Perceived Body Weight And Actual Body Mass Index (Bmi) Among Urban Poor Communities In Accra, Ghana

ABSTRACT Body weight (especially overweight and obesity) are of global concern because of their health implications. Notwithstanding the health implications, the prevalence rates continue to increase in both developed and Low and Middle Income Countries (LMICs). The rising prevalence of overweight/obesity is attributed to changing behavioural practices such as reduced physical activity and poor dietary behaviours. Being underweight is associated with high levels of malnutrition, starvation and poverty. Weight misperception can also result in large numbers of individuals with weight problems failing to understand the need for weight control. The aim of the study was to examine the association between perceived body weight and actual body mass index among residents in Ga Mashie and Agbogbloshie. It was a crosssectional survey conducted on a sample of 700 adults aged 15-59 years. It was based on data from the second round of Edulink Urban and Poverty survey conducted in three poor urban communities in Accra, Ghana in 2011 by the Regional Institute for Population Studies (RIPS). Chi-square tests and multinomial logistic regression analysis were used to analyse the data. Basically, respondents misperceived their weight. Although misperception of body weight was found in each category of body weight, underestimation and overestimation of weight was higher among respondents who perceived themselves as underweight and obese (79% and 45% respectively). Even though a sizeable proportion of respondents misperceived their weight, they did nothing about it. The perceptions people had about their weight influenced their dietary behaviour but not their engagement in physical activity. While 78.3% of those who perceived themselves to be overweight belonged to the category of high dietary diversity score, 93.5% of them were physically inactive. Perceived weight remained significantly associated with actual BMI even after controlling for some confounders. Other variables that predicted BMI were sex, age, marital status and type of occupation. Males are less likely to be overweight or obese compared to females (Odds Ratio=0.415 (p=0.002) and 0.072 (p=0.000) respectively). Based on the finding of the study, it is recommended that there should be a continual sensitization of the regenerative health policy on healthy diet especially among women. Intervention programs should also target adults.