  |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
03-July-2009, Vol 122 No 1298
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
New Zealand’s Pacific population continues to be one
of the fastest growing and significant ethnic groups in New
Zealand,1 with approximately 6.9% residents
identifying as being of Pacific descent.2 Yet,
Pacific peoples residing in New Zealand also continue to be socially
disadvantaged relative to other New
Zealanders.3
Findings from the National Children’s Nutrition Survey
and 2006/07 New Zealand Health Survey (NZHS) showed that approximately 55-60% of
Pacific children are overweight or obese; substantially higher than other New
Zealand children.4,5
These statistics have immediate negative implications for
child health including increased risk of exhibiting type 2 diabetes, risk
factors for cardiovascular disease, experiencing obstructive sleep apnoea,
orthopaedic complications, a reduced quality of life, and negative psychosocial
effects stemming from discrimination and preoccupation with
weight.6–8 Further, obesity in childhood
is associated with significant long-term developmental consequences, including
an increased risk of exhibiting overweight/obesity and associated morbidities in
later life.9
Obesity is predominantly the result of an “energy
gap”, the excess of energy intake over energy
expenditure.10 These energy gaps can be
relatively small (e.g. ~20–30 kcal/day), and therefore could be easily
reduced by small increases in physical activity (PA).
Participation in regular PA is fundamental to obesity
prevention and treatment in children.11 PA also
confers other important benefits in children, including improved bone
health12 and cognitive
function,13 a decreased risk of developing type
2 diabetes,14,15 and a reduced risk of
exhibiting cardiovascular disease risk
factors.16
Conversely, physical inactivity (often quantified using
television or screen time) is associated with a multitude of both short-term and
long-term negative health outcomes.17-19
Consequently, PA promotion for obesity prevention in Pacific children is a
public health priority in New Zealand. The Healthy Eating – Healthy
Action strategy highlighted children and young people, as well as Pacific
peoples, as two groups that will achieve the greatest benefits from
participating in more PA, and the strategy calls for research to address
physical in/activity and obesity in Pacific
peoples.20
Nationwide assessment of children’s PA levels to date
has been conducted using self-report, or parental proxy-report measures only,
both of which are inherently biased.21
Notwithstanding this, it appears that Pacific children may actually be more
physically active than their European and Māori counterparts, but at
increased risk of developing obesity and other lifestyle related diseases due to
increased participation in sedentary pursuits such as television
watching.4
Culturally appropriate, effective, and integrated programmes
are urgently required at national and community levels to combat the rising
problem of obesity, and to promote healthy lifestyles and well-being for Pacific
children and families. Accurate quantification of even small changes in PA and
sedentary behaviour is fundamental to understanding associates of PA and health
gain, and informing effective programme development.
The Pacific Islands Families (PIF) study offers a unique
opportunity to meet this research need. The PIF study follows a cohort of
Pacific infants born at Middlemore Hospital, South Auckland, between 15 March
and 17 December 2000.
General aims of the PIF study are to:
In-depth information on parent
and child health and social, demographic, cultural, and lifestyle factors has
been collected from mothers, fathers, and children when the children were 6
weeks, 12 and 24 months, and 4 and 6
years.22,23
At the 6-year PIF measurement phase (2006), an additional
study (Pacific Islands Families: Child and Parental Physical Activity and Body
Size [PIF:PAC]) was conducted, using accelerometry to gather a precise and
objective measurement of children’s PA and to identify supports for and
barriers to children’s activity. The current paper provides a detailed
description of the PIF:PAC study design, methodology, and study
population.
MethodsDesign—The PIF:PAC was a
separate nested sub-study, designed to investigate PA levels, sedentary
behaviours, and associates of activity behaviours in children and mothers
participating in the PIF study’s 6-year measurement wave. Existing PIF
protocols for home visits and data collection, entry, accuracy, storage, and
security were adhered to.23 Additional
protocols and measures specific to the PIF:PAC are described below.
Aims—Specific study aims were
to:
Participant recruitment and
enrolment—A comprehensive description of the PIF recruitment
process has been published elsewhere.23 The
full PIF cohort comprises 1398 children and their families. High retention has
been achieved to date, with 910 children and 1066 mothers participating in the
4-years measurement phase. The full cohort were revisited in 2006 when the
children had their sixth birthdays, with the exception of those who had
withdrawn from the study over the past 6 years, and those not currently living
in New Zealand.
The nature of this longitudinal study allows for those
who have not participated in some earlier assessments, to still be eligible for
subsequent assessments. The PIF:PAC study ran concurrently with measurement of
the substantive PIF variables at 6 years. Due to funding constraints, the first
393 (386 families) eligible PIF participants only were also invited to
participate in the PIF:PAC component of the study at the first home visit (see
Figure 1).
Participation in the PIF:PAC study was not required for
ongoing involvement with the longitudinal PIF study. Mothers willing to
participate provided informed consent, and assent was given by participating
children. Mothers consenting to participate in the PIF:PAC study were visited at
their home on a separate occasion (home visit 2).
At this second home visit, maternal body size was
measured and accelerometers provided to participating mothers and children, and
written and verbal instructions on their use given. Families were visited
approximately 8 days later to collect the accelerometers and pedometers and
gather compliance information and participant feedback.
Measures—Figure 1 outlines the
measurements specific to the PIF:PAC study that were taken at the 6-year data
collection point; detailed descriptions are also provided below.
Child and maternal physical
activity—Children and their mothers were visited at their homes
by a trained PIF researcher and asked to wear an Actical accelerometer (Mini
Mitter, Bend, OR) on a purpose-made elastic waistband for 8 consecutive days
(including water activities).
Participants were asked to wear the elastic waistband
with the accelerometer sitting above the right
hip;24 appropriate placement was demonstrated
to the participants and written instructions for accelerometer use were also
provided. Accelerometers were set to collect data in 60-second epochs. The
Actical accelerometer is lightweight, water resistant, and contains a
piezoelectric sensor that detects movement and acceleration over all planes of
movement. This monitor has been validated using indirect calorimetry in both
children and adults,24 and is the only
accelerometer that measures omni-directional movement, making it most suitable
for use with children.25
A measurement period of at least 7 days is recommended
to gather a reliable estimate of usual activity, and also enables the comparison
of activity during week days and weekend
days.26 Accelerometers were collected by the
same researcher approximately 8 days after the initial home visit; information
on accelerometer problems and participant compliance was recorded at this
time.
Figure 1. Measurement battery for the PIF:PAC
study
 Maternal barriers and facilitators of physical
activity—Questions to the mothers about barriers and facilitators
of PA participation were included within the standard parent interview protocol
within the substantive PIF Study. Items from the Obstacles to Action
survey27 were used, to allow comparisons with
nationally representative data. This survey was implemented in a nationwide
study to identify at-risk groups for insufficient activity, and specific
motivators and barriers for physical activity and inactivity for differing
groups using a comprehensive range of questions related to attitudes and
opinions, individual health, health behaviours, and demographics. Thirty-six
items from the physical activity section of this survey (copied verbatim from
sections 6 and 8) were utilised in the PIF:PAC study. The interviewer read the
survey questions to the parents and recorded their responses.
Maternal body
size—Mothers’ height was measured to the nearest 0.1 cm
using a stadiometer, and weight was assessed using Seca scales to the nearest
0.1kg with the parent in light clothing. Body mass index was calculated as
weight (kg) / height (m)2. Ethnic-specific
cut-offs for overweight (26 kg/m2) and obesity
(32 kg/m2) were used to determine weight
status.28 Waist circumference was measured at
the mid axillary line (halfway between the top of the hipbone and the lower rib)
to the nearest 0.1 cm and thresholds for high trunk fat mass
applied.29
Two serial measurements were made for each body size
measurement, and the average calculated. If the difference between two readings
exceeded 0.5 cm, 0.5 kg, or 1 cm for height, weight, or waist circumference,
respectively, a third recording was taken, and the average of the two closest
readings taken. These measurements were taken to complement the anthropometric
measures (height, weight, waist circumference, body fatness) that were taken of
the children during the child assessment at their school as part of the PIF
study.
For the children, gender-specific thresholds for high
trunk fat mass developed with New Zealand children aged 5.9 years using waist
circumference values were applied.30
International gender-specific thresholds for 6-year-old children were used to
define overweight and obesity.31
Standard PIF study measures:
The following information has been obtained at the first 4 measurement
points (6 weeks, 12 and 24 months, 4 years) through interviews and direct child
assessments:
These structured interviews were
repeated with mothers and fathers of the children, and the children themselves
at the 6-year measurement phase. Additional data were collected at this time
regarding child body size, nutritional practices, and physical activity (basic
step-based information and questionnaire), as below:
Data accuracy and
security—Standard PIF study protocols include a variety of
systems to ensure data accuracy and consistency, including: manual coding of
each interview (to identify potentially spurious data at the time of data
entry), accompanying interviewers to gauge rapport and conduct, participant
coding to ensure no individual can be identified from the data, and post
interview random phone checks with participants (to confirm specific interview
details). All data collected are treated as sensitive information.
Participants own their data and have the right to
withdraw this information from the study at any time. Additional checks were
completed to determine accelerometer data quality and accuracy for the PIF:PAC
study. This involved manual scanning of activity graphs for each participant and
identifying corrupt (e.g. constant accelerometer count values for extended
periods, scrambled data, and so on) or empty (i.e. no data due to unit not being
worn) files and potentially erroneous data (e.g. 0 activity counts for >30
minutes, activity counts exceeding 12,000/60-second epoch, activity counts for
>16 hour time periods).
In cases where files were corrupt or empty, these data
were excluded from further analyses. Where data were questionable, information
was confirmed using the participant compliance information.
Statistical analyses—For the
purposes of the current paper, descriptive statistics (frequencies and
percentages) of basic sociodemographic variables measured were calculated for
participants of the PIF:PAC study and compared with those for the full PIF
cohort using the PIF study baseline data (6 weeks).
Ethics—Ethical approval for the
PIF and related studies has been obtained from the Auckland Branch of the
National Ethics Committee, the Royal New Zealand Plunket Society and the South
Auckland Health Clinical Board. Conduct of the study complied with the ethical
standards for human experimentation as established by the Helsinki
Declaration.
ResultsFrom the original PIF cohort at 6 weeks (N=1376), 1001
mothers participated in the 6-year PIF measurement phase (32% attrition from
those eligible at baseline; see Figure 2). Due to funding constraints, the first
386 families were also invited to participate in the PIF:PAC study. Of those
invited, 254 (66%) mothers consented for themselves and their child(ren) to
participate in the PIF:PAC study, and child assent was gathered for all of these
children.
Figure 2. Recruitment characteristics of the
PIF:PAC study and PIF study at each measurement phase
 Table 1 contains descriptive statistics for basic baseline
(6-weeks postpartum) demographic factors of the full PIF cohort and those
participating in the PIF:PAC study. The characteristics of participants in the
PIF:PAC study were broadly similar to the characteristics measured in the
overall cohort. Of those consenting to participate in the PIF:PAC study (254
mothers, 261 children), usable accelerometer data were gathered for 173 (68%)
mothers and 200 (77%) children. Seven of these children were twins; in these
cases, only the first born twin was included.
Table 1. Frequencies (%) of demographic factors
for maternal participants enrolled in the PIF (N=1376) and PIF:PAC (N=254)
studies using baseline PIF data (6 weeks)
Note: 247 mothers in the PIF:PAC study
had singletons and 7 had twins – where twins were measured, only the first
born twin was included; a1 observation invalid;
bincludes mothers identifying equally with two
or more Pacific groups; cincludes non-Pacific
mothers who were eligible through the Pacific ethnicity of the father.
Considerable data were lost due to participant
non-compliance, researchers not being able to contact the mothers for the second
home visit, and accelerometer hardware failure. Descriptive statistics of the
accelerometer data collected and missing data are provided in Table 2.
Table 2. Descriptive information for
accelerometer data collected in children and their mothers participating in the
PIF:PAC study at 6 years (N=254 mothers, 254 children)
Height and weight measurements were taken for 238 (94%)
participating mothers and 248 (98%) children. Using gender-specific and
ethnic-specific body mass index values to classify weight status, 97% of
mothers, 58% of girls, and 61% of boys were considered overweight or obese (see
Table 3). Waist circumference was measured in 230 mothers and 227 children; of
these, a high trunk fat mass was found in 97% of mothers, 53% of girls, and 60%
of boys. Barriers to and facilitators of physical activity were also assessed in
254 (100%) mothers participating in the PIF:PAC study (data not reported
here).
Table 3. Body size measurements of children and
their mothers participating in the PIF:PAC study at 6 years
(N=254)
BMI=body mass index status using ethnic-specific
thresholds for adults,28 and international
thresholds for children.31
DiscussionInternationally, the PIF:PAC is one of the first large-scale
epidemiological studies to use accelerometry for objective PA measurement in
children and their mother. With the developed world currently suffering from an
obesity epidemic, results from this study will provide much sough after evidence
in the relationship between familial PA. Furthermore, the combination of PA and
longitudinal data from the PIF on both mothers and children will allow important
and timely investigations into PA, obesity, and factors contributing to the
health and wellbeing of Pacific peoples; a population carrying an abnormally
high overweight/obesity load.
Concordant with nationally representative data, a high
prevalence of overweight and obesity was observed in Pacific children and their
mothers participating in the PIF:PAC study. Body size measures revealed
relatively similar results, with 61 or 60% of boys, 58 or 53% of girls, and 96
or 97% of mothers categorised as having high body mass index or waist
circumference, respectively.
Overweight and obesity prevalence in our sample was
approximately 10% higher than that found for boys and adult females in the
NZHS,5 and identical to that found for girls.
The exceedingly high body size measured in the mothers is of concern,
particularly considering that if ethnic-specific thresholds had been applied in
the NZHS, it is likely an even greater difference between obesity prevalence in
our mothers and the nationally representative sample would have been
found.
By using accelerometers in the PIF:PAC study, we have been
able to gather a substantial amount of complex and precise information about the
PA patterns, intensities, inactivity levels, and duration and timing of
in/activity in Pacific children and their mothers that will be explored and
described for the first time. Feasibility of using accelerometers with a
sub-sample of the PIF cohort was demonstrated; participants were largely
compliant, providing detailed PA information over an average of 3–4 days
for an average of 13–14 hours.
Application of apposite longitudinal modelling techniques of
the activity data will allow us to investigate PA patterning of individuals
based on a large amount of data (e.g. 1 day of wear with 10 hours data using 60
second epochs is equal to 600 data points). The method of accelerometer data
treatment as well as the longitudinal nature of the PIF study data will also
enable the use of powerful repeated measure multivariable regression methods to
explore relationships between physical activity, body size, and the associates
and determinants of each in great detail.
Demographic information measured in our sample was largely
similar to that found in the representative sample participating in the PIF
study cohort at 6 weeks; as such findings resulting from this study will be
broadly generalisable to the wider Pacific population.
While undoubtedly a strength of the study, the use of
accelerometers for research purposes can also prove problematic. Activities such
as swimming and cycling are not well captured by hip-mounted
accelerometers.32 Though most participants were
compliant, 12% did not wear the accelerometers at all during the measurement
phase, and others only wore the units sporadically (resulting in a range of
number of days worn and hours worn per day).
Given that 66% of the eligible families participated, it is
also possible that the additional burden associated with wearing an
accelerometer may have discouraged involvement in the study. The substantial
data collected necessitates manual cleaning and processing that is time
consuming, and there is no agreed best-practice approach for data reduction and
interpretation. Although researchers within our team have devised a robust
accelerometer data treatment protocol using generalised estimating equation
methods (not presented here), this method currently requires statistical support
and considerable resources to complete.
Culturally appropriate, effective, and integrated programmes
are urgently required at national and community levels to combat the rising
problem of obesity, and to promote healthy lifestyles and well-being for Pacific
children and their families. The PIF:PAC study affords the opportunity to better
understand the effects of an active (or inactive) childhood.
Findings from the PIF:PAC study, in combination with the
substantive PIF study findings, will provide ethnic-specific information on
relationships between PA participation and health outcomes in Pacific children.
This valuable and timely information can assist stakeholders in their promotion
of Pacific health and wellbeing in New Zealand.
The information may be used to inform school and community
programmes, create the opportunity to conduct and evaluate randomised prevention
trials, and improve the delivery of service and professional practice as it
pertains to Pacific family life both locally, and throughout New Zealand.
Competing interests: None known.
Author information: Melody Oliver,
Postdoctoral Research Fellow1; Philip J
Schluter, Professor of Biostatistics2,3; Janis
Paterson, Co-Director of Pacific Islands Families
Study2,4; Gregory Kolt, Head of School of
Biomedical and Health Sciences5 and Adjunct
Professor1; Grant M Schofield, Director of
Centre for Physical Activity and Nutrition
Research1
Acknowledgements: The
Pacific Islands Families: Child and Parental Physical Activity and Body Size
study (PIF:PAC) was partially funded by Sport and Recreation New Zealand. Sport
and Recreation New Zealand also provided permission to use parts of the New
Zealand Physical Activity and Nutrition Questionnaire for this study. Melody
Oliver is supported by a National Heart Foundation of New Zealand Research
Fellowship (grant number 1280). The authors also gratefully acknowledge the
families who participated in the study, the Pacific Peoples Advisory Board, and
other members of the PIF research team.
Correspondence: Dr Melody Oliver (mail
#A-24), Centre for Physical Activity and Nutrition Research, Auckland University
of Technology, Private Bag 92006, Auckland 1142, New Zealand. Email: melody.oliver@aut.ac.nz
References:
This article was
corrected on 9 October 2009 as per the Erratum at http://www.nzmj.com/journal/122-1304/3842/content.pdf
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|