Early life physical activity

LASA246 (in C)

Contact: Marjolein VIsser

Hypotheses that chronic diseases such as diabetes mellitus, cardiovascular diseases or osteoporosis in older persons may originate from determinants in adulthood or even in childhood or from pre- or postnatal period (Barker hypothesis; Lucas et al., 1999), are widespread. Positive associations have been suggested between early life physical activity and late life cardiovascular health (Twisk et al, 02) and late life cognition (Dik et al, 2003). These associations may originate from a direct pathway between early life physical activity and late life factors, for example from stimulation of growth factors early in life (see Dik et al.). In addition, the associations may also originate from an indirect pathway via late life activity, because early life physical activity may be associated with an active lifestyle later in life (tracking) (Malina, 1996).

Data on lifetime physical activity were collected during the main interview of the LASA-C measurement (1995/1996), and are available for 2,222 subjects.

The questionnaire consists of two parts, the first part asking about the highest physical demand that the respondents had performed almost daily, and the second part asking whether the respondents had participated in sports or any other physical activity that caused them to sweat or that made them exhausted. Both parts are described in more detail below. The questions were asked retrospectively, separately for the periods: 1) between the age of 15 to 25 years; 2) between the age of 26 to 35 years; 3) between the age of 36 to 50 years; 4) after age of 50 years.

Physical demands (question 1)
Physical demands were assessed by asking 'What is the highest physical demand that you had performed almost daily?'. Answers were coded as 'low' (e.g. sitting, office-work), 'moderate' (e.g. salesman, postman), 'high' (e.g. loader/unloader, household for more than four persons), and 'very high' (e.g. farmer, construction worker).

Sports or any other physical activity (question 2)
Lifetime physical activity was assessed by asking 'Did you participate in sports or any other physical activity that caused you to sweat or that made you exhausted?'. Answer categories were: 1) never; 2) sometimes; 3) < 1 hour per week; 4) 1 to 2 hours per week; 5) >2 hours per week (question 2b). If the activity was >2 hours per week, the number of hours per week was registered. Then, if the number of hours per week was 5 hours or more per week, it was registered which activities (maximal 2 activities) the respondent had performed (question 2c). The additional data showed that in this high activity group about 54% was physically active during work compared with 46% who were active in sports.

Previous use in LASA
The definitions and terminology differ between the articles:

Physical demands (question 1)

Sports or any other physical activity (question 2)

Per age period, physical activity was categorized into 4 categories (according to Miranda’s definitions). The answers of the four age periods were summed.

Proposed terminology and definitions for future use
Sports or any other physical activity (question 2)

Question 2 (clphya63-66) Question 2b (clphya67-70) [age period] physical activity
1. never   1. no regular physical activity
2. sometimes       1 or 2
3. <1 hour per week           2. low
4. 1 to 2 hours per week     2. low
5. >2 hours per week   3-9 hours per week 

≥10 hours per week
3. moderate

4. high


Methodological considerations
See article Dik et al. (2003):

First, lifetime physical activity was asked retrospectively with several questions. The questions were not derived from an existing questionnaire. This questionnaire was specifically constructed for the LASA-interview, and has not been validated. The level of physical activity was estimated over a ten-year period many years ago, and it is likely that information bias has occurred. Probable misclassification of the physical activity levels may be non-differential (Falkner et al, 2001), leading to probable underestimation of the associations. However, differential missclassification (i.e. misclassification of the exposition is dependent on the outcome measure or vice versa) may also be possible, for example in cognitively impaired persons, and may lead to an underestimation or overestimation of the associations. The reproducibility of long-term recall in the elderly, including physical activity, is described by Cumming et al (1994).

Second, lifetime physical activity included both sports and occupational activities. Especially in the group with >10 hours per week physical activity (question 2b), the proportion of occupational activities was high (~54%: determined from question 2c). For the other (lower) activity groups, sports and occupational activities could not be separated.

Third, activity levels in the different lifetime periods may overlap, plus may overlap with the physical activity level (measured with LAPAQ) at the 1995/96 interview. The phenomenon that early life physical activity is associated with an active lifestyle later in life, is known as 'tracking' (see Malina, 1996).

Fourth, be alert on possible differences in men and women with regard to total time spent on activities, intensity of the activities, or recall of the activities. Furthermore, the degree of sweating and exaustion may depend on many factors such as sex, body composition and region.


  1. Cumming RG, Klineberg RJ. A study of the reproducibility of long-term recall in the elderly. Epidemiology 1994;5:116-9.
  2. Dik MG, Deeg DJH, Visser M, Jonker C. Early life physical activity and cognition at old age. J Clin Exp Neuropsychol. 2003 Aug;25(5):643-53.
  3. Falkner KL, McCann SE, Trevisan M. Participants characteristics and quality of recall of physical activity in the distant past. Am J Epidemiol 2001;154:865-72.
  4. Lucas A, Fewtrell MS, Cole TJ. Fetal origins of adult disease - the hypothesis revisited. BMJ 1999;319:245-249.
  5. Malina RM. Tracking of physical activity and physical fitness across the lifespan. Res Q Exerc Sport 1996;67(3Suppl):S48-57.
  6. Twisk J, van Mechelen W, Kemper H, Eds. The relationship between physical activity and physical fitness in youth and cardiovascular health in later life. What longitudinal studies can tell. Int J Sports Med 2002;23(1Suppl):S1-S50.
  7. Pluijm SMF, Visser M, Puts MTE, Dik MG, Schalk BMW, van Schoor NM, Schaap L, Bosscher RJ, Deeg DJH. Unhealthy lifestyles during the life course: association with physical decline in late life. Aging Clin Exp Res. 2007 Feb;19(1):75-83.
  8. Verweij LM, Van Schoor NM, Deeg DJH, Dekker J, Visser M. Physical activity and incident clinical knee osteoarthritis in older adults. Arthritis & Rheumatism (Arthritis Care & Research) 2009;61, 2, 152-157.
  9. Peeters GMEE, van Schoor NM, Pluijm SMF, Verweij L, Lips P. Which activities should older persons do at what level to prevent recurrent falling? in progress


Syntax: early life physical work demands
*Start with LASAB016 and LASAC246.

*if 'longest job' is missing, recode as value from 'current job' because in all cases 'current' is also 'longest job' (bjob8).

if (bljtype=-2) bljtype=bcjtype.

  (MISSING=Copy) (1=2)  (2=2)  (4=4)  (5=1) (6=2) (8=3) (9=3) (11=1) (13=2) (15=2) (16=1) (17=3) (18=2)  INTO  bcjob
.VARIABLE LABELS bcjob 'job at B classified into demands of C'.

if missing(clphya54) clphya54=bcjob.
freq clphya54.

*new variable name.
compute bjobdeman=clphya54.
VARIABLE LABELS bjobdeman 'physical demands during work at B'.
VALUE LABELS bjobdeman 1 'low' 2 'medium' 3 'high' 4 'very high'.
recode bjobdeman (-1=sysmis) (-2=sysmis).

*remake original variable
if missing(clphya53) clphya54=-9.
recode clphya54 (-9=sysmis).

Syntax: physical activity at young age (15-26)
*Same set up can be used for other age periods.

*The variables clphya67 thru 70 contain the number of hours spent a week sweating or out of breath.
*The variables clphya63 thru 66 are categorized as:
*3=less than 1 hour a week
*4=1-2 hours a week
*5=more than 2 hours a week

recode clphya63 (-3=-3) (-2=-2) (-1=-1) (1=1) (2=1) (3=2) (4=2) (5=5) into cpa15t25.

recode clphya67 (-3=-3) (-2=-2) (-1=-1) (0 thru 9=3) (10 thru highest=4) into cpa15t25b.

if (cpa15t25=5) cpa15t25= cpa15t25b.
VARIABLE LABELS cpa15t25 'age 15-25 physical activity'.
VALUE LABELS cpa15t25 -3 'na, wrong skip' -2 'na, see CLPHYA51' -1 'na, asked' 1 'no regular PA' 2 'low' 3 'moderate' 4 'high'.