Table 3.
Intake of food groups by gender (Diff = male – female)
| Food group | Mean intake (servings per day) | Univariate | Multivariate† | |||
|---|---|---|---|---|---|---|
| Female | Male | Mean difference (95% CI) | p-value | Mean difference | p- value | |
| Vegetables | 3.63 | 3.25 | −0.38 (−0.65 to −0.10) | 0.01 | −0.47 (−0.86 to −0.07) | 0.02 |
| Cereals | 3.78 | 3.51 | −0.27 (−0.43 to −0.11) | 0.001 | −0.21 (−0.44 to 0.001) | 0.05 |
| Sweet | 1.44 | 1.31 | −0.12 (−0.22 to −0.02) | 0.01 | −0.18 (−0.32 to −0.04) | 0.01 |
| Oil | 1.62 | 1.58 | −0.04 (−0.16 to 0.07) | 0.48 | −0.09 (−0.25 to 0.07) | 0.28 |
| Fish | 0.92 | 0.91 | −0.01 (−0.07 to 0.06) | 0.88 | 0.01 (−0.08 to 0.10) | 0.75 |
| Dairy | 0.29 | 0.28 | −0.004 (−0.05 to 0.04) | 0.87 | −0.02 (−0.08 to 0.04) | 0.56 |
| Eggs | 0.25 | 0.26 | 0.001 (−0.04 to 0.04) | 0.96 | 0.02 (−0.04 to 0.07) | 0.58 |
| Pulses | 1.24 | 1.31 | 0.07 (−0.04 to 0.18) | 0.19 | 0.06 (−0.08 to 0.20) | 0.38 |
| Fruits | 3.13 | 3.21 | 0.08 (−0.16 to 0.33) | 0.5 | 0.05 (−0.27 to 0.38) | 0.75 |
| Tuber | 1.39 | 1.48 | 0.09 (−0.04 to 0.21) | 0.17 | 0.08 (−0.09 to 0.25) | 0.34 |
| Alcoholic beverage* | 0.01 | 0.01 | 0.002 (−0.01 to 0.01) | 0.65 | −0.005 (−0.02 to 0.01) | 0.5 |
| Non-alcoholic beverage* | 1.07 | 1.18 | 0.11 (0.02 to 0.19) | 0.01 | 0.05 (−0.06 to 0.21) | 0.35 |
| Meat | 0.55 | 0.7 | 0.14 (0.08 to 0.20) | <.0001 | 0.14 (0.06 to 0.21) | 0.001 |
| Dietary diversity score | 10.99 | 11.07 | 0.08 (−0.03 to 0.18) | 0.17 | 0.06 (−0.08 to 0.20) | 0.42 |
1 We considered alcohol intake as a special sub-group of beverages, but in computing dietary diversity, total beverage intake was considered.
2 Multivariate estimates adjusted for demographic factors including age (<30, 30 – 50 and >50 years), educational attainment (primary/none, secondary, tertiary), occupation (business/professional, skilled formal, skilled informal, unskilled, unemployed), marital status (never married, not currently married and married/cohabiting), household assets (0 – 1, 2 – 3 and 4 - 5), household size (1 - 4, 5-6, More than 6), district (Ilala, Kinondoni and Temeke), and season of ART initiation [December – March (long rains), April – May (harvest), June – September (post-harvest) and October – November (short rain)]; social factors such as disclosure of status (yes, no), smoking history (yes, no), amount spent on food (US$1 or less, more than US$1), and social support (never, much less than desired, less than desired, as much as desired). The estimates were also adjusted for clinical factors including hemoglobin (Greater than 11mg/dl, 8.5 to 11mgdl, less than 8.5mg/dl), WHO clinical stage of HIV infection (1 - 4), body mass index (less than 18.5, 18.5 to 24.99, above 25 kg/m2), mid-upper arm circumference (less than or equal to 23cm if male or 22cm if female, above 23cm if male or 22cm if female), loss of appetite (yes, no), presence of mouth sores (yes, no) or oral thrush (yes, no) and history of tuberculosis (yes, no).
3 Dietary diversity score was a simple count of the food groups usually consumed in the last month: beverages (alcoholic and non-alcoholic), cereal, dairy, eggs, fish, meat, oil, pulses, sweet, tubers, vegetables and fruits based on the guidelines of the Food and Agricultural Organization (FAO)
4 Mean difference and confidence intervals were estimated from linear regression models using female gender as the reference. Difference estimates less than zero mean that intake was greater among females while estimates greater zero suggest that intake was greater among males.