Diet and nutritional status in elderly Zimbabweans

Age and Ageing, Nov, 1997 by Theresa J. Allain, Adrian O. Wilson, Z. Alfred R. Gomo, Donald J. Adamchak, Jonathon A. Matenga

Introduction

In the developed world the commonest nutritional problem in adulthood is obesity [1, 2], although in elderly people undernutrition is an increasingly recognized problem [3] and is associated with increased mortality [4] and worse morbidity after acute illnesses [5]. Little is known about the nutritional status of elderly people in the developing world. Although undernutrition has always been considered to be the predominant problem in these areas, the main focus has been on the nutritional status of children and of pregnant and lactating women [6, 7]. In Africa there have been few studies on diet or nutritional status in the elderly population [8-10].

The pattern of nutritional disorders in old age in the developing world is further complicated by sociological changes which are taking place, with a shift from traditional to Westernized lifestyles. Changes in diet and its implications for health are part of this transition [11, 12].

The aim of this study was to document the diet and nutritional status of healthy, elderly Zimbabweans in rural and urban settings. Factors, such as age, gender, income and level of education, which might influence nutritional status, were investigated and ranges of biochemical and haematological variables which may reflect nutritional status were established.

Method

Subjects and location

This project was part of a larger survey of health in elderly people carried out in Zimbabwe between October 1994 and March 1995. The geographical characteristics and sampling method of the study population have already been described in detail [13]. In summary, 278 subjects aged over 60 years (154 female and 124 male) were recruited using a random cluster sample design within a rural district in north-eastern Zimbabwe called Uzumba-Maramba-Pfungwe and in high-density residential areas in Bindura (population 21 167) and Marondera (population 39 384). Uzumba-Maramba-Pfungwe lies between 245 and 100 km from Harare and the two urban locations are approximately 80 km from Harare. Within the rural area diet and physical measurements were not homogeneous, Uzumba district having some features similar to the urban population. Uzumba has a high level of rural development with higher income per capita, more households with protected wells and ventilated pit latrines and more years of education of its population when compared with the poorer rural areas. These differences are described in more detail elsewhere [14]. For this reason, in some analyses, subjects living in Uzumba have been regarded as a separate group (designated rural area II), with Maramba/Pfungwe as rural area I.

Selected subjects were visited in their home by one of two specially trained research nurses who carried out a structured interview and physical examination, in private, in order to complete a detailed questionnaire. The following morning a fasting, venous blood sample was taken. No subject refused to participate. One subject failed to complete the questionnaire as she was called to a funeral. Physical data were incomplete for nine subjects and blood analyses absent or incomplete for 23 subjects.

Dietary history

Diet was assessed by a food frequency questionnaire (Appendix 1) [15]. The questionnaire was developed with the hospital dietician in the main teaching hospital in Harare and was piloted in 1988 [8]. It was designed to assess how frequently subjects were eating foods from the major food groups and the extent to which their diets contained `Westernized' style foodstuffs such as fried food, tinned food, refined sugar or other processed foods. Approximate alcohol intake was calculated in mg alcohol/week by quantifying the frequency of drinking, the type of alcoholic drink and the volume consumed per sitting.

Physical examination

Subjects were weighed in light clothing to the nearest 0.5 kg and their height recorded to the nearest 0.5 cm. These data were used to calculate the body mass index (BMI; height/[weight.sup.2]). Skinfold thickness (SFT) was measured using Harpenden skinfold callipers (British Indicators Ltd) at the triceps and subscapular sites using defined anatomical landmarks [16]. The mean of 3 measurements at each site was calculated. All SFT analyses were carried out with both triceps and the sum of triceps and subscapular. Results were similar whether triceps or triceps and subscapular measurements were used. Association with other variables was stronger for the combined SFT so data are presented for SFT (triceps and subscapular) unless otherwise stated. The waist and hip circumference were measured to the nearest 0.5 cm at the umbilicus, iliac crest and greater trochanter so that the waist: hip ratio (WHR) could be calculated. ,Results were very similar whether the iliac crest or greater trochanter ratio was taken; however, as the WHR based on the greater trochanter measurement consistently showed stronger associations with other parameters, results are presented for the greater trochanter WHR only.

Biochemical and haematological analysis