Background
The overall health benefits of physical activity have long been established. In 1953, Morris and colleagues showed that bus conductors in London experienced half the coronary heart disease mortality rates of their driver counterparts who spent their working days sitting behind the wheel [
1]. Physical activity (PA) diminishes the risk of cardiovascular risk factors, and its role, notably in the prevention and management of type 2 diabetes, is well documented [
2,
3]. Today, lack of physical activity is the fourth leading risk factor for global mortality [
4] and is one of the leading public health indicators [
5]. Despite the known hazards of being sedentary, physical activity levels continue to fall worldwide [
4]. In one article of the Lancet series dedicated to physical activity, Hallal and colleagues [
6] note that roughly three out of ten adults aged 15 years and over are below the recommended physical activity threshold, and urge governments and populations to take physical inactivity as serious public health issue. The Middle East and North Africa (MENA) region, as a whole, is burdened with high prevalence rates of physical inactivity [
6] reaching 82.3% among Syrian adults [
7] and 96.1% in Saudi Arabia [
8]. In parallel, the region has the highest comparative prevalence of diabetes (4%–21%) [
9]. According to the International Diabetes Federation [
10], six out of the world’s top ten countries for highest prevalence of diabetes are from the MENA region.
Lebanon, a small-middle income country of the MENA region, is characterized by high urbanization rate (87%), high literacy rate (89%), and life expectancy approaching 73 years for men and 76 years for women. With westernization and socioeconomic changes, non-communicable diseases have long emerged as the leading cause of morbidity and mortality [
11]. Obesity rates have increased from 17·4% in 1997 to 28·2% in 2009 among adults aged 20 years and older [
12] and close to one third of the adult population suffer from metabolic syndrome [
13]. Yet, national data on the distribution of physical activity and diabetes are limited to a few small-scale surveys and remain deficient. Also, physical activity is influenced by a diverse range of factors, and the influence of these factors on the pattern of physical activity in the general population and among those with diabetes has not been studied. Using data from a recent nationally representative cross-sectional survey, this study intended to address these issues in order to inform public health policies and plans. The study aimed to answer the following questions: firstly, what is the prevalence of physical activity among the total population, and among those with diabetes and risk of developing diabetes; and secondly, what are the factors associated with the level of physical activity performed as stratified by diabetes status?
Results
The mean age of the study sample was 44.7 years (SD±14.9) with a slightly larger proportion of females than males (53.6% vs. 46.4%, respectively) (Table
1). The majority of participants had attained middle or high level education (75.4%), resided outside the capital city of Beirut (89.2%), and owned at least one car (71.5%). Close to 38% were current smokers. Their mean BMI was 27.8 kg/m
2 (SD=5.4), with around 29% being obese (BMI≥30 Kg/m
2). Hypertension was the most prevalent health condition (31.8%). This was followed by hyperlipidemia (18.5%) and heart disease (8.0%). Around one third of individuals were advised to exercise in the six months prior to the survey.
Table 1
Baseline characteristics of study participants and rates of physical activity, Lebanon, 2009
Total Sample | 2195 | | 1026 | 46.7 | 682 | 31.1 | 487 | 22.2 |
Socio-demographic
| | | | | | | | |
Age | | | | | | | | |
25–44 | 1245 | 56.7 | 608 | 48.8 | 348 | 28.0 | 289 | 23.2 |
45–64 | 664 | 30.3 | 271 | 40.8 | 239 | 36.0 | 154 | 23.2 |
≥65 | 286 | 13.0 | 147 | 51.4 | 95 | 33.2 | 44 | 15.4 |
Gender | | | | | | | | |
Male | 1019 | 46.4 | 520 | 51.0 | 286 | 28.1 | 213 | 20.9 |
Female | 1176 | 53.6 | 506 | 43.0 | 396 | 33.7 | 274 | 23.3 |
Governorate | | | | | | | | |
Beirut | 236 | 10.8 | 139 | 58.9 | 64 | 27.1 | 33 | 14.0 |
Other | 1959 | 89.2 | 887 | 45.3 | 618 | 31.5 | 454 | 23.2 |
Education | | | | | | | | |
Low | 540 | 24.6 | 237 | 43.9 | 184 | 34.1 | 119 | 22.0 |
Middle | 952 | 43.4 | 418 | 43.9 | 300 | 31.5 | 234 | 24.6 |
High | 703 | 32.0 | 371 | 52.8 | 198 | 28.2 | 134 | 19.1 |
Owns a car | | | | | | | | |
None | 547 | 24.9 | 215 | 39.3 | 197 | 36.0 | 135 | 24.7 |
1 or more | 1648 | 71.5 | 811 | 49.2 | 485 | 29.4 | 352 | 21.4 |
Health-related
| | | | | | | | |
Smoking | | | | | | | | |
Never | 1217 | 55.4 | 557 | 45.8 | 401 | 32.9 | 259 | 21.3 |
Past | 150 | 6.8 | 78 | 52.0 | 41 | 27.3 | 31 | 20.7 |
Current | 828 | 37.7 | 391 | 47.2 | 240 | 29.0 | 197 | 23.8 |
BMI | | | | | | | | |
Normal | 719 | 32.8 | 336 | 46.7 | 216 | 30.0 | 167 | 23.2 |
Overweight | 832 | 37.9 | 359 | 43.1 | 275 | 33.1 | 198 | 23.8 |
Obese | 638 | 29.1 | 328 | 51.4 | 190 | 29.8 | 120 | 18.8 |
Hypertension | | | | | | | | |
No | 1489 | 67.8 | 698 | 46.9 | 447 | 30.0 | 344 | 23.1 |
Yes | 699 | 31.8 | 324 | 46.4 | 234 | 33.5 | 141 | 20.2 |
Hyperlipidemia | | | | | | | | |
No | 1790 | 81.5 | 815 | 45.5 | 553 | 30.9 | 422 | 23.6 |
Yes | 405 | 18.5 | 211 | 52.1 | 129 | 31.9 | 65 | 16.0 |
Heart disease | | | | | | | | |
No | 2019 | 92.0 | 926 | 45.9 | 628 | 31.1 | 465 | 23.0 |
Yes | 176 | 8.0 | 100 | 56.8 | 54 | 30.7 | 22 | 12.5 |
Advice to exercise | | | | | | | | |
No | 1544 | 70.3 | 724 | 46.9 | 454 | 29.4 | 366 | 23.7 |
Yes | 651 | 29.7 | 302 | 46.4 | 228 | 35.0 | 121 | 18.6 |
Table
1 also presents physical activity rates according to the various socio-demographic and health characteristics of the study sample (row %s). Close to half of the survey participants reported low or no physical activity (46.7%), 31.1% reported moderate levels, while 22.2% reported HEPA levels. Physical activity, notably HEPA, varied with age being lowest among older adults aged 65 and above. Compared to their counterparts, physical activity was slightly higher among females, those living outside Beirut, and among participants with low-to-middle educational levels. Obese respondents and those suffering from co-morbid conditions including hypertension, hyperlipidemia and heart disease were less likely to be performing HEPA. Respondents receiving medical advice to exercise were more likely to engage in moderate levels of physical activity.
Survey participants with diagnosed diabetes made up 8.5% of the study sample (Table
2). An additional 26.3% and 4.1% of the subjects without diabetes had 3–4 and 5–6 diabetes risk factors, respectively. HEPA prevalence rates decreased consistently with increasing number of diabetes risk factors (p-value<0.001) and was significantly lower among those with diabetes than those without diabetes (9.6% vs. 23.4%, p-value<0.001).
Table 2
Rates of physical activity among Lebanese adults diagnosed with diabetes and at risk for developing diabetes, 2009
Diabetes | 187 | 8.5 | 104 | 55.6 | 48.0–63.0 | 65 | 34.8 | 28.0–42.0 | 18 | 9.6 | 5.8–14.8 | <0.001 |
No Diabetes | 2008 | 91.5 | 922 | 45.9 | 43.7–48.1 | 617 | 30.7 | 28.7–32.8 | 469 | 23.4 | 21.5–25.3 |
0 diabetes risk factors* | 229 | 10.4 | 109 | 47.6 | 40.9–54.3 | 54 | 23.6 | 18.2–29.6 | 66 | 28.8 | 23.0–35.1 | <0.001 |
1 to 2 diabetes risk factors | 1111 | 50.6 | 513 | 46.2 | 43.2–49.1 | 337 | 30.3 | 27.6–33.1 | 261 | 23.5 | 21.0–26.1 |
3 to 4 diabetes risk factors | 577 | 26.3 | 256 | 44.4 | 40.3–48.5 | 197 | 34.1 | 30.3–38.2 | 124 | 21.5 | 18.2–25.1 |
5 to 6 diabetes risk factors | 91 | 4.1 | 44 | 48.4 | 37.7–59.1 | 29 | 31.9 | 22.5–42.5 | 18 | 19.8 | 12.2–29.5 |
Total | 2195 | 100 | 1026 | 46.7 | 44.6–48.9 | 682 | 31.1 | 29.1–33.0 | 487 | 22.1 | 20.5–24.0 | |
Results of the multivariable logistic regression model for the total sample and stratified by diabetes status are shown in Table
3. In the total sample, physical activity was significantly higher among the middle-age group (45–64 years) (OR=1.46, 95% CI 1.17–1.81), females (OR=1.30, 95% CI 1.09–1.55), and among respondents residing in governorates outside the capital city of Beirut (OR=1.61, 95% CI 1.21–2.62). In contrast, participants with high levels of education (OR=0.75, 95% CI 0.57–0.98) and those who owned at least one car (OR=0.71, 95% CI 0.57–0.88) were significantly less likely to be physically active. Whilst hypertension did not associate with physical activity, obesity (OR=0.78, 95% CI 0.62–0.99) and pre-existing health conditions significantly decreased the odds of physical activity (OR=0.66, 95% CI 0.51–0.85 for hyperlipidemia; OR=0.61, 95% CI 0.43–0.87 for heart disease). A significant association was also noted for receiving health professional advice to engage in physical activity (OR=1.25, 95% CI 1.01–1.54).
Table 3
Correlates of physical activity in the total sample and stratified by diabetes status among Lebanese adults: results of the logistic regression, 2009
Age | | | | | | |
25–44 | 1.00 | | 1.00 | | 1.00 | |
45–64 | 1.46 | 1.17–1.81 | 1.46 | 1.64–1.84 | 1.51 | 0.53–4.34 |
≥65 | 0.98 | 0.71–1.35 | 0.92 | 0.64–1.32 | 1.42 | 0.45–4.46 |
Gender | | | | | | |
Male | 1.00 | | 1.00 | | 1.00 | |
Female | 1.30 | 1.09–1.55 | 1.38 | 1.14–1.66 | 0.64 | 0.31–1.33 |
Governorate | | | | | | |
Beirut | 1.00 | | 1.00 | | 1.00 | |
Other | 1.61 | 1.21–2.13 | 1.63 | 1.21–2.18 | 1.57 | 0.55–4.47 |
Education | | | | | | |
Low | 1.00 | | 1.00 | | 1.00 | |
Middle | 1.02 | 0.80–1.29 | 0.95 | 0.74–1.22 | 1.36 | 0.64–2.89 |
High | 0.75 | 0.57–0.98 | 0.72 | 0.54–0.96 | 0.47 | 0.17–1.31 |
Owns a car | | | | | | |
No | 1.00 | | 1.00 | | | |
One or more | 0.71 | 0.57–0.88 | 0.69 | 0.55–0.86 | 0.96 | 0.45–2.06 |
Smoking | | | | | | |
Never | 1.00 | | 1.00 | | 1.00 | |
Past | 0.83 | 0.58–1.20 | 0.85 | 0.70–1.03 | 0.83 | 0.31–2.25 |
Current | 0.87 | 0.72–1.06 | 0.88 | 0.59–1.32 | 1.02 | 0.48–2.11 |
BMI | | | | | | |
Normal | 1.00 | | 1.00 | | 1.00 | |
Overweight | 1.16 | 0.94–1.43 | 1.16 | 0.93–1.44 | 1.08 | 0.40–2.87 |
Obese | 0.78 | 0.62–0.99 | 0.81 | 0.64–1.04 | 0.50 | 0.19–1.31 |
Hypertension | | | | | | |
No | 1.00 | | 1.00 | | 1.00 | |
Yes | 0.96 | 0.74–1.25 | 1.17 | 0.93–1.47 | 0.85 | 0.41–1.78 |
Hyperlipidemia | | | | | | |
No | 1.00 | | 1.00 | | 1.00 | |
Yes | 0.66 | 0.51–0.85 | 0.69 | 0.52–0.93 | 0.67 | 0.34–1.36 |
Heart Disease | | | | | | |
No | 1.00 | | 1.00 | | 1.00 | |
Yes | 0.61 | 0.43–0.87 | 0.81 | 0.54–1.22 | 0.29 | 0.13–0.70 |
Advice to Exercise | | | | | | |
No | 1.00 | | 1.00 | | 1.00 | |
Yes | 1.25 | 1.01–1.54 | 1.17 | 0.93–1.46 | 3.68 | 1.55–8.74 |
Findings of the regression analysis varied according to diabetes status. Owing to the relatively large proportion of subjects without diabetes in our study sample (91%), the direction and strength of the associations among the non-diabetic subjects were, overall, similar to those estimated for the total sample. Among diabetic subjects, we note that the results were, in the majority, not significant and with wide confidence intervals. However, the association of physical activity with heart disease (OR=0.29, 95% CI 0.13–0.70) and with receiving professional advice to exercise were further enhanced among subjects with diabetes (OR=3.68, 95% CI 1.55–8.74), compared to those without diabetes.
Discussion
Findings of this study indicate that close to one third (31%) of Lebanese adults, 25 years of age and older, participate in moderate levels of physical activity and only 22% engage in health enhancing energy expenditure levels. These results are lower than those reported in the International Prevalence Study conducted in 20 countries worldwide using the same instrument as ours (median 33% and 40% for moderate and vigorous activity, respectively) [
20]. Compared to their counterparts, physical activity was lower in subgroups with urban characteristics and lifestyle, namely those with high levels of education, who owned at least one car, and among those who resided in Beirut, the capital city. Given the clear evidence in the literature of the health benefits of physical activity in the prevention and management of diabetes, it is disturbing to note that HEPA rates decreased consistently with increasing number of diabetes risk factors and were only 10% among subjects known to have diabetes.
During the last few years, different subgroups of the Lebanese adult population have been studied with regard to physical activity with varying results. For example, Sibai and colleagues found approximately 16% of HEPA levels in a sample of 499 adult population attending health centers in Lebanon [
13]. Tohme and colleagues [
21] estimated the prevalence rate of daily exercise for 30 minutes in a nationally representative sample of 2,125 subjects at 20.4%. A cross sectional study conducted among 346 adults 18 years and over selected from large stores across all governorates yielded a much higher estimate (55%) [
22]. Findings from the WHO STEPwise surveys in countries of the MENA region also show a wide range of prevalence estimates of physical activity, varying between 13.2% and 78.4% [
23]. The outcomes of these studies are difficult to compare with findings from the present study owing to differences in sampling design, sample characteristics, as well as the tool, operational definition and the scoring method employed to measure physical activity.
We found inverse associations between physical activity and various indicators of socio-economic status (SES), as illustrated by education (OR=0.75), car ownership (OR=0.71), and residence in Beirut, the highly urbanized metropolitan capital city of the country (OR=0.62). Reviews of SES correlates with physical activity show mostly positive associations; this is particularly evident for education [
24,
25]. Yet, this evidence has largely been noted for leisure-time physical activity with diverse and often crude measures being employed [
6,
24]. In comparison, our assessment of physical activity was based on the IPAQ that includes energy expenditure from several domains, and captures expenditure from lifestyle activities integrated in daily routines of work, household chores and transportation, additional to leisure time activities. Thus, higher education, car ownership and residence inside the capital city can mean more sedentary jobs, reduced walking and less physically active lifestyles. Support for our results is found in studies from sub-Saharan Africa where urban dwellers had a significantly lower physical activity energy expenditure than rural dwellers (44.2 ± 21.0 vs. 59.6 ± 23.7 kJ) [
26], and in selected developing countries of the Asian-Pacific and the MENA regions [
14,
27]. Similarly, cross-country comparisons at the ecological level show that physical activity is more common in low income countries than in high income countries [
6,
28].
Consistent with findings from the literature [
3], the presence of obesity and co-morbidities associated inversely with physical activity among our study participants [
3]. Also, HEPA levels declined consistently as the number of diabetes risk factors increased, and was lowest among people with diabetes. Given the benefits of regular moderate physical exercise as a form of intervention management for improving HbA1c levels and cardiovascular risk factors in diabetic patients, it is concerning that the majority of Lebanese adults with diabetes remain either sedentary or insufficiently active (55.6%) and only 9.6% participate in health enhancing activities. Our results are similar to those found among 406 consecutive patients attending the Diabetes clinic in Dundee, UK (9%) [
29], yet much lower than findings from other patient groups in the National Health and Nutrition Examination Survey (28.2%) and in the Medical Expenditure Panel Survey (39%) in the US [
30,
31].
Patients receiving advice from a health professional to alter their lifestyle and exercise were found to be significantly more likely to engage in physical activity than non-diabetic subjects receiving the same advice, net of the effect of other covariates (OR=3.68 and OR=1.17, respectively). This finding merits further consideration. Post hoc analysis of our data shows a positive gradient between receiving advice and the presence of diabetes risk factors (p-value for trend <0.01) with health professionals advising most patients with or at highest risk for diabetes (76.5% and 71.4%, respectively) and least those without diabetes (13.1%). While this implies the recognition by health professionals of the importance of physical activity among diabetes patients on one end of the spectrum, it also suggests a missed opportunity for primary disease prevention in the clinic for the majority non-diabetic population, on the other end. Given the crucial role of physical activity in the prevention and management of diabetes, these results have important implications to health care providers and policy makers with regard to diabetes control guidelines and protocols. Whereas studies are needed to examine contextual factors that influence behavioral change of patients and non-patients in Lebanon, it is the time to start targeting interventions that promote physical activity in the general as a life-style to be adopted. The challenge remains to alert physicians towards primary disease prevention services for the non-diabetic majority and to re-direct policy attention to diabetes control from the high-risk emphasis towards the population-based public health approach.
The findings of this study are considered in light of the following limitations. Because of the cross-sectional nature of the study, the temporality of certain associations cannot be established with confidence. Also, the study lacked sufficient power to detect significant correlates of physical activity among subjects with diabetes (n=187). Whereas BMI and hypertension were objectively measured using calibrated equipment and standardized techniques, data on diabetes status are self-reported, introducing possible misclassification bias. However, the amount of bias is minimal as post-hoc analysis of our data showed that those who reported having diabetes were mostly on medication and/or were following a diet (84%). Also, diabetes is an overt disease, and so diabetic individuals are likely to be referred to a physician for diagnosis and follow-up. Furthermore, studies from Lebanon and elsewhere show substantial agreement between self-report and medical record diagnosis for several medical conditions including diabetes (kappa range 0.71–0.78) and heart disease (kappa range 0.83–0.87) [
32‐
34]. Although the IPAQ’s validity has originally been assessed for individuals aged up to 69 years of age, nonetheless, studies in the developed world and in the region have used and validated it on samples aged over 70 years [
35,
36].
In spite of the above caveats, this is the first study that we are aware of to assess the prevalence of physical activity and its correlates with a focus on diabetes and diabetes risk factors in a large national sample in Lebanon and the MENA region. Estimates of physical activity from national surveys such as the current one can provide valuable information to guide national policies and serve as a benchmark for monitoring and evaluation thereafter. Furthermore, our assessment of physical activity based on the IPAQ captures more contributions towards total physical activity levels and does not ignore routine activities undertaken by housewives, for example, who constitute a good percent of the Lebanese female population (67.5%). The ability to quantify benefit from several domains including routine daily pursuits, whether at work or home, may be more applicable to health promotion initiatives in low-income countries such as Lebanon where access to leisure-time activity is hampered by the cost of private club memberships and the lack of public and age-friendly outdoor spaces to exercise.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
AMS contributed towards study design, study logistics, data collection, data analysis, and interpretation and manuscript drafting. RT contributed toward study logistics, data collection, analysis and interpretation. SA contributed to conception of the hypothesis, the analysis and write-up of the paper. NH contributed to study logistics, data collection and data interpretation. CC contributed to the conception of the hypothesis, analysis and write-up. All authors provided critical insight, and revisions to the manuscript. All authors read and approved the final version of the manuscript submitted for publication.