Contents - Previous - Next

Results

BMI profile

The average measurements of heights and weights and BMI of adult men and women surveyed during the two periods 1974-79 and 1988-90 are given in Table 1 and their distributions according to different BMI classes are set out in Table 2. Results show that while the men were taller and heavier than their women counterparts, there was no significant difference between the average BMI values of men and women. Slightly higher coefficients of variation in BMI of women compared with men could be attributed to relatively more women being obese than men. Since the gender difference was not significant, a common BMI criterion of classification was used as reference for both men and women for studying the relationship of BMI with other parameters.

Table 2 shows that ±48.3% of men and 46.6% of women belonged to the 'normal' BMI category (18.5-25.0) and another 2.7% men and 4.1% women to the obese category (BMI >25.0). The remaining were distributed in CED categories of different grades. In other words, nearly one-half (49%) of the adult population is suffering from some grade of CED. Thus, in the context of the Indian rural population, CED is of primary significance rather than the obesity or overweight, as found in Western and in the affluent sections of other populations. Since comparable data from the NNMB repeat survey report were available for two distinct periods, 1974-79 and 1988-90, an attempt was made to find change, if any, in BMI status of the population between the two periods by plotting frequency distribution curves for men and women separately (Figs 1 and 2). The slight shift towards the right seen in the figures is suggestive of some improvement in the nutritional status of the adult population over the period.

Adult BMI and child nutrition

For this, the anthropometric data of preschool children and adults belonging to the same household were used. The BMI was calculated for each adult individual in the household separately and the average for the household was arrived at (HH BMI). The preschool children were classified into different nutritional grades on the basis of SD classification for weight-for-age using NCHS reference values (WHO, 1983). The bivariate distribution was derived between adult HH BMI values and the child nutritional category, based on weight-for-age (Table 3). Results show that the house-holds with lower BMI values had larger proportions of underweight children. The households with BMI value <18.5 had 85% of the children malnourished whereas those with average HH BMI >18.5 had ±62% malnourished children. A similar trend was seen during the period 1988-90. Thus the adult BMIs and the children's nutritional status were associated (x2: P < 0.01).

Table 1. Mean anthropometric measurements of Indian adults

Table 2. Distribution of adults according to BMI classification

There was also an association between the 'stunting' (height-for-age status) of children and the CED status of their households (Table 4) but this relationship was more marked in relation to the degree of wasting (low weight-for-height) in the children (Table 5).

Table 3. Nutritional status (weight-for-age) of preschool children according to the BMI status of adults in the family

Table 4. Nutritional status (height-for-age) of preschool children according to the BMI status of adults in the family

Table 5. The relationship between the wasting (weight/height) of preschool children and the BMI of the adults in their families

BMI and socio-economic status

The occupational status per capita income of households reflects their poverty levels. The relationship between BMI and these socioeconomic variables, given in Tables 6 and 7, shows that the landless households with agricultural labourers and those with the lowest monthly per capita income had lower values of BMI compared with those with better occupations (cultivators, artisans etc.) and incomes, i.e. more than 60 rupees.

BMI and energy consumption

Analysis of energy intake data (assessed through a household food consumption survey) and the BMI of adults showed a consistent relationship - the higher energy intakes were seen in households with the better BMI status (Table 8). Table 8 also shows that the percentage of households consuming <1800 kcal per capita progressively decreased from 42% in the <16 BMI group to 24% in the normal BMI group.

Table 6. Mean BMI according to occupation

Table 7. Mean BMI according to income

Table 8. Energy intake of adults according to BMI status

NNMB Surveys, 1990-91.

Table 9. Average BMI, height, weight, arm circumference, triceps skinfold thickness (TST) according to energy intake

Table 9 provides the different anthropometric indices of adults, including BMI, according to the energy intake status of households. A weak but consistent relationship was observed between energy intake status and the body weight and the BMI of the adults. In respect of height, arm circumference and triceps skinfold thickness, such a relationship was not present. The prevalence of CED in relation to individual stature (short, average or tall) examined in Table 10 suggests that the order of energy deficiency is more or less the same in all stature groups (Norgan, 1990). In other words, the data clearly indicate that persons with better heights in undernourished populations are also suffering from various grades of energy deficiency.

BMI, birth weight and maternal variables

Table 11 compares different maternal parameters such as age, parity, haemoglobin and anthropometric status with the birth weights of their infants born (Naidu et al., 1991). Risk ratios were calculated for BMI groups taking the incidence of low birth weight among mothers with BMI values between 20 and 25 as reference. The results show that 63% of women examined were 'normal' (i.e. with a BMI between 18.5 and 25) while the rest belonged to different CED groups (33.4%) or a small obese group (3.6%). Except for body weight the mean values of most of the maternal parameters studied varied within a narrow range across different BMI classes. The mean birth weights, however, showed definite differences between the BMI classes. The birth weights invariably increased with increasing BMI of mothers. The mean birth weight was 2510g in grade III or severe CED group (BMI <16) while it was (2800 g in the normal BMI group and 2972 ± in those with grade I obesity (BMI 25-30). The incidence of LBW (<2500g) was highest (53%) in the severe CED group and gradually declined as the BMI status of mothers improved. Its incidence was lowest, i.e. (15% in mothers with a BMI of 25-30. The analysis also showed that the risk (odds ratio) of LBW was more in the extreme BMI range, i.e. in those with CED III and obesity. It is noteworthy that the BMI group of 18.5-20 is as safe as that with a BMI of 20-25 group as far as the proportion of babies is concerned.

Continue

Contents - Previous - Next