Levels and correlates of physical activity, inactivity and body mass index among Saudi women working in office jobs in Riyadh city

Prevalence of obesity in this sample of women working in office settings (26 %) was
lower than the 43 % previously reported by the WHO 5] for women in the KSA. It also fell below the 44 % of women reported as obese in a
large (n?=?17,232) cross sectional household survey conducted between 1995–2000 19]. Another household survey in 2013 (n?=?10,735) also reported obesity prevalence among women as 33.5 %, greater than the
present sample 20]. The variation in prevalence from these and the present study may be the result of
the large cross sectional studies including subjects from different cities in the
KSA and from rural areas which are known to differ in their BMI. Education level,
a significant predictor of obesity 21], also differed between these studies with one study 20] reporting only 21 % of participants having a college degree or higher while in the
present study it was 80 %. The varying employment status of the subjects may also
affect obesity levels as it has been reported that a greater proportion of unemployed
women in the KSA are obese as compared to employed women 21]. It is therefore difficult to compare the prevalence of the present study with those
reported previously due to the difference in the characteristics of the study groups.

In the present sample, participants had a significantly greater risk of overweight
and obesity when family income was less than 10,000 Saudi Riyals (2,667 USD). In fact
it was one of only three predictors for high BMI. A positive association between socioeconomic
status (SES) and BMI and between SES and overweight has been demonstrated in a sample
of over 500,000 women from 54 low to middle-income countries 22]. In developing countries, the burden of obesity tends to shift towards the groups
with lower SES as the country’s gross national product improves. This shift occurs
at an earlier stage of economic development for women than it does for men 23] resulting in a particularly greater danger for women. Whether the association found
in this study is only in employed Saudi women or exists in the general public is unclear
and warrants further research.

The greater proportion of overweight and obese women over the age of 35 years, married
and having children is to be expected 24], 25], however it is alarming to find that 44 % of women 18 to 25 years of age were overweight
or obese which is greater than has been previously reported 24]. Over a third of single women and 51.4 % of women without children were also overweight
or obese. As a number of these women are expected to marry and bear children, their
risk for increased body weight will also increase along with a higher chance for the
co morbidities of obesity.

Working in the public sector was also found to be a predictor of overweight/obese
independent of other variables (Table 4). No studies were found on this subject however as most government workers in the
KSA have tenure, it may be attributed to a more relaxed culture in government workplaces
which were also observed to be more flexible about the presence of food in work areas.
Government workers may also be under less social pressure to present a fit body image
to clients as compared to those working in the private sector. Further investigation
into other variables in the workplace such as availability and types of food consumed
and social factors is warranted.

The total physical activity level reported by this sample showed that 52.1 % of the
respondents were insufficiently active, exceeding previous global and local reports
4], 13], 26]. Hallal et al. 26] reported on PA levels in adults from 122 countries and found 31 % to be physically
inactive (range 17–43 %). Bauman et al. 13], in a comparative international study of population PA prevalence, used a self-report
survey, the International Physical Activity Questionnaire (IPAQ), across 20 countries
and reported low PA levels in females ranging from 6 to 49 %. The IPAQ has been shown
to over report physical activity 27] yet Saudi Arabia participated in this study and reported only 37.3 % of their female
sample (n?=?344) as having low level PA. The occupational status of that sample is unreported
and only 38.3 % of all the Saudi participants in that study had an education level
above high school unlike the present sample which exceeded 80 %. As higher educational
level is associated with more sedentary professions, this may be an important variable
explaining the greater prevalence of inactivity in the present study. Al-Hazzaa 4] in a sample including 365 adult females in Riyadh assessed PA using the short form
IPAQ and found 34.3 % to be in the lowest PA level, however less than 20 % were employed.
The sample was also older, the majority (44 %) being 30–44 years, while the present
study was predominately 26–35 years (55 %). Amin et al. (2012) 28] reported on a sample from primary care centers in AlHassa (n?=?967) using a survey derived from the IPAQ and found approximately 80 % of respondents
were insufficiently active. Only a third of the females in his sample were employed
and less than 40 % had over high school level education levels.

Predictors of low physical activity included ‘working seven or more hours per day
’ (Table 6) and it is reasonable to assume that the long working hours in office based jobs
would reduce the time spent in PA. This is consistent with the subjects reporting
“not having enough time” as the main reason for not being active (Table 7). Longer working hours, particularly in sustained postures are known to cause fatigue
even when physical exertion has not been expended 29] and may also explain the second highest reported barrier for PA being “too tired”
(Table 7).

Working in the private sector was a second predictor of low physical activity independent
of work hours (Table 6). This merits further investigation however the private sector in the KSA is known
to have a more competitive work environment and greater work load compared to public
sector jobs which offer early tenure and tend to have more employees for similar work
loads. This may contribute to more sitting and greater fatigue after work for private
sector employees.

Surprisingly psychosocial variables often found to correlate with PA, self-efficacy
and social support, were not found to be significant in this sample. Social support
was found to be an important variable in one previous study in KSA 8] and in numerous international studies 30]–32]. In this study however no correlation was found between physical activity social
support scores and BMI or PA. In addition, only 21.5 % of respondents mentioned “no
one to exercise with” as a factor for their inactivity. Self- efficacy, which has
been found to correlate to PA in the majority of international literature 32]–34] and in one study on women in KSA 10], was not significantly correlated to PA in this sample. Major barriers to PA noted
by both low and highly active respondents; ‘not having time’ and being ‘too tired’
may be interpreted as a lack of self- efficacy and it is possible that using a survey
targeting physical activity self-efficacy may have been more sensitive to differences among respondents than
the ‘general self- efficacy’ survey used in this study.

Environmental factors including ‘distance to facilities’, ‘lack of transportation’
and ‘facilities being too expensive’ were barriers to activity for less than 30 %
of this sample and PA being ‘socially unacceptable’ was mentioned by the least number
of respondents as a reason for inactivity (2 %) (Table 7). This is in contrast to previous studies in KSA 8], 9] which reported lack of facilities/resources and traditions as the most important
barriers. This may be explained by the present study sample being predominately educated
professionals whose awareness of the importance of physical activity may override
local traditions. Their professions may also allow them more access to facilities
than the predominately unemployed samples used in the previous studies.

Interestingly there was no significant difference in the proportion of normal weight
and overweight/obese respondents reporting low PA level in this sample which is similar
to previous results on female Saudi university students 35].

This study was subject to several limitations including the use of a survey as the
main study instrument which is subject to recall bias and social desirability effect.
Therefore, despite using a reliable and valid physical activity questionnaire in this
survey, it is recommended that future studies may consider the use of more objectives
instruments for the assessments of physical activity in Saudi working women, such
as accelerometer. Secondly, the cross sectional design limits the ability to make
causal inferences. Another limitation is that the sample was from Riyadh city which
decreases generalizability of the findings to other regions in the country.