Prevalence and Factors Associated With Overweight or Obesity Among Primary and Secondary School Teachers In Moshi Municipality, Kilimanjaro, Tanzania.

Background: Overweight and obesity are global public health problems. Overweight or obesity is estimated to cause 4 million deaths, 3.9% of all Years of Potential Life Lost (YPLL) and 3.8% of all Disability Adjusted Life Years (DALYs). Overweight or obesity also leads to conditions such as diabetes and hypertension. It has been shown that teachers have a greater occurrence of overweight or obesity than the general population. This study aimed to determine the prevalence of overweight or obesity among primary and secondary school teachers and to assess factors associated with these problems in Moshi Municipality, Kilimanjaro region, Tanzania. Methods: This was a cross sectional study in which data was collected using a modied WHO-STEPS Survey tool. Data cleaning and analysis were conducted using STATA version 14.0. Frequency and percentage were used to summarize categorical data while mean/median with their respective measures of dispersion were used to summarize continuous data. The Modied Poisson regression model was used to determine factors associated with overweight or obesity among primary and secondary school teachers. The effect was estimated by Prevalence Ratio with 95% condence intervals and P-value less than 5% considered to be statistically signicant. Results: A total of 968 teachers were included in this study with median (IQR) age of 37, (32-48) years. Most of the participant were female (69.7%) about (51.6%) were secondary school teachers. Overall, the prevalence of overweight or obesity was 72.7%. Teachers who were 35+ years had a 27% (aPR: 1.27; 95%CI: 1.1 – 1.4; p<0.0001) greater prevalence of overweight or obesity than those below the age of 35. Female teachers had a 43% higher prevalence of overweight or obesity than male teachers (aPR: 1.43; 95%CI: 1.3 – 1.7; p<0.0001). Persons with a positive history of smoking had only 1% higher prevalence of overweight or obesity, however this was not statistically signicant. Conclusion: early intervention to design lifestyle interventions targeting overweight or obesity among teachers - especially female teachers and those aged 35+.


Introduction
Overweight/obesity is abnormal, excess fat accumulation that may impair health 1

. The World Health
Organization (WHO) de nes Body Mass Index (BMI) as a simple index of weight and height that is commonly used to classify adults as overweight or obesity 1 . It is calculated by dividing a person's weight in kilograms by the square of height in meters (kg/m 2 ). A BMI of < 18.5 kg/m 2 is classi ed as underweight, BMI of 18.5-24.9 kg/m 2 is classi ed as normal weight, BMI of 25-29.9 kg/m 2 is classi ed as overweight and BMI ≥ 30 kg/m 2 is classi ed as obesity 1 .
Overweight/obesity is a global public health problem 2 . In 2016 overweight or obesity was estimated to cause 4 million deaths, 3.9% of Years of Potential Life Lost (YPLL) and 3.8% of Disability Adjusted Life Years (DALYs) 2 . According to WHO, in 2016 about 2.01 billion adults (39%) aged 18 + were overweight, of these 678 million (33%) were obese 2 . Low-and middle-income countries (LMICs) are highly affected by overweight and obesity 3 . In Africa, the prevalence of overweight is estimated to be 27% and the prevalence of obesity is 8% among adults aged 20 years or more 4 . In Tanzania overweight and obesity is an increasing health problem and the prevalence of overweight and obesity is 26% in the general population 5 . The prevalence of overweight and obesity is high in urban settings compared to rural settings. Generally, the prevalence of overweight or obesity is higher among women (36%) than among men (13%) 6 .
Overweight or obesity leads to high morbidity and mortality 7 . It increases the risk of developing cardiovascular diseases (CVDs) and increases risk of developing co-morbidities like hypertension and diabetes [8][9] . It also increases the risk of developing different types of cancers, e.g., endometrial and colon cancer 10 . Furthermore, overweight or obesity increases the burden of disease and health care costs among individuals/family and community in the health system 7 . Generally, the problem of overweight and obesity in developing countries is neglected. There is a misconception that an overweight or obese person is "healthy".
Teachers are a very important group in the community. Improving the health of teachers also improves their effectiveness as role models. Students are more likely to eat healthier foods if their teachers are of normal weight and are regularly seen to eat healthy foods 11 . Studies have shown that teachers have a higher occurrence of overweight or obesity compared to general population 12 . Excess intake of energy dense foods and little or no physical exercises are believed to cause this observation 12 . Despite several studies on overweight and obesity that have been carried out in Tanzania, few studies have been done in special in uential groups such as primary and secondary school teachers. This study aimed to determine the prevalence of overweight and obesity and to assess factors associated with these problems among primary and secondary school teachers in Moshi Municipality, Kilimanjaro region, Tanzania.

Study Population and Procedures.
The study population included primary and secondary school teachers who were enrolled in the primary study (January 2018). The study excluded participants who were pregnant at the time of the study. Teachers were informed about the study aims and follow-up schedule and those agreeing to participate gave signed consent. Face-to-face interviews using a WHO STEPS Instrument 27 were conducted by trained research assistants who were medical students at Kilimanjaro Christian Medical University College (KCMUCo). The interviews were conducted in the Swahili language. The information collected included social demographic characteristics, economic characteristics and behavioural characteristics such as history of alcohol use, history of smoking and fruit and vegetable intake per day. After the interviews, physical and clinical examinations were conducted including, anthropometric measures such as weight (kg), height (cm) and waist circumference (cm). Blood pressure and blood sugar levels were also measured. A total of 997 primary and secondary school teachers were enrolled. Analysis was conducted on 968 teachers for whom complete information of weight and height were available.

Data Processing, Categorization, and Analysis.
The data were cleaned and analysed using STATA version 14.2. Descriptive statistics were used to summarize data. Proportions were used for categorical variables and mean or median with respective measures of dispersion were used for numerical variables. The Prevalence Ratio (PR) with 95% Con dence Interval (CI) was used to measure the strength of association between overweight or obesity and exposure variables (socio demographic, economic and behavioural characteristics). The Modi ed Poisson regression model was performed to control for confounders. A value of less than 0.05 was considered as statistically signi cant.

Categorization of Variables.
A teacher was considered overweight or obese if their BMI was ≥ 25.0 kg/m 2 . The age of participants was categorized into, 20-34, 35-44, 45-54 and 55 & above. Income per month was categorized into < 50,000Tsh, 50,000-1,000,000Tsh, and > 1,000,000Tsh. Education level was categorized into, Diploma, Bachelor's Degree and Master's Degree or above. The school level at which they teach was recorded as either primary or secondary. Smoking status was categorized as a) has smoked in the past (yes or no), or b) is currently smoking (yes or no). Alcohol use status, whether one has used alcohol in the past (yes or no), is currently using alcohol (yes or no). Fruit and vegetable intake were independently assessed by recording the average number of days in a week one had fruit and vegetable intake, and later were categorized into, 0-3 days or 4-7 days. Physical activity was assessed by asking the number of minutes in a week that one is engaged in moderate or vigorous activity. Those who were active less than 150 minutes per week were regarded not physically active, and those who were active for 150 minutes or more were regarded to be physically active.

Ethical Consideration.
The permission to conduct this study was sought from the Kilimanjaro Christian Medical University College ethical committee. The study was granted ethical clearance certi cate number 1208. Participants who were enrolled gave signed consent. Participants who were found to be overweight or obese received counselling according to the Tanzanian National Treatment Guidelines. Numbers instead of names were used in all the questionnaires used for interview.

Results
Social Demographic characteristics of participants A total of 968 participants were recruited in the study, 69.7% were female, 60.9% had university degree, 63.3% had monthly income of 500,000-1,000,000/=TZS and 75.2% were married or cohabiting. The median age was 37, (IQR) (32-48) years. Table 1 shows other social demographic characteristics of the participants.   (Fig. 1).
There was a correlation between overweight or obesity and age, sex, educational level, type of school teaching, marital status and smoking status. Whereby the burden of Overweight or Obesity increased with age, ranging from 57.3% (18-24 yrs.) to 85.9% (55-60yrs) P-value < 0.0001 (Table 3). And, teachers who ever smoked had signi cantly 21% higher prevalence of overweight or obesity compared to those who had never smoked (PR: 0.79; 95%CI: 0.6-0.9; p 0.042), Table 4.  (Table 4).

Discussion
The ndings from this study showed a prevalence of 72.7% for overweight or obesity among primary and secondary school teachers. This was higher than the ndings of the Tanzania STEPS survey report, which reported the prevalence of 26% 23 . Several factors may explain the observed differences, including that the current study recruited a homogenous population (teachers) from a single district (Moshi municipal -urban). Whereas Tanzania STEPS survey recruited a diverse population representative of the entire Tanzanian population. Also, the current study had higher representation from females (69%), who have been shown to have a higher risk for overweight or obesity than males 24,25,16,26 .
The high prevalence of overweight or obesity among primary and secondary school teachers found in this study (72.7%), was comparable to other studies conducted among teachers, where it was 74.3% in Malaysia 17 , 70% in India 16 , and 85% in Cape Town, South Africa 8 . The observed similarities in all these studies may be explained partially by the fact that they were all conducted in urban populations and that they all recruited mainly female participants. Other studies conducted among teachers, showed a relatively lower prevalence of overweight or obesity; A study from Ghana reported it to be 61%, this study however, recruited only females who were still of reproductive age 11 . A study from Ibadan, Nigeria, reported it to be 55.3%. The fact that 66% of participants in this study were male, could partially explain the lower prevalence 18 . A prevalence of 47.2% was reported by a study from Brazil, in this study it was noted that male teachers had higher prevalence than female teachers 19 .
The striking observation from all the discussed studies above, is they all report higher prevalence of overweight or obesity among teachers than the general population in their respective countries. This study also observed a higher prevalence of overweight or obesity among teachers married or cohabiting (76.5%) than the separated/divorced/single (61.4%). Such a higher prevalence was also observed in Ghana, Tanzania and Nigeria 11,20 . The possible explanation is that married teachers' pay less attention on their appearance, including being physically attractive as compared to individuals who are still searching for their partners. Also, cultural factors and lay beliefs further may contribute to this observation as obesity is culturally embraced and seen as a sign of wealth and good life in most of our communities, and therefore a sign of successful marriage (partnership).
This study has also shown an increase in the prevalence of overweight or obesity with increasing age among schoolteachers. These ndings are consistent with previous studies from Tanzania 6 25 . This accumulated fat provides energy for the growth of the fetus and for the mother to breastfeed 25 . When this accumulated fat is not shed after childbirth and after breastfeeding, over time, the overall fat mass increases and puts the woman at risk of overweight or obesity and its related problems 25 . On the other hand, a study from Brazil 19 , reported a higher prevalence of overweight or obesity among males than their female counerparts 19 . Observations, such as this, from Brazil, may explain the multidimensional nature of risk factors for overweight or obesity, including socialcultural roles.

Limitations Of The Study
The limitations of this study include recruiting teachers from public schools only, therefore the information obtained cannot be generalised to those in private schools. The study was conducted in an urban setting, and therefore the information obtained cannot be generalized to those in rural setting. The study also, excluded teachers who were on sick leave.

Conclusions
There is high prevalence of overweight or obesity among primary and secondary school teachers in Moshi Municipality. Female teachers and those aged 35 years or more are at increased risk. The preponderance of overweight or obesity among the females and those aged 35 years or more points to the need for group speci c or targeted interventions to combat the threat. In developing interventions for weight control, barriers to lifestyle change at the personal, environmental and socio-economic levels should be targeted and stakeholders at different levels should be involved. Policies to regulate dietary habit, provide environments that encourage physical activity behavior, such as creating walkways, and support health services should be formulated and implemented. The permission to conduct this study was sought from Kilimanjaro Christian Medical University College ethical committee. The study was granted ethical clearance certi cate number 1208. Participants who were enrolled gave a signed consent which also gave permission for data collected to be published in scienti c journals. Participants who were found to be overweight or obese received counselling according to Tanzania National Treatment Guidelines. Numbers instead of names were used in all interview reports.

Availability of data and Material
The dataset used for analysis of this study is available from the corresponding author on reasonable request.