Tracking of Cardiorespiratory Fitness from Childhood to Mid-adulthood

High cardiorespiratory fitness (CRF) in adulthood is important for survival from major chronic diseases and preserving good health. We examined how childhood CRF tracks, or persists, into adulthood. Among a cohort of 748 school children followed over 34 years, we found child CRF correlated with young- (r=0.30) and mid-adulthood (r=0.16) CRF. High cardiorespiratory fitness (CRF) in adulthood is important for survival from major chronic diseases and preserving good health. We examined how childhood CRF tracks, or persists, into adulthood. Among a cohort of 748 school children followed over 34 years, we found child CRF correlated with young- (r=0.30) and mid-adulthood (r=0.16) CRF. Cardiorespiratory fitness (CRF) is a physical fitness component representing the body’s ability to supply oxygen to skeletal muscle for physical work1American College of Sports Medicine. ACSM's health-related physical fitness assessment manual: Lippincott Williams & Wilkins; 2013.Google Scholar. Higher CRF levels are associated with broad health benefits in adulthood, including increased survival and reduced chronic disease risk2Ross R. Blair S.N. Arena R. Church T.S. Després J.-P. Franklin B.A. et al.Importance of assessing cardiorespiratory fitness in clinical practice: A case for fitness as a clinical vital sign: A scientific statement from the American Heart Association.Circulation. 2016; 134: e653-e699Crossref PubMed Google Scholar,3Raghuveer G. Hartz J. Lubans D.R. Takken T. Wiltz J.L. Mietus-Snyder M. et al.Cardiorespiratory fitness in youth: An important marker of health: A scientific statement from the American Heart Association.Circulation. 2020; 142: e101-e118Crossref PubMed Scopus (191) Google Scholar. Indeed, CRF is an important risk factor for cardiovascular disease that remains an independent and additive marker of risk over traditional risk factors2Ross R. Blair S.N. Arena R. Church T.S. Després J.-P. Franklin B.A. et al.Importance of assessing cardiorespiratory fitness in clinical practice: A case for fitness as a clinical vital sign: A scientific statement from the American Heart Association.Circulation. 2016; 134: e653-e699Crossref PubMed Google Scholar. As such, physical activity and exercise training interventions are targeted to adults to help raise or maintain adequate CRF levels4Ozemek C. Laddu D.R. Lavie C.J. Claeys H. Kaminsky L.A. Ross R. et al.An update on the role of cardiorespiratory fitness, structured exercise and lifestyle physical activity in preventing cardiovascular disease and health risk.Prog. Cardiovasc. Dis. 2018; 61: 484-490Crossref PubMed Scopus (134) Google Scholar. Because CRF clusters within families5Sarzynski M.A. Rice T.K. Després J.P. Pérusse L. Tremblay A. Stanforth P.R. et al.The HERITAGE Family Study: A review of the effects of exercise training on cardiometabolic health, with insights into molecular Transducers.Med. Sci. Sports Exerc. 2022; 54: S1-s43Crossref PubMed Scopus (11) Google Scholar, early life might provide an important window in the development of CRF3Raghuveer G. Hartz J. Lubans D.R. Takken T. Wiltz J.L. Mietus-Snyder M. et al.Cardiorespiratory fitness in youth: An important marker of health: A scientific statement from the American Heart Association.Circulation. 2020; 142: e101-e118Crossref PubMed Scopus (191) Google Scholar, such that prevention and intervention programs aimed at raising CRF levels in childhood might have the benefit of persistently high CRF (or tracking) and improving associated health outcomes6Dwyer T. Magnussen C.G. Schmidt M.D. Ukoumunne O.C. Ponsonby A.L. Raitakari O.T. et al.Decline in physical fitness from childhood to adulthood associated with increased obesity and insulin resistance in adults.Diabetes Care. 2009; 32: 683-687Crossref PubMed Scopus (107) Google Scholar across the life course. However, studies of CRF tracking from childhood to adulthood have been limited by small sample sizes7Mikkelsson L. Kaprio J. Kautiainen H. Kujala U. Mikkelsson M. Nupponen H. School fitness tests as predictors of adult health-related fitness.Am. J. Hum. Biol. 2006; 18: 342-349Crossref PubMed Scopus (30) Google Scholar, short follow-up durations 8Sorić M. Jembrek Gostović M. Gostović M. Hočevar M. Mišigoj-Duraković M. Tracking of BMI, fatness and cardiorespiratory fitness from adolescence to middle adulthood: the Zagreb Growth and Development Longitudinal Study.Ann. Hum. Biol. 2014; 41: 238-243Crossref PubMed Scopus (23) Google Scholar, 9Blasquez Shigaki G. L. Barbosa C.C. Batista M.B. Romanzini C.L.P. Gonçalves E.M. Serassuelo Junior H. et al.Tracking of health-related physical fitness between childhood and adulthood.Am. J. Hum. Biol. 2020; 32e23381Crossref PubMed Scopus (15) Google Scholar, 10Trudeau F. Shephard R.J. Arsenault F. Laurencelle L. Tracking of physical fitness from childhood to adulthood.Can. J. Appl. Physiol. 2003; 28: 257-271Crossref PubMed Google Scholar, different measurement methods at baseline and follow-up7Mikkelsson L. Kaprio J. Kautiainen H. Kujala U. Mikkelsson M. Nupponen H. School fitness tests as predictors of adult health-related fitness.Am. J. Hum. Biol. 2006; 18: 342-349Crossref PubMed Scopus (30) Google Scholar,8Sorić M. Jembrek Gostović M. Gostović M. Hočevar M. Mišigoj-Duraković M. Tracking of BMI, fatness and cardiorespiratory fitness from adolescence to middle adulthood: the Zagreb Growth and Development Longitudinal Study.Ann. Hum. Biol. 2014; 41: 238-243Crossref PubMed Scopus (23) Google Scholar,10Trudeau F. Shephard R.J. Arsenault F. Laurencelle L. Tracking of physical fitness from childhood to adulthood.Can. J. Appl. Physiol. 2003; 28: 257-271Crossref PubMed Google Scholar,11Van Oort C. Jackowski S.A. Eisenmann J.C. Sherar L.B. Bailey D.A. Mirwald R. et al.Tracking of aerobic fitness from adolescence to mid-adulthood.Ann. Hum. Biol. 2013; 40: 547-553Crossref PubMed Scopus (11) Google Scholar, and the availability of measurements from only two time-points7Mikkelsson L. Kaprio J. Kautiainen H. Kujala U. Mikkelsson M. Nupponen H. School fitness tests as predictors of adult health-related fitness.Am. J. Hum. Biol. 2006; 18: 342-349Crossref PubMed Scopus (30) Google Scholar,11Van Oort C. Jackowski S.A. Eisenmann J.C. Sherar L.B. Bailey D.A. Mirwald R. et al.Tracking of aerobic fitness from adolescence to mid-adulthood.Ann. Hum. Biol. 2013; 40: 547-553Crossref PubMed Scopus (11) Google Scholar. We aimed to address these gaps by determining the tracking of CRF from childhood to young- and mid-adulthood using data collected from a prospective cohort over 34 years. In 1985, 8,498 children and adolescents aged 7–15 years participated in the Australian Schools Health and Fitness Survey (ASHFS). The ASHFS used a two-stage probability sampling frame (school, then individual) achieving a nationally representative sample of primary and secondary school children across Australia12Dwyer T. Gibbons L.E. The Australian Schools Health and Fitness Survey. Physical fitness related to blood pressure but not lipoproteins.Circulation. 1994; 89: 1539-1544Crossref PubMed Google Scholar. Due to time and economic constraints, all participants that were aged 9, 12, or 15 years had additional technical measurements, including a submaximal graded exercise test. Students of these ages were selected to approximate pre-pubertal, pubertal, and post-pubertal growth stages and formed a subset of N=2,626. Although not originally intended as a cohort study, participants in ASHFS were traced and enrolled into the Childhood Determinants of Adult Health Study and subsequently invited to attend clinics held across Australia when they were adults (young adulthood – data collection from 2004–06 when aged 27–36 years; and mid-adulthood – data collection from 2014–19 when aged 37–49 years). Participants in the adult follow-ups undertook a similar suite of measures that were collected from them as children, including a submaximal graded exercise test. The sample of participants included in this analysis (n = 748) were those aged 9-, 12-, or 15-years who had their CRF levels estimated from the submaximal graded exercise tests performed in ASHFS and at least one adult time-point, 32.5% (n = 243) of participants had their CRF estimated at all three time-points. Figure 1 (online) provides a flowchart of participation. The ASHFS was approved by the State Directors General of Education with participant consent obtained from a parent and assent obtained from the child. The Southern Tasmania Health and Medical Human Research Ethics Committee and the Tasmania Health and Medical Human Research Ethics Committee approved the adult follow-up studies, when participants gave written informed consent. CRF in childhood, young-, and mid-adulthood was estimated using a submaximal graded exercise test – physical working capacity at a heart rate of 170 beats per minute (PWC170). The PWC170 test is a widely used method of estimating CRF with reasonable criterion validity and test-retest reliability13Bland J. Pfeiffer K. Eisenmann J.C. The PWC170: comparison of different stage lengths in 11–16 year oldsareyou我也很.Eur. J. Appl. Physiol. 2012; 112: 1955-1961Crossref PubMed Scopus (34) Google Scholar, 14Gore C.J. Booth M.L. Bauman A. Owen N. Utility of pwc75% as an estimate of aerobic power in epidemiological and population-based studies.Med. Sci. Sports Exerc. 1999; 31: 348-351Crossref PubMed Scopus (43) Google Scholar, 15Rodríguez-Marroyo J.A. Pernía R. Villa J.G. Foster C. Reliability and seasonal changes of submaximal variables to evaluate professional cyclists.Int. J. Sports Physiol. Perform. 2017; 12: 1356-1362Crossref PubMed Scopus (6) Google Scholar, 16Ahmed S. Darwish F. Association between sub-maximal and maximal measures of aerobic power in female adolescents.Egypt. J. Occup. Med. 2014; 38: 227-236Crossref Google Scholar. The test was performed using Monark bicycle ergometers (Monark Exercise AB, Vansbro, Sweden) in childhood (model 818E), young adulthood (model 828E), and mid-adulthood (model 928G3r). At each time-point, participants pedalled at a cadence of 60 revolutions per minute. The test included successive workloads that increased resistance stepwise. Three 3-minute workloads were used in childhood and three 4-minute workloads were used at each adult follow-up. In the final minute of each workload, a research technician recorded the workload setting and the participant’s heart rate. In childhood, heart rate was measured using a stethoscope and a stopwatch. Polar heart rate monitors (Polar Electro Oy, Kempele, Finland) were used in young adulthood (model A5) and mid-adulthood (model RS300X). For each participant, heart rate was regressed on workload and extrapolated (or interpolated if final heart rate was >170 beats per minute) to estimate workload in watts at heart rate of 170 beats per minute (or, PWC170). As the absolute workload achieved in the PWC170 test is a function of muscle mass, measures of PWC170 not attributable to lean body mass were created by regressing PWC170 on lean body mass (methods outlined below) and using the residuals added to the grand mean17Quan H.L. Blizzard C.L. Sharman J.E. Magnussen C.G. Dwyer T. Raitakari O. et al.Resting heart rate and the association of physical fitness with carotid artery stiffness.Am. J. Hypertens. 2013; 27: 65-71Crossref PubMed Scopus (43) Google Scholar. Body mass was measured using regularly calibrated bathroom scales to the closest 0.5 kg in childhood and with Heine scales (Heine, Dover, New Hampshire, USA) to the closest 0.1 kg in adulthood. Height was measured to the closest 0.1 cm employing a KaWe height tape (KaWe Krichner & Wilhelm, Aspeg, Germany) in childhood and a Leicester height measure (Invicta, Leicester, UK) in adulthood. Body mass index was computed as body mass (kg) / [height (m)]2Ross R. Blair S.N. Arena R. Church T.S. Després J.-P. Franklin B.A. et al.Importance of assessing cardiorespiratory fitness in clinical practice: A case for fitness as a clinical vital sign: A scientific statement from the American Heart Association.Circulation. 2016; 134: e653-e699Crossref PubMed Google Scholar. Skinfolds at four sites (biceps, triceps, subscapular, and supra-iliac) were assessed in childhood to the nearest 0.2 mm with Holtain skinfold callipers (Holtain, Crymych, UK), and in adulthood to the nearest 1 mm with Slim Guide callipers (Michigan, USA). Body density was calculated from the log of the sum of four skinfolds using age-specific regression equations18Durnin J.V.G.A. Rahaman M.M. The assessment of the amount of fat in the human body from measurements of skinfold thickness.Br. J. Nutr. 1967; 21: 681-689Crossref PubMed Google Scholar, 19Brook C.G. Determination of body composition of children from skinfold measurements.Arch. Dis. Child. 1971; 46: 182-184Crossref PubMed Google Scholar, 20Durnin J.V.G.A. Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 Years.Br. J. Nutr. 1974; 32: 77-97Crossref PubMed Google Scholar. Body fat was determined from body density21Siri WE. The gross composition of the body. In: Lawrence JH, Tobias CA, editors. Advances in Biological and Medical Physics. 4: Elsevier; 1956. p. 239-279.Google Scholar and lean body mass was approximated as the difference between total body mass and fat mass. Statistical analyses were made using Stata 17.0 (StataCorp, College Station, TX). Participant characteristics are presented as mean and standard deviation stratified by sex. CRF in this study was represented by age- and sex-standardised PWC170 adjusted for lean body mass. Two different statistical approaches were used to quantify the tracking of CRF from childhood to adulthood. Firstly, rank correlations were quantified between CRF measured in childhood and young- or mid-adulthood – a common technique used to assess tracking between two time-points. Secondly, generalized estimating equations (GEE) were used to estimate tracking as stability coefficients, described previously by Twisk and colleagues22Twisk J.W. Kemper H.C. van Mechelen W. Post G.B. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study.Am. J. Epidemiol. 1997; 145: 888-898Crossref PubMed Scopus (184) Google Scholar. This approach allows childhood CRF to be regressed on the change of CRF from the first adult follow-up to the last, with the coefficients at different time points tested simultaneously. The stability coefficients are comparable to the average value of rank correlations between child CRF and each adulthood CRF23Twisk J.W. Kemper H.C. Mellenbergh D.J. van Mechelen W. Factors influencing tracking of cholesterol and high-density lipoprotein: the Amsterdam Growth and Health Study.Prev. Med. 1996; 25: 355-364Crossref PubMed Scopus (34) Google Scholar. Strengths of this approach are that all measurements are considered in the tracking estimate and it is robust to unequal time between follow-ups, as is the case for participants in this study22Twisk J.W. Kemper H.C. van Mechelen W. Post G.B. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study.Am. J. Epidemiol. 1997; 145: 888-898Crossref PubMed Scopus (184) Google Scholar. The stability coefficient ranges between 0 and 1 if the correlations between the measurements is positive (as is assumed in tracking analyses), with a coefficient closer to 1 representing stronger stability22Twisk J.W. Kemper H.C. van Mechelen W. Post G.B. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study.Am. J. Epidemiol. 1997; 145: 888-898Crossref PubMed Scopus (184) Google Scholar. Each model (rank correlation or stability coefficient) was adjusted for duration of follow-up as the young-adult follow-up occurred across a two-year period, and the mid-adulthood follow-up occurred across a and stratified by and baseline 12, and 15 years We tested for by and duration of follow-up. for data at we an approach by et L. and 2012; PubMed Google Scholar. of baseline variables that to follow-up school and baseline were and then probability was used with on the probability of in follow-up using these baseline Participant characteristics are in duration of follow-up from childhood to young adulthood was years to and from childhood to mid-adulthood was years to characteristics at childhood, young-, and adulthood of mass body mass, physical capacity at a heart rate of 170 beats per standard and physical capacity at a heart rate of 170 beats per minute adjusted for lean body mass. in a physical capacity at a heart rate of 170 beats per standard and physical capacity at a heart rate of 170 beats per minute adjusted for lean body mass. rank correlations for tracking of CRF from childhood to young- and the we found that CRF from childhood to young adulthood = from childhood to mid-adulthood = We a correlation of between CRF in young and when duration of follow-up was years This of stronger correlations over of follow-up was across different and we in correlations between and by baseline were and correlation coefficients for tracking of cardiorespiratory fitness from childhood to young- and to young to in a The stability coefficients and for the tracking of CRF from childhood to adulthood are presented in with a stability coefficient of for the We of by or baseline but duration of follow-up to with a stability coefficient estimating equations (GEE) stability coefficients for cardiorespiratory fitness tracking from childhood to stability stability coefficients adjusted for to in a research that CRF to from childhood to adulthood, comparison across studies is to in the methods used to J.W. Kemper H.C. van Mechelen W. Post G.B. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study.Am. J. Epidemiol. 1997; 145: 888-898Crossref PubMed Scopus (184) Google Scholar. study to this body of on the to with one of the follow-up with studies are a of correlations from studies found that CRF from childhood and adolescence to adulthood over a mean duration of follow-up of A. M. R. Tracking of physical fitness levels from childhood and adolescence to adulthood: a review and 2022; PubMed Scopus Google Scholar. childhood to young adulthood correlation (r=0.30) is as it within the estimated from the and mean duration of follow-up is However, the not duration of follow-up to tracking in study of tracking as duration of follow-up the two studies included in the with a duration of follow-up had sample but correlation was within This that the not have been represented by studies with duration of as of a major positive that the have been by including a data for of were comparable with the of the rank J.W. Kemper H.C. Mellenbergh D.J. van Mechelen W. Factors influencing tracking of cholesterol and high-density lipoprotein: the Amsterdam Growth and Health Study.Prev. Med. 1996; 25: 355-364Crossref PubMed Scopus (34) Google Scholar. study had a sample and duration of follow-up studies, and provide into the of duration of follow-up on CRF the to factors from childhood to adulthood help the value of for such factors in childhood that or factors that might be to lifestyle changes that cohort stronger tracking from childhood to adulthood for cholesterol = = C.G. R. Ukoumunne O.C. Dwyer T. A. Factors the stability of blood and levels from to adulthood: from the Childhood Determinants of Adult Health Med. PubMed Scopus Google Scholar, body mass index = = R.J. Schmidt M.D. B.A. H.C. et and obesity from childhood to adulthood: a follow-up of participants in the Australian Schools Health and Fitness J. PubMed Scopus Google Scholar, and = Schmidt M.D. Dwyer T. Magnussen C.G. Tracking of and power from to young adulthood: Longitudinal from the Childhood Determinants of Adult Health Sci. Med. 2017; PubMed Scopus Google Blizzard L. Schmidt M.D. Dwyer T. et across the life The tracking and of between childhood and mid-adulthood in an Australian Sci. Med. PubMed Scopus Google Scholar we for CRF. However, CRF to blood pressure = R. Dwyer T. A. Magnussen C.G. Factors tracking of blood pressure from childhood to adulthood: The Childhood Determinants of Adult Health PubMed Scopus Google Scholar and physical activity = = K. Magnussen C. Dwyer T. A. and the tracking of physical activity and cardiorespiratory fitness from childhood to adulthood.Am. J. Epidemiol. 2009; PubMed Scopus Google Scholar. The for CRF tracking factors is but be related to measurement or a CRF is M.A. Factors levels of physical activity in Med. 32: PubMed Google Scholar, as and data J.W. Kemper H.C. van Mechelen W. Post G.B. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study.Am. J. Epidemiol. 1997; 145: 888-898Crossref PubMed Scopus (184) Google Scholar, interventions aimed at raising CRF in children and CRF into adulthood be physical activity interventions have been to a to CRF levels in childhood and C. A. G. Ponsonby A.L. M. et of interventions for of physical activity on fitness and cardiometabolic in children and a review of 2013; PubMed Scopus Google Scholar. important life stage from childhood to mid-adulthood that physical activity these CRF M.A. Factors levels of physical activity in Med. 32: PubMed Google Blizzard L. Dwyer T. C. Childhood and predictors of time physical activity the from adolescence to adulthood: a cohort J. Nutr. PubMed Scopus Google Scholar and as data have Blizzard L. H. et al.Tracking of cardiorespiratory fitness and physical activity from to young adulthood: from the prospective Google Scholar. levels of S. N. M. et for Cardiorespiratory Fitness Among A Med. 2022; PubMed Scopus Google Scholar and achieving K. Magnussen C. Dwyer T. A. and the tracking of physical activity and cardiorespiratory fitness from childhood to adulthood.Am. J. Epidemiol. 2009; PubMed Scopus Google Scholar a role in CRF tracking and development from childhood to adulthood, into factors is of this study the for if to follow-up was probability with of baseline variables were on the that was to to follow-up was it is that be and the be a but they are not have been to tracking as the of CRF levels over but this is not when are as is the case for childhood S. N. M. et for Cardiorespiratory Fitness Among A Med. 2022; PubMed Scopus Google Scholar. was not might have the and at Study were the follow-up childhood to a and measurement of CRF using a at all three time-points. In CRF is an important health marker in childhood and a clinical of cardiometabolic health in adulthood, tracking data in using CRF levels as a of This the of factors life to or maintain CRF We the of study and including and The Childhood Determinants of Adult Health Study was by and from the of and and the Schools the Heart the Health and Medical Research and the Family The of with with