Dietary Intakes of Africans in Transition in the North West Province

Abstract

The North West province of South Africa is experiencing a rapid influx of the predominantly African population from relatively underdeveloped rural to developed urban areas. Little is known of the effects of urbanisation on the dietary intakes and health status of the population. The objective of this study was to develop, evaluate and implement a culture sensitive quantitative food frequency questionnaire (QFFQ) and a food portion photograph book (FPPB) for the assessment of the dietary intakes of the African population of North West province as part of the Transition, Health and Urbanisation in South Africa (THUSA) project. A preliminary study, using qualitative and quantitative research methods was undertaken among 32 female subjects representing rural(R), farm worker(F), informal settlement (IS), middle class urban (MCU) and upper class urban (UCU) strata. This information was used to develop a FPPB comprising near life-size, colour photographs of37 foods. The FPPB was evaluated in a sample of 169 African subjects. The 145-item QFFQ was administered by trained Tswana speaking field workers. The reproducibility of the QFFQ was tested by repeat interviews on a sub sample of 144 THUSA volunteers. Relative validity was tested by comparison to intakes obtained from seven￾day weighed food records of 74 THUSA subjects. Urinary nitrogen excretion was used to validate protein intake in 88 subjects. Completeness of the urinary collections was verified using para-amino benzoic acid (P ABA). Basal metabolic rate (BMR.) was used to estimate underreporting. Data was analysed as nutrient and food intakes. Data from the reproducibility and relative validation studies were analysed using Spearman rank correlation coefficients, paired t-tests, quintile distributions, the Bland-Altman technique and structural equation models (SEM). The QFFQ and FPPB were used to collect dietary information from 890 subjects from the above strata of urbanisation.Reported mean intakes and distributions were compared among strata. Reproducibility was satisfactory for energy and macronutrients and was not affected by age, gender or stratum. The QFFQ tended to underestimate intakes compared to the weighed records. Quintile distributions were similar for both methods. A poor correlation was obtained between urinary nitrogen excretion and dietary intake. Validity coefficients using SEM ranged between 0,2 and 0, 7. Possible underreporting was identified in 40% of the total sample but was not lV affected by age, gender or stratum of urbanisation. Comparison of intakes between strata showed that maize was the staple food of the R F, IS and MCU and met the prudent diet recommendations. Micronutrient intakes were low. The diet of the UCU showed a decrease in carbohydrate and increase in fat content and improved rnicronutrient intakes. There was a trend of movement away from a high carbohydrate, low fat diet in the R and F strata to a lower carbohydrate, higher fat diet associated with the chronic diseases of lifestyle in the urban strata. The QFFQ was found to be a reproducible, relatively valid, culture sensitive instrument for the assessment of the dietary intakes of the population of the North West province.