Socioeconomic and behavioural determinants of overweight/obesity among adults in Botswana: a cross-sectional study.

OBJECTIVE: To undertake a comprehensive assessment of socioeconomic and behavioural determinants of overweight/obesity among adult population in Botswana.
DESIGN: The study adopted a cross-sectional design by selecting adult respondents in 3 cities and towns, 15 urban villages and 15 rural areas across Botswana using a multistage probability sampling technique.
SETTING: The study was conducted in selected rural and urban areas of Botswana. PARTICIPANTS: The study sample consisted of 1178 adult males and females aged 15 years and above. PRIMARY OUTCOME MEASURES: Objectively measured overweight/obesity.
RESULTS: Prevalence of overweight/obesity in the study population was estimated at 41%. The adjusted OR (AOR) of overweight/obesity were highest among women (AOR=2.74, 95% CI 1.92 to 3.90), in ages 55-64 years (AOR=5.53, 95% CI 2.62 to 11.6), among individuals with secondary (AOR=1.70, 95% CI 1.11 to 2.61) and tertiary education (AOR=1.99, 95% CI 1.16 to 3.38), smokers (AOR=2.16, 95% CI 1.22 to 3.83) and people with poor physically activity (AOR=1.46, 95% CI 1.03 to 3.24). These were statistically significant at 5% level.
CONCLUSION: Women, older adults, people with high education level, smokers and people who reported poor physical activity were found to have higher odds of being overweight/obesity. These findings suggest the need for broad based strategies encouraging physical activity among different socioeconomic groups. © Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

To our knowledge, this is the first cross-sectional study in Botswana to comprehensively assess both the socioeconomic and behavioural determinants of overweight/obesity among adults.

Data were collected from a large and randomly selected representative population. The data contained information on potential confounding factors, with a low proportion of missing information making the study more comparable.

As the cross-sectional design was employed, the causal relationship between an explanatory and an outcome variable was not established.

Introduction

The disease profile in Botswana is changing, with most deaths and disability in the foreseeable future likely to be accounted for by non-communicable diseases (NCDs). Children, adults and the elderly are all vulnerable to the risk factors that contribute to NCDs especially overweight/obesity, due to unhealthy diets, low physical activity, exposure to tobacco smoke and/or the effects of the harmful use of alcohol.1–3 The key drivers of NCDs are ageing, rapid urbanisation and the globalisation of unhealthy lifestyles. Globalisation of unhealthy lifestyles like unhealthy diets may show up in individuals as raised blood pressure, increased blood glucose, elevated blood lipids and overweight/obesity.1

Several studies have recently shown that many low/middle-income countries (LMICs) are experiencing an increase in the prevalence of overweight/obesity.4–6 This prevalence is predicted to continue to increase in the future. Much of the increase in prevalence levels for overweight/obesity is expected in countries such as India, some parts of South-East Asia, China, most of South America and some parts of sub-Saharan Africa (SSA).7 8 Until recently, SSA was minimally affected by the overweight/obesity epidemic due to under-nutrition and a major burden of HIV and tuberculosis.9 Meanwhile, research on overweight/obesity has started to gain significance in SSA10–14 due to its increasing magnitude.

It has been shown by previous studies that socioeconomic and behavioural factors have a significant association with overweight/obesity.15 16 These studies have shown that socioeconomic and demographic factors such as sex, age, education level, work status, place of residence and wealth status are determinants of overweight/obesity. Studies have found that women are more likely to be overweight/obese than men, while the odds of being overweight/obesity has been seen to increase with age.17–19 It was has also been observed that being employed in a high-income job, residing in urban areas, having high education and wealth status are significantly associated with overweight/obesity.20

The review of literature suggests significant association between a number of behavioural factors and overweight/obesity. For example, significant positive association was found between physical inactivity and overweight/obesity.21 Further, studies have noted that the association between physical inactivity and health outcomes such as overweight/obesity is moderated by other lifestyle factors, especially sedentary lifestyle.15 20 21 In most developing countries, there has been a substantial shift from jobs with high-energy expenditure such as farming, mining and forestry to more-sedentary sectors of manufacturing, services and office-based work.22 This shift has contributed to risky lifestyle behaviours condoning alcohol consumption, poor diets and smoking.23 24

The Botswana WHO STEPS surveys conducted in 2007 and 2014 have shown rising levels of overweight/obesity among adult population of Botswana. The estimated prevalence of overweight/obesity was 18.7% in 200725 and increased to 38.6% in 2014.26 Despite evidence of the high and rising prevalence of overweight/obesity in Botswana, there is little evidence examining the association between socioeconomic and behavioural factors, and overweight/obesity. To the best of our knowledge, this is the first cross-sectional study which takes into account both socioeconomic and behavioural factors together for understanding the determinants of overweight/obesity. The study would be significant in the context that the reduction of overweight/obesity prevalence should be a key goal of healthcare policy because of its associated causes with many NCDs. Keeping this view, the main aim of this study was to examine the association between socioeconomic and behavioural factors, and overweight/obesity in Botswana.

Methods

Sampling, recruitment and ethics

The study used cross-sectional design by selecting respondents in 3 cities and towns, 15 urban villages and 15 rural areas across Botswana using a multistage probability sampling technique. A list of districts, localities and enumeration areas (EAs) together with their households was prepared based on the 2011 Botswana population and housing census that form a sampling frame. At the first stage, population was stratified into cities and towns, urban villages and rural settlements. A listing of all 26 census districts in each stratum was made and from these districts a total of all 4845 EAs were listed for cities and towns, urban villages and rural settlements. At the second stage, localities in cities and towns, urban villages and rural villages were randomly selected. At the third and fourth stage, a random selection of EAs and households, respectively, was made. Lastly, individuals aged 15 years and over were selected for interview from the list of households with persons 15 years and over. The NCD study questionnaire was developed using the instruments from the WHO Study on Global Ageing and Adult Health (SAGE), and WHO STEPS Surveys.

All ethical clearance procedures were followed before the start of the study. The study proposal along with the necessary documents were submitted to and approved by the Institutional Review Board of the University of Botswana. Approval for conducting the study was also obtained from the Government of Botswana through the Ministry of Health and Wellness.

As part of the study protocol, all respondents were requested to sign a consent form and were told that participation in the study was voluntary. For participants aged 15–17 years, written consent was obtained from the parents or guardians. All individuals aged 15 years and above who had successfully completed the NCD study questionnaire were included in the analysis for this study. Privacy and confidentiality of highest standard were maintained by treating all respondents as anonymous and no names of respondents are mentioned or implied when presenting findings of the study.

Definitions and measurement of variables

The outcome variable for the study is overweight/obesity. Body mass index (BMI) was used to classify overweight/obesity. BMI was derived from weight and height: weight (kg) / (height (m) x height (m)).1 The Charder MS7301 250 kg digital scale and the Muac measuring tape were used for anthropometric measurements. Weight was measured, to the nearest 0.1 kg, while height was measured in metres. BMI was categorised into: underweight (BMI<18.5 kg/m2), normal weight (18.5≤BMI<25 kg/m2), overweight (25≤BMI<30 kg/m2) and obese (BMI≥30 kg/m2).1 Overweight and obese were used to create a binary outcome variable which was coded as: being overweight/obese (BMI ≥25 kg/m2) =1; not overweight/obese=0 (BMI <25 kg/m2).

Information on NCD risk factors was collected through self-reports. Respondents were asked questions on NCD risk factors such as tobacco use, alcohol consumption, physical activity and fruit and vegetable consumption. For tobacco use, respondents were asked whether they currently smoke tobacco products and their response was coded as yes=1 and no=0. Alcohol consumption was measured based on the intensity of alcohol consumed in the last 30 days. Respondents who had consumed alcohol in the last 30 days were asked about the number of standard alcohol drinks they had each day in the last 7 days and if they reported to have had three or more drinks per day (of ~60 g alcohol) it was considered to be excessive drinking=1 and 0=otherwise.

Based on WHO recommendations poor physical activity was calculated using an average typical types of activity undertaken.13 It was calculated based on the time taken while doing physical activity in the last 7 days. Respondents were asked whether they do any moderate to rigorous intensity activities for at least 10 min continuously. This was considered for the domains of work and walking (includes at work and at home, walking to travel from place to place, and any other walking for recreation, sport, exercise or leisure). Based on the time taken on work and walking, the respondents were grouped into four categories of: no activity, low, moderate and high activity levels to show the intensity of their physical activity. The resultant variable was coded such that no and low activity (<10 min of physical activity)=1 and moderate and high activity (≥10 min of physical activity) were coded=0.

Poor fruit and vegetables consumption was created when an individual reported daily consumption of less than the recommended five servings of fruits and vegetables. Respondents reported the number of servings for fruits/vegetables they had in a typical day, and if the servings were less than five in a day, they were considered to be having poor fruit/vegetable consumption.13 27 28

A wealth index (WI) was constructed to be a measure of wealth status. WI is a composite measure of ownership of consumer durables, housing characteristics and access to public services.16 Information on a range of durable assets was collected during the survey (eg, car, refrigerator, television), housing characteristics (eg, material of dwelling floor and roof, main cooking fuel), access to basic services (eg, electricity supply, source of drinking water, sanitation facilities) and ownership of livestock (eg, cattle, goats, sheep, horses, chickens). Further to collection of information on durable assets, information on ownership of land was also collected. Principal component analysis was employed to derive the WI variable, which had five categories from the first to the fifth quintile (poorest to richest).

Control variables

Age and place of residence were used as control variables. These variables were conceptualised to have an association with the outcome variable. Therefore, as control variables they are included in the net effects model, so that the association between the independent variables of interest becomes isolated and discernible.

Statistical analysis

Descriptive and multivariable logistic analysis were used. Descriptive analysis provided bivariate association of explnatory and outcome variable for further exploration of key determinants of overweight and obesity in Botswana. In the multivariable binary logistic regression, the outcome variable was coded as ‘1’ if respondents were overweight/obese and ‘0’ if they were not overweight/obese. Results of logistic regression analysis were presented together with adjusted ORs (AORs), 95% CI and 5% statistical significance level. χ2 test was used for bivariate analysis. Data analysis was done using SPSS V.25 programme. Since this was a population-based study representative of the population, selection bias was minimal and missing cases were few. In order to control for cluster effects, complex sample module in SPSS has been used.

Patient and public involvement

No patients were involved in developing the research question, outcome measures and overall design of the study.

Results

Sample characteristics

Table 1 presents socioeconomic characteristics of the respondents. The sample constituted a high proportion of females (69.1%) than males (30.9%). The sample age distribution suggests a relatively young population, with over half (59%) of the sample being <39 years of age, and almost three-quarters (73.5%) being <50 years of age.

Table 1

Sociodemographic characteristics

Variable	N	%
Sex
Male	364	30.9
Female	813	69.1
Total	1177	100
Age in years
<24	270	26.4
25–34	302	29.5
35–44	196	19.2
45–54	130	12.7
55–64	75	7.3
65+	50	4.9
Total	1023	100
Locality type
Cities/towns	355	30.2
Urban villages	534	45.4
Rural settlements	288	24.5
Total	1177	100
Marital status
Never married	864	73.8
Currently married	199	17.0
Formerly married	108	9.2
Total	1171	100
Highest level of education attained
Primary or less	410	35.5
Junior secondary	314	27.2
Senior secondary	200	17.3
Tertiary and over	230	19.9
Total	1154	100
Work status in last 12 months
Public sector	122	10.5
Private sector	182	15.7
Self-employed	130	11.2
Not employed	436	37.5
Homemaker-student	218	18.8
Retired-other	74	6.4
Total	1162	100
Wealth status
Lowest	234	19.9
Second	237	20.1
Middle	235	19.9
Fourth	237	20.1
Highest	235	19.9
Total	1178	100

More than two-fifths (45.4%) of the population resided in urban villages; while one-third of them resided in cities and towns (30.2%), and a quarter of them lived in rural areas (24.5%). Almost three-fourth (73.8%) of respondents was never married; over a third (35.5%) had primary education or less; over a quarter (27.2%) had junior secondary education while just under a fifth had senior secondary education (17.3%) and tertiary education (19.9%). Close to two-fifth (37.5%) respondents were not employed; over a quarter were employed in either the public or private sector (26.2%).

Table 2 shows behavioural characteristics of the study population. It shows that prevalence of alcohol consumption was estimated at 17.3% while tobacco smoking was estimated at 11.5%. Poor physical activity was also high in the study population with 48.9% of respondents indicating poor physical activity. Poor fruit and vegetable intake was estimated at 82.5%.

Table 2

Behavioural characteristics of the study population

Variable	N	%
Smoking
Yes	136	11.5
No	1042	88.5
Total	1178	100
Alcohol consumption
Yes	204	17.3
No	974	82.7
Total	1178	100
Poor physical activity
Yes	576	48.9
No	602	51.1
Total	1178	100
Poor fruit/vegetable intake
Yes	1045	82.5
No	133	17.5
Total	1178	100

Prevalence of overweight/obesity

Estimated prevalence of overweight/obesity in the study population was 41.3% (table 3). There were significant gender differences in prevalence of overweight/obesity. It was found that women than men (48.6% among women, 25.5% among men) were overweight/obese. Individuals residing in rural areas (46.8%), public sector employees (55.2%) and those with low education (45%) showed high prevalence of overweight/obesity. Meanwhile, it was found that wealth status was not significantly associated with overweight/obesity. Among the behavioural factors, the prevalence of overweight/obesity was greater among non-smoker (42.9%). Alcohol consumption, poor physical activity and poor fruits and vegetable consumption did not show any significant association with overweight/obesity.

Table 3

Prevalence of overweight/obesity by socioeconomic and behavioural characteristics of the study population

Variable	N	%	P value
Sex
Male	364	25.5	≤0.001*
Female	813	48.6
Age (years)			≤0.001*
≤24	270	21.8
25–34	302	36.8
35–44	196	53.0
45–54	130	54.5
55–64	75	61.8
65+	50	43.8
Marital status			≤0.001*
Never married	864	35.5
Currently married	199	62.2
Formerly married	108	54.1
Education			0.019*
Primary or less	410	45.4
Secondary	514	36.7
Tertiary or higher	230	44.6
Residence			0.008*
Cities and towns	355	37.9
Urban villages	534	41.0
Rural villages	288	46.8
Work status			≤0.001*
Public sector	122	52.2
Private sector	182	36.8
Self-employed	130	55.0
Not employed	436	45.8
Home-maker/student	218	29.4
Retired/other	74	40.3
Wealth status			0.765
Lowest	234	38.8
Second	237	43.2
Middle	235	41.1
Fourth	237	39.7
Highest	235	44.1
Alcohol consumption			0.445
No	766	33.3
yes	412	37.7
Smoking			0.004*
Yes	136	29.7
No	1042	42.9
Poor physical activity			0.547
Yes	376	41.5
No	771	39.6
Poor fruit/vegetable consumption			0.832
No	133	41.9
yes	1045	43.0
Overall	1178	41.3

*χ2 test statistically significant at p<0.05

Determinants of overweight/obesity

Three models were run to assess determinants of overweight/obesity (table 4). Model I is the crude model and it assessed the association between overweight/obesity and each of the socioeconomic and demographic variables and behavioural factors independently, model II assessed the association between overweight/obesity adjusting for socioeconomic factors only, while model III examined the association between overweight/obesity adjusting for both socioeconomic and behavioural factors. It was observed that gender had a significant association with overweight/obesity when controlling for both socioeconomic and behavioural factors. Women were observed to have more than two times (AOR=2.74, 95% CI 1.92 to 3.90) greater odds for overweight/obesity than men.

Table 4

ORs from multivariable logistic regression for overweight/obesity among adults in the study population, Botswana

Variable	Model I	Model II	Model III
	OR (CI)	OR (CI)	OR (CI)
Sex
Male	1.00	1.00	1.00
Female	2.75* (2.08 to 3.64)	3.01* (2.12 to 4.26)	2.74* (1.92 to 3.90)
Age (years)
≤24	1.00	1.00	1.00
25–34	2.08* (1.42 to 3.03)	1.77* (1.12 to 2.77)	1.83* (1.17 to 2.88)
35–44	4.02* (2.67 to 6.08)	3.70* (2.23 to 6.12)	3.87* (2.34 to 6.42)
45–54	4.28* (2.70 to 6.78)	3.04* (1.66 to 5.55)	3.26* (1.78 to 5.98)
55–64	5.78* (3.26 to 10.2)	4.91* (2.35 to 10.2)	5.53* (2.62 to 11.6
65+	2.78* (1.46 to 5.28)	2.74* (1.15 to 6.50)	2.88* (1.21 to 6.86)
Marital status
Never married	0.46* (0.30 to 0.71)	0.68 (0.34 to 1.34)	0.66 (0.33 to 1.32)
Currently married	1.39 (0.84 to 2.29)	1.34 (0.66 to 2.71)	1.28 (0.63 to 2.61)
Formerly married	1.00	1.00	1.00
Education
Primary or less	1.00	1.00	1.00
Secondary	0.69* (0.53 to 0.91)	0.49* (0.29 to 0.83)	1.70* (1.11 to 2.61)
Tertiary or higher	0.96 (0.69 to 1.35)	0.87 (0.58 to 1.32)	1.99* (1.16 to 3.38)
Residence
Cities and towns	1.00	1.00	1.00
Urban villages	1.13 (0.86 to 1.50)	1.06 (0.74 to 1.50)	1.07 (0.75 to 1.52)
Rural villages	1.44* (1.04 to 1.99)	1.37 (0.88 to 2.12)	1.37 (0.88 to 2.12)
Work status
Public sector	1.61 (0.89 to 2.93)	1.30 (0.64 to 2.65)	1.22 (0.59 to 2.50)
Private sector	0.86 (0.49 to 1.51)	1.20 (0.61 to 2.36)	1.19 (0.60 to 2.35)
Self-employed	1.18 (0.65 to 2.13)	1.24 (0.61 to 2.51)	1.29 (0.63 to 2.62)
Not employed	1.25 (0.75 to 2.08)	1.30 (0.70 to 2.40)	1.30 (0.70 to 2.41)
Home-maker/student	0.61 (0.35 to 1.07)	0.96 (0.48 to 1.92)	0.96 (0.48 to 1.92)
Retired/other	1.00	1.00	1.00
Wealth status
Lowest	0.80 (0.55 to 1.17)	0.60 (0.34 to 1.05)	1.03 (0.64 to 1.68)
second	0.96 (0.66 to 1.40)	0.67 (0.40 to 1.11)	1.12 (0.68 to 1.85)
middle	0.88 (0.60 to 1.28)	0.73 (0.45 to 1.19)	1.20 (0.71 to 2.01)
Fourth	0.83 (0.57 to 1.21)	0.79 (0.50 to 1.23)	1.53 (0.87 to 2.69)
Highest	1.00	1.00	1.00
Smoking		N.I
Yes	0.56* (0.37 to 0.83)		2.16* (1.22 to 3.83)
No	1.00		1.00
Poor physical activity		N.I
Yes	1.08 (0.83 to 1.40)		1.46* (1.03 to 3.24)
No	1.00		1.00
Poor fruit/vegetable consumption		N.I
Yes	0.82 (0.55 to 1.22)		0.80 (0.28 to 2.24)
No	1.00		1.00
Alcohol consumption		N.I
Yes	0.66* (0.47 to 0.91)		1.23 (0.53 to 2.83)
No	1.00		1.00

Model I: Crude model; Model II: socioeconomic variables included; Model III: socioeconomic+behavioural characteristics included.

*Statistically significant at p<0.05.

N.I, not included.

The odds of being overweight/obese increased with age after controlling for socioeconomic and behavioural factors and were highest among individuals aged 55–64 years (AOR=5.53, 95% CI 2.62 to 11.6). However, it declined at ages 65+years (AOR=2.88, 95% CI 1.21 to 6.86). Education was also found to be a significant factor associated with overweight/obesity when controlling for socioeconomic and behavioural factors. Individuals who had secondary (AOR=1.70, 95% CI 1.11 to 2.61) and tertiary or higher education (AOR=1.99, 95% CI 1.16 to 3.38) were found to have higher probability of being overweight/obese than individuals with primary or less education. Work status and wealth status were not significantly associated with overweight/obesity.

The association between smoking and alcohol consumption with overweight/obesity was found to be negative in the crude model. However, when controlling for other socioeconomic and demographic factors, the odds of overweight/obesity increased two times (AOR=2.16, 95% CI 1.22 to 3.83) among smokers than non-smokers. Poor physical activity was also significantly associated with overweight/obesity (AOR=1.46, 95% CI 1.03 to 3.24) adjusting for socioeconomic and behavioural factors.

Discussion

The prevalence of overweight/obesity in the study population was 41.3%. Botswana also seems to have joined with other Southern African countries where the national prevalence rates ranges between 30% and 60% among adults age 15 years and over.10 26–31 Nutrition transition from high-fibre, low-calorie diets rich in fruits and vegetables to refined, energy-dense foods high in fat, calories, sweeteners and salt12 and low physical activity coupled with transition from agrarian to a sedentary industrial society contributes to a greater risk of overweight/obesity in Botswana.32

Gender differences in overweight/obesity found in this study is a well-established finding and is consistent with most previous studies.10 33 Congruent with most studies in developing countries women were found to have higher odds of overweight/obesity than men.34–36 This is contrary to findings from some developed countries such as United States (US) and United Kingdom (UK) where it was found that adult men tend to be overweight/obese than adult women.37–40 In developing countries such as Botswana, adult women are generally noted to be overweight/obese than adult men at all ages.26 34 38 39

In the context of Botswana, gender differences in overweight/obesity can be explained by sociocultural dynamics and complex sociocultural pathways that encourage weight gain among women.26 39 For example, cultural values favour larger body size among women as a sign of fertility, healthfulness or prosperity in Botswana.26 In this study, women reported lower levels of physical activity which may also contribute to higher prevalence of overweight/obesity among them.

Overweight/obesity was highest among individuals aged 55–64 years but declined at ages 65+ years. This finding corroborates data from large scale surveys41–44 which show that mean body weight and BMI gradually increases during most of adult life and reach peak values at 50–59 years of age in both men and women and after the age of 60 years, mean body weight and BMI tend to decrease.33 45 46 Similarly in South Africa it was found that overweight/obesity tends to increase with age.21

Education differences were also observed for overweight/obesity. Individuals who had high educational attainment had higher odds of overweight/obesity than those with low education. This is consistent with other studies that have shown that education is associated with overweight/obesity via socioeconomic status, literacy and health behaviours.5 47 48

There was no significant association observed between overweight/obesity and work status and for wealth status. Other studies in neighbouring countries with comparable Human Development Index (HDI) as Botswana including Namibia and South Africa27 28 have also shown no wealth or work status differences in overweight/obesity. There are several reasons to such scenario. These reasons include; rapid and disorganised urbanisation, increasing sedentary lifestyles, easy access to and consumption of unhealthy food and high energy drinks among both the poor and non-poor in these countries.21 27 28 Further to this a general observation has been made by Ford et al 5 that overweight/obesity is no longer an issue of affluence in majority of LMICs.

Further, the study noted that the odds of overweight/obesity were two times greater among smokers than non-smokers. Tuovinen et al 23 based on literature review observed that heavy smokers tend to have greater body weight than do light smokers or non-smokers, which likely reflect a clustering of other risky behaviours (eg, low level of physical activity and poor diet) that are conducive to weight gain.23 24 On the other hand, some studies have shown that smoking tends to suppress appetite leading to low BMI among smokers.10 21 30

Other important finding in this study is that overweight/obesity was significantly associated with poor physical activity. This is consistent with other studies which have shown that a decline in physical activity leads to adiposity.5 The interpretation of this finding is that Botswana has undergone a transition from an agrarian to industrial society due to urbanisation subsequently leading to industrial modernity. As a result there is an increase in poor physical activity in the general population.24 Poor physical activity leads to overweight/obesity through energy imbalance between calories consumed and calories expended.5 Generally formal physical activity that is planned, structured, repetitive and purposeful is not common in Botswana. Thus in order to increase physical activity and reduce overweight/obesity there is need for a population-based, multi-sectoral, multi-disciplinary and culturally relevant approach.

This study benefits from data collected from a large and randomly selected sample of respondents. The data contained information on potential confounding factors, with a low proportion of missing information making the study more comparable. Given that the recall period for behavioural risk factors such as smoking, alcohol consumption, poor physical activity and poor fruit and vegetable consumption was 30 days or less there were minimal chances of recall bias. The main limitation is that since the study used a cross-sectional design, it was not possible to establish the causal relationship between explanatory variables and overweight/obesity. Furthermore, the NCDs study sample was not designed to be representative of the whole of Botswana. However, the findings of this study give an indication of emerging patterns of overweight/obesity in the country.

Conclusion

In conclusion our findings indicate high prevalence of overweight/obesity in the adult population of Botswana. Overweight/obesity was found to be significantly high among women, older adults and people with higher education levels. Behavioural factors such as smoking and poor physical activity were found to be key correlates of overweight/obesity. These results demonstrate that socioeconomic and behavioural factors play an important role in prevalence of overweight/obesity. Botswana's multisectoral national strategic plan implementation framework for combating NCDs should focus on developing a community based and people centred awareness about the harmful impact of smoking, alcohol, physically inactivity and poor quality of diet facilited by comprehensive national monitoring and evaluvation framework.