Skip to main content

Table 3 Association between father’s past recommended vigorous physical activity a (VPA) frequency and youth’s VPA

From: Multivariable analysis of the association between fathers’ and youths’ physical activity in the United States

 

Panel 1: Whether youth participates in weekly VPA

Odds ratio(95% confidence interval)

Model 1

Model 2

Model 3

Model 4

All (N = 887)

1.862***

1.928***

1.921***

2.054***

(1.195 - 2.902)

(1.234 - 3.11)

(1.228 - 3.004)

(1.300 - 3.246)

Male (N = 453)

1.537

1.567

1.589

1.590

(0.835 - 2.830)

(0.849 - 2.893)

(0.859 - 2.939)

(0.855 - 2.959)

Female (N = 434)

1.872+

1.964**

1.920**

2.368***

(0.976 - 3.592)

(1.026 - 3.762)

(1.012 - 3.642)

(1.251 - 4.483)

 

Panel 2: Youths’ weekly VPA frequency (i.e. days/week)

Incident rate ratio (95% confidence interval)

Model 1

Model 2

Model 3

Model 4

All (N = 887)

1.140+

1.154+

1.156+

1.184**

(0.980 - 1.327)

(0.993 - 1.342)

(0.994 - 1.345)

(1.015 - 1.382)

Male (N = 453)

0.987

0.997

0.996

0.983

(0.805 - 1.211)

(0.814 - 1.221)

(0.813 - 1.221)

(0.807 - 1.198)

Female (N = 434)

1.310**

1.322**

1.309**

1.427***

(1.044 - 1.644)

(1.053 - 1.660)

(1.045 - 1.641)

(1.140 - 1.786)

Control measures:

    

Child characteristicsb

YES

YES

YES

YES

Family characteristicsc

YES

YES

YES

NO

Contextual measuresd

YES

YES

NO

NO

Mother’s past recommended VPAe

YES

NO

NO

NO

  1. Notes: Odd ratios (estimated coefficients “βi” transformed to “eβi”) and exponentiated confidence intervals in parenthesis that are robust and clustered using family identifiers are reported from the logistic proability regression analysis for the “Whether Youth Participates in Any Weekly VPA” outcome.
  2. Rate ratios (estimated coefficients “βi” transformed to “eβi” ) and exponentiated confidence intervals in parenthesis that are robust and clustered using family identifiers are reported from negative binomial count data regression analysis for the “Number of Days per Week Youth Participates in VPA” outcome.
  3. All regressions are weighted.
  4. *** denotes significance at p≤0.01; ** denotes significance at p≤0.05; + denotes significance at p≤0.10.
  5. a The independent variable of interest, “Father’s Past Recommended Vigorous Physical Activity Frequency”, is a binary indicator of lagged (t-2 = 2001) recommended weekly number of times of VPA, which takes a value of “1” if father’s past (t-2 = 2001) recommended VPA is ≥3, and “0” otherwise.
  6. bChild characteristics include the following measures: Age, gender, race
  7. cFamily Characteristics include the following measures: Mother’s Education, Mother’s Work Hours, Family Income
  8. dContextual Measures include the following: Local Area Physical Activity-Related Facility Availability, Local Area Median Household Income, Local Area Degree of Urbanization
  9. e “Mother’s Past Recommended VPA” is the lagged (t-2 = 2001) recommended (≥3 times/week) weekly number of times of VPA participation of child’s mother.
  10. McFadden’s Adjusted R-squares for Model 1 (i.e. fully-adjusted model) from the logistic regression analyses reported in Panel 1 are 0.169, 0.137, and 0.247 for the full, male, and female sub-samples, respectively.
  11. McFadden’s Adjusted R-squares for Model 1 (i.e. fully-adjusted model) from the negative binomial regression analyses reported in Panel 2 are 0.025, 0.015, and 0.043 for the full, male, and female sub-samples, respectively.
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us