Faced with the high prevalence of
childhood overweight, there is need
for prevention.1,2 Preventive measures
should be based on a detailed analysis
of determinants.3 In children, parental
overweight, social status, lifestyle,
and early life events are known risks
for overweight.4–9 By using linear or
logistic regression analyses, genetic,
lifestyle, and environmental factors
altogether explained up to 30% of the
variance in BMI. The unexplained variance
has led some authors to speculate
on missing and as yet unidentified
determinants of childhood overweight.
Alternatively, the statistical approach
may camouflage the true effect size;
thus, the population-based risks rather
than the relative risks should be
calculated. An attributable risk (AR)
considers both the strength of the association
between a determinant and
the outcome as well as the exposure
prevalence in the population. To illustrate
this point, a risk factor may be
strongly associated with overweight
but may have a low prevalence in the
population. In that case, the risk factor
is less relevant compared with a risk
factor of lower impact affecting a
large proportion of the population.10
ARs can also be used to estimate the
effectiveness of specific interventions
on population health. ARs of childhood
overweight have been calculated in
only a few studies.10–13 This may be
partly because calculation of ARs for
multiple risk factors is a complex issue
and sophisticated methods to calculate
the joint AR have been established only
recently.14,15 Furthermore, methods of
partitioning the joint AR to partial risks
are still in progress and so far there is
no standardization.16–18
The Interdisciplinary Consortium on
Obesity Prevention in Children and
Adolescents (PreVENT) is part of the
German Competence Network on Obesity.
PreVENT aimed to create a portfolio
of preventive measures against
Faced with the high prevalence of
childhood overweight, there is need
for prevention.1,2 Preventive measures
should be based on a detailed analysis
of determinants.3 In children, parental
overweight, social status, lifestyle,
and early life events are known risks
for overweight.4–9 By using linear or
logistic regression analyses, genetic,
lifestyle, and environmental factors
altogether explained up to 30% of the
variance in BMI. The unexplained variance
has led some authors to speculate
on missing and as yet unidentified
determinants of childhood overweight.
Alternatively, the statistical approach
may camouflage the true effect size;
thus, the population-based risks rather
than the relative risks should be
calculated. An attributable risk (AR)
considers both the strength of the association
between a determinant and
the outcome as well as the exposure
prevalence in the population. To illustrate
this point, a risk factor may be
strongly associated with overweight
but may have a low prevalence in the
population. In that case, the risk factor
is less relevant compared with a risk
factor of lower impact affecting a
large proportion of the population.10
ARs can also be used to estimate the
effectiveness of specific interventions
on population health. ARs of childhood
overweight have been calculated in
only a few studies.10–13 This may be
partly because calculation of ARs for
multiple risk factors is a complex issue
and sophisticated methods to calculate
the joint AR have been established only
recently.14,15 Furthermore, methods of
partitioning the joint AR to partial risks
are still in progress and so far there is
no standardization.16–18
The Interdisciplinary Consortium on
Obesity Prevention in Children and
Adolescents (PreVENT) is part of the
German Competence Network on Obesity.
PreVENT aimed to create a portfolio
of preventive measures against
การแปล กรุณารอสักครู่..