than this study. According to the meta-analysis implemented by Bahrami et al. (30), which reported the mean age at menarche in Iran as 12.81 years, it can be said that the age at menarche in girls of Qaen is as the country average.
In the studies of different countries, the mean age at menarche has been 12.4 in European girls (31), 12.96 in Indonesia (9), 12.9 in Korea (3), and 12.54 in the United States (32). Given the many hereditary, geographical, cultural, and socioeconomic factors affecting the age at menarche, the observed differences can be justified, although different methods of data collection and implementation of studies can also affect these differences.
The concern about the age at menarche in different societies is its decreasing trend as a risk factor for breast cancer and cardiovascular disease (24-26). However, due to the lack of a similar study in the city of Qaen, it was not possible to compare the results and to investigate the trend of menarche age, however, the study of Mohammad et al., reported the mean age at menarche in the eastern regions of Iran in 1990 as 13.81 and 13.67 in 1999, which is more than the present study (33). The lower mean age at menarche in elementary school students, in comparison to secondary school students in the present study, can also be indicative of a decreasing trend in age at menarche. Studies in different parts of the world have confirmed the decline of age at menarche and have mentioned it as a result of changes in lifestyle, industrialization, and the improvement of the socioeconomic status of the societies and the promotion of nutrition and public health (3, 8, 14, 32, 33).
The results of this study showed that there was a significant positive correlation between age at menarche and height of students, which was also significant in the regression model, after adjusting for other variables, and with increasing age at menarche, the height also increased. It seems that the results of various studies in this area have been different with regard to the height measured before or after the menarche. In the studies of Farahmand et al., in Tehran (16), and Hossain et al., in Bangladesh (34), in which height was measured after the menarche, as in the present study, there was a positive correlation between age at menarche and height. However, in studies of Krzyzanowska et al. (23), and Freedman et al. (35), which the height was measured before reaching the menarche, this relationship was negative, and the taller girls had reached the menarche earlier. Although this association has not been significant in some studies (22, 36), it seems that girls who have reached the menarche age sooner and reach sexual maturity, skeletal maturation also occurs earlier in them and their maximum growth is seen earlier as the reason of the closure of epiphyseal growth plates (35, 37). Therefore, girls who have reached menarche earlier than their peers are more likely to be higher in adolescents; however, girls who later reach menarche will be higher due to more longitudinal growth in long bones (38, 39).
In the present study, there was no significant relationship between age at menarche and weight of students, which was consistent with the results of most studies in Iran (11, 16, 36). Kim et al. (38), in a study on Korean girls aged 10 to 18, concluded that girls who had reached menarche were significantly heavier than girls who had not reached menarche. In the Krzyzanowska et al.’s study on British girls (23), there was a negative relationship between age at menarche and childhood weight, and heavier girls had reached menarche sooner.
In the present study, the age at menarche had a significant relationship with the birth order of students and the family size, and with the increase in the birth order and the number of family members, the age at menarche was increased, which was similar to the study of Krzyzanowska et al., in the United Kingdom (23). This relation could be due to a lower social economic level, a crowded family, and less attention to children in more populous families.
According to the results, age at menarche was significantly and directly related to the maternal age. As the maternal age increased, the age at menarche also increased. In the Cho et al.’s study in South Korea (3), there was a direct relationship between maternal age and age at menarche. Due to limited articles in this field, the study of the validity and cause of this relationship requires further studies.
Another important factor for the students’ menarche age was the duration of their sleep in a day. Thus, age at menarche increases with sleep increasing. Murata and Araki (40) also reported a positive relationship between age at menarche and sleep duration in a study of Japanese girl students and mentioned the reduced the duration of sleep as one of the factors affecting the early menarche in societies. It seems that further studies, especially longitudinal ones, are needed in order to better understand the effect of the duration of sleep on age at menarche.
One of the most important limitations of this study is its cross-sectional nature, the self-reported age at menarche, and its recall bias. For further investigation of the factors affecting the age at the menarche and confirming the causal relationships, longitudinal studies are suggested in the future.
5.1. Conclusions
In conclusion, the findings showed that in this study, the mean age at menarche was 12.78 years which was as average as Iranian studies. There was a significant correlation between age at menarche and BMI, and lower age at menarche was related to higher BMI.
