Fruit and vegetable consumption and prevalence of diet-related chronic non-communicable diseases in Zanzibar, Tanzania: a mixed methods study

Publikation: Bidrag til tidsskriftKonferenceabstrakt i tidsskriftfagfællebedømt

Abstract Background Non-communicable diseases (NCDs) are the leading cause of death in developed countries and account for roughly a third of deaths in developing countries. According to the 2004 Food and Agricultural Organization and WHO joint report on fruit and vegetables for health, low consumption of fruit and vegetables is associated with NCDs. In Zanzibar, the incidence of diabetes has increased from 252 new cases in 2006, to 373 in 2008, in an adult population of just over a million people and hypertension is the second commonest cause of death. We explored the association between fruit and vegetable consumption and prevalence of diet-related NCDs in Zanzibar. Methods We used mixed methods research. The quantitative part of the study is a secondary analysis of data for obesity, hypertension, diabetes, and fruit and vegetable consumption previously collected in the Zanzibar NCD STEPS survey (n=2800, age 25–65 years, done from June to July, 2011). We calculated frequency, percentage, and 95% CIs for age, sex, marital status, level of education, income, tobacco use, alcohol use, obesity (body mass index), hypertension (systolic and diastolic blood pressure), and diabetes (fasting blood glucose). We used the independent sample t test to compare fruit and vegetable intake and sex with the following variables: income, waist-to-hip ratio, body mass index, systolic blood pressure, diastolic blood pressure, fasting blood glucose, and age. These variables had p<0·05 in two by two tables. We also did the independent sample t test for obesity and blood pressure and obesity and fruit and vegetable consumption in rural and urban areas. We did ANOVA to assess the association between hypertension and fruit and vegetable intake. We used SPSS (version 20) for the analyses. The qualitative component includes ten household assessments of fruit and vegetable preparation and consumption and 20 market assessments of fruit and vegetable purchases. We did a systematic analysis of transcribed, coded data to classify themes and sub-themes. Findings Mean daily fruit intake was 72 g (SD 64, range 0–480) and mean daily vegetable intake was 56 g (SD 48, range 0–272). Both are below the minimum of five servings (5×80 g) of fruit and vegetables per day recommended by WHO. Three times more women than men were obese (318/1736 vs 65/1064, 18·3% vs 6·1%). 383 men (38%) versus 535 women (33%) had hypertension and 22 (2·2%) versus 46 (2·8%) had diabetes. The prevalence of diabetes, hypertension, and obesity was higher in urban sites than in rural sites. People from rural areas earned on average 4·5 times less than did those in urban areas (mean annual income 83 708 Tanzanian shillings vs 383 065 Tanzanian shillings [€42 vs €197]) and had less (quantity and quality) choice of fruits and vegetables. Systolic blood pressure varied significantly depending on fruit consumption (p=0·06), but diastolic blood pressure did not (p=0·125). Neither systolic (p=0·757) nor diastolic (p=0·118) blood pressure varied significantly with vegetable consumption. People from urban areas were more likely to have electricity, running water, an electric cooker, and a fridge, and had a more diverse diet and ate more fruits and vegetables, than did people from rural areas. Furthermore, fruit seems to be considered only as a snack and not used in food preparation. Interpretation The prevalence of NCDs was higher in urban areas than in rural areas and onsumption of fruit and vegetables was generally low. This low consumption can be explained by the low availability and high price of fruit and vegetables. Further research using other measures, such as 24 h recall, should be done to assess fruit and vegetable consumption more precisely and assess its association with NCDs in Zanzibar.
OriginalsprogEngelsk
TidsskriftThe Lancet Global Health
Vol/bind380
Sider (fra-til)60302-5
Antal sider4
ISSN2214-109X
StatusUdgivet - 21 okt. 2012

ID: 120029265