Background
In most of the developing world, ownership of modern household assets such as television sets, refrigerators, microwave ovens and washing machines is becoming common. Ownership of these household assets, however, promotes sedentary behaviour which has implications for obesity and non-communicable disease conditions such as hypertension, diabetes, stroke and other cardiovascular conditions. Additionally, increased household asset ownership is an indicator of socio-economic status and is also associated with obesity promoting dietary behaviours especially in urban areas. Very few studies have examined the relationship between household asset ownership and obesity in sub-Saharan Africa where asset ownership is becoming a norm. This paper examined the relationship between ownership of different types of household assets and Body Mass Index (BMI) among a nationally representative sample of Ghanaian women.

Methods
The study analysed secondary data from the 2008 Ghana Demographic and Health Survey (GDHS) involving a total of 4916 women aged 15–49 years. The analytical sample consist of 4010 (weighted sample) non-pregnant females who had valid data on all the variables used in the analyses. The BMI of the women was used as the dependent variable with three categories of normal weight, overweight and obese based on the standard World Health Organization (WHO) classifications. Multinomial logistic regression models were specified to examine the relationship between ownership of household assets and BMI.

Results
Obesity was more common among women whose household owned a television-15.60 %, DVD/VCD-18.58 %, computer-20.70 %, refrigerator-17.16 % and washing machine-27.43 %, but less common among women whose household owned a motorcycle/scooter-7.74 % and a bicycle-7.92 %. Household ownership of DVD was significantly associated with increased odds of obesity (OR?=?1.59, P?<?0.01) while ownership of a motor cycle/scooter (OR?=?2.05, P <0.001), a refrigerator (OR?=?1.33, P?<?0.05), and a television set (OR?=?1.27, P?<?0.10) were associated with higher odds of overweight. Ownership of each additional household asset was also associated with 15 % and 25 % higher odds of overweight and obesity respectively.

Conclusion
The findings indicate that at least one asset among the various domains of household assets examined is associated with either overweight or obesity among Ghanaian women. Also, increased household asset ownership is associated with increased odds of overweight and obesity. Interventions that aim at reducing sedentary and unhealthy dietary behaviours in the phase of increasing asset ownership maybe helpful in addressing the rising prevalence of obesity among Ghanaian women.