


* Lifespan Health Research Center, Departments of Community Health and Pediatrics, Wright State University, School of Medicine, Dayton, Ohio
Pennsylvania State University College of Medicine Hershey Medical Center, Hershey, Pennsylvania
School of Public Health, University of Minnesota, Minneapolis, Minnesota
| ABSTRACT |
|---|
|
|
|---|
Objective. The objectives of this study were to estimate the distribution of age at menarche for all US girls and for non-Hispanic white, black, and Mexican American girls in the Third National Health and Nutrition Examination Survey and to test for racial differences.
Design. Menstrual status data were collected from 2510 girls aged 8.0 to 20.0 years. The Third National Health and Nutrition Examination Survey followed a complex, stratified, multistage probability cluster design. SUDAAN was used to calculate proportions of girls reaching menarche at an age. Ages at menarche were estimated by probit analysis at the ages at which 10%, 25%, 50%, 75%, and 90% of the girls attained menarche.
Results. Less than 10% of US girls start to menstruate before 11 years, and 90% of all US girls are menstruating by 13.75 years of age, with a median age of 12.43 years. This age at menarche is not significantly different (0.34 years earlier) than that reported for US girls in 1973. Age at menarche for non-Hispanic black girls was significantly earlier than that of white girls at 10%, 25%, and 50% of those who had attained menarche, whereas Mexican American girls were only significantly earlier than the white girls at 25%.
Conclusion. Overall, US girls are not gaining reproductive potential earlier than in the past. The age at menarche of non-Hispanic black girls is significantly earlier than that of non-Hispanic white and Mexican American girls.
Key Words: menarche age menstruation NHANES III race
Abbreviations: NHES, National Health Examination Survey NHANES, National Health and Nutrition Examination Survey
| INTRODUCTION |
|---|
|
|
|---|
National data for the timing of menarche for non-Hispanic white and non-Hispanic black girls was documented 30 years ago in the National Health Examination Survey (NHES).7 Age at menarche was not reported from the first and second National Health and Nutrition Examination Surveys (NHANES I and II)8,9 because the girls were 12 years and older at the time of the examinations. This age is not young enough to provide a good estimate for the population.10 The population of US girls today includes an appreciable proportion of Mexican Americans, but age at menarche data were not collected during the Hispanic National Health and Nutrition Examination Survey.11 More recently, race-specific mean ages at menarche were reported by Herman-Giddens et al1 from a large sample of white girls and black girls, but the upper age limit of these samples was "through 12 years of age." These results do not include any girls who would have attained menarche after 13 years old, effectively excluding 15% or more of all girls.7
Menarcheal data are available from the Third National Health and Nutrition Examination Survey (NHANES III).12 The aims of the present analysis were 1) to estimate the distribution of age at menarche from NHANES III for all US girls and separately for non-Hispanic white, black, and Mexican American girls and 2) to determine the differences in these estimates among these racial-ethnic groups. These results will provide clinically meaningful information regarding the timing of the onset of menstruation and its distribution among a national sample of non-Hispanic white, black, and Mexican American US girls.
| METHODS |
|---|
|
|
|---|
Menstrual status data were collected with 3 different questionnaires that were administered depending on the age of the girl at the time of the NHANES III examination.12 Separate questionnaires were used for females aged 2 months to 11 years, 10 to 16 years, and 17+ years. The questionnaire for girls 2 months to 11 years was completed by a parent or guardian. The relevant question on this form was "How old was (name) when her periods or menstrual cycles started, or havent they started yet?" Girls aged 10 to 16 years and those 17+ years completed the questionnaires by themselves. The relevant question on these 2 forms was "How old were you when your periods or menstrual cycles started?" If an age was given, this was also coded as a "yes" response. If the response was "havent started yet," this was coded as a "no" response. These "yes-no" data were used in the analysis instead of the ages provided by the participants.
These "yes-no" data were from 2510 girls aged 8.0 to 20.0 years and included 1518 girls who had attained menarche and 992 who had not. There were 427 non-Hispanic white girls who had attained menarche and 283 who had not, 576 non-Hispanic black girls who had attained menarche and 341 who had not, and 515 Mexican American girls who had attained menarche and 368 who had not. The sample sizes in the present study were larger than that required for reliable estimates for the total population.15
Data Analysis
The 3-month age groupings were calculated based on the chronological ages of the girls at the time of the examination. The percentage of girls in each 3-month age grouping with a "yes" response indicating that they had attained menarche was calculated for all girls and separately for the non-Hispanic white, black, and Mexican American girls. The midpoint of a 3-month age group was used as the representative age for that group; for example, the 3-month age group, 96 to 98 months, had a midpoint of 97 months. The percentage of girls in these age groups who reached menarche was computed using Proc Crosstab of SUDAAN (Research Triangle Institute, Research Triangle Park, NC)16 that adjusts for the individual sampling weights and the design effects in NHANES III.
These adjusted percentages were used in a SAS 8.2 probit analysis (SAS Institute, Cary, NC) to determine the age at menarche for all girls and for each race-specific group by estimating the age at which 10%, 25%, 50%, 75%, and 90% of the girls attained menarche.17 In the probit analysis, the proportion of girls who had reached menarche was transformed into "y", a normal equivalent deviate, ie, p =
-
y f(
)d
where y = a + bx, x is the age in months and "a" and "b" were the parameters to be estimated from the data. Age at menarche was estimated when p equaled 10%, 25%, 50%, 75%, and 90%. The ages at these values of p are equivalent to the corresponding percentiles of ages of menarche, eg, 10% = 10th percentiles age.
The fiducial limits of these estimates were computed in SAS using Fiellers theorem.18 The ranges for the fiducial limits indicate the level of precision for these estimated ages from the probit analysis. The more narrow the range, the higher the precision of the estimated age. Because comparisons were made among the 3 racial groups of girls, the fiducial limits were calculated at a 98.3% confidence level. The 3 sets of fiducial limits yielded a comparison-wise confidence level of 95%, ie, (1-
) = 0.983 and (1-
)3 = 0.95. Intervals were adjusted for multiple comparisons between races for each sex, stage and sexual maturity indicator to test for significant differences with an overall
of 0.05.
| RESULTS |
|---|
|
|
|---|
|
The median age at menarche was 12.06 years for the non-Hispanic black girls, 12.25 years for the Mexican American girls and 12.55 years for the non-Hispanic white girls. Ten percent of non-Hispanic black girls were menstruating at 10.52 years compared with a corresponding age of 10.81 years for the Mexican American girls and 11.32 years for the non-Hispanic white girls. Ninety percent of non-Hispanic black girls were menstruating by 13.60 years compared with a corresponding age of 13.69 years for the Mexican American girls and 13.78 years for the non-Hispanic white girls.
Some significant differences occurred in the ages where 10%, 25%, and 50% of the girls in each racial group had started menstruating (Table 1). At 10%, the age of the non-Hispanic black girls was significantly earlier than that of the white girls. At 25%, the ages of both the non-Hispanic black and Mexican American girls were significantly earlier than the age of the white girls. At the median or 50%, only the non-Hispanic black girls were significantly earlier than the white girls. Racial differences in ages at the higher percentages were not significant. These racial differences in the ages where 10%, 25%, and 50% of the girls had attained menarche are more clearly observed in Fig 1. The race-specific plots of the probit analyses demonstrate that the distributions of the ages at menarche are closer between the non-Hispanic black and Mexican American girls than between these 2 groups and the non-Hispanic white girls.
|
0.3 to 0.7 years at 10% and 90% and slightly more narrow at
0.2 to 0.4 years at 25%, 50%, and 75% of all girls and for the race-specific analyses. | DISCUSSION |
|---|
|
|
|---|
We used the dichotomous yes-no data on the presence/absence of menarche for each girl in a probit analysis because such information is subject to less error or bias than alternative more complex recall methods.7,19 The ranges for the fiducial limits indicate the level of precision about these estimated ages at menarche. The more narrow the range, the higher the precision of the estimated age. The range of the fiducial limits was 0.2 to 0.3 years for the estimated age at menarche, indicating high precision. For the race-specific analysis, the range of the fiducial limits of these estimates was only slightly wider at
0.4 to 0.7 years. Because of the inability of SUDAAN to calculate fiducial limits, the present estimates may be biased.
The findings show the range of ages when 10% to 90% of all US girls and of the race-specific groups had attained menarche (Table 1). Ages are not presented at 5% and 95% because the percentage of girls with yes-no data, respectively, at these ages was very small. As a result, the estimated ages at these percentages were less precise than those at 10% and 90% because of the wider range of the fiducial limits. These findings indicate that <10% of US girls start to menstruate before
11 years, and 90% are menstruating by
13.75 years of age. This distribution of ages indicated that 80% of all US girls start to menstruate between 11.00 and 13.75 years of age with a median (mean) age of 12.43 years.
The ages at menarche presented in Table 1 for the non-Hispanic white, black, and Mexican American girls demonstrate the value of race-specific information. The ages at menarche for the non-Hispanic black and Mexican American girls are earlier than those for the entire group of girls and for the non-Hispanic white girls. The age at menarche for the entire group of girls is very similar to that for the non-Hispanic white girls, but it is not representative of the non-Hispanic and Mexican American girls. The ages at menarche for non-Hispanic black girls are significantly earlier than that of non-Hispanic white girls at 10%, 25%, and 50% of those who had attained menarche (Fig 1). The Mexican American girls were also significantly earlier than the white girls at 25% of the sample. Therefore, before
13.0 years of age, non-Hispanic black, and Mexican American girls are different from non-Hispanic white girls in the timing of menarche. In addition, earlier maturation among non-Hispanic black and Mexican American girls compared with white girls has also been reported for Tanner stages from NHANES III.20
National data for age at menarche have been reported previously from the NHES, but these findings are 30 years old, were collected from non-Hispanic white and non-Hispanic black girls only, and the distribution of age at menarche for 10% to 90% of the samples were not reported. The median age at menarche from these earlier NHES data were 12.77 years as calculated from a logit analysis.7 The NHES median age for all girls is not significantly later (only 0.34 years) than the NHANES III median age for all US girls presented in this report. The NHES medians for non-Hispanic white and black girls were 12.80 years and 12.52 respectively, and there was no statistically significant difference between these medians.7 The 30-year difference in medians between NHES and NHANES III is 0.25 years for non-Hispanic white girls and 0.46 years for non-Hispanic black girls, and neither of these differences is significant.
Recently, Herman-Giddens et al1 reported menarche data from a very large independent sample of white (n = 15 439) and black (n = 1638) girls. Median age at menarche was 12.88 years for white girls and 12.16 for black girls from a probit analysis. These girls, however, were between 3 through 12 years of age so girls who were later in their pubertal development, ie, older than 13 years of age were not included and as a result these findings are subject to bias. As shown in Table 1,
25% of all US girls regardless of race reach menarche after 13 years of age. Thus, the selection criteria of Herman-Giddens focused on girls with an early age at menarche and thereby excluded approximately a quarter of the possible sample of girls.
| CONCLUSION |
|---|
|
|
|---|
There is no consensus definition of a normal range for age at menarche because it can be affected by numerous nonbiological and biological factors affecting the population. These data indicate that age at menarche occurs between 11 and 13.75 years of age with a median (mean) of 12.43 years for
80% of US girls. Race-specific data are clearly important because the distribution of the age at menarche depends on a girls race. The 80% range for non-Hispanic black girls extends from
10.5 years to 13.6 years. From this distribution of ages, one can reasonably assume that US girls, depending on their race, who attain menarche before 10.5 to 11.0 years are relatively "early" and those who attain menarche after 13.6 to 13.75 years are relatively "late."
Concern has been expressed that an early onset of sexual development among girls could be related to the current epidemic of childhood obesity and its associated comorbidity at these ages.1 However, the median (mean) age at menarche for all US girls has not changed significantly in 30 years with a shift of only
4 months in that period. Only 10% of all US girls, regardless of race, start to menstruate before
11 years of age. Girls in the United States are not gaining reproductive potential earlier than in the immediate past in regard to their menstrual status. Non-Hispanic black girls, however, exhibit a 5.5-month earlier age at menarche than 30 years ago, and they do have an earlier age at onset in their Tanner stages20 as well. There are no suitable comparative data for Mexican American girls from previous decades, so it is not possible to make an assessment if menarche in these girls is earlier or later than in the past.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Reprint requests to (W.C.C.) Lifespan Health Research Center, Department of Community Health, Wright State University School of Medicine, 3171 Research Blvd, Kettering, OH 45420. E-mail: cameron.chumlea{at}wright.edu
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Lee "A Kotex and a Smile": Mothers and Daughters at Menarche Journal of Family Issues, October 1, 2008; 29(10): 1325 - 1347. [Abstract] [PDF] |
||||
![]() |
K. Casazza, M. I. Goran, and B. A. Gower Associations among Insulin, Estrogen, and Fat Mass Gain over the Pubertal Transition in African-American and European-American Girls J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2610 - 2615. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Keeney and D. A. Vernik Analysis of the Biological Clock Decision Decision Analysis, September 1, 2007; 4(3): 114 - 135. [Abstract] [PDF] |
||||
![]() |
A. H. Haider, D. T. Efron, E. R. Haut, D. C. Chang, C. N. Paidas, and E. E. Cornwell III Mortality in Adolescent Girls vs Boys Following Traumatic Shock: An Analysis of the National Pediatric Trauma Registry Arch Surg, September 1, 2007; 142(9): 875 - 880. [Abstract] [Full Text] [PDF] |
||||
![]() |
American Academy of Pediatrics, Committee on Adolescence, American College of Obstetricians and Gynecologist, and Committee on Adolescent Health Care Menstruation in Girls and Adolescents: Using the Menstrual Cycle as a Vital Sign Pediatrics, November 1, 2006; 118(5): 2245 - 2250. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Z. Kasa-Vubu, W. Ye, K. T. Borer, A. Rosenthal, and T. Meckmongkol Twenty-Four Hour Growth Hormone and Leptin Secretion in Active Postpubertal Adolescent Girls: Impact of Fitness, Fatness, and Age at Menarche J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3935 - 3940. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. F. Nichols, M. J. Rauh, M. J. Lawson, M. Ji, and H.-S. Barkai Prevalence of the Female Athlete Triad Syndrome Among High School Athletes Arch Pediatr Adolesc Med, February 1, 2006; 160(2): 137 - 142. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. K. Danielson, M. Palta, C. Allen, and D. J. D'Alessio The Association of Increased Total Glycosylated Hemoglobin Levels with Delayed Age at Menarche in Young Women with Type 1 Diabetes J. Clin. Endocrinol. Metab., December 1, 2005; 90(12): 6466 - 6471. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. B Pierce and D. A Leon Age at menarche and adult BMI in the Aberdeen Children of the 1950s Cohort Study Am. J. Clinical Nutrition, October 1, 2005; 82(4): 733 - 739. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. L. Rayner, R. R. Enoch, and S. E. Fenton Adverse Effects of Prenatal Exposure to Atrazine During a Critical Period of Mammary Gland Growth Toxicol. Sci., September 1, 2005; 87(1): 255 - 266. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. A. Graham Economic, Racial, and Cultural Influences on the Growth and Maturation of Children Pediatr. Rev., August 1, 2005; 26(8): 290 - 294. [Full Text] [PDF] |
||||
![]() |
K. E. Remsberg, E. W. Demerath, C. M. Schubert, Wm. C. Chumlea, S. S. Sun, and R. M. Siervogel Early Menarche and the Development of Cardiovascular Disease Risk Factors in Adolescent Girls: The Fels Longitudinal Study J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2718 - 2724. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Jacobson, L. K. Midyett, and W. V. Moore Navigating Recent Articles on Girls' Puberty: Where Should Our Patients Go for Evaluation? Pediatrics, January 1, 2005; 115(1): 194 - 195. [Full Text] [PDF] |
||||
![]() |
B. Crosby, M. K. LeBourgeois, and J. Harsh Racial Differences in Reported Napping and Nocturnal Sleep in 2- to 8-Year-Old Children Pediatrics, January 1, 2005; 115(1/S1): 225 - 232. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. W Demerath, J. Li, S. S Sun, W C. Chumlea, K. E Remsberg, S. A Czerwinski, B. Towne, and R. M Siervogel Fifty-year trends in serial body mass index during adolescence in girls: the Fels Longitudinal Study Am. J. Clinical Nutrition, August 1, 2004; 80(2): 441 - 446. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Z. Kasa-Vubu, M. Sowers, W. Ye, N. E. Carlson, and T. Meckmongkol Differences in Endocrine Function With Varying Fitness Capacity in Postpubertal Females Across the Weight Spectrum Arch Pediatr Adolesc Med, April 1, 2004; 158(4): 333 - 340. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. E. Herman-Giddens, P. B. Kaplowitz, and R. Wasserman Navigating the Recent Articles on Girls' Puberty in Pediatrics: What Do We Know and Where Do We Go from Here? Pediatrics, April 1, 2004; 113(4): 911 - 917. [Full Text] [PDF] |
||||
![]() |
T. M. Plant and M. L. Barker-Gibb Neurobiological mechanisms of puberty in higher primates Hum. Reprod. Update, January 1, 2004; 10(1): 67 - 77. [Abstract] [Full Text] [PDF] |
||||
![]() |
A.-S. Parent, G. Teilmann, A. Juul, N. E. Skakkebaek, J. Toppari, and J.-P. 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., October 1, 2003; 24(5): 668 - 693. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Rose Timing of Puberty in US Girls AAP Grand Rounds, April 1, 2003; 9(4): 39 - 39. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||