Physical activity, sedentary behavior time and lipid levels in the Observation of Cardiovascular Risk Factors in Luxembourg study

Participants

Data was obtained from the Observation of Cardiovascular Risk Factors in Luxembourg
(ORISCAV-LUX), a nationwide cross-sectional study conducted, to gather information
on the prevalence of cardiovascular risk factors, and lifestyle health-related factors
such as PA and nutritional status, among the general adult population of Luxembourg.
A representative random sample of 1432 individuals, stratified by sex, age (18–69
years) and district of residence, selected from the national health insurance registry
was recruited between November 2007 and January 2009. This sample size was estimated
to ensure statistical power, i.e. a statistical precision of at least 2 % for the
estimation of the prevalence of the risk factors at the 95 % confidence level. Institutionalized
subjects (n?=?12), pregnant women (n?=?21), subjects with serious mental and/or physical
handicap (n?=?5), prisoners (n?=?1), people outside the determined age range (n?=?2)
and those deceased before recruitment (n?=?5) were excluded. Description of the study
design and recruitment have been published in detail previously 39], 40]. A total of 1432 participants completed the recruitment procedure. After eliminating
those with missing data on lipids, sedentary behavior time, PA or covariates, data
from 1331 participants were available for the present analysis.

Procedure and measures

Demographics and health information

Information on demographic and socioeconomic characteristics, including age, gender,
education, occupation, and income was obtained from a detailed self-administered questionnaire.
Education level was classified into three levels, based on the highest diploma obtained:
‘primary’ (less than 12 years of education), ‘secondary’ (approximately 12 to 13 years
of education) and ‘tertiary’ (more than 13 years). Participants were required to indicate
their occupation, and based on the type of work performed, each participant was categorized
into one of three groups with regard to their occupation: ‘sedentary’, ‘moderately
active’, and ‘active’. For example, manual laborers were placed in the active category,
while scientific professionals were placed in the sedentary category. Economic status
was ascertained by asking participants to select the category best representing total
monthly household income. Body weight was measured using a digital column scale (Seca®
701, Hamburg, Germany), recorded to the nearest 0.1 kg, with subject barefoot and
wearing light clothing. Standing body height was recorded to the nearest 0.2 cm with
a portable wall stadiometer (Seca, Germany). BMI was calculated as weight in kg divided
by height in metres squared (kg/m
²
).

Lipid levels

Blood samples were collected after a minimum 8-hour fast from the antecubital vein
to measure total cholesterol (mg/dL), HDL-cholesterol (HDL, mg/dL) and LDL-cholesterol
(LDL, mg/dL), and triglycerides (mg/dL), by using Roche (Switzerland) reagents on
a P module of a Modular analyzer (Roche, Switzerland) 41]. All biochemical analyses were carried out within two hours of blood sampling in
the core laboratory of ‘Centre Hospitalier du Luxembourg (CHL)’. The CHL laboratory
applies strict internal and external standard quality control techniques.

Physical activity and sedentary behavior time

PA and sedentary behaviour time was assessed using the short-form International Physical
Activity Questionnaire (IPAQ) 42], designed and validated to measure PA in large populations. The IPAQ is conceived
in a way to distinguish the time spent in performing four types of physical behaviors
(vigrous physical activity, moderate physical activity, walking and sitting).

Physical activity

Participants were required to report the number of days in the preceding week (including
the weekend) that they engaged in both moderate and intense PA, and the amount of
time spent in these activities on one of these days (in hours). Examples of moderate
activities provided included carrying light loads, vacuuming or slow cycling, whilst
intense activities included carrying heavy loads, digging or playing soccer.

Mean daily moderate and intense PA time (in hours per day) was calculated by multiplying
self-reported time spent engaging in each with the reported number of days per week
in which these activities were undertaken, and dividing by seven. These values were
summed to obtain total mean PA time (moderate plus intense). Moderate, intense and
total PA time were then categorized into three groups: 0.5 hours per day, 0.5-1 hour
per day, and 1 hour per day.

Sedentary behavior time

Sedentary behaviour time was also obtained from the IPAQ. Participants were required
to report screen time, i.e., how much time they spent watching television (including videos/DVD), and in
front of a computer (including internet and video games), during a typical workday,
and during a day off. Participants also reported overall sitting time, i.e., how much time they spent sitting during a normal weekday (distinguished form
a weekend day), including time spent sitting at place of work, on transportation,
reading, visiting friends’, sitting or laying down to watch television or use a computer.
All responses were given in hours per day, with reference to the preceding seven days.

Total screen time for a workday and a day off were calculated by summing television time and computer
time. The screen time variables were then divided into three categories: ?1, 1-4, ?4 hours per day. The
overall sitting time variable was divided into five categories: 0–2, 2-4, 4-6, 6-10, and 10 hours
per day.

In addition, individuals were categorized into one of four groups to reflect their
mean weekly behavior with regard to activity level, referred to as physical activity status. Participants who reported moderate or intense PA on at least one day per week were
classified into low, medium or high tertiles of PA, calculated from the total PA time
variable (hours per day). The remainding participants who did not engage in moderate
or intense PA during the preceding week were classified as ‘sedentary’. This evaluation
method of overall PA has been used previously 29].

Other covariates

Detailed data regarding cigarette smoking were obtained from the questionnaire. Each
participant was classified as current smoker, ex-smoker or non-smoker. A validated,
semi-quantified food frequency quesionnaire (FFQ) was used to assess the frequency
of consumption of 134 food and beverage items over the previous three months 43], 44]. Participants were asked how frequently they consumed one standardized portion of
each food, with six frequency response categories ranging from “never or rarely” to
“2 or more times per day”. Energy and nutrient intake data were compiled to obtain
total carbohydrate (g/day), total protein (g/day), total fat (g/day), total fiber
(g/d), alcohol (g/day), and total energy intake (Kcal/day). Total carbohydrate, protein,
total and saturated fat as proportions of total enery intake were subsequently calculated
and expressed as % E.

Statistical analyses

For the descriptive analyses, Chi-squared tests and ANOVA were performed to compare
the socio-demographic characteristics, blood lipid levels and dietary characteristics
of participants according to the overall activity level classification, ranging from
sedentary to a high level of PA.

General linear modelling with polynomial trend analyses was used to compare each lipid
biomarker (as continuous variables) across increasing categories of screen time (workday and day off), total sitting time, and PA time (moderate and intense). Two covariate sets were employed:

1) Basic: age, education, gender.

2) Extended: basic model in addition to income, profession, carbohydrate % E, protein
% E, saturated fat % E, alcohol (ml/day), and smoking.

Total PA (hours/day) was included in the extended model for analyses with the sedentary
behavior time variables only.

The same statistical procedures, assessing relations between sedentary behavior time,
PA time and blood lipids, were performed separately in the sample stratified by BMI
status: normal weight (BMI??25 kg/m
²
, n?=?589), and overweight/obese (BMI???25 kg/m
²
, n?=?741).

All statistical anlyses were performed with PASW for Windows® version 21.0 software
(formerly SPSS Statistics Inc. Chicago, Illinois). P??0.05 was considered statistically
significant.

Ethical statement

All participants gave informed written consent to take part in the study. The study
design and information collected were approved by the National Research Ethics Committee
(Comité d’Ethique de Recherche, CNER) and the National Commission for Private Data
Protection (Commission Nationale pour la Protection des Données, CNPD).