Trajectories of Physical Activity Predict the Onset of Depressive Symptoms but Not Their Progression: A Prospective Cohort Study
Table 1
Descriptive statistics of the original sample (–3596).
Variables
Measurement year
Mean ± SD/%
Range
Covariates
Age
1980
3596
10.44 ± 4.99
3–18
Childhood negative emotionality
1980
3177
1.06 ± 0.11
1–2
Parental education†
1980
3540
1.90 ± 0.77
1–3
Parental income
1980
3453
4.80 ± 1.94
1–8
Symptoms of depression
1992
2330
2.14 ± 0.60
1–4.57
Symptoms of depression
1997
2099
2.15 ± 0.67
1–4.57
Symptoms of depression
2001
2097
2.07 ± 0.67
1–4.62
Symptoms of depression
2007
2056
2.06 ± 0.68
1–4.67
Participants’ education‡
2007
2022
2.11 ± 0.90
1–3
Participants’ income
2007
2146
3.50 ± 1.56
1–8
Social support
2007
2055
4.15 ± 0.80
1.08–5.00
Body mass index
2007
2170
26.00 ± 4.75
16.56–58.82
Smoking status
2007
2224
3.81 ± 1.53
1–5
Physical activity indices
Physical activity
1980
2224
9.05 ± 1.83
5–14
Physical activity
1983
2116
9.03 ± 1.88
5–14
Physical activity
1986
2320
8.90 ± 2.01
5–14
Physical activity
1989
2619
8.63 ± 2.10
5–14
Physical activity
1992
2192
9.08 ± 1.92
5–14
Physical activity
2001
2442
8.86 ± 1.96
5–15
Physical activity
2007
2166
8.81 ± 1.81
5–15
Physical activity
2011
1910
9.02 ± 1.88
5–15
Physical activity trajectory groups
Lightly physically active
1980–2011
371
10.4%
Moderately physically active
1980–2011
3046
85.5%
Highly physically active
1980–2011
147
4.1%
Dependent variable
Symptoms of depression (BDI-II)
2012
1724
5.04 ± 6.60
0–58
The original sample size was 3596, and 1764 (49.1%) of the participants were males and 1832 (50.9%) were females. †The frequencies of parents’ educational levels were as follows: low, (34.7%), average, (40.3%), high, and (25.0%). ‡The frequencies of participants’ educational levels were as follows: low, (35.3%), average, (18.6%), high, and (46.1%). Physical activity indices ≤ 7 indicate low, >7 to 10 < moderate, and ≥10 high levels of physical activity. Factors scores, which were predicted from physical activity indices (1980–2011) (see Supplementary Table 3), were used in LCGA.