Pubertal timing and adult obesity and cardiometabolic risk in women and men: a systematic review and meta-analysis

Abstract

Obesity has complex multifactorial aetiology. It has been suggested by many, but not all, reports that earlier pubertal maturation may increase adult obesity risk. We conducted a systemic review and meta-analysis in both women and men, and hypothesised that any association between pubertal timing and adult obesity is likely to be confounded by childhood adiposity. In addition, we investigated whether pubertal timing is related to other cardiometabolic risk and long-term cardiovascular morbidity/mortality. Literature search was undertaken using MEDLINE, EMBASE, Web of Knowledge and TRIP databases, with a hand search of references. Both authors independently reviewed and extracted pre-defined data from all selected papers. Meta-analyses were conducted using Review Manager (RevMan) 5.0.24. A total of 48 papers were identified. Out of 34 studies, 30 reported an inverse relationship between pubertal timing and adult body mass index (BMI), the main adiposity measure used. Meta-analysis of 10 cohorts showed association between early menarche (menarche <12 vs 12 years) and increased adult BMI, with a standardised mean difference of 0.34 kg m−2 (95% confidence interval: 0.33–0.34). Heterogeneity was large (I2=92%) but reduced significantly when grouped by outcome age. Late menarche (menarche 15 vs <15 years) was associated with decreased adult BMI, with a standardised mean difference of −0.26 kg m−2 (95% confidence interval: −0.36, −0.21) (seven cohorts). Only eight papers included data on childhood BMI; the majority reported that childhood BMI only partially attenuated association between early menarche and later obesity. Although not suitable for meta-analysis, data on cardiometabolic risk factors and puberty suggested negative association between earlier pubertal timing and cardiovascular mortality, hypertension, metabolic syndrome (MetS) and abnormal glycaemia. Earlier pubertal timing is predictive of higher adult BMI and greater risk of obesity. This effect appears to be partially independent of childhood BMI. Earlier pubertal development appears to also be inversely correlated with risk of other cardiometabolic risk factors and cardiovascular mortality. Further work is needed to examine potential mechanisms and the level at which interventions may be targeted.

Introduction

Obesity has become an international epidemic, with a complex multifactorial aetiology. Both modifiable and less changeable risk factors need to be identified to better target public health and clinical interventions. It is especially important to understand epidemiological factors in early life, predicting long-term obesity and cardiometabolic risk.

A systematic review in 1999 noted that a number of studies raised the possibility that earlier pubertal maturation is correlated with early adulthood obesity.1 The review included few studies with data past early adulthood or with male cohorts. Since that time a large number of epidemiological studies have been published, without further review of their findings.

Mechanisms explaining an effect of pubertal timing on later obesity in women have been postulated. One possibility is that this association merely reflects earlier timing of puberty in girls already obese in childhood,2, 3 especially in those with rapid infancy weight gain.4, 5 However, existing studies have inconsistently considered whether the association between pubertal timing and later obesity is mediated by childhood body mass index (BMI). In contrast, there are suggestions that increasing childhood obesity in boys correlates with delayed puberty, but there are few longitudinal studies looking at adult adiposity gains.6

Although studies have largely focused on the risks associated with early puberty, there is some evidence that late puberty may protect against obesity and cardiometabolic risk when compared with normal pubertal timing.7

Recent studies have also reported that other increasingly prevalent conditions associated with obesity, including type 2 diabetes, hypertension, hyperlipidaemia, metabolic syndrome (MetS) and even overall cardiovascular mortality, may also be associated with early pubertal maturation.8 However, these findings have been inconsistent and require systematic review.6 The impact of childhood adiposity on the risk of these later adult morbidities has not been fully elucidated.9

We identified the following objectives and plan to address them:

(1) To examine whether puberty timing is correlated with adult adiposity in men and women, we reviewed observational, population-based studies, which assessed adult (average cohort age 25 years) adiposity and measurements of pubertal status.

(2) To examine whether pubertal timing is related to other cardiometabolic risk factors, as well as long-term cardiovascular morbidity and mortality, we reviewed observational studies of adult cohorts, with outcomes such as hypertension, hyperlipidaemia, type 2 diabetes and cardiovascular mortality, and measures of pubertal status (see Table 1).

Table 1 Outcomes eligible for inclusion in systematic review

We therefore undertook a systematic review and meta-analysis to evaluate the impact of early or late pubertal timing on long-term adult obesity and cardiometabolic risk in both men and women. We also examined whether associations were explained by childhood BMI and considered the reporting of pubertal maturation assessment.

Methods

We present our method and findings, including recommendations from the Meta-analysis Of Observational Studies in Epidemiology (MOOSE),10 and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines11 and reference to the PRISMA statement.12 The protocol and criteria for the analysis of studies, with inclusion and exclusion criteria, were agreed and documented prospectively by two reviewers.

Ethics

No ethical approval was needed for this systematic review of previously published observational trials.

Searching

We used the following databases: MEDLINE (1950–2010), EMBASE (1980–2010), ISI Web of Knowledge and the TRIP database. Databases were accessed in May 2010. Additional citations were found by hand-searching through reference lists of obtained and reviewed articles. One title was identified from hand-searching, but the citation could not be found after searching or contacting the author. Non-English papers were included and translated if relevant from title and/or abstract. Published abstracts were included if they provided sufficient data to judge eligibility.

The search strategy is summarised in Figure 1 and search terms used are listed in Supplementary Appendix 4. Keywords and exploded MeSH (Medical Subject Headings) terms were used, in English. The following search strategy was added to identify prognostic studies: ‘incidence (MeSH) or ‘mortality’ (MeSH), or ‘mortality’, or ‘prognos*’ or ‘predict*’ or ‘course’, and combined with ‘cohort studies’.13

Figure 1
figure1

Eligible studies included in systematic review.

Selection

Inclusion criteria

Eligible studies were required to:

  • Include a population with a mean age of study population more than 25 years of age. There was no restriction on country of origin and study date.

  • Include a measure of pubertal timing, whether reported in adolescence or in adulthood (including age at menarche, age at peak height velocity, skeletal age, calculated maturity score and pubertal staging).

  • Report at least one measure of adiposity and/or other cardiometabolic risk factors/diseases as an outcome (see Table 1).

  • Use any observational study design, including population-based cohort studies and case–control studies, whether retrospective, prospective or cross-section.

All eligible studies were included in the systematic review. In addition, studies were included in meta-analysis if they included the following:

For obesity outcomes: Raw data, allowing calculation of mean BMI (with standard deviation scores), or reported/calculated odds ratios (ORs).

For other outcomes: Raw data, allowing calculation of comparable OR or risk ratio.

Exclusion criteria

  • Cohorts included only specific conditions, for example, women with breast cancer or polycystic ovarian syndrome as they might not be representative of the total population.

  • Any study without a measure of pubertal status.

  • Outcomes of only height or stroke.

Validity assessment

Validity and quality of individual studies were assessed and also graded using the Oxford Centre for Evidence-based Medicine levels of evidence (2011). Details were recorded for quality of outcome measurements, reporting of pubertal timing and confounding factors.

Data abstraction

Each article was assessed separately by the two authors, with data extracted independently into an excel spreadsheet using pre-defined criteria. Any differences were discussed and resolved. When data were suitable for meta-analysis, raw data or ORs were extracted from articles and entered into the meta-analysis programme (Review Manager (RevMan) 5.0.24).

Where there were queries regarding the data, in three cases, authors were contacted for further information. Authors either did not respond or did not provide additional useful information. Where duplicate publications reported data from the same cohort, the paper describing a larger sample size or more reported samples of interest was included in our review. Each cohort was included only once in the meta-analysis.

Study characteristics

Each paper was appraised using the ‘PICO with quality’, strategy (Population, Intervention, Comparison, Outcome, Quality). Potential confounder information was collected, particularly whether data on childhood adiposity were recorded and whether analyses were adjusted for childhood adiposity.

Information was collected on:

  1. 1)

    Study population quality and variability, including: study design, sample population, country and setting of study, subject number and gender, ethnicity and socioeconomic group and possible confounding variables, including menstrual and pregnancy history, smoking, alcohol and exercise, author contact details, funding and key conclusions by the authors.

  2. 2)

    Details of pubertal/ physical assessment, such as age of assessment and definition of early pubertal development. The variable used for pubertal status, as well as the timing of recording was documented.

  3. 3)

    Type of outcome measure, including definition, age measured, whether self-reported or measured, number of subjects, and unadjusted and adjusted associations.

We assessed clinical heterogeneity by looking at the population location, age of menarche/pubertal maturation and age at outcome.

Analysis and data synthesis

A descriptive analysis was carried out for all outcome variables.

Meta-analyses were carried out using RevMan 5.0.24. Raw data (for studies reporting age at menarche) were grouped into ‘early menarche’ (age at menarche <12 years) and ‘not-early menarche’ (age at menarche 12 years), as well as late (age at menarche >15 years) and ‘not-late’ menarche (age at menarche 15 years). These definitions were selected to reflect those groupings used in publications, across the whole time period reviewed. Data were frequently only presented for ‘early’/’late’ menarche vs ‘not early/late’, and hence it was not possible to report three groups: ‘early’, ‘normal’ and ‘late’ menarche. Extracted measures for outcome variables included means, percentages, ORs, relative risks and correlation coefficients.

Shown or calculated ORs, or means with standard deviations, were imputed into a generic inverse-variance meta-analysis, using a random-effects model. For entry into an inverse-variance meta-analysis, ORs were converted into ln OR and the standard error of ln OR. Where standard deviations were not shown, they were calculated from other statistics available. Percentages were converted into numbers as far as data allowed.

It was anticipated that meta-analysis might be difficult, not only because of the differences in presented results but also because of the large variations between study populations. For this reason, it was planned that subgroups would be analysed according to population differences, for example, study date or participant age.

Results

Flow of included studies

A total of 48 studies were eligible for inclusion into the systematic review (Figure 1). Out of these 48 eligible studies, 37 included an adiposity outcome, 9 blood pressure, 6 hyperlipidaemia, 9 diabetes or impaired glucose tolerance, 6 MetS and 10 cardiovascular morbidity and mortality.

Study characteristics

Characteristics of included studies

Characteristics of included outcome variables are shown Table 1. The majority of studies were prospective observational longitudinal cohort studies, with smaller numbers of cross-sectional and case–control studies, the latter relating to cardiovascular morbidity and mortality. Publications spanned over 35 years (1974–2010). All, but one study, were published in English. The single non-English study was translated from Mandarin.14

The 48 publications include 37 papers on adiposity (11 also including cardiometabolic risk factors, 1 cardiovascular morbidity and mortality, 1 all risk factors), with 3 additional papers on cardiometabolic risk and 8 on solely cardiovascular morbidity and mortality. The 48 publications included 52 different studies.

Participants

Sample size ranged from 177 to 101 415 adult participants. In total, there were 704 869 participants in 46 separate cohorts and 6 case–control studies. Adult outcomes were reported between 1934 and 2005, and with the exception of two papers, all were reported after 1971.

Intervention/pubertal timing

Various measures of pubertal status were reported, including timing of menarche, peak height velocity, skeletal ages, pubertal staging and general maturity factor. The majority (36 of the 46 cohorts) reported retrospective menarcheal timing, recalled in adulthood. Ten used menarcheal data reported contemporaneously in adolescence.

Outcomes

Data synthesis—qualitative and quantitative data for individual outcomes

Outcomes with some data available for meta-analysis are presented first.

(i) Measures of adiposity:

In all, 37 papers described the effect of pubertal timing on adult adiposity, most frequently reporting BMI (Supplementary Appendix 1)

(a) BMI—

Out of 34 papers, 30 reported on BMI outcomes, and showed an inverse relationship between pubertal timing and adult BMI. The majority of studies were American or European, with one Japanese cohort and four recent Chinese papers. In all, 26 were solely female cohorts, 6 included data on men and women and 2 included only men.

The quality of the data was heterogeneous. In total, 11 papers calculated BMI from adult self-reported height and weight measurements. Of the 30 papers using menarcheal age for pubertal timing, 22 used adult retrospective report. The population characteristics were less varied, with mean menarcheal age similar across publications. Date of outcome measurements was in the past three decades in the great majority of studies. Age when BMI measurements were taken/reported varied from 18 to 92 years.

Focusing first on women, it was possible to include 10 cohorts in a meta-analysis of the standardised mean difference in BMI between ‘early’ and ‘not-early’ puberty. These were all United Kingdom- or United States-based studies, with the exception of one Austrian and one Chinese study.14, 15 Early menarche (menarche <12 vs 12 years) was associated with increased adult BMI, with a standardised mean difference of 0.34 kg m−2 (95% confidence interval (CI): 0.33–0.34), although with high heterogeneity (I2=92%). As there were a considerable number of studies, we were able to undertake a subgroup analyses splitting cohorts by mean age at outcome less or more than 40 years (Figure 2). A significant increase in BMI was associated with early puberty in both subgroups, although the increased BMI was even more evident in the younger populations. No differences were seen when cohorts were divided into the most recent and older studies.

Figure 2
figure2

Meta-analysis for early menarche and adult BMI. (a) Early menarche with mean age at outcome 40 years. (b) Early menarche with mean age at outcome <40 years.

Five studies reported odds of obesity (BMI 30 kg m−2) for age of menarche <12 years compared with menarche 12 years and could be included in a meta-analysis (Figure 3). This found that early menarche was associated with increased risk of adult obesity, with a pooled OR of 2.00 (95% CI: 1.79–2.24) with moderate heterogeneity. These data did not include adjustment for childhood BMI.

Figure 3
figure3

Early menarche and odds ratios for adult obesity.

Late menarche

Seven studies provided data sufficient to be included in a meta-analysis comparing adult BMI in groups with late menarche (menarche 15 years) compared with those with menarche <15 years. We found that late menarche was associated with decreased adult BMI by a standardised mean difference of −0.24 kg m−2 (confidence interval: −0.30, −0.19), although with moderately high heterogeneity (see Figure 4).

Figure 4
figure4

Meta-analysis for late menarche and adult BMI.

Effect of childhood BMI

Only eight studies included a measure of childhood obesity as a possible confounder in their models. There was large variation in the reported confounding effect of childhood BMI, ranging from no real reduction in reported correlation16 to a 28-fold reduction in the β-coefficient.17 Five of the studies reported that a significant association between early puberty and higher adult BMI remained when childhood adiposity was included in the model, although the strength of the association was generally reduced.16, 18, 19, 20, 21 Owing to the nature of the data, it was not possible to include these in a separate meta-analysis.

(b) Other measures of adiposity—In total, 10 papers reported on waist circumference, of which six found an inverse association with pubertal timing. A total of 17 studies reported on another adiposity outcome, including waist-to-hip ratio (7 studies, only 2 studies reporting an association between puberty and risk of later obesity), skinfold thickness measurements (7 studies, 4 reporting an association), percentage body fat (6 studies, 4 reporting an association) and weight gain (1 study, showing an association). Of note, as with BMI, all studies reported increased risk of adiposity with earlier puberty (mainly reported as earlier menarche), decreased risk with later puberty or no association. No studies reported that earlier pubertal maturation was associated with reduced risk.

Childhood adiposity was only included in one paper reporting other measures of adult adiposity, where the association of pubertal timing with waist circumference was lost when correcting for childhood BMI.22 It was not possible to include these data in a meta-analysis.

Ethnicity

Only one study reported findings separately by ethnic group. Freedman et al.18 noted a correlation between pubertal timing and adult skinfold thicknesses in the white population, but not for the black women.18 There was a significant association for BMI and this disappeared when childhood BMI was accounted for only in the black subgroup.

Men

Eight studies examined the effect of pubertal timing on later obesity in male cohorts. These studies formed a highly heterogeneous group, largely because they used a wide variety of indicators of pubertal timing. Five papers measured pubertal timing by peak height velocity, but the methods of reporting differed and therefore data were not suitable for meta-analysis. Male puberty was additionally assessed by pubertal staging, skeletal age and a calculated variable ‘general maturity factor’.23 Five out of eight reports showed that BMI in men is inversely associated with pubertal maturation.19, 20, 23, 24, 25 The remaining three studies found no significant differences between maturity groups of boys.26, 27, 28

(ii) CV morbidity and mortality:

Ten studies reported risk of cardiovascular disease or death and age of menarche (Supplementary Appendix 2). Data sources and the validity of cardiovascular disease diagnosis varied markedly.

Morbidity—Eight studies focused on cardiovascular disease morbidity, largely acute myocardial infarction in case–control or cohort studies. Only 3 studies found age of menarche to be associated with cardiovascular disease.29, 30, 31 However, meta-analysis suggested an association between early menarche (<12 years) and increased risk of later cardiovascular disease, with pooled risk ratio 1.15 (1.02, 1.28) with acceptable heterogeneity (Figure 5). Only five out of the eight studies reporting morbidity could be included in the meta-analysis because of the presentation of available data.

Figure 5
figure5

Meta-analysis for early menarche and adult cardiovascular morbidity.

Mortality—Four studies reported cardiovascular deaths in relation to age at menarche. Three studies reported that early menarche was associated with higher mortality risk. When adjusted for adult BMI, this association was retained in two studies.29, 32 Data did not allow meta-analysis.

The following data for lipid, blood pressure, MetS and diabetes outcomes were not suitable for meta-analysis because of the paucity of data and heterogeneity of reporting. Further summaries and details are presented below.

(iii) Lipids:

Six papers evaluated the relationship between pubertal timing and adult lipid values. Two were in Chinese populations and all but one described female cohorts (Supplementary Appendix 3). It is difficult to make overall conclusions, but with the data available it seems unlikely that there is a relationship between pubertal timing and lipidaemia later in life.

Only three studies identified an association between a lipid measurement and pubertal timing. Triglycerides showed a relationship in two out of four reports (although only in men in one study),27, 33 and high-density lipoprotein (HDL) in one out of five studies.34 The following lipids showed no association: cholesterol (in one study),29 low-density lipoproteins (LDL; two studies),22, 34 LDL/HDL ratio (one study)35 and lipoprotein, not further specified (one study).36

Of note, both of the Chinese cohorts did show an association for either triglycerides or HDL, which was retained, even when correcting for adult adiposity. However, the results were not consistent, with one study showing a correlation for HDL34 and the other no relationship.33 No studies adjusted for childhood adiposity.

(iv) Blood pressure:

Nine papers reported blood pressure outcomes, either as mean systolic and diastolic pressures (four papers),22, 24, 27, 29 risk of hypertension (two papers)21, 36 or both (three papers).23, 33, 34 Three cohorts included both men and women, and the remainder were female cohorts. Two cohorts were ethnic Chinese. Six studies reported some association between increased risk of higher blood pressure and earlier pubertal timing, with some differences between men and women, but no consistent relationships. For example, one study reported an association between pubertal timing and both systolic and diastolic blood pressure in female cohorts, but less of a relationship in male cohorts,23 whereas Hardy et al.24 show this association in men but not in women. In four papers, this relationship between pubertal timing and blood pressure was retained after correction for adult BMI (in another it was lost, and in one paper there was no correction). Only one study considered childhood BMI and showed that there was a slight strengthening of association between later pubertal timing and lower blood pressure for men, when childhood BMI and height were included in the model.24

(v) Metabolic syndrome:

Six papers reported on the association of pubertal timing and the prevalence of MetS. MetS was defined either using the American Heart Association (AHA) (three papers)33, 37, 38 or International Diabetes Federation (two papers) definitions,22, 34 or as a clustering of metabolic risk factors (one paper).36 In four cohorts (the two studies using the International Diabetes Federation definition, one using AHA, one as a clustering of risk factors), there was an association between increased prevalence of MetS and earlier pubertal maturation. In the other two cohorts, which used the AHA definition, there was no association. Out of these studies, three included adult adiposity data, and in only one the relationship was retained—a Chinese cohort, using the AHA definition for MetS.33 No papers provided childhood BMI data.

(vi) Diabetes:

Four papers reported upon the relationship between prevalence of type 2 diabetes and pubertal timing. All these studies were from the United Kingdom and the United States. The prevalence of diabetes was self-reported in two studies and validated by medical records in the two other papers. Two studies reported an increased risk of type 2 diabetes in those with earlier pubertal maturation (one with self-reported cases).21, 29 However, in both cases this association was lost attenuated when correcting for adult BMI.

Six papers reported adult glucose measurements or a measure of insulin resistance (mostly HOMA-IR (Homeostasis Model of Assessment—Insulin Resistance)) in relation to pubertal timing. Five of these reports showed some relationship with pubertal timing, but the data were very varied, for example, with correlation shown between pubertal timing and fasting glucose in one report,39 insulin and HOMA in another22 and glycated haemoglobin in a third.29 Supplementary Appendix 3 shows further details.

Discussion

This is the first systematic review to summarise data on the association of pubertal timing and adiposity and cardiometabolic outcomes in mid–late adulthood in both men and women.

We found strong evidence for an association between earlier pubertal maturation and greater adult adiposity in women. In meta-analysis, early menarche (<age 12 years, approximately the earliest quintile) was associated with an increased adult BMI by a mean of 0.34 kg m−2 and a doubling of risk for adult obesity compared with those with normal or later menarche. In a sensitivity analysis, the effect of earlier menarche on adult BMI was larger on younger women (those <40 years at the time of outcome measurements). This association might be stronger in younger women, as they are temporally closer to puberty. In older women, the increased importance of other lifestyle factors over time may mean that the association with puberty is lessened. The evidence for a similar association in men was equivocal.

The effect of pubertal timing on later adiposity was not confined to early puberty. In women, we found that menarche later than age 15 years, approximately equivalent to the latest timing decile, was negatively associated with adiposity, decreasing the mean adult BMI by 0.24 kg m−2 compared with normal and earlier puberty.

Evidence for an association of pubertal timing with other cardiometabolic outcomes is markedly weaker than for adiposity. The evidence is most convincing for cardiovascular disease in women, as we found that early menarche was associated with an increased disease risk approximately 15% in meta-analysis. A number of studies suggest that higher blood pressure, metabolic syndrome and abnormal glycaemia share this association with early puberty. However, it remains unclear whether this relationship merely reflects the common association of both pubertal timing and cardiometabolic risk factors with obesity.

We are aware of only two systematic reviews that have addressed similar questions, both now over a decade old. Parsons et al.1 in a descriptive review concluded that early menarche was likely to be associated with later obesity, supporting our findings. De Kleijn et al.40 in a review of female reproductive history and cardiovascular disease risk in postmenopausal women identified four studies reporting age of menarche and cardiovascular risk, and concluded that there was no consistent relationship between age of menarche and cardiovascular risk. They did not distinguish between cardiovascular morbidity and mortality. In contrast, through meta-analysis of a larger number of studies, we found that early menarche modestly increased risk of cardiovascular disease.

We were unable to assess adequately whether the observed association between early puberty and adult obesity and cardiometabolic risk was mediated by childhood BMI because of the limited number of studies, including childhood adiposity data. While five papers reported that pubertal timing predicted later adiposity independently of childhood BMI, three found that the association was entirely explained by childhood BMI. It may be that the role of childhood BMI in mediating this relationship differs between individuals and populations, especially since the secular trend in earlier pubertal timing in populations with greater childhood obesity does not seem to be universal.

The associations we identified are biologically plausible and identifiable across a large number of studies; however, they are not necessarily causal. Recent reviews discussing the association between childhood obesity and pubertal timing emphasise the uncertainty of direct causation, the complexity of this association and the inter-relationship between multiple factors. It seems likely that pre-, peri- and early postnatal exposures, such as nutrition, may be important influencing factors for both childhood adiposity gains and the initiation of puberty, but postulating single causality is probably simplistic.6, 41

It may be that adiposity in childhood and puberty influence each other to some extent, for example, through hormonal mechanisms, but that pubertal timing has an additional influence on adult disease, whether this is directly causal or not. Therefore, the associations found are useful, but further evaluation is needed to identify potential causal pathways. Recent genome-wide association studies concerning the timing of puberty have found that multiple loci found to be associated with earlier menarche have been identified targets for adult BMI, with variants also associated with childhood BMI.42

Strengths and limitations

The strengths of this systematic review lie in its robust methodology and inclusion of a wide range of adiposity and cardiometabolic outcomes.

Our findings are subject to a number of limitations. The majority of studies did not include childhood or adolescent data and were therefore limited to adult report of menarcheal timing and unable to adjust for childhood BMI. This lack of data meant that we were unable to more robustly examine effects of childhood adiposity on the associations of pubertal timing with adult outcomes, or examine the influence of the timing of adiposity development in childhood.

Clearly, adult report of pubertal timing is subject to various biases; however, high agreement has been shown between contemporary report and adult recall of age at menarche.43, 44 The presentation of data from the majority of papers reviewed only allowed for comparison of ‘early’ vs ‘not-early’ menarche, and ‘late’ vs ‘not late’. This dichotomisation may potentially either strengthen or obscure real associations.

The evidence that these associations are causal is limited. The findings that pubertal timing influences adult cardiovascular risk are biologically plausible, repeated in multiple studies and relate to longitudinal data with a lengthy time period between the exposure (pubertal timing) and the outcome (adult cardiometabolic risk). However, we recognise that assessment of causality is limited in systematic reviews, and other study designs are necessary to determine if these associations are causal.

Lack of ethnic diversity limits the generalisability of our findings beyond European and American white populations. The great majority of included studies involved white populations and defined early puberty consistently as menarche at or before 12 years. However, only one study included separate analyses for black subjects, in whom mean age at menarche is generally earlier than white populations,18 and five studies involved Chinese or Japanese populations, in whom mean age at menarche is later.33, 34, 37 Although early menarche was associated with adult adiposity in all of these East Asian studies, it remains unclear whether the association of pubertal timing and adult adiposity is consistent across all ethnic groups. Indeed, Freedman et al.18 reported that pubertal timing was associated with adult adiposity in white but not black women.

Conclusion

Early puberty is associated with increased risk of adult obesity and some associated cardiometabolic disorders in women. The balance of evidence is that these associations are at least partially independent of childhood adiposity, suggesting that they are not epiphenomena resulting from the association of early puberty and childhood obesity.

While pubertal timing is not malleable at a population level, adolescents with early normative pubertal timing are increasingly recognised as a potential intervention target given they are at high-risk group for a range of other mental and physical health outcomes.45 Targeted obesity prevention interventions in adolescents with early puberty may contribute to reducing adult obesity burden.

References

  1. 1

    Parsons TJ, Power C, Logan S, Summerbell CD . Childhood predictors of adult obesity: a systematic review. Int J Obes Relat Metab Disord 1999; 23 (Suppl 8): S1–S107.

  2. 2

    Kaplowitz P . Pubertal development in girls: secular trends. Curr Opin Obstet Gynecol 2006; 18: 487–491.

  3. 3

    Sorensen K, Aksglaede L, Petersen JH, Juul A . Recent changes in pubertal timing in healthy Danish boys: associations with body mass index. J Clin Endocrinol Metab 2010; 95: 263–270.

  4. 4

    Ong KK, Emmett P, Northstone K, Golding J, Rogers I, Ness AR et al. Infancy weight gain predicts childhood body fat and age at menarche in girls. J Clin Endocrinol Metab 2009; 94: 1527–1532.

  5. 5

    Ong KK . Early determinants of obesity. Endocr Dev 2010; 19: 53–61.

  6. 6

    Burt Solorzano CM, McCartney CR . Obesity and the pubertal transition in girls and boys. Reproduction 2010; 140: 399–410.

  7. 7

    Hartge P . Genetics of reproductive lifespan. Nat Genet 2009; 41: 637–638.

  8. 8

    Golub MS, Collman GW, Foster PM, Kimmel CA, Rajpert-De Meyts E, Reiter EO et al. Public health implications of altered puberty timing. Pediatrics 2008; 121 (Suppl 3): S218–S230.

  9. 9

    Walvoord EC . The timing of puberty: is it changing? Does it matter? J Adolesc Health 2010; 47: 433–439.

  10. 10

    Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2010; 283: 2008–2012.

  11. 11

    von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370: 1453–1457.

  12. 12

    Moher D, Liberati A, Tetzlaff J, Altman DG . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009; 62: 1006–1012.

  13. 13

    Altman DG . Systematic reviews of evaluations of prognostic variables. BMJ 2001; 323: 224–228.

  14. 14

    Mi J, Chen FF, Wang YF, Cheng H, Hou DQ, Zhao XY . Impact of early menarche on adiposity during late puberty and mid-life. Zhonghua Liu Xing Bing Xue Za Zhi 2007; 28: 833–837.

  15. 15

    Kirchengast S, Gruber D, Sator M, Huber J . Impact of the age at menarche on adult body composition in healthy pre- and postmenopausal women. Am J Phys Anthropol 1998; 105: 9–20.

  16. 16

    Pierce MB, Leon DA . Age at menarche and adult BMI in the Aberdeen children of the 1950s cohort study. Am J Clin Nutr 2005; 82: 733–739.

  17. 17

    Laitinen J, Power C, Jarvelin MR . Family social class, maternal body mass index, childhood body mass index, and age at menarche as predictors of adult obesity. Am J Clin Nutr 2001; 74: 287–294.

  18. 18

    Freedman DS, Khan LK, Serdula MK, Dietz WH, Srinivasan SR, Berenson Gs . The relation of menarcheal age to obesity in childhood and adulthood: the Bogalusa heart study. BMC Pediatr 2003; 3: 3.

  19. 19

    Power C, Lake JK, Cole TJ . Body mass index and height from childhood to adulthood in the 1958 British born cohort. Am J Clin Nutr 1997; 66: 1094–1101.

  20. 20

    Sandhu J, Ben-Shlomo Y, Cole TJ, Holly J, Davey Smith G . The impact of childhood body mass index on timing of puberty, adult stature and obesity: a follow-up study based on adolescent anthropometry recorded at Christ’s Hospital (1936–1964). Int J Obes (Lond) 2006; 30: 14–22.

  21. 21

    He C, Zhang C, Hunter DJ, Hankinson SE, Buck Louis GM, Hankinson SE et al. Age at menarche and risk of type 2 diabetes: results from 2 large prospective cohort studies. Am J Epidemiol 2010; 171: 334–344.

  22. 22

    Kivimaki M, Lawlor DA, Smith GD, Elovainio M, Jokela M, Keltikangas-Jarvinen L et al. Association of age at menarche with cardiovascular risk factors, vascular structure, and function in adulthood: the Cardiovascular Risk in Young Finns study. Am J Clin Nutr 2008; 87: 1876–1882.

  23. 23

    Hulanicka B, Lipowicz A, Koziel S, Kowalisko A . Relationship between early puberty and the risk of hypertension/overweight at age 50: evidence for a modified Barker hypothesis among Polish youth. Econ Hum Biol 2007; 5: 48–60.

  24. 24

    Hardy R, Kuh D, Whincup PH, Wadsworth ME . Age at puberty and adult blood pressure and body size in a British birth cohort study. J Hypertens 2006; 24: 59–66.

  25. 25

    van Lenthe FJ, Kemper CG, van Mechelen W . Rapid maturation in adolescence results in greater obesity in adulthood: the Amsterdam Growth and Health Study. Am J Clin Nutr 1996; 64: 18–24.

  26. 26

    Beunen G, Malina RM, Lefevre J, Claessens AL, Renson R, Simons J et al. Size, fatness and relative fat distribution of males of contrasting maturity status during adolescence and as adults. Int J Obes Relat Metab Disord 1994; 18: 670–678.

  27. 27

    Sun SS, Schubert CM . Prolonged juvenile States and delay of cardiovascular and metabolic risk factors: the Fels Longitudinal study. J Pediatr 2009; 155: e1–e6.

  28. 28

    Taeymans J, Hebbelinck M, Borms J, Clarys P, Abidi H, Duquet W . Tracking of adult adiposity in early, average and late maturing children: a thirty year longitudinal growth study. J Sports Med Phys Fitness 2008; 48: 326–334.

  29. 29

    Lakshman R, Forouhi NG, Sharp SJ, Luben R, Bingham SA, Khaw KT et al. Early age at menarche associated with cardiovascular disease and mortality. J Clin Endocrinol Metab 2009; 94: 4953–4960.

  30. 30

    Cooper GS, Ephross SA, Weinberg CR, Baird DD, Whelan EA, Sandler DP et al. Menstrual and reproductive risk factors for ischemic heart disease. Epidemiology 1999; 10: 255–259.

  31. 31

    La Vecchia C, Decarli A, Franceschi S, Gentile A, Negri E, Parazzini F . Menstrual and reproductive factors and the risk of myocardial infarction in women under fifty-five years of age. Am J Obstet Gynecol 1987; 157: 1108–1112.

  32. 32

    Jacobsen BK, Oda K, Knutsen SF, Fraser GE . Age at menarche, total mortality and mortality from ischaemic heart disease and stroke: the Adventist Health Study, 1976-88. Int J Epidemiol 2009; 38: 245–252.

  33. 33

    Heys M, Schooling CM, Jiang C, Cowling BJ, Lao X, Zhang W et al. Age of menarche and the metabolic syndrome in China. Epidemiology 2007; 18: 740–746.

  34. 34

    Feng Y, Hong X, Wilker E, Li Z, Zhang W, Jin D et al. Effects of age at menarche, reproductive years, and menopause on metabolic risk factors for cardiovascular diseases. Atherosclerosis 2008; 196: 590–597.

  35. 35

    Emaus A, Espetvedt S, Veierod MB, Ballard-Barbash R, Furberg AS, Ellison PT et al. 17-Beta-estradiol in relation to age at menarche and adult obesity in premenopausal women. Hum Reprod 2008; 23: 919–927.

  36. 36

    Frontini MG, Srinivasan SR, Berenson GS . Longitudinal changes in risk variables underlying metabolic Syndrome X from childhood to young adulthood in female subjects with a history of early menarche: the Bogalusa Heart Study. Int J Obes Relat Metab Disord 2003; 27: 1398–1404.

  37. 37

    Cho GJ, Park HT, Shin JH, Kim T, Hur JY, Kim YT et al. The relationship between reproductive factors and metabolic syndrome in Korean postmenopausal women: Korea National Health and Nutrition Survey 2005. Menopause 2009; 16: 998–1003.

  38. 38

    Santos AC, Ebrahim S, Barros H . Gender, socio-economic status and metabolic syndrome in middle-aged and old adults. BMC Public Health 2008; 8: 62.

  39. 39

    Saquib N, Kritz-Silverstein D, Barrett-Connor E . Age at menarche, abnormal glucose tolerance and type 2 diabetes mellitus: The Rancho Bernardo Study. Climacteric 2005; 8: 76–82.

  40. 40

    de Kleijn MJ, van der Schouw YT, van der Graaf Y . Reproductive history and cardiovascular disease risk in postmenopausal women: a review of the literature. Maturitas 1999; 33: 7–36.

  41. 41

    Cheng G, Buyken AE, Shi L, Karaolis-Danckert N, Kroke A et al. Beyond overweight: nutrition as an important lifestyle factor influencing timing of puberty. Nutr Rev 2012; 70: 133–152.

  42. 42

    Elks CE, Perry JR, Sulem P, Chasman DI, Franceschini N, He C et al. Thirty new loci for age at menarche identified by a meta-analysis of genome-wide association studies. Nat Genet 2010; 42: 1077–1085.

  43. 43

    Cooper R, Blell M, Hardy R, Black S, Pollard TM, Wadsworth ME et al. Validity of age at menarche self-reported in adulthood. J Epidemiol Community Health 2006; 60: 993–997.

  44. 44

    Must A, Phillips SM, Naumova EN, Blum M, Harris S, Dawson-Hughes B et al. Recall of early menstrual history and menarcheal body size: after 30 years, how well do women remember? Am J Epidemiol 2002; 155: 672–679.

  45. 45

    Patton GC, Viner R . Pubertal transitions in health. Lancet 2007; 369: 1130–1139.

  46. 46

    Keim SA, Branum AM, Klebanoff MA, Zemel BS . Maternal body mass index and daughters’ age at menarche. Epidemiology 2009; 20: 677–681.

  47. 47

    Ong KK, Northstone K, Wells JC, Rubin C, Ness AR, Golding J et al. Earlier mother’s age at menarche predicts rapid infancy growth and childhood obesity. PLoS Med 2007; 4: e132.

  48. 48

    Harris MA, Prior JC, Koehoorn M . Age at menarche in the Canadian population: secular trends and relationship to adulthood BMI. J Adolesc Health 2008; 43: 548–554.

  49. 49

    Cui R, Iso H, Toyoshima H, Date C, Yamamoto A, Kikuchi S et al. Relationships of age at menarche and menopause, and reproductive year with mortality from cardiovascular disease in Japanese postmenopausal women: the JACC study. J Epidemiol 2006; 16: 177–184.

  50. 50

    Must A, Naumova EN, Phillips SM, Blum M, Dawson-Hughes B, Rand WM et al. Childhood overweight and maturational timing in the development of adult overweight and fatness: the Newton Girls Study and its follow-up. Pediatrics 2005; 116: 620–627.

  51. 51

    Newby PK, Dickman PW, Adami HO, Wolk A . Early anthropometric measures and reproductive factors as predictors of body mass index and obesity among older women. Int J Obes (Lond) 2005; 29: 1084–1092.

  52. 52

    Wen W, Gao YT, Shu XO, Yang G, Li HL, Jin F et al. Sociodemographic, behavioral, and reproductive factors associated with weight gain in Chinese women. Int J Obes Relat Metab Disord 2003; 27: 933–940.

  53. 53

    Lahmann PH, Lissner L, Gullberg B, Berglund G . Sociodemographic factors associated with long-term weight gain, current body fatness and central adiposity in Swedish women. Int J Obes Relat Metab Disord 2000; 24: 685–694.

  54. 54

    van Lenthe FJ, Kemper HC, van Mechelen W, Post GB, Twisk JW, Welten DC et al. Biological maturation and the distribution of subcutaneous fat from adolescence into adulthood: the Amsterdam Growth and Health Study. Int J Obes Relat Metab Disord 1996; 20: 121–129.

  55. 55

    Troisi RJ, Wolf AM, Mason JE, Klingler KM, Colditz GA . Relation of body fat distribution to reproductive factors in pre- and postmenopausal women. Obes Res 1995; 3: 143–151.

  56. 56

    Helm P, Munster K, Schmidt L . Recalled menarche in relation to infertility and adult weight and height. Acta Obstet Gynecol Scand 1995; 74: 718–722.

  57. 57

    Burke GL, Savage PJ, Manolio TA, Sprafka JM, Wagenknecht LE, Sidney S et al. Correlates of obesity in young black and white women: the CARDIA Study. Am J Public Health 1992; 82: 1621–1625.

  58. 58

    den Tonkelaar I, Seidell JC, van Noord PA, Baanders-van Halewijn EA, Ouwehand IJ . Fat distribution in relation to age, degree of obesity, smoking habits, parity and estrogen use: a cross-sectional study in 11,825 Dutch women participating in the DOM-project. Int J Obes 1990; 14: 753–761.

  59. 59

    Kaye SA, Folsom AR, Prineas RJ, Potter JD, Gapstur SM . The association of body fat distribution with lifestyle and reproductive factors in a population study of postmenopausal women. Int J Obes 1990; 14: 583–591.

  60. 60

    Frisancho AR, Housh CH . The relationship of maturity rate to body size and body proportions in children and adults. Hum Biol 1988; 60: 759–770.

  61. 61

    Bjorkelund C, Lissner L, Andersson S, Lapidus L, Bengtsson C . Reproductive history in relation to relative weight and fat distribution. Int J Obes Relat Metab Disord 1996; 20: 213–219.

  62. 62

    Garn SM, LaVelle M, Rosenberg KR, Hawthorne VM . Maturational timing as a factor in female fatness and obesity. Am J Clin Nutr 1986; 43: 879–883.

  63. 63

    Sherman B, Wallace R, Bean J, Schlabaugh L . Relationship of body weight to menarcheal and menopausal age: implications for breast cancer risk. J Clin Endocrinol Metab 1981; 52: 488–493.

  64. 64

    Bertuccio P, Tavani A, Gallus S, Negri E, La Vecchia C . Menstrual and reproductive factors and risk of non-fatal acute myocardial infarction in Italy. Eur J Obstet Gynecol Reprod Biol 2007; 134: 67–72.

  65. 65

    Rosenberg L, Palmer JR, Rao RS, Adams-Campbell LL . Risk factors for coronary heart disease in African American women. Am J Epidemiol 1999; 150: 904–909.

  66. 66

    Palmer JR, Rosenberg L, Shapiro S . Reproductive factors and risk of myocardial infarction. Am J Epidemiol 1992; 136: 408–416.

  67. 67

    Colditz GA, Willett WC, Stampfer MJ, Rosner B, Speizer FE, Hennekens CH et al. A prospective study of age at menarche, parity, age at first birth, and coronary heart disease in women. Am J Epidemiol 1987; 126: 861–870.

  68. 68

    Oliver MF . Ischaemic heart disease in young women. BMJ 1974; 4: 253–259.

  69. 69

    Cooper GS, Ephross SA, Sandler DP . Menstrual patterns and risk of adult-onset diabetes mellitus. J Clin Epidemiol 2000; 53: 1170–1173.

Download references

Author information

Correspondence to P Prentice.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Author contributions

PP and RV both contributed substantially to the conception and design of the review, and the acquisition, analysis and interpretation of data. Both authors wrote and approved the paper for publication.

Supplementary Information accompanies the paper on International Journal of Obesity website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Prentice, P., Viner, R. Pubertal timing and adult obesity and cardiometabolic risk in women and men: a systematic review and meta-analysis. Int J Obes 37, 1036–1043 (2013). https://doi.org/10.1038/ijo.2012.177

Download citation

Keywords

  • puberty
  • cardiovascular disease
  • metabolic syndrome

Further reading