Descriptive and univariable model findings As shown in table 2, this study found high BMI prevalence of 35.9 percent. This study found that 25.3% and 18.4% of women were poorest and poorer respectively. For the middle-class, richer and richest women accounted for 20.7%, 18.4% and 17.2% respectively. Out of the poorest and poorer women, 12.8% and 25.8% had high BMI respectively. The middle-class (39.6%), richer (52.1%), and richest (58.6%) women were having high BMI greater than the national high BMI prevalence (Figure 1).
INSERT FIGURE 1 HERE
Besides, women were between the ages of 15 and 49 years. Women aged 15-24 years were 35.5% whereas women who were within the ages of 25 to 34 years and 35-49 years constituted 30.6% and 33.9% respectively. Concerning high BMI, nearly half (49.6%) of women aged 35-49 years had high BMI. Roughly two-fifths (43.8%) and 15.9% of women aged 25-34 years and 15-24 years respectively were having high BMI.
Moreover, most women were Christians (75.2%) while about 1 in 5 women (19.7%) were Muslim and just 5.1% were traditional or other believers. Regarding high BMI, about one-third of Muslims (34.0%), nearly two-fifth of Christians (37.1%) and one-quarter (25.4%) of traditional or other believers were having high BMI. Also, the majority (45.2%) of women were currently married while formerly married women were the least (9.8%). Further, more than half (52.1%) of formerly married women had high BMI whereas close to one-fifth (19.4%) of never-married women were having high BMI. The high BMI prevalence among currently married women and those cohabiting were 43.5% and 37.4% respectively.
In this study, most women were Akans (41.3%). More than two-fifth of Akan (45.0%), Ga (45.9%) and other ethnic groups (46.4%) were having high BMI but only about 1 in 5 (22.4%) of women who belong to the northern tribe had high BMI. About 37% of Ewe women were having high BMI.
For place of residence, rural dwellers constituted a little over half of the women (51.0%). One-quarter (25.4%) of women who reside in rural areas had high BMI while 46.7% of urban residents were having high BMI.
Close to three-fourth (72.6%) of women were employed while about one-quarter (27.4%) of women were unemployed. Approximately 2 in 5 (40.4%) employed women had high BMI. Of the unemployed women, 23.6% were having high BMI.
A majority (56.7%) of the women had secondary or higher education while primary educated women were the least (18.8%). Nearly, one-quarter (24.5%) of the women did not have formal education. Secondary/higher educated women had the greatest prevalence of high BMI (41.0%) followed by primary ((33.4%) and uneducated women (25.7%).
From the univariable model, all the study variables were candidates of the multivariable logistic regression model because they all had a p-value less than 0.25 (Table 2).
INSERT TABLE 2 HERE
Multivariable model findings
The model diagnostics used in this study assures accurate model estimates. The final model was selected because of lower AIC when compared to the empty model (4999.3 vs 6201.4). Also, area under the ROC curve (Figure 2) and Hosmer-Lemeshow goodness of fit test (p-value=0.1067) indicate a good fit.
INSERT FIGURE 2 HERE
From table 3, the results based on wealth status revealed that poorer (AOR=2.18, 95%CI: 1.66-2.85) and middle-class women (AOR=4.44, 95%CI: 3.24-6.09) had a higher propensity of having high BMI than poorest women. Also, the odds of having high BMI was 7.75 (95%CI: 5.53-10.86) and 11.03 (95%CI: 8.07-15.06) times greater among richer and richest women respectively than the poorest.
Besides, the analyses from this research revealed that women aged 25-34 years were 3.18 (95%CI: 2.56-3.96) times more likely to have high BMI than women between the ages of 15-24 years. Likewise, the odds of having high BMI was 4.56 (95%CI: 3.44-6.03) times more probable among women aged 35-49 years when compared to those aged 15-24 years.
Concerning marital status, women who were cohabiting and formerly married were 1.69 (95%CI: 1.25-2.27) and 2.31 (95%CI: 1.66-3.23) times respectively more likely to have high BMI relative to women who never married. Among married women, the likelihood of having high BMI was 2.11 (95%CI: 1.63-2.72) times greater than never-married women.
The likelihood of having high BMI among women with primary education was 47% (95%CI: 1.15-1.86) higher than uneducated women. Likewise, women who have attained at least secondary education were 1.59 (95%CI: 1.27-2.01) times more likely to have high BMI when compared to women without formal education.
INSERT TABLE 3 HERE
Household wealth-associated inequality in high BMI
As depicted in Figure 3, the concentration curve falls below the diagonal line indicating a concentration of high BMI among wealthier women.
INSERT FIGURE 3 HERE
From Table 4, the overall concentration index (concentration index = 0.24, 95%CI: 0.22- 0.26) shows a significant household wealth inequality in high BMI among Ghanaian women exist. The positive value of the concentration index indicates household wealth inequality in high BMI was concentrated among wealthier women. Women aged 15-24 years (concentration index = 0.24, 95%CI:0.18-0.30), 25-34 years (concentration index= 0.21, 95%CI: 0.18-0.24) and 35-49 years (concentration index = 0.24, 95% CI: 0.22-0.26) had significant aggregation of household wealth inequalities in high BMI but the inequality difference between the groups was not significant (p-value= 0.6). Concerning marital status, the degree of household wealth inequality in high BMI among never married (concentration index = 0.27, 95%CI: 0.21-0.33), cohabitating (concentration index = 0.21, 95%CI: 0.15-0.27), formerly married (concentration index = 0.16, 95CI: 0.11-0.21) and married (concentration index = 0.27, 95%CI: 0.24-0.30) was significant with a strong inequality difference between the groups (p-value=0.004). Furthermore, the decomposition of the concentration index revealed a significant concentration of high BMI among the wealthier women who were uneducated (Concentration index = 0.32, 95%CI: 0.26-0.38), primary educated (Concentration index = 0.25, 95%CI: 0.20-0.30) and at least secondary educated (Concentration index = 0.19, 95%CI: 0.17-0.21). Also, a significant difference in household wealth inequality was detected among the women’s education groups (p-value=0.003).