Physical Activity Across Adulthood and Physical Performance in Midlife

Findings from a British Birth Cohort
  • Rachel Cooper
    Correspondence
    Address correspondence to: Rachel Cooper, PhD, MRC Unit for Lifelong Health and Ageing, 33 Bedford Place, London, WC1B 5JU, United Kingdom
    Affiliations
    MRC Unit for Lifelong Health and Ageing, Division of Population Health, University College London, London, United Kingdom
    Search for articles by this author
  • Gita D. Mishra
    Affiliations
    MRC Unit for Lifelong Health and Ageing, Division of Population Health, University College London, London, United Kingdom

    School of Population Health, University of Queensland, Herston, Australia
    Search for articles by this author
  • Diana Kuh
    Affiliations
    MRC Unit for Lifelong Health and Ageing, Division of Population Health, University College London, London, United Kingdom
    Search for articles by this author

      Background

      Evidence, mainly from cross-sectional studies, suggests that physical activity is a potentially important modifiable factor associated with physical performance and strength in older age. It is unclear whether the benefits of physical activity accumulate across life or whether there are sensitive periods when physical activity is more influential.

      Purpose

      To examine the associations of leisure-time physical activity across adulthood with physical performance and strength in midlife, and to test whether there are cumulative benefits of physical activity.

      Methods

      Using data on approximately 2400 men and women from the UK Medical Research Council National Survey of Health and Development, followed up since birth in March 1946, the associations of physical activity levels during leisure time self-reported prospectively at ages 36, 43, and 53 years with grip strength, standing balance, and chair rise times, assessed by nurses at age 53 years (in 1999), were examined in 2010.

      Results

      There were independent positive effects of physical activity at all three ages on chair rise performance, and at ages 43 and 53 years on standing balance performance, even after adjusting for covariates. These results were supported by evidence of cumulative effects found when using structured life course models. Physical activity and grip strength were not associated in women and, in men, only physical activity at age 53 years was associated with grip strength.

      Conclusions

      There are cumulative benefits of physical activity across adulthood on physical performance in midlife. Increased activity should be promoted early in adulthood to ensure the maintenance of physical performance in later life.

      Introduction

      Maintaining physical performance and muscle strength with age is important given that lower levels in older populations are associated with increased risk of subsequent health problems, loss of independence, and shorter survival times.
      • Guralnik J.M.
      • Simonsick E.M.
      • Ferrucci L.
      • et al.
      A short physical performance battery assessing lower-extremity function—association with self-reported disability and prediction of mortality and nursing-home admission.
      • Cooper R.
      • Kuh D.
      • Hardy R.
      Mortality Review Group
      Objectively measured physical capability levels and mortality: systematic review and meta-analysis.
      • Cooper R.
      • Kuh D.
      • Cooper C.
      • et al.
      Objective measures of physical capability and subsequent health: a systematic review.
      As the global population ages, there is a growing need to identify modifiable factors across life that influence physical performance and strength in later life. Such factors may influence the peak achieved in earlier life or the timing and rate of subsequent decline.
      • Kuh D.
      A life course approach to healthy aging, frailty, and capability.
      • Kuh D.
      • Ben-Shlomo Y.
      Introduction.
      It is therefore necessary to elucidate whether the effects of these factors accumulate across life or are more influential during sensitive periods when intervention to maintain or improve performance and strength is likely to be most beneficial.
      Observational and intervention studies suggest that physical activity is a potentially important modifiable factor associated with physical performance and strength in older age.
      • Brach J.S.
      • Simonsick E.M.
      • Kritchevsky S.
      • Yaffe K.
      • Newman A.B.
      The association between physical function and lifestyle activity and exercise in the health, aging and body composition study.
      • Balzi D.
      • Lauretani F.
      • Barchielli A.
      • et al.
      Risk factors for disability in older persons over 3-year follow-up.
      • Hillsdon M.M.
      • Brunner E.J.
      • Guralnik J.M.
      • Marmot M.G.
      Prospective study of physical activity and physical function in early old age.
      • Lang I.A.
      • Guralnik J.M.
      • Melzer D.
      Physical activity in middle-aged adults reduces risks of functional impairment independent of its effect on weight.
      • Forrest K.Y.
      • Zmuda J.M.
      • Cauley J.A.
      Patterns and correlates of muscle strength loss in older women.
      • Martin H.J.
      • Syddall H.E.
      • Dennison E.M.
      • Cooper C.
      • Aihie Sayer A.
      Relationship between customary physical activity, muscle strength and physical performance in older men and women: findings from the Hertfordshire Cohort Study.
      • Patel K.V.
      • Coppin A.K.
      • Manini T.M.
      • et al.
      Midlife physical activity and mobility in older age—The InCHIANTI Study.
      • Cress M.E.
      • Buchner D.M.
      • Questad K.A.
      • Esselman P.C.
      • de Lateur B.J.
      • Schwartz R.S.
      Exercise: effects on physical functional performance in independent older adults.
      • Pahor M.
      • Blair S.N.
      • Espeland M.
      • et al.
      Effects of a physical activity intervention on measures of physical performance: results of the Lifestyle Interventions and Independence for Elders Pilot (LIFE-P) study.
      • Rantanen T.
      • Parkatti T.
      • Heikkinen E.
      Muscle strength according to level of physical exercise and educational-background in middle-aged women in Finland.
      • Rantanen T.
      • Guralnik J.M.
      • Sakari-Rantala R.
      • et al.
      Disability, physical activity, and muscle strength in older women: The Women's Health and Aging Study.
      • Rejeski W.J.
      • Marsh A.P.
      • Chmelo E.
      • et al.
      The lifestyle interventions and independence for elders pilot (LIFE-P): 2-year follow-up.
      • Visser M.
      • Pluijm S.M.
      • Stel V.S.
      • Bosscher R.J.
      • Deeg D.J.H.
      Physical activity as a determinant of change in mobility performance: the Longitudinal Aging Study Amsterdam.
      However, many of the existing observational studies are cross-sectional, so findings could be explained by reverse causality. In addition, few studies have examined the effect of physical activity earlier in life on physical performance and strength in mid to late adulthood and those that have are limited by the retrospective ascertainment of prior physical activity levels.
      • Patel K.V.
      • Coppin A.K.
      • Manini T.M.
      • et al.
      Midlife physical activity and mobility in older age—The InCHIANTI Study.
      It is therefore unclear whether the beneficial effects of physical activity accumulate over life or whether there are sensitive periods when physical activity is more beneficial.
      Using data from a British birth cohort, the aim of the present study was to examine whether leisure-time physical activity levels at three ages across adulthood were associated with physical performance and strength in midlife. Another aim was to examine whether there was evidence of cumulative benefits of physical activity across adulthood or of sensitive periods when the impact of physical activity was greater than during other periods.

      Methods

      The Medical Research Council National Survey of Health and Development (NSHD) is a socially stratified sample of all births that occurred during 1 week in March 1946 across England, Scotland, and Wales. This cohort of 5362 men and women has been followed up prospectively over 20 times across life from birth onwards. In 1999, when study participants were aged 53 years, 3035 were contacted successfully, of whom 2984 received a home visit from a nurse and 2956 successfully completed at least one of the physical performance or strength tests. Of those 2327 participants not successfully contacted in 1999, a total of 469 had died (8.7% of the original cohort); 948 had refused to participate (17.7%); 580 were abroad (10.8%); and 330 could not be traced (6.2%).
      • Wadsworth M.
      • Kuh D.
      • Richards M.
      • Hardy R.
      Cohort profile: the 1946 National Birth Cohort (MRC National Survey of Health and Development).
      The survey collects data on many aspects of health and lifestyle, including physical activity. The data collection in 1999 received ethical approval from the UK Multicentre Research Ethics Committee (MREC), and informed consent was given by participants to each set of questions and measures undertaken.

      Ascertainment of Physical Performance and Strength

      Physical performance and strength were assessed during the home visits at age 53 years using three objective measures: grip strength, chair rises, and standing balance. Trained nurses conducted these tests using standardized protocols as described elsewhere.
      • Kuh D.
      • Bassey E.J.
      • Butterworth S.
      • Hardy R.
      • Wadsworth M.E.J.
      Grip strength, postural control, and functional leg power in a representative cohort of British men and women: associations with physical activity, health status, and socioeconomic conditions.
      In summary, grip strength was measured isometrically using an electronic handgrip dynamometer.
      • Kuh D.
      • Hardy R.
      • Butterworth S.
      • et al.
      Developmental origins of midlife grip strength: findings from a birth cohort study.
      Two values were recorded for each hand and the highest were used in analyses. Chair rise time was measured as the time taken to rise from a sitting to a standing position with straight back and legs and then sit down again as fast as possible ten times. For high scores to indicate good performance, the reciprocal of the time taken (multiplied by 100) was used. Standing balance time was measured as the longest time, up to a maximum of 30 seconds, for which participants could maintain a one-legged stance in a standard position with their eyes closed. The distribution of these times was positively skewed and so they were normalized using a natural logarithm transformation (ln(seconds)).

      Ascertainment of Physical Activity Levels

      Physical activity levels were ascertained at ages 36, 43, and 53 years during interviews with nurses at the study participants' homes. Different measures for physical activity were included in the surveys administered at the respective ages. At age 36 years, questions were asked about the frequency and duration of participation in 27 different leisure-time activities in the preceding month, based on the Minnesota leisure-time physical activity questionnaire.
      • Taylor H.L.
      • Jacobs D.R.
      • Schucker B.
      • Knudsen J.
      • Leon A.S.
      • Debacker G.
      A questionnaire for the assessment of leisure time physical activities.
      • Kuh D.J.L.
      • Cooper C.
      Physical activity at 36 years: patterns and childhood predictors in a longitudinal study.
      At age 43 years, participation in any sports, vigorous leisure activities, or exercises and how many months in the year and how often in these months each of the activities was done were reported. At age 53 years, participation in any sports, vigorous leisure activities or exercises in study participant's spare time, not including getting to and from work, in the past 4 weeks and the number of occasions on which these activities were undertaken was reported. At each age, participants were categorized as inactive (reported no participation); moderately active (participated in relevant activities one to four times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years); or most active (participated in relevant activities five or more times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years).

      Covariates

      Factors that could confound the main associations were identified a priori. Height (cm) and weight (kg) were measured by nurses during the home visit at age 53 years. Own occupational class at age 53 years (or if not available, the most recent measure in adulthood [n=74]) was categorized using the Registrar General's Social Classification into three groups: I or II (high); IIINM or IIIM (medium); IV or V (low). Educational level attained at age 26 years was categorized into five groups: (1) degree or higher; (2) A levels, usually attained at age 18 years, or their equivalents; (3) O levels, usually attained at age 16 years, or their equivalents; (4) certificate of secondary education, clerical course, or equivalent; and (5) none. Health status at age 53 years was coded as a binary variable and identified those who reported being diagnosed with one or more of the following health conditions in the previous 10 years: diabetes, cancer, epilepsy, or cardiovascular disease (which was defined as having a heart attack or stroke ever, aortic stenosis or valvular disease in the past 10 years, doctor-diagnosed angina or Rose angina Grade I or II, or intermittent claudication). Smoking status at age 53 years was categorized as current, ex, or never smoker.

      Statistical Analyses

      The associations between physical activity at each age and each of the physical performance and strength measures were tested using multiple linear regression models. In these and all subsequent models where there was evidence of an interaction between physical activity and gender, analyses were stratified by gender, but where not, analyses were gender-adjusted. Models were adjusted for current height and weight, then also for occupational class, educational level, smoking status, and health problems and finally also for physical activity at the other two ages with this final set of adjustments testing the independence of the effect of physical activity at each age from the effects of physical activity at other ages.
      In a subsequent stage of analyses, at each of the three ages those classified as inactive were assigned a value of 0, those as moderately active a value of 1, and those as most active a value of 2. The scores from each age were then summed to create a lifetime physical activity score ranging from 0 (inactive at all three ages) to 6 (most active at all three ages). The association between this score (categorized into four groups: 0, 1 or 2, 3 or 4, and 5 or 6) and each of the outcomes was then tested with adjustments made for the same covariates as in previous models.
      This model tested whether there were cumulative effects of physical activity across adulthood, assuming the effects of physical activity at each of the three ages were the same. To test whether an accumulation model such as this, an accumulation model that allows the size of the effect of physical activity at different ages to vary or a sensitive periods model best fit the data, the structured approach described by Mishra et al.
      • Mishra G.
      • Nitsch D.
      • Black S.
      • De Stavola B.
      • Kuh D.
      • Hardy R.
      A structured approach to modelling the effects of binary exposure variables over the life course.
      then was applied. This involved comparing a series of nested models representing the two different accumulation models and a sensitive period model with a fully saturated model that assumes that all of the possible trajectories of physical activity across adulthood are associated with physical performance and strength (Appendix A, available online at www.ajpmonline.org). Large p-values indicate that the nested model fits the data as well as the saturated model and, therefore, that the hypothesis for the nested model is supported by the data. In these analyses, physical activity at each of the three ages were entered separately as linear ordinal terms with adjustments made for the same covariates as in previous models.
      The analyses presented are not weighted and are based on the sample with complete data on physical activity at all three ages, all covariates and at least one of the outcome measures (n=2442). Analyses were also rerun (1) with inclusion of sample weights to allow for the stratified sampling design; (2) restricted to the sample who were inactive at age 53 years; and (3) on maximum available samples, but there were no differences in findings.

      Results

      Men were stronger, had better physical performance levels at age 53 years, and were more likely to be active at ages 36 and 43 years than women (Table 1). Physical activity levels at each age were strongly associated with levels at the other two ages (p<0.01 from chi-square tests). Eighteen percent of the study participants were inactive, and 10% were most active at all three ages.
      Table 1Characteristics of the sample (N=2442), n (%) unless otherwise specified
      Men (n=1189
      Except for grip strength: men=1155, women=1205; chair rising: men=1119, women=1171; standing balance: men=1129, women=1182
      )
      Women (n=1253
      Except for grip strength: men=1155, women=1205; chair rising: men=1119, women=1171; standing balance: men=1129, women=1182
      )
      STRENGTH AND PHYSICAL PERFORMANCE AT AGE 53 YEARS (M [SD])
      Grip strength (kg)47.8 (12.0)27.8 (8.0)
      Chair rise time ((1/s) ×100)
      Reciprocal of time taken for 10 chair rises × 100 (e.g., a value of 5 = 20 seconds to complete 10 chair rises)
      5.3 (1.7)5.0 (1.6)
      Standing balance time (seconds)
      Geometric M and SD
      5.5 (2.2)4.5 (2.0)
      PHYSICAL ACTIVITY AT AGE
      36 years
       Inactive363 (30.5)510 (40.7)
       Moderately active319 (26.8)309 (24.7)
       Most active507 (42.6)434 (34.6)
      43 years
       Inactive562 (47.3)689 (55.0)
       Moderately active281 (23.6)290 (23.1)
       Most active346 (29.1)274 (21.9)
      53 years
       Inactive555 (46.7)621 (50.0)
       Moderately active240 (20.2)210 (16.8)
       Most active394 (33.1)422 (33.7)
      OTHER CHARACTERISTICS AT AGE 53 YEARS
      Weight (kg; M [SD])83.5 (13.2)71.8 (14.3)
      Height (cm; M [SD])174.6 (6.4)161.7 (5.9)
      Occupational class
       I or II (High)614 (51.6)457 (36.5)
       III (Medium)458 (38.5)534 (42.6)
       IV or V (Low)117 (9.8)262 (20.9)
      Educational level
       Degree or higher181 (15.2)65 (5.2)
       A levels or equivalent341 (28.7)299 (23.9)
       O levels or equivalent184 (15.5)327 (26.1)
       CSE, clerical course or equivalent71 (6.0)118 (9.4)
       None412 (34.7)444 (35.4)
      Disabling/life-threatening health conditions
       None1058 (89.0)1104 (88.1)
       One or more131 (11.0)149 (11.9)
      Smoking status
       Never426 (35.8)621 (50.0)
       Ex500 (42.1)356 (28.4)
       Current263 (22.1)276 (22.0)
      Note: The sample includes those with data on covariates (Appendix C, available online at www.ajpmonline.org inlcudes comparison with sample with missing data on covariates), physical activity at all three ages, and at least one of the physical performance or strength measures. A levels are those usually attained at age 18 years; O levels are those usually attained at age 16 years.
      CSE, Certificate of Secondary Education
      a Except for grip strength: men=1155, women=1205; chair rising: men=1119, women=1171; standing balance: men=1129, women=1182
      b Reciprocal of time taken for 10 chair rises × 100 (e.g., a value of 5 = 20 seconds to complete 10 chair rises)
      c Geometric M and SD
      Physical activity levels at ages 36 and 43 years were not associated with grip strength (Table 2), and there was no evidence that these associations differed by gender (p=0.75 and 0.74 from tests of gender interaction, respectively). There was no association between physical activity at age 53 years and grip strength in women, but among men, those who were active at age 53 years had stronger grip strength than those who were inactive (p=0.01 from test of gender interaction), and this was maintained after adjustments. When using the structured approach, these findings were confirmed; the only nested model found to fit the data as well as the saturated model was that representing a sensitive period for physical activity at age 53 years among men (p=0.98).
      Table 2Associations between physical activity levels across adulthood and grip strength at age 53 years (n=2360)
      Physical activity at agen (%)Difference in mean grip strength (kg; 95% CI)
      Model 1Model 2Model 3
      36 years
       Inactive843 (35.7)0.000.000.00
       Moderately active606 (25.7)0.07 (−0.97, 1.12)0.02 (−1.03, 1.07)−0.24 (−1.30, 0.82)
       Most active911 (38.6)0.69 (−0.26, 1.63)0.64 (−0.31, 1.60)0.25 (−0.78, 1.27)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      0.300.330.66
      43 years
       Inactive1210 (51.3)0.000.000.00
       Moderately active550 (23.3)0.55 (−0.46, 1.56)0.53 (−0.48, 1.55)0.10 (−0.95, 1.14)
       Most active600 (25.4)0.47 (−0.52, 1.46)0.47 (−0.53, 1.48)−0.27 (−1.36, 0.82)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      0.470.490.81
      53 years
      Men (n=1155)
        Inactive536 (46.4)0.000.000.00
        Moderately active235 (20.4)3.30 (1.49, 5.11)3.35 (1.50, 5.20)3.44 (1.53, 5.34)
        Most active384 (33.3)2.53 (0.98, 4.08)2.83 (1.24, 4.42)2.95 (1.26, 4.64)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.001<0.001<0.001
      Women (n=1205)
        Inactive596 (49.5)0.000.000.00
        Moderately active199 (16.5)−0.16 (−1.41, 1.09)−0.30 (−1.58, 0.98)−0.30 (−1.59, 0.99)
        Most active410 (34.0)0.67 (−0.32, 1.67)0.59 (−0.44, 1.62)0.54 (−0.54, 1.62)
        p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      0.310.340.41
      Note: Model 1: adjusted for gender (if no evidence of gender interaction in unadjusted model), current height and weight; Model 2: Model 1 plus adult SEP (own occupation and education), smoking and health problems at age 53 years; Model 3: Model 2 plus physical activity levels at the other two ages (including gender × physical activity at age 53 years interaction). Inactive = no participation in relevant activities; moderately active = participated in relevant activities one to four times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years; most active = participated in relevant activities five or more times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years
      a p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      There were graded associations between physical activity levels at all three ages and chair rise performance (Table 3) and no evidence of differences in association by gender (p-values from tests of gender interaction ≥ 0.30). These associations were maintained after adjustments. When the effects of physical activity at each of the three ages were adjusted mutually for each other, the associations of physical activity at each age were shown to be largely independent of each other.
      Table 3Associations between physical activity levels across adulthood and chair rise performance at age 53 years (n=2290)
      Physical activity at agen (%)Difference in mean reciprocal chair rise time (1/s ×100) (95% CI)
      Model 1Model 2Model 3
      36 years
       Inactive797 (34.8)0.000.000.00
       Moderately active592 (25.9)0.22 (0.05, 0.39)0.17 (−0.001, 0.34)0.09 (−0.09, 0.26)
       Most active901 (39.3)0.60 (0.44, 0.75)0.52 (0.37, 0.68)0.34 (0.18, 0.51)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.001<0.001<0.001
      43 years
       Inactive1153 (50.4)0.000.000.00
       Moderately active541 (23.6)0.42 (0.25, 0.58)0.37 (0.20, 0.53)0.26 (0.09, 0.43)
       Most active596 (26.0)0.64 (0.47, 0.80)0.56 (0.39, 0.72)0.35 (0.17, 0.52)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.001<0.001<0.001
      53 years
       Inactive1067 (46.6)0.000.000.00
       Moderately active437 (19.1)0.37 (0.19, 0.55)0.29 (0.10, 0.47)0.18 (−0.01, 0.36)
       Most active786 (34.3)0.55 (0.40, 0.70)0.47 (0.32, 0.62)0.29 (0.13, 0.45)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.001<0.0010.002
      Note: Model 1: adjusted for gender (if no evidence of gender interaction in unadjusted model), current height, and weight; Model 2: Model 1 plus adult SEP (own occupation and education), smoking and health problems at age 53 years; Model 3: Model 2 plus physical activity levels at the other two ages. Inactive = no participation in relevant activities; moderately active = participated in relevant activities one to four times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years; most active = participated in relevant activities five or more times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years.
      a p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      Physical activity levels at all three ages were positively associated with standing balance (Table 4), with no evidence of differences in associations by gender (p-values from tests of gender interaction ≥ 0.09; Table 4). These associations were maintained after adjustments. When the effects of physical activity at each of the three ages were adjusted mutually for each other, the associations with standing balance performance were maintained for physical activity at ages 43 years and 53 years whereas the association of physical activity at age 36 years with balance attenuated.
      Table 4Associations between physical activity levels across adulthood and standing balance performance at age 53 years (n=2311)
      Physical activity at agen (%)Difference in mean ln(standing balance time [s]) (95% CI)
      Model 1Model 2Model 3
      36 years
       Inactive822 (35.6)0.000.000.00
       Less active596 (25.8)0.10 (0.01, 0.18)0.05 (−0.03, 0.13)0.01 (−0.07, 0.09)
       Most active893 (38.6)0.21 (0.14, 0.28)0.14 (0.07, 0.21)0.05 (−0.02, 0.13)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.0010.0010.35
      43 years
       Inactive1166 (50.5)0.000.000.00
       Less active553 (23.9)0.24 (0.16, 0.32)0.19 (0.12, 0.27)0.16 (0.08, 0.24)
       Most active592 (25.6)0.31 (0.23, 0.39)0.24 (0.16, 0.32)0.19 (0.11, 0.27)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.001<0.001<0.001
      53 years
       Inactive1092 (47.3)0.000.000.00
       Less active435 (18.8)0.25 (0.16, 0.33)0.17 (0.09, 0.26)0.13 (0.04, 0.21)
       Most active784 (33.9)0.23 (0.16, 0.30)0.15 (0.08, 0.23)0.08 (0.01, 0.16)
      p-value
      p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      <0.001<0.0010.01
      Note: Model 1: adjusted for gender (if no evidence of gender interaction in unadjusted model), current height and weight; Model 2: Model 1 plus adult SEP (own occupation and education), smoking and health problems at age 53 years; Model 3: Model 2 plus physical activity levels at the other two ages. Inactive = no participation in relevant activities; moderately active = participated in relevant activities one to four times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years; most active = participated in relevant activities five or more times: in the previous month at age 36 years, per month at age 43 years, and in the previous 4 weeks at age 53 years.
      a p-value from likelihood ratio test comparing a model with physical activity at specified age included to a model with physical activity not included
      Graded associations of the lifetime physical activity score with chair rise and standing balance performance were found (Table 5) suggesting that the benefits of physical activity for chair rise and standing balance performance are cumulative across adulthood. This was supported by the findings when using the structured approach that showed that the only nested models to fit the data as well as the fully saturated model were an accumulation model, assuming similar effect sizes at each age (p=0.48), for chair rises and an accumulation model, allowing for differences in effect size at each age (p=0.23), for standing balance.
      Table 5Associations between lifetime physical activity score and physical performance at age 53 years
      Lifetime physical activity scoreChair rise performance (n=2290)Standing balance (n=2311)
      n (%)Difference in M (1/s ×100) (95% CI)n (%)Difference in M ln(s) (95% CI)
      0396 (17.3)0.00416 (18.0)0.00
      1–2749 (32.7)0.24 (0.04, 0.44)744 (32.2)0.07 (−0.02, 0.16)
      3–4676 (29.5)0.52 (0.31, 0.72)687 (29.7)0.20 (0.10, 0.29)
      5–6469 (20.5)0.93 (0.70, 1.15)464 (20.1)0.31 (0.21, 0.42)
      p-value
      p-value from likelihood ratio test comparing a model with the physical activity score included to a model with the score not included
      <0.001<0.001
      Note: Effect estimates presented are adjusted for: gender; current height and weight; adult SEP (own occupation and education); smoking; and health problems at age 53 years
      Lifetime physical activity score derived by assigning those classified as inactive a value of 0, those as moderately active a value of 1, and those as most active a value of 2 at each age and then summing the values for the three ages whereby an individual with a physical activity score of 0 has been categorized as inactive at all three ages, whereas an individual with a physical activity score of 6 has been categorized as most active at all three ages.
      a p-value from likelihood ratio test comparing a model with the physical activity score included to a model with the score not included

      Discussion

      In a nationally representative British population, evidence was found of cumulative benefits of physical activity across adulthood for physical performance in midlife. These associations were robust to adjustment for a range of potential confounding factors. There was also evidence to suggest that higher current physical activity levels were associated with stronger grip strength in men.
      The results with respect to physical performance support other study findings
      • Lang I.A.
      • Guralnik J.M.
      • Melzer D.
      Physical activity in middle-aged adults reduces risks of functional impairment independent of its effect on weight.
      • Patel K.V.
      • Coppin A.K.
      • Manini T.M.
      • et al.
      Midlife physical activity and mobility in older age—The InCHIANTI Study.
      and extend the existing literature by providing evidence of the cumulative benefits of physical activity across adulthood. The most consistent evidence of a cumulative effect of physical activity was found in relation to chair rising. This could be due to the fact that the types of leisure-time physical activity that NSHD study participants undertake are beneficial for lower body strength and power, essential components of good performance in the chair rising test. Likewise, high levels of lifetime physical activity have a positive impact on cardiorespiratory fitness,
      • Wagner E.H.
      • LaCroix A.Z.
      • Buchner D.M.
      • Larson E.B.
      Effects of physical-activity on health-status in older adults 1. Observational studies.
      • Buchner D.M.
      • Beresford S.A.
      • Larson E.B.
      • LaCroix A.Z.
      • Wagner E.H.
      Effects of physical-activity on health-status in older adults 2. Intervention studies.
      which is required to successfully complete the chair rising test. It is also possible that the association operates through body weight; while the association was maintained after adjustment for current weight there could be residual confounding by lifetime weight change. In the NSHD, both standing balance and chair rising ability are more strongly associated than grip strength with neuromuscular speed and control and cognitive performance,
      • Kuh D.
      • Cooper R.
      • Hardy R.
      • Guralnik J.
      • Richards M.
      Lifetime cognitive performance is associated with midlife physical performance in a prospective national birth cohort study.
      which may be influenced by lifetime physical activity.
      The current study is not the only one to find inconsistent evidence of effects of physical activity on strength.
      • Martin H.J.
      • Syddall H.E.
      • Dennison E.M.
      • Cooper C.
      • Aihie Sayer A.
      Relationship between customary physical activity, muscle strength and physical performance in older men and women: findings from the Hertfordshire Cohort Study.
      • Chang M.
      • Jonsson P.V.
      • Snaedal J.
      • et al.
      The effect of midlife physical activity on physical performance among older adults: AGES-Reykjavik study.
      However, some observational studies
      • Forrest K.Y.
      • Zmuda J.M.
      • Cauley J.A.
      Patterns and correlates of muscle strength loss in older women.
      • Rantanen T.
      • Parkatti T.
      • Heikkinen E.
      Muscle strength according to level of physical exercise and educational-background in middle-aged women in Finland.
      • Rantanen T.
      • Guralnik J.M.
      • Sakari-Rantala R.
      • et al.
      Disability, physical activity, and muscle strength in older women: The Women's Health and Aging Study.
      have found associations between physical activity and grip strength among women. The findings in this paper may be in contrast to other studies because of the younger age of the sample. Differences in findings could be also due to differences in the types of activity undertaken.
      Intervention studies
      • Cress M.E.
      • Buchner D.M.
      • Questad K.A.
      • Esselman P.C.
      • de Lateur B.J.
      • Schwartz R.S.
      Exercise: effects on physical functional performance in independent older adults.
      • Buchner D.M.
      • Beresford S.A.
      • Larson E.B.
      • LaCroix A.Z.
      • Wagner E.H.
      Effects of physical-activity on health-status in older adults 2. Intervention studies.
      • Munn J.
      • Herbert R.D.
      • Hancock M.J.
      • Gandevia S.C.
      Resistance training for strength: effect of number of sets and contraction speed.
      demonstrating a beneficial effect of physical activity on strength have implemented specific training regimes designed to improve strength and performance which may not be equivalent to the type of physical activity generally undertaken. These results suggest that among the NSHD participants, recreational physical activity of the correct type, intensity, or frequency to beneficially affect upper body strength, as demonstrated in trials,
      • Buchner D.M.
      • Beresford S.A.
      • Larson E.B.
      • LaCroix A.Z.
      • Wagner E.H.
      Effects of physical-activity on health-status in older adults 2. Intervention studies.
      is not being taken sufficiently often especially among women. Men are more likely than women to undertake leisure-time activities such as racquet sports, martial arts, and weight training
      • Stamatakis E.
      • Chaudhury M.
      Temporal trends in adults' sports participation patterns in England between 1997 and 2006: the Health Survey for England.
      that improve and maintain their upper body strength; this may explain the gender difference in association that was found.
      A strength of the present study is the prospective collection of information on physical activity at multiple times across adulthood. Although these measures were self-reported, they have each been found to be correlated in expected directions with other health behaviors and BMI. Further, the Minnesota leisure-time physical activity questionnaire
      • Taylor H.L.
      • Jacobs D.R.
      • Schucker B.
      • Knudsen J.
      • Leon A.S.
      • Debacker G.
      A questionnaire for the assessment of leisure time physical activities.
      on which the assessment at age 36 years was based was shown to have a high 1-month reliability and was correlated with a treadmill estimation of oxygen uptake and body composition.
      • Jacobs D.R.
      • Ainsworth B.E.
      • Hartman T.J.
      • Leon A.S.
      A simultaneous evaluation of 10 commonly used physical-activity questionnaires.
      Nonetheless, improvements in objective measures of physical activity will be needed to validate these relationships.
      Another strength is the availability of objective measures of physical performance and strength. These are expected to be less subject to bias than self-reported measures used in some other studies.
      • Hillsdon M.M.
      • Brunner E.J.
      • Guralnik J.M.
      • Marmot M.G.
      Prospective study of physical activity and physical function in early old age.
      Assessing these outcomes in midlife makes it less likely that findings will be explained by comorbidities. In older populations, these are likely to have an impact on physical activity as well as performance and strength and to confound associations.
      To enable physical activity across adulthood to be examined, participation in sports, recreational, and leisure-time activities was the main explanatory factor selected as information on this had been ascertained at all three ages. It is acknowledged that this information was not ascertained using the same instrument at each age; however, these measures are associated. Further, these measures do not take into consideration work time, active travel, or domestic physical activity. However, the amount of activity that people undertake as part of their daily lives is decreasing, and participation in leisure-time activity is increasingly necessary to ensure that the recommended levels of physical activity are met.
      The Chief Medical Officer
      At least five a week: evidence on the impact of physical activity and its relationship to health.
      • Stamatakis E.
      • Ekelund U.
      • Wareham N.J.
      Temporal trends in physical activity in England: the Health Survey for England 1991 to 2004.
      Leisure-time physical activity is therefore likely to be a reasonably representative indicator of overall physical activity levels and an important target for intervention.
      The NSHD cohort was established using a sampling frame that ensured that it was nationally representative of the population born in England, Scotland, and Wales in 1946. Since then, losses to follow-up due to death, emigration, loss of contact, and permanent refusal have occurred. Despite this, at age 53 years, the sample remained representative of the national population born at a similar time in most respects,
      • Wadsworth M.
      • Kuh D.
      • Richards M.
      • Hardy R.
      Cohort profile: the 1946 National Birth Cohort (MRC National Survey of Health and Development).
      • Wadsworth M.E.J.
      • Butterworth S.L.
      • Hardy R.J.
      • et al.
      The life course prospective design: an example of benefits and problems associated with longevity.
      and so findings should be generalizable to the generation of postwar baby boomers currently reaching old age. In addition, 514 of the 2956 people with at least one valid outcome measure were excluded from analyses because of missing data.
      When the characteristics of those excluded were compared with those included, there were no differences in the distributions of most key characteristics (Appendix B, available online at www.ajpmonline.org). However, those excluded were more likely to be inactive at age 53 years, have no educational qualifications, and be current smokers than those included. It is not expected that these differences would introduce substantial bias. There was also a small proportion of study participants who were unable to perform one or more of the outcome assessments (grip strength n=69, chair rising n=154, standing balance n=113). These people were more likely to have health problems, poorer performance on the other tests, and a lower educational level (results not shown). Those people unable to perform the chair rise and standing balance tests were also more likely to be inactive, so by excluding these people the associations found may be weaker than they would have been if these individuals had been included.
      The findings in relation to chair rising and standing balance performance suggest that promotion of leisure-time physical activity across adulthood would have beneficial effects on physical performance later in life and hence the functional health and quality of life of the aging population, especially as the size of the differences in performance detected may be clinically relevant. Promotion of leisure-time activity is likely to become increasingly important in younger populations as people's daily routines become more sedentary.
      The Chief Medical Officer
      At least five a week: evidence on the impact of physical activity and its relationship to health.
      That not all people categorized as most active in the current study will meet the recommended level of at least 30 minutes of moderate-intensity physical activity five times a week
      The Chief Medical Officer
      At least five a week: evidence on the impact of physical activity and its relationship to health.
      (e.g., only approximately 35% of people in the most active group at age 53 years reported this frequency of activity) suggests that only low levels of physical activity need to be achieved for there to be beneficial effects on physical performance. Findings in relation to grip strength suggest that specific exercises and activities may need to be promoted to ensure that people undertake physical activity that is beneficial for upper body strength.
      It has been proposed that the associations of physical performance and strength with mortality rates and cardiovascular disease
      • Cooper R.
      • Kuh D.
      • Hardy R.
      Mortality Review Group
      Objectively measured physical capability levels and mortality: systematic review and meta-analysis.
      • Cooper R.
      • Kuh D.
      • Cooper C.
      • et al.
      Objective measures of physical capability and subsequent health: a systematic review.
      • Hamer M.
      • Kivimaki M.
      • Lahiri A.
      • et al.
      Walking speed and subclinical atherosclerosis in healthy older adults: the Whitehall II study.
      may be explained by the fact that performance and strength are acting as markers of lifetime physical activity.
      • Hamer M.
      • Kivimaki M.
      • Lahiri A.
      • et al.
      Walking speed and subclinical atherosclerosis in healthy older adults: the Whitehall II study.
      • Harwood R.H.
      • Conroy S.P.
      Slow walking speed in elderly people.
      The results with respect to physical performance provide support for this explanation but the findings with respect to grip strength suggest that this is unlikely to fully explain the grip strength–mortality associations.
      • Cooper R.
      • Kuh D.
      • Hardy R.
      Mortality Review Group
      Objectively measured physical capability levels and mortality: systematic review and meta-analysis.

      Conclusion

      Increased activity should be promoted early in adulthood to ensure the maintenance of physical performance in later life. When promoting physical activity, it may be necessary to encourage people to participate in specific types of activity in order for a beneficial effect on upper body strength to be seen.
      RC and DK are supported by the UK Medical Research Council. GM receives support from the Australian National Health and Medical Research Council ( APP1000986 ). RC also receives support from the HALCyon programme funded by the New Dynamics of Ageing ( RES-353-25-0001 ).
      No financial disclosures were reported by the authors of this paper.

      Supplementary data

      References

        • Guralnik J.M.
        • Simonsick E.M.
        • Ferrucci L.
        • et al.
        A short physical performance battery assessing lower-extremity function—association with self-reported disability and prediction of mortality and nursing-home admission.
        J Gerontol. 1994; 49: M85-M94
        • Cooper R.
        • Kuh D.
        • Hardy R.
        • Mortality Review Group
        Objectively measured physical capability levels and mortality: systematic review and meta-analysis.
        Br Med J. 2010; 341: c4467
        • Cooper R.
        • Kuh D.
        • Cooper C.
        • et al.
        Objective measures of physical capability and subsequent health: a systematic review.
        Age Ageing. 2011; 40: 14-23
        • Kuh D.
        A life course approach to healthy aging, frailty, and capability.
        J Gerontol A Biol Sci Med Sci. 2007; 62: 717-721
        • Kuh D.
        • Ben-Shlomo Y.
        Introduction.
        in: Kuh D. Ben-Shlomo Y. A life course approach to chronic disease epidemiology. 2nd ed. Oxford University Press, Oxford2004
        • Brach J.S.
        • Simonsick E.M.
        • Kritchevsky S.
        • Yaffe K.
        • Newman A.B.
        The association between physical function and lifestyle activity and exercise in the health, aging and body composition study.
        J Am Geriatr Soc. 2004; 52: 502-509
        • Balzi D.
        • Lauretani F.
        • Barchielli A.
        • et al.
        Risk factors for disability in older persons over 3-year follow-up.
        Age Ageing. 2010; 39: 92-98
        • Hillsdon M.M.
        • Brunner E.J.
        • Guralnik J.M.
        • Marmot M.G.
        Prospective study of physical activity and physical function in early old age.
        Am J Prev Med. 2005; 28: 245-250
        • Lang I.A.
        • Guralnik J.M.
        • Melzer D.
        Physical activity in middle-aged adults reduces risks of functional impairment independent of its effect on weight.
        J Am Geriatr Soc. 2007; 55: 1836-1841
        • Forrest K.Y.
        • Zmuda J.M.
        • Cauley J.A.
        Patterns and correlates of muscle strength loss in older women.
        Gerontology. 2007; 53: 140-147
        • Martin H.J.
        • Syddall H.E.
        • Dennison E.M.
        • Cooper C.
        • Aihie Sayer A.
        Relationship between customary physical activity, muscle strength and physical performance in older men and women: findings from the Hertfordshire Cohort Study.
        Age Ageing. 2008; 37: 589-593
        • Patel K.V.
        • Coppin A.K.
        • Manini T.M.
        • et al.
        Midlife physical activity and mobility in older age—The InCHIANTI Study.
        Am J Prev Med. 2006; 31: 217-224
        • Cress M.E.
        • Buchner D.M.
        • Questad K.A.
        • Esselman P.C.
        • de Lateur B.J.
        • Schwartz R.S.
        Exercise: effects on physical functional performance in independent older adults.
        J Gerontol A Biol Sci Med Sci. 1999; 54: M242-M248
        • Pahor M.
        • Blair S.N.
        • Espeland M.
        • et al.
        Effects of a physical activity intervention on measures of physical performance: results of the Lifestyle Interventions and Independence for Elders Pilot (LIFE-P) study.
        J Gerontol A Biol Sci Med Sci. 2006; 61: 1157-1165
        • Rantanen T.
        • Parkatti T.
        • Heikkinen E.
        Muscle strength according to level of physical exercise and educational-background in middle-aged women in Finland.
        Eur J Appl Physiol Occup Physiol. 1992; 65: 507-512
        • Rantanen T.
        • Guralnik J.M.
        • Sakari-Rantala R.
        • et al.
        Disability, physical activity, and muscle strength in older women: The Women's Health and Aging Study.
        Arch Phys Med Rehabil. 1999; 80: 130-135
        • Rejeski W.J.
        • Marsh A.P.
        • Chmelo E.
        • et al.
        The lifestyle interventions and independence for elders pilot (LIFE-P): 2-year follow-up.
        J Gerontol A Biol Sci Med Sci. 2009; 64: 462-467
        • Visser M.
        • Pluijm S.M.
        • Stel V.S.
        • Bosscher R.J.
        • Deeg D.J.H.
        Physical activity as a determinant of change in mobility performance: the Longitudinal Aging Study Amsterdam.
        J Am Geriatr Soc. 2002; 50: 1774-1781
        • Wadsworth M.
        • Kuh D.
        • Richards M.
        • Hardy R.
        Cohort profile: the 1946 National Birth Cohort (MRC National Survey of Health and Development).
        Int J Epidemiol. 2006; 35: 49-54
        • Kuh D.
        • Bassey E.J.
        • Butterworth S.
        • Hardy R.
        • Wadsworth M.E.J.
        Grip strength, postural control, and functional leg power in a representative cohort of British men and women: associations with physical activity, health status, and socioeconomic conditions.
        J Gerontol Med Sci. 2005; 60A: 224-231
        • Kuh D.
        • Hardy R.
        • Butterworth S.
        • et al.
        Developmental origins of midlife grip strength: findings from a birth cohort study.
        J Gerontol Med Sci. 2006; 61: 702-706
        • Taylor H.L.
        • Jacobs D.R.
        • Schucker B.
        • Knudsen J.
        • Leon A.S.
        • Debacker G.
        A questionnaire for the assessment of leisure time physical activities.
        J Chronic Dis. 1978; 31: 741-755
        • Kuh D.J.L.
        • Cooper C.
        Physical activity at 36 years: patterns and childhood predictors in a longitudinal study.
        J Epidemiol Community Health. 1992; 46: 114-119
        • Mishra G.
        • Nitsch D.
        • Black S.
        • De Stavola B.
        • Kuh D.
        • Hardy R.
        A structured approach to modelling the effects of binary exposure variables over the life course.
        Int J Epidemiol. 2009; 38: 528-537
        • Wagner E.H.
        • LaCroix A.Z.
        • Buchner D.M.
        • Larson E.B.
        Effects of physical-activity on health-status in older adults.
        Annu Rev Public Health. 1992; 13: 451-468
        • Buchner D.M.
        • Beresford S.A.
        • Larson E.B.
        • LaCroix A.Z.
        • Wagner E.H.
        Effects of physical-activity on health-status in older adults.
        Annu Rev Public Health. 1992; 13: 469-488
        • Kuh D.
        • Cooper R.
        • Hardy R.
        • Guralnik J.
        • Richards M.
        Lifetime cognitive performance is associated with midlife physical performance in a prospective national birth cohort study.
        Psychosom Med. 2009; 71: 38-48
        • Chang M.
        • Jonsson P.V.
        • Snaedal J.
        • et al.
        The effect of midlife physical activity on physical performance among older adults: AGES-Reykjavik study.
        Gerontologist. 2008; 48: 105
        • Munn J.
        • Herbert R.D.
        • Hancock M.J.
        • Gandevia S.C.
        Resistance training for strength: effect of number of sets and contraction speed.
        Med Sci Sports Exerc. 2005; 37: 1622-1626
        • Stamatakis E.
        • Chaudhury M.
        Temporal trends in adults' sports participation patterns in England between 1997 and 2006: the Health Survey for England.
        Br J Sports Med. 2008; 42: 901-908
        • Jacobs D.R.
        • Ainsworth B.E.
        • Hartman T.J.
        • Leon A.S.
        A simultaneous evaluation of 10 commonly used physical-activity questionnaires.
        Med Sci Sports Exerc. 1993; 25: 81-91
        • The Chief Medical Officer
        At least five a week: evidence on the impact of physical activity and its relationship to health.
        Department of Health, London2004
        • Stamatakis E.
        • Ekelund U.
        • Wareham N.J.
        Temporal trends in physical activity in England: the Health Survey for England 1991 to 2004.
        Prev Med. 2007; 45: 416-423
        • Wadsworth M.E.J.
        • Butterworth S.L.
        • Hardy R.J.
        • et al.
        The life course prospective design: an example of benefits and problems associated with longevity.
        Soc Sci Med. 2003; 57: 2193-2205
        • Hamer M.
        • Kivimaki M.
        • Lahiri A.
        • et al.
        Walking speed and subclinical atherosclerosis in healthy older adults: the Whitehall II study.
        Heart. 2010; 96: 380-384
        • Harwood R.H.
        • Conroy S.P.
        Slow walking speed in elderly people.
        Br Med J. 2009; 339: 1153-1154