Advertisement

Is sexual maturity occurring earlier among U.S. children?

      Abstract

      Purpose

      To compare the onset and completion of sexual maturation among U.S. children between 1966 and 1994.

      Methods

      Tanner stages were from 3042 non-Hispanic white boys, 478 black boys, 2625 white girls, and 505 black girls (NHES 1966–70), from 717 Mexican-American boys and 712 Mexican-American girls (HHANES 1982–84) and from 259 non-Hispanic white boys, 411 black boys, 291 white girls, 415 black girls, 576 Mexican-American boys and 512 Mexican-American girls (NHANES III 1988–1994). Proportions of entry into a stage, probit analysis estimated medians and selected percentiles for ages at entry were calculated using SUDAAN.

      Results

      NHANES III (1988–1994) non-Hispanic white boys entered stage 2, 3, and 4 genital development and stages 3 and 4 pubic hair earlier than NHES (1966–1970) white boys, but they entered stage 5 genital development significantly later. NHANES III (1988–1994) Mexican-American boys were in stage 2, 3 and 4 genital development earlier than HHANES (1982–1984) boys, but entry into stage 5 genital and pubic hair development was not significant. NHANES III (1988–1994) white girls entered stage 5 pubic hair later than NHES (1966–1970) white girls. NHANES III (1988–1994) Mexican-American girls entered stage 2 breast and pubic hair development earlier than HHANES (1982–1984) girls, entered stage 4 breast and pubic hair development earlier but entered stage 5 pubic hair later than the HHANES (1982–1984) girls.

      Conclusion

      Persuasive evidence of a secular trend toward early maturation is not found between 1966 and 1994 in non-Hispanic black boys and non-Hispanic black and white girls. Some evidence of this trend is found in non-Hispanic white boys between 1966 and 1994 and in Mexican-American boys and girls between 1982 and 1994.

      Keywords

      In the 1970s, Damon [
      • Troiano R.P.
      • Flegal K.M.
      • Kuczmarski R.J.
      • et al.
      Overweight prevalence and trends for children and adolescents the National Health and Nutrition Examination Surveys, 1963 to 1991.
      ] noted that the secular trend in early sexual maturation reported during the middle of the 20th Century was abating, but there were no national data to support or confirm this finding. Presently, there is concern among pediatric health care professionals and the public that sexual maturation is once again occurring earlier among U.S. children. This concern is based, in part, on the report that the prevalence of Tanner stage 2 breast and/or pubic hair development at 8 years of age was 48% for black girls and 15% for white girls who were observed in the early 1990s [
      • Herman-Giddens M.E.
      • Slora E.J.
      • Wasserman R.C.
      • et al.
      Secondary sexual characteristics and menses in young girls seen in office practice a study from the Pediatric Research in Office Settings Network.
      ]. The timing of sexual maturation for U.S. children was first documented nationally only about 30 years ago. It was based on Tanner stages for sexual maturity indicators, but only for children 12 years of age and older collected as part of Cycle III of the National Health Examination Survey or NHES (1966–1970) in the late 1960s [
      • Harlan W.R.
      • Grillo G.P.
      • Cornoni-Huntley J.
      • Leverton P.E.
      Secondary sex characteristics of boys 12 to 17 years of age the U.S. Health Examination Survey.
      ,
      • Harlan W.R.
      • Harlan E.A.
      • Grillo G.P.
      Secondary sex characteristics of girls 12 to 17 years of age the U.S. Health Examination Survey.
      ]. Results from this survey indicate that non-Hispanic black girls were significantly more advanced in their sexual maturity than non-Hispanic white girls, but there were no racial differences among the boys [
      • Harlan W.R.
      • Grillo G.P.
      • Cornoni-Huntley J.
      • Leverton P.E.
      Secondary sex characteristics of boys 12 to 17 years of age the U.S. Health Examination Survey.
      ,
      • Harlan W.R.
      • Harlan E.A.
      • Grillo G.P.
      Secondary sex characteristics of girls 12 to 17 years of age the U.S. Health Examination Survey.
      ].
      Sexual maturity data were not collected during the 1970s in the first and second National Health and Nutrition Examination Surveys (NHANES I and II) [

      National Health and Nutrition Examination Survey (NHANES). Plan and Operation of the Health and Nutrition Examination Survey United States 1971–1973, Series 1, DHEW Publication No. (PHS) 79-1310. Washington, DC: Vital and Health Statistics Programs and Collection procedure; 1977 Publication no. 10b.

      ,
      • McDowell A.
      • Engel A.
      • Massey J.T.
      • Maurer K.
      Plan and operation of the Second National Health and Nutrition Examination Survey, 1976–1980.
      ]. Sexual maturity data were again collected in the early 1980s from children 10 years of age and older during the Hispanic Health and Nutrition Examination Survey (HHANES, 1982–1984) [
      National Center for Health Statistics
      Plan and operation of the Hispanic Health and Nutrition Examination Survey, 1982–84.
      ] and in the early 1990s from children 8 years of age and older in the NHANES III (1988–1994) [
      National Health and Nutrition Examination Survey (NHANES I)
      ]. Analysis of the NHANES III (1988–1994) Tanner stage data by Sun and colleagues confirmed that non-Hispanic black children continue to have an earlier sexual maturation than most non-Hispanic white and now Mexican-American children [
      • Sun S.S.
      • Schubert C.M.
      • Chumlea W.C.
      • et al.
      National estimates of the timing of sexual maturation and racial differences among U.S. children.
      ], and that most non-Hispanic black girls reach menarche significantly earlier than most white girls [
      • Chumlea W.C.
      • Schubert C.M.
      • Roche A.F.
      • et al.
      Age at menarche and racial comparisons in U.S. girls.
      ].
      In this article, we question whether a secular trend toward early sexual maturation has occurred among U.S. children as possibly indicated in the report by Herman-Giddens and colleagues [
      • Herman-Giddens M.E.
      • Slora E.J.
      • Wasserman R.C.
      • et al.
      Secondary sexual characteristics and menses in young girls seen in office practice a study from the Pediatric Research in Office Settings Network.
      ]. This question is addressed using sexual maturity data from a series of national surveys representing the U.S. pediatric population over a 30-year period.

      Methods

      Samples

      The NHES Cycle III was conducted between 1966 and 1970 [
      • Bryant E.E.
      • Baird J.T.
      • Miller H.W.
      ], HHANES (1982–1984) was conducted between 1982 and 1984 [
      National Center for Health Statistics
      Plan and operation of the Hispanic Health and Nutrition Examination Survey, 1982–84.
      ], and NHANES III (1988–1994) was carried out between 1988 and 1994 [
      • Ezzati T.M.
      • Massey J.T.
      • Waksberg J.
      • et al.
      Sample design Third National Health and Nutrition Examination Survey.
      ]. The NHES (1966–1970) and NHANES III (1988–1994) used a complex, stratified, multistage probability cluster design to obtain representative samples of the noninstitutionalized U.S. civilian population at the time the data were collected [
      • Bryant E.E.
      • Baird J.T.
      • Miller H.W.
      ,
      • Ezzati T.M.
      • Massey J.T.
      • Waksberg J.
      • et al.
      Sample design Third National Health and Nutrition Examination Survey.
      ]. The HHANES (1982–1984) also used a complex, multistage, stratified probability cluster design, but sampled only from Mexican-Americans, Cuban-Americans, and Puerto Ricans living in selected geographical areas of the United States [
      National Center for Health Statistics
      Plan and operation of the Hispanic Health and Nutrition Examination Survey, 1982–84.
      ]. Sexual maturity data were available from non-Hispanic white and non-Hispanic black children in the NHES (1966–1970) at 12 to 18 years of age, from Mexican-American children in the HHANES (1982–1984) at 10 to 18 years of age, and from non-Hispanic white, non-Hispanic black, and Mexican-American children in the NHANES III (1988–1994) at 8 to 19 years of age.
      To address the question of secular trends toward early sexual maturation, inter-survey comparisons for matching age ranges were made for maturational stages for each maturity indicator by race and gender. Differences in the age ranges of the children in each of the surveys necessitated that the NHES - NHANES III (1988–1994) comparisons use data only from children 12 to 18 years of age. Likewise, for the HHANES (1982–1984) - NHANES III (1988–1994) comparisons, data only from Mexican-American children from 10 to 18 years old were analyzed. As a result of these comparisons, young children in NHANES were not included in the analysis. The number of children in these comparable age groups is presented in Table 1 for each survey.
      Table 1Numbers of non-Hispanic white, non-Hispanic black, and Mexican-American children in the NHES (1966–1970), HHANES (1982–1984), and NHANES III (1988–1994) by sex and ethnic groups
      Age groupsNational surveys
      12 to 18 yearsNHES (1966–1970)NHANES III (1988–1994)
       Non-Hispanic black boys478411
       Non-Hispanic white boys3042259
       Non-Hispanic black girls505415
       Non-Hispanic white girls2625291
      10 to 18 yearsNHANES (1980–1982)NHANES III (1988–1994)
       Mexican-American boys717576
       Mexican-American girls712512

      Sexual maturity stages

      Sexual maturity was assessed in each national survey during physical examinations by physicians who had received standardized training [
      National Health and Nutrition Examination Survey (NHANES I)
      ,
      • Villarreal S.F.
      • Martorell R.
      • Mendoza F.
      Sexual maturation of Mexican-American adolescents.
      ]. Tanner stages [
      • Tanner J.
      ] were assigned as maturity indicators for the development of secondary sex characteristics, pubic hair in each gender, breast development in girls, and genital development (penis, testes, and scrotum) in boys. Each indicator has 5 stages, extending from prepuberty (stage 1) to full maturity (stage 5). The onset of puberty is denoted by the age at entry into stage 2 breast development in girls or into stage 2 genital development in boys. Completion of sexual maturation is denoted by the age at entry into stage 5 breast development in girls and into stage 5 genital development in boys. Written descriptions and photographs describing the stages of the sexual maturity indicators were available for reference during the examinations. The criteria used to characterize sexual maturity assessments in these surveys have been well established over the past 30 years [
      • Tanner J.
      ,
      • Marshall W.A.
      • Tanner J.M.
      Variations in the pattern of pubertal changes in boys.
      ,
      • Marshall W.A.
      • Tanner J.M.
      Variations in pattern of pubertal changes in girls.
      ].

      Data analysis

      Expected ages at entry into stage 2 breast development in girls and stage 2 genital development in boys indicating the onset of puberty [
      • Tanner J.
      ,
      • Marshall W.A.
      • Tanner J.M.
      Variations in the pattern of pubertal changes in boys.
      ,
      • Marshall W.A.
      • Tanner J.M.
      Variations in pattern of pubertal changes in girls.
      ] precede the earliest ages of children with sexual maturity data available in the NHES (1966–1970) and HHANES (1980–1982). Therefore, for the earliest ages available between the studies, only the proportions of children with a maturation of stage 2 or higher were computed and compared using 12 to 13 years for the non-Hispanic white children and black children in the NHES (1966–1970) and NHANES III (1988–1994), and at 10 to 11 years for the Mexican-American children in the HHANES (1982–1984) and NHANES III (1988–1994). In the comparison of these proportions, however, the survey with the greater proportion of children with stage 2 or higher exhibited a more advanced onset of sexual maturation than the other survey. The comparisons of proportions of children for the age at entry for each maturity indicator stage between surveys were made using SUDAAN t-tests.
      The existence of secular trends between surveys was determined further for each gender- and race-specific group by comparing the ages at which 10%, 25%, 50%, 75%, and 90% of the children entered into stages 3 through 5 for each maturity indicator. Children were grouped into 3-month age groups based on chronological age at the time of examination, starting at age 12 years for the non-Hispanic white and non-Hispanic black children in the NHES (1966–1970) and NHANES III (1988–1994), and starting at age 10 years for the Mexican-American children in HHANES (1982–1984) and NHANES III (1988–1994).
      Age at entry into sexual maturity stages 3 through 5 was based on the proportion of children who entered a specified stage within each age group for a given maturity indicator. This was computed using PROC CROSSTAB of SUDAAN [
      • Shah B.V.
      • Barnwell B.G.
      • Bieler G.S.
      ], which adjusts for the individual sampling weight and the design effects of the complex sampling design of each survey. Probit analysis was applied to these proportions across age groups for each gender and race group using SAS 8.2 [
      SAS I
      ]. Probit analysis is designed to analyze the relation between response and exposure. In the context of sexual maturity, the response is reaching a specific stage of a particular genital organ and the exposure is the age. For example, the onset of puberty refers to reaching stage 2 breast development in girls and genitalia in boys. The interest is then at what age a child reaches the onset of puberty. Therefore, the response is a child reaches the onset of puberty and the exposure is age. The response ranges from no child (0%) reaching the onset of puberty to all children reaching the onset of puberty (100%). The median age at the onset of puberty is the age that produces a response in 50% of children. Similarly, 10%, 25%, 75%, and 90% are used for the ages that produce a response in 10%, 25%, 75%, and 90% of children. The precision is greater for the median age than the other proportions. The group mean has been previously used to estimate the onset of puberty. To compute the group mean, the children were grouped according to their stages of a specific genital organ and the mean age of the children within each maturity stage was calculated. Therefore, the group mean is, in general, larger than the median age at entry computed from the probit analysis.
      In the probit analysis, the proportion of children who entered a maturity stage was transformed into “y,” a normal equivalent deviate, i.e., p=yf(θ)dθ where, y = a + bx, and x was the age in months and a and b were the parameters to be estimated from the data. Age at entry was estimated when p is equal to 10%, 25%, 50%, 75%, and 90%. The fiducial limits computed in SAS using Fieller’s theorem [
      • Finney D.J.
      ] indicate the precision of the estimated ages at entry. The more narrow the range of the fiducial limits about the estimated age, the higher its precision and the closer that age is to the true value for the population. The fiducial limits for each age for each racial group were calculated at a 95% confidence level for each survey.

      Results

      Boys

      Onset of puberty: stage 2

      At 12 to 13 years of age, there were no significant differences between the NHES (1966–1970) and NHANES III (1988–1994) in the proportion of non-Hispanic black boys in stage 2 or higher genital development (Table 2). However, at 12 to 13 years of age, significantly more non-Hispanic white boys in NHANES III (1988–1994) were in stage 2 or higher genital development than were NHES (1966–1970) white boys (Table 2). There were no significant differences between the NHES (1966–1970) and NHANES III (1988–1994) in the proportions of non-Hispanic black boys or white boys in stage 2 or higher pubic hair development (Table 2).
      Table 2Proportions of non-Hispanic white, non-Hispanic black, and Mexican-American children in the NHES (1966–1970), HHANES (1982–1984), and NHANES III (1988–1994) in stage 2 or higher for genital and pubic hair development for boys and breast and pubic hair development for girls
      Proportions (%) in Stage 2 and higher
      12 to 13 years non-Hispanic childrenNHESNHANES III
       Genital development
        Black boys7889
        White boys7391
      p Value < .05.
       Pubic hair
        Black boys7182
        White boys6871
       Breast development
        Black girls9498
        White girls9493
       Pubic hair
        Black girls9595
        White girls9089
      10 to 11 years Mexican-American ChildrenHHANESNHANES III
       Genital development
        Mexican-American boys1146
      p Value < .05.
       Pubic hair
        Mexican-American boys74
       Breast development
        Mexican-American girls4070
      p Value < .05.
       Pubic hair
        Mexican-American girls2462
      p Value < .05.
      low asterisk p Value < .05.
      At 10 to 11 years of age, there was a significantly greater proportion of the NHANES III (1988–1994) Mexican-American boys in stage 2 or higher genital development than there were HHANES (1982–1984) boys. There were no significant differences between the HHANES (1982–1984) and NHANES III (1988–1994) in the proportion of Mexican-American boys at 10 to 11 years of age in stage 2 or higher pubic hair development (Table 2). There was a significantly greater proportion of the NHANES III (1988–1994) Mexican-American girls in stage 2 or higher breast development and pubic hair than there were HHANES (1982–1984).

      Age at entry: stages 3–5

      There were no significant differences between the NHES (1966–1970) and the NHANES III (1988–1994) non-Hispanic black boys in their ages at entry for stages 3 through 5 genital development (Table 3). The NHANES III (1988–1994) non-Hispanic white boys had earlier ages at entry for stage 3 genital development at the 50th and 75th percentiles than the NHES (1966–1970) white boys (Table 3), but significantly later ages at entry into stage 5 genital development at the 50th through the 90th percentiles than did the white boys in NHES (1966–1970) (Table 3). The NHANES III (1988–1994) Mexican-American boys also had earlier ages at entry than did the HHANES (1982–1984) boys for stage 3 genital development at the 10th and 50th percentiles and for stage 4 genital development at the 10th through the 75th percentiles (Table 3).
      Table 3Comparisons between National Surveys [NHES (1966–1970), NHANES (1988–1994), HHANES (1982–1984)] for ages of entry at selected percentiles for a Tanner stage in years and the 95% fiducial limits (FL) in years for genital development in boys
      Boys’ genitalia development10th percentile95% FL25th percentile95% FL50th percentile95% FL75th percentile95% FL90th percentile95% FL
      Non-Hispanic Blacks
       Stage 3
        NHES10.9610.11–11.5011.8111.20–12.2112.7412.38–13.0213.6813.43–13.9614.5314.22–14.97
        NHANES III10.869.62–11.5111.5810.67–12.0612.3711.82–12.7013.1612.85–13.4613.8713.55–14.36
       Stage 4
        NHES11.9611.43–12.3512.7912.42–13.0813.7213.47–13.9414.6414.40–14.9315.4715.16–15.90
        NHANES III11.8311.06–12.3212.6012.05–12.9713.4613.10–13.7514.3114.01–14.6715.0814.72–15.61
       Stage 5
        NHES12.6711.92–13.1613.6613.16–14.0314.7614.42–15.1115.8615.48–16.3916.8516.34–17.64
        NHANES III11.9510.63–12.7413.3312.48–13.8814.8614.38–15.3116.3915.89–17.1217.7717.06–18.95
      Non-Hispanic Whites
       Stage 3
        NHES11.4511.16–11.6912.2112.00–12.3813.0512.91–13.1713.8913.78–14.0214.6514.50–14.83
        NHANES III10.779.06–11.5411.5610.35–12.1412.4411.73–12.8413.3112.92–13.7414.1013.68–14.86
       Stage 4
        NHES12.4512.14–12.6913.1312.91–13.3113.8813.72–14.0414.6414.48–14.8315.3215.11–15.59
        NHANES III11.9111.10–12.4112.6612.09–13.0413.4913.12–13.8114.3214.00–14.6415.0714.67–15.67
       Stage 5
        NHES13.2913.10–13.4614.1314.00–14.2615.0714.96–15.1816.0015.87–16.1516.8416.67–17.05
        NHANES III13.0011.91–13.6714.3413.68–14.8215.8315.38–16.3617.3216.73–18.2518.6617.84–20.06
      Mexican-Americans
       Stage 3
        NHES11.9011.11–12.3712.5612.01–12.9513.2912.90–13.7014.0313.63–14.6114.6914.20–15.51
        NHANES III10.9610.56–11.2611.7211.44–11.9612.5612.34–12.8013.4113.15–13.7414.1713.84–14.63
       Stage 4
        NHES12.7312.4–12.9913.5113.28–13.7214.3814.19–14.5915.2515.02–15.5416.0315.73–16.42
        NHANES III12.0611.67–12.3812.8712.58–13.1213.7613.52–14.0114.6614.39–14.9815.4615.13–15.90
       Stage 5
        NHES13.9513.49–14.3114.9614.63–15.2616.0815.76–16.4617.2016.77–17.7918.2117.64–19.01
        NHANES III13.2612.58–13.7514.3713.90–14.7615.6015.23–16.0216.8416.39–17.4517.9517.36–18.80
      The NHANES III (1988–1994) non-Hispanic black boys had earlier ages at entry for stage 3 pubic hair at 50th and 75th percentiles than did the NHES (1966–1970) black boys (Table 4). The NHANES III (1988–1994) non-Hispanic white boys had earlier ages at entry for stage 3 pubic hair at the 10th through the 75th percentiles and for stage 4 pubic hair at the 10th through the 50th percentiles than did the NHES (1966–1970) white boys (Table 4). There were no significant differences between the HHANES (1982–1984) and NHANES III (1988–1994) Mexican-American boys in their ages at entry for stages 3 through 5 pubic hair development (Table 4).
      Table 4Comparisons between National Surveys [NHES (1966–1970), NHANES (1988–1994), HHANES (1982–1984)] for ages of entry at selected percentiles for a Tanner stage in years and the 95% fiducial limits (FL) in years for public hair development in boys
      Boys’ pubic hair development10th percentile95% FL25th percentile95% FL50th percentile95% FL75th percentile95% FL90th percentile95% FL
      Non-Hispanic Blacks
       Stage 3
        NHES11.5810.98–12.0012.3911.96–12.7013.2813.01–13.5114.1713.94–14.4414.9714.67–15.38
        NHANES III11.1410.17–11.7011.8711.18–12.2912.6912.27–12.9913.5113.23–13.8214.2513.93–14.72
       Stage 4
        NHES12.3511.88–12.6913.1412.82–13.4014.0313.80–14.2414.9114.68–15.2015.7115.40–16.12
        NHANES III12.3811.68–12.8213.0512.55–13.3913.7913.45–14.0914.5314.23–14.9115.2014.83–15.74
       Stage 5
        NHES13.2012.75–13.5514.1513.84–14.4115.2014.95–15.4616.2515.94–16.6317.1916.79–17.74
        NHANES III12.9011.68–13.5913.9913.20–14.5115.2114.71–15.7116.4215.90–17.2317.5216.82–18.75
      Non-Hispanic Whites
       Stage 3
        NHES11.8811.70–12.0312.5512.42–12.6613.2913.20–13.3814.0413.95–14.1314.7114.59–14.84
        NHANES III11.009.55–11.6911.7410.71–12.2512.5511.95–12.9313.3713.01–13.7914.1113.70–14.81
       Stage 4
        NHES12.6712.53–12.7913.3213.22–13.4114.0413.96–14.1114.7614.67–14.8515.4115.29–15.53
        NHANES III11.9511.16–12.4312.6712.12–13.0513.4813.12–13.7914.2913.97–14.6915.0114.62–15.61
       Stage 5
        NHES13.6813.51–13.8214.4914.37–14.6015.3915.30–15.4916.3016.18–16.4317.1116.95–17.29
        NHANES III13.5412.79–14.0414.5114.01–14.8915.5915.23–15.9616.6716.26–17.2317.6417.10–18.45
      Mexican-Americans
       Stage 3
        NHES11.9811.40–12.3712.6612.25–12.9713.4113.10–13.7214.1613.84–14.5914.8414.44–15.44
        NHANES III11.6311.25–11.9312.3212.03–12.5613.0812.84–13.3313.8413.56–14.1814.5214.17–14.99
       Stage 4
        NHES12.7912.36–13.1213.5613.25–13.8314.4214.17–14.6915.2714.98–15.6516.0415.67–16.56
        NHANES III12.6412.27–12.9413.3213.04–13.5614.0813.85–14.3114.8314.58–15.1315.5115.20–15.91
       Stage 5
        NHES13.9713.61–14.2614.8614.60–15.1015.8615.61–16.1416.8516.53–17.2617.7417.32–18.30
        NHANES III13.8213.28–14.2214.7514.37–15.0715.7915.49–16.1116.8316.48–17.2817.7617.30–18.40

      Girls

      Onset of puberty: stage 2

      At 12 to 13 years of age, comparisons for the proportions of non-Hispanic black girls and white girls in the NHES (1966–1970) and NHANES III (1988–1994) in stage 2 or higher breast and pubic hair development were not significantly different (Table 2). At 10 to 11 years of age, however, the proportion of the NHANES III (1988–1994) Mexican-American girls in stage 2 or higher for breast and pubic development were significantly larger than the proportion of the HHANES (1982–1984) girls (Table 2).

      Age at entry: stages 3–5

      The NHANES III (1988–1994) non-Hispanic black and non-Hispanic white girls had no significant differences in the ages at entry for stages 3, 4, and, 5 for breast or pubic hair development compared with the NHES (1966–1970) girls except for the stage 5 pubic hair in the NHANES III (1988–1994) white girls (Table 5, Table 6). The ages at entry into stage 5 pubic hair for the NHANES III (1988–1994) white girls at the 50th and 75th percentiles were significantly later than the NHES (1966–1970) white girls (Table 6).
      Table 5Comparisons between National Surveys [NHES (1966–1970), NHANES (1988–1994), HHANES (1982–1984)] for ages of entry at selected percentiles for a Tanner stage in years and the 95% fiducial limits (FL) in years for breast development in girls
      Girls’ breast development10th percentile95% FL25th percentile95% FL50th percentile95% FL75th percentile95% FL90th percentile95% FL
      Non-Hispanic Blacks
       Stage 3
        NHES7.899.0310.2911.559.41–12.2512.6911.79–13.21
        NHANES III6.327.779.3911.0112.4710.43–13.21
       Stage 4
        NHES8.039.657.44–10.8011.4610.09–12.2113.2612.60–13.7514.8914.38–15.62
        NHANES III8.079.6711.469.73–12.3113.2412.42–13.8414.8514.25–15.80
       Stage 5
        NHES10.138.45–11.1412.1011.06–12.7614.2913.80–14.7116.4815.95–17.2618.4517.60–19.84
        NAHES III9.607.91–10.8611.479.95–12.3413.5412.77–14.1015.6215.08–16.3717.4916.68–18.89
      Non-Hispanic Whites
       Stage 3
        NHES9.6310.528.14–11.4311.529.98–12.1512.5211.75–12.9413.4213.01–13.99
        NHANES III10.397.52–11.3211.078.93–11.7811.8210.49–12.3212.5811.93–12.9613.2512.87–13.90
       Stage 4
        NHES10.229.55–10.7311.4310.95–11.8012.7812.49–13.0214.1313.94–14.3315.3515.10–15.66
        NHANES III10.298.59–11.2511.5310.31–12.2312.9012.18–13.3914.2813.84–14.7415.5215.02–16.29
       Stage 5
        NHES11.6111.08–12.0313.4413.13–13.7015.4715.26–15.6917.5017.16–17.9219.3318.81–19.98
        NHANES III11.299.45–12.3113.1212.03–13.7915.1614.61–15.7117.1916.51–18.3219.0217.99–20.90
      Mexican-Americans
       Stage 3
        NHES10.149.04–10.7711.0210.27–11.4911.9811.52–12.412.9512.53–13.5413.8313.29–14.73
        NHANES III9.5310.411.3712.3311.12–17.7313.2012.11–25.14
       Stage 4
        NHES11.2210.74–11.6012.4012.07–12.6913.7213.46–13.9815.0414.74–15.3916.2215.83–16.72
        NHANES III10.7410.08–11.2211.8311.36–12.1913.0412.70–13.3714.2513.88–14.7015.3314.86–15.99
       Stage 5
        NHES12.4912.05–12.8513.6613.35–13.9414.9714.70–15.2516.2715.94–16.6817.4517.00–18.01
        NHANES III11.9211.35–12.3613.2612.87–13.6014.7414.42–15.0916.2315.82–16.7417.5717.02–18.29
      ∼ = unstable estimates.
      Table 6Comparisons between National Surveys [NHES (1966–1970), NHANES (1988–1994), HHANES (1982–1984)] for ages of entry at selected percentiles for a Tanner stage in years and the 95% fiducial limits (FL) in years for pubic hair development in girls
      Girls’ pubic hair development10th percentile95% FL25th percentile95% FL50th percentile95% FL75th percentile95% FL90th percentile95% FL
      Non-Hispanic Blacks
       Stage 3
        NHES7.949.0810.3411.609.61–12.2812.7311.89–13.24
        NHANES III8.009.0811.487.54–12.2412.5511.12–13.15
       Stage 4
        NHES9.037.10–10.1310.368.94–11.1911.8510.97–12.3913.3312.87–13.7014.6614.26–15.22
        NHANES III7.619.1510.8712.5910.66–13.4214.2813.45–15.74
       Stage 5
        NHES10.479.15–11.3212.1511.30–12.7214.0213.58–14.3915.8915.47–16.4517.5816.93–18.56
        NHANES III10.197.99–11.3912.1310.75–12.9414.2913.62–14.8416.4515.81–17.4318.3917.41–20.14
      Non-Hispanic Whites
       Stage 3
        NHES7.949.0810.3411.609.61–12.2812.7311.89–13.24
        NHANES III8.009.0810.2811.487.54–12.2412.5511.12–13.15
       Stage 4
        NHES10.6910.23–11.0511.6611.33–11.9312.7412.54–12.9213.8313.68–13.9814.8014.62–15.02
        NHANES III10.639.15–11.4411.6210.56–12.2312.7312.08–13.1613.8313.43–14.2514.8314.39–15.48
       Stage 5
        NHES10.479.15–11.3212.1511.30–12.7214.0213.58–14.3915.8915.47–16.4517.5816.93–18.56
        NHANES III12.5010.90–13.4714.2613.34–14.8716.1315.55–16.9217.9917.14–19.5919.6718.44–22.13
      Mexican-Americans
       Stage 3
        NHES10.589.24–11.2511.3710.46–11.9012.2411.67–12.7613.1112.6–13.9113.8913.27–15.11
        NHANES III9.698.08–10.4810.629.53–11.2111.6511.01–12.1512.6812.17–13.4213.6012.97–14.79
       Stage 4
        NHES11.6911.27–12.0312.7412.44–13.0013.9213.68–14.1615.0914.81–15.4116.1415.78–16.59
        NHANES III10.629.84–11.1811.8211.28–12.2413.1512.77–13.5314.4814.07–15.0115.7915.23–16.58
       Stage 5
        NHES13.0212.61–13.3614.0913.79–14.3515.2715.01–15.5416.4516.13–16.8317.5117.09–18.05
        NHANES III13.2012.58–13.6714.6414.23–15.0116.2415.83–16.7317.8417.27–18.6219.2718.51–20.37
      ∼ = unstable estimates.
      The NHANES III (1988–1994) Mexican-American girls had significantly earlier ages at entry for stage 4 breast development at the 50th and 75th percentiles than did the HHANES (1982–1984) girls (Table 5). The NHANES III (1988–1994) Mexican-American girls also had significantly earlier ages at entry into stage 4 pubic hair development at the 10th through the 50th percentiles, but significantly later ages at entry for stage 5 pubic hair at the 50th through the 90th percentiles than did the HHANES (1982–1984) girls (Table 6).

      Discussion

      The present study compares the sexual maturity data for U.S. children available from three national health surveys collected between 1966 and 1994 to determine the occurrence of any secular trends toward early sexual maturation. The non-Hispanic white, non-Hispanic black, and Mexican-American children in these national surveys were representative of the pediatric population of the United States at the time the surveys were conducted. In these analyses, data from NHANES III (1988–1994) were truncated to match the ages of the children in NHES (1966–1970) and HHANES (1982–1984) by eliminating children younger than 12 or 10 years of age, respectively. As a result, the median ages at entry for NHANES III (1988–1994) in this analysis do not represent the timing of sexual maturation as reported previously [
      • Sun S.S.
      • Schubert C.M.
      • Chumlea W.C.
      • et al.
      National estimates of the timing of sexual maturation and racial differences among U.S. children.
      ], but are estimates from similar samples in the NHANES III (1988–1994) data to make comparisons between nationally representative samples of children from previous decades [NHES (1966–1970) and HHANES (1980–1982)].
      All analyses took into account the complex, stratified, multistage, probability cluster design effect and the sample weights of each survey to account for the unequal probability of selection. The medians and selected percentiles for ages at entry into each stage within each indicator were calculated using probit analysis. This approach differs from calculating the mean age for being in a maturity stage. The mean age for being in a maturity stage generally occurs later than the median age at entry for all stages within all maturity indicators [
      • Sun S.S.
      • Schubert C.M.
      • Chumlea W.C.
      • et al.
      National estimates of the timing of sexual maturation and racial differences among U.S. children.
      ]. Other investigators [
      • Herman-Giddens M.E.
      • Slora E.J.
      • Wasserman R.C.
      • et al.
      Secondary sexual characteristics and menses in young girls seen in office practice a study from the Pediatric Research in Office Settings Network.
      ,
      • Herman-Giddens M.E.
      • Wang L.
      • Koch G.G.
      Secondary sexual characteristics in boys.
      ,
      • Reynolds E.L.
      • Wines J.V.
      Individual differences in physical changes associated with adolescence in girls.
      ] report mean ages of being in a stage within each indicator.

      Boys

      Between the NHES (1966–1970) and NHANES III (1988–1994), there appears to be no secular trend in sexual maturity for non-Hispanic black boys. There were no statistically significant differences in the proportion of non-Hispanic black boys in stage 2 or higher or in the ages at entry into stages 3 to 5 genital or pubic hair development between these two national surveys (Table 2, Table 3, Table 4). The NHANES III (1988–1994) non-Hispanic white boys were in stage 2 or higher genital development in greater proportion and they entered stage 3 genital development and stages 3 and 4 pubic hair development earlier than the NHES (1966–1970) white boys. These results indicate an earlier onset and progression through the stages of sexual maturation for the NHANES III (1988–1994) white boys. However, the non-Hispanic white boys were significantly later than the NHES (1966–1970) white boys in their ages at entry into stage 5 genital development, and there were no significant differences in the ages at entry into stage 5 pubic hair. It appears that the NHANES III (1988–1994) non-Hispanic white boys start puberty early but finish it later than the NHES (1966–1970) white boys. Factors affecting onset of puberty may differ from those affecting progression of maturation. These two surveys were conducted nearly 30 years apart, and the inter-survey differences in the onset and progression of sexual maturity stages may reflect a trend for the non-Hispanic white boys.
      The NHANES III (1988–1994) Mexican-American boys were also in stage 2 or higher genital development in greater proportion, and they entered stages 3 and 4 genital development earlier than the HHANES (1982–1984) boys (Table 2, Table 3, Table 4). These results also indicate an earlier onset and progression through sexual maturation for the NHANES III (1988–1994) Mexican-American boys compared with the HHANES (1982–1984) boys. However, there was no significant difference in the timing of pubic hair stages in the HHANES-NHANES III (1988–1994) comparisons. Furthermore, the completion of sexual maturation or entry into stage 5 genital and pubic hair development was not significantly different for the Mexican-American boys in the HHANES-NHANES III (1988–1994) comparisons. The HHANES-NHANES III (1988–1994) comparisons covered only a 6- to 10-year period, and the design of HHANES (1982–1984) included only Hispanics living in selected geographical areas of the United States. This early onset and progression through sexual maturity for genital development in these Mexican-American boys is over a relatively short period of time and should be interpreted cautiously. There is no persuasive evidence of a secular trend toward early maturation in non-Hispanic black boys between 1966 and 1994. We found some evidence of secular trend toward early maturation in non-Hispanic white boys between 1966 and 1994 and in Mexican-American boys between 1982 and 1994.

      Girls

      There was also no apparent secular trend in sexual maturation for non-Hispanic black girls or white girls between the NHES (1966–1970) and the NHANES III (1988–1994) (Table 2, Table 5, Table 6). There were no differences in the proportions of NHANES III (1988–1994) non-Hispanic black girls and white girls in stage 2 or higher breast and pubic hair development at 12 to 13 years of age compared with NHES (1966–1970) girls (Table 2). At these ages, however, about 90% of all girls had reached stage 2 or higher breast and pubic hair development. There was no early breast or pubic hair development for the non-Hispanic black girls compared with the black girls in the NHES. However, the lower percentiles for stages 3 and 4 breast and pubic hair development in the NHES-NHANES III (1988–1994) comparisons for the black girls were unstable. There was also no early breast or pubic hair development for the NHANES III (1988–1994) non-Hispanic white girls compared with the NHES (1966–1970) white girls. Similar to the late stage 5 genital development in the NHANES III (1988–1994) white boys, the NHANES III (1988–1994) white girls completed sexual maturity later than the NHES (1966–1970) white girls (Table 6).
      For the NHANES III (1988–1994) Mexican-American girls, there were greater proportions in stage 2 or higher breast and pubic hair development than the HHANES (1982–1984) girls, and stage 4 breast and pubic hair development were significantly earlier than the HHANES (1982–1984) girls. Like the Mexican-American boys, there was an early onset and progression through sexual maturity for breast and pubic hair development in these Mexican-American girls. Unlike the boys, however, the NHANES III (1988–1994) Mexican-American girls, like the NHANES III (1988–1994) white girls, completed sexual maturity with stage 5 pubic hair later than the HHANES (1982–1984) girls (Table 6). The HHANES-NHANES III (1988–1994) comparisons covered only a short time period, and this finding of an early onset and progression but later completion of sexual maturation in Mexican-American girls should also be interpreted cautiously. We analyzed the menstrual status data from 2510 girls aged 8.0 to 20.0 years in the NHANES III [
      • Chumlea W.C.
      • Schubert C.M.
      • Roche A.F.
      • et al.
      Age at menarche and racial comparisons in U.S. girls.
      ]. The median age at menarche for all U.S. girls is 12.43 years. The median age at menarche from earlier NHES data was 12.77 years [
      • MacMahon B.
      ]. The NHES median age for all girls is 0.34 years later than the NHANES III median age for all U.S. girls. 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. The NHES medians for non-Hispanic white and black girls were 12.80 and 12.52 years, respectively. 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. Although age at menarche in the U.S. girls appears to become earlier from 1966 to 1994, we did not find persuasive evidence of a secular trend toward early maturation in non-Hispanic white or black girls between 1966 and 1994. There is some evidence of secular trend toward early maturation in Mexican-American girls between 1982 and 1994, however.

      Limitations

      Tanner stage data for sexual maturity indicators were used to compare the timing of sexual maturity separately for boys and girls between the NHES (1966–1970) and NHANES III (1988–1994) for non-Hispanic whites and non-Hispanic blacks and between HHANES (1982–1984) and NHANES III (1988–1994) for Mexican Americans. In these national surveys, maturity stages were assigned by physicians after visual assessments only using the same criteria of Tanner [
      • Tanner J.
      ]. In clinical practice, the entry for boys into puberty is determined by a palpated enlargement of the testes. Assessment of Tanner stage 2 genital development does not include palpation of the testes [
      • Tanner J.
      ], and thus it may be an uncertain indicator of pubertal onset. In girls, the timing of the onset of puberty is denoted by entry into stage 2 breast development [
      • Tanner J.
      ]. The clinical evaluation and accuracy of stage 2 stage breast development in girls is also difficult without palpation and is complicated by the presence of obesity. The development of secondary sex characteristics is a continuous process, and Tanner stages are an arbitrary and subjective maturity grading system that is overlaid on this process. Survey and clinical assessment differences in Tanner stage assessments may have contributed, in part, to an over-interpretation of the possibility of a secular trend reported for girls by Herman-Giddens et al [
      • Herman-Giddens M.E.
      • Slora E.J.
      • Wasserman R.C.
      • et al.
      Secondary sexual characteristics and menses in young girls seen in office practice a study from the Pediatric Research in Office Settings Network.
      ] that the present study does not support.
      These inter-survey comparisons were hampered by the inability to match maturity stage data at young ages. There were too few young children in the NHES (1966–1970) and HHANES (1982–1984) to calculate medians and selected percentiles for age at entry into stage 2 for any of the maturity indicators. Only NHANES III (1988–1994) collected data from children as young as 8 years of age, which is necessary to determine if there were shifts in the ages at entry into stage 2 for any indicators. Thus, inter-survey comparisons were possible only using the proportion of children in Tanner stage 2 or higher for each maturity indicator by race and gender at 12 to 13 years for the NHES-NHANES III (1988–1994) comparisons and 10 to 11 years for the HHANES-NHANES III (1988–1994) comparisons. For later stages of the maturity indicators, insufficient numbers of girls in the NHES (1966–1970), the HHANES (1982–1984) and in the NHANES III (1988–1994) resulted in unstable estimates for ages at entry and also prevented inter-survey comparisons at some selected percentiles. Furthermore, instability occurred in the estimated ages at selected percentiles for some stages 3 and 4 because a large proportion of children had already progressed to or beyond these stages over the comparable age ranges between the surveys.
      The reliability of Tanner stage assessments by others, including the physicians in NHANES III (1988–1994), has been high [
      • Herman-Giddens M.E.
      • Slora E.J.
      • Wasserman R.C.
      • et al.
      Secondary sexual characteristics and menses in young girls seen in office practice a study from the Pediatric Research in Office Settings Network.
      ,
      • Reiter E.O.
      • Lee P.A.
      Have the onset and tempo of puberty changed?.
      ,
      • Matsudo S.M.M.
      • Matsudo V.K.R.
      Self-assessment and physician assessment of sexual maturation in Brazilian boys and girls concordance and reproducibility.
      ]. However, there is no formal report of reliability for Tanner stages from any of the national health surveys. There is almost no reliability data from the assessment of sexual maturity indicators in these national surveys. In each survey, the assessments, although made by physicians, included an unknown number of assessors, so observer bias and errors are possible both within and between surveys despite efforts to standardize data collection and control for measurement errors.

      Conclusions

      There is no persuasive evidence that sexual maturation for U.S. children has become earlier, but some evidence of earlier sexual maturation was found in non-Hispanic white boys and Mexican-American boys and girls. There is the indication of earlier onset and progression of sexual maturation for non-Hispanic white boys, but the completion of sexual maturation is later in the NHANES III (1988–1994) boys. The completion of sexual maturity is also later in the NHANES III (1988–1994) non-Hispanic white girls. The onset and progression of sexual maturation in Mexican-American boys and girls does occur earlier between 1982 and 1994. However, the completion of sexual maturation in the Mexican-American boys was not earlier and it was later for the Mexican-American girls. For white boys and Mexican-American boys and girls, the onset of puberty and its progression into mid-puberty are earlier than in the past. However, the completion of sexual maturation is not earlier and appears later for non-Hispanic white boys and girls and Mexican-American girls.
      From our analyses of inter-survey comparisons using national health survey data collected between 1966 and 1994, we conclude that there is insufficient evidence and data to support a secular trend toward early sexual maturation in U.S. children from 1966 to 1994, but there is some evidence of secular trend toward early maturation in non-Hispanic white boys from 1966 to 1994 and in Mexican-American boys from 1982 to 1994. These same national data sets have been used to report a significant increase in the prevalence of overweight and obesity among U.S. children [
      • Troiano R.P.
      • Flegal K.M.
      • Kuczmarski R.J.
      • et al.
      Overweight prevalence and trends for children and adolescents the National Health and Nutrition Examination Surveys, 1963 to 1991.
      ]. Despite this increased prevalence of obesity, our findings indicated that these children do not appear to have any conclusive shift toward an earlier sexual maturation during the same time period. This implies that the increased prevalence of childhood obesity is not accounted for with an earlier sexual maturation.

      Acknowledgments

      This work was supported by the 1999 Beta Research Program Award from Wright State University School of Medicine, Dayton OH and by grant HD-38356 from the National Institutes of Health, Bethesda MD.

      References

        • Troiano R.P.
        • Flegal K.M.
        • Kuczmarski R.J.
        • et al.
        Overweight prevalence and trends for children and adolescents.
        Arch Pediatr Adolesc Med. 1995; 149: 1085-1091
        • Herman-Giddens M.E.
        • Slora E.J.
        • Wasserman R.C.
        • et al.
        Secondary sexual characteristics and menses in young girls seen in office practice.
        Pediatrics. 1997; 99: 505-512
        • Harlan W.R.
        • Grillo G.P.
        • Cornoni-Huntley J.
        • Leverton P.E.
        Secondary sex characteristics of boys 12 to 17 years of age.
        J Pediatr. 1979; 95: 293-297
        • Harlan W.R.
        • Harlan E.A.
        • Grillo G.P.
        Secondary sex characteristics of girls 12 to 17 years of age.
        J Pediatr. 1980; 96: 1074-1078
      1. National Health and Nutrition Examination Survey (NHANES). Plan and Operation of the Health and Nutrition Examination Survey United States 1971–1973, Series 1, DHEW Publication No. (PHS) 79-1310. Washington, DC: Vital and Health Statistics Programs and Collection procedure; 1977 Publication no. 10b.

        • McDowell A.
        • Engel A.
        • Massey J.T.
        • Maurer K.
        Plan and operation of the Second National Health and Nutrition Examination Survey, 1976–1980.
        Vital Health Stat 1. 1981; Series 1: 1-144
        • National Center for Health Statistics
        Plan and operation of the Hispanic Health and Nutrition Examination Survey, 1982–84.
        Vital Health Stat 1. 1985; : 1-429
        • National Health and Nutrition Examination Survey (NHANES I)
        Analytic and reporting guidelines:the third health and nutrition examination survey (1988–94). National Centers for Health Statistics Centers for Disease Control and Prevention, 1997 (CD-ROM:)
        • Sun S.S.
        • Schubert C.M.
        • Chumlea W.C.
        • et al.
        National estimates of the timing of sexual maturation and racial differences among U.S. children.
        Pediatrics. 2002; 110: 911-919
        • Chumlea W.C.
        • Schubert C.M.
        • Roche A.F.
        • et al.
        Age at menarche and racial comparisons in U.S. girls.
        Pediatrics. 2003; 111: 110-113
        • Bryant E.E.
        • Baird J.T.
        • Miller H.W.
        Sample design and estimation procedures for a national health examination survey of children. National Center for Health Statistics; Vital and Health Statistics, Washington, DC1971 (Series 2, Publication no. 43.)
        • Ezzati T.M.
        • Massey J.T.
        • Waksberg J.
        • et al.
        Sample design.
        Vital Health Stat 2. 1992; : 1-35
        • Villarreal S.F.
        • Martorell R.
        • Mendoza F.
        Sexual maturation of Mexican-American adolescents.
        Am J Hum Biol. 1989; 1: 87-95
        • Tanner J.
        Growth at adolescence. Book Oxford, UK1962: 121-130
        • Marshall W.A.
        • Tanner J.M.
        Variations in the pattern of pubertal changes in boys.
        Arch Dis Child. 1970; 45: 13-23
        • Marshall W.A.
        • Tanner J.M.
        Variations in pattern of pubertal changes in girls.
        Arch Dis Child. 1969; 44: 291-303
        • Shah B.V.
        • Barnwell B.G.
        • Bieler G.S.
        SUDAAN User’s Manual. Research Triangle Institute, Research Triangle Park, NC1997
        • SAS I
        SAS/Stat User’s Guide, Version 8. 8.2. SAS Institute Inc, Cary, NC2000
        • Finney D.J.
        Statistical Method in Biological Assay. 3rd edn. Macmillan Publishing Co., Inc, New York, NY1978
        • Herman-Giddens M.E.
        • Wang L.
        • Koch G.G.
        Secondary sexual characteristics in boys.
        Arch Pediatr Adolesc Med. 2001; 155: 1022-1028
        • Reynolds E.L.
        • Wines J.V.
        Individual differences in physical changes associated with adolescence in girls.
        Am J Dis Child. 1948; 75: 329-350
        • MacMahon B.
        Age at menarche, United States. Series 11: Vital and Health Statistics, NCHS DHEW Publication (HRA) 74-1615. 1973 (Report No. 133.)
        • Reiter E.O.
        • Lee P.A.
        Have the onset and tempo of puberty changed?.
        Arch Pediatr Adolesc Med. 2001; 155: 988-989
        • Matsudo S.M.M.
        • Matsudo V.K.R.
        Self-assessment and physician assessment of sexual maturation in Brazilian boys and girls.
        Am J Hum Biol. 1994; 6: 451-455