Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Nutritional status and age at secondary sterility in rural Bangladesh

M. Kabir and A. Karim
Department of Statistics, Jahangirnagar University, Savar, Dhaka, Bangladesh

This paper suggests that there may be an inverse relationship between age at menopause and nutritional status, as determined by anthropometry, in poorly nourished populations.


Menarche and menopause demarcate the limits of potential reproductive lifespan in the female. Gray (1) presented a comprehensive review of many of the biological factors related to the timing of menarche and menopause. The relationship of nutritional status to menopause has been inferred from evidence that a decline in age at menopause is occurring in many developing countries and that variations in age at menopause are associated with socio-economic status (2\. However, age at menopause has received less attention than age at menarche, and many studies of age at menopause have suffered from methodological limitations and biases that restrict interpretation of the results. It is inherently difficult to estimate age at menopause because its onset is not a clearly defined event, the age range is broad compared to that at menarche, and the distribution of age at menopause tends to be negatively skewed (1).

According to Henry (3), in a population that does not use contraception, a married couple who remain childless for some years may be considered sterile. This sterility may be male or female, or both partners may be sterile. When a couple remains childless for several years, e.g., five years after marriage, the sterility is primary. If a couple has no further children for some years after a live birth, the sterility is secondary. Five years is considered the time period for defining such sterility (3). Sterility often results from a deteriorating general health status of both husband and wife, and venereal disease is an important cause of sterility.

Bangladesh is one of the poorest countries in the world, with very low per capita income. About 80 per cent of its population lives below the poverty level. Although it is mainly an agricultural country, per capita availability of land is only 0.38 acre, and per capita food intake is only 13 ounces per day. Because of the scarcity of food more than half the population suffers from malnutrition and the vast majority of these are women and children. The purpose of this paper is to investigate age in cases of secondary sterility among Bangladeshi women in relation to their nutritional status.


The data used here were derived from a three-year prospective survey by the International Centre for Diarrhoeal Disease Research-Bangladesh (ICDDR-B) at Matlab in the Comilla District. Matlab is located on the flood plain of the Meghna River basin 40 miles south of Dhaka, the capital of Bangladesh. The field station of ICDDR-B in Matlab has provided a unique opportunity for prospective studies of fertility dynamics in a population not using contraceptives.

A sample of 2,324 women was followed prospectively from October 1975 to April 1978. All currently married women between 15 and 50 years old in 14 villages were selected for study. Women entered the study at time of marriage and left it if the marriage ended. Women were excluded from the study if they migrated out of the area, or if they reached menopause. Women who were surgically sterilized were also excluded from the detailed analysis.

Five female field workers belonging to the locality were selected and trained extensively for the work. One female supervisor was assigned to perform routine checking in the field. At the beginning of the study, a short questionnaire was used to collect retrospective vital events and socioeconomic status of the sample women. The nutritional status of women was assessed by anthropometric measurements, i.e., height, weight, and arm circumference.

Beam balance scales were used to weigh the women. All weights were taken with the women dressed as usual with saris but without shoes or heavy ornaments. Heights were measured in centimetres with a stadiometer. A plastic tape was used to measure left upper arm circumference. After completion of the study the records were brought to the statistical branch of the Dhaka office for processing and detailed analysis. It was expected that the secondary sterility of married women might vary with different socio-economic status.


Socio-economic Characteristics

Among the women interviewed, about 76 per cent had no formal education. Nearly 89 per cent were Muslim and 11 per cent Hindu. About 40 per cent of the respondents' husbands worked in the agricultural sector. They were mainly cultivators and agricultural labourers. Other significant occupations were mainly fishing, dairy work, small businesses, and services.

Nutritional Status of the Women

Many studies indicate a significant impact of nutritional status on demographic variables (2-6). In the present study we considered maternal nutritional status one of the most important socio-economic factors. In order to estimate the nutritional differentials of secondary sterility, the sample women were classified according to nutritional status. The indicators in this study were height, weight, and arm circumference.

The women were classified into two groups by height- those with height less than 148 cm and those 148 cm and above. Similarly, the women were classified according to weight- less than 41 kg and 41 kg and above. The women were classified into two groups by arm circumference- below 21.9 cm and 21.9 cm and above.

About 47 per cent were found to be under 148 cm, and an equal proportion were 148 cm tall or taller. Similarly, 44 per cent of the women weighed less than 41 kg, and 41 per cent weighed 41 kg or more. The remaining 14 per cent were not included in the study because they refused to be weighed. About 42 per cent had an arm circumference less than 21.9 and 41 per cent, 21.9 cm or above. About 17 per cent of the women refused to allow arm circumference measurements and were excluded from the study.

TABLE 1. Distribution of Currently Married Women by Age and Sterility Status

Age Number of
Observed Proportion
of Fertile Women
Expected Value,c F (x)
15-19 514 0.975 0.97
20-24 403 0.963 0.96
25-29 358 0.955 0.96
30-34 437 0.915 0.92
35-39 283 0.792 0.78
40-44 202 0.559 0.57
45-49 74 0.311 0.30
50-54 - 0(d)  

a. Total N - 2,271.
b. Women giving birth within the preceding 60 months.
c. From curve fitted with fourth degree polynomial.
d. Assumed.


The overall estimates of secondary sterility of married women are shown in table 1. Approximately 98 per cent of the women who were 15 to 19 years old were fertile. This proportion decreased gradually up to age 30 to 34 years, and thereafter declined sharply; only 31 per cent aged 45 to 49 remained fertile.

Because of the random variations and errors in the observed values, a curve is fitted by the least-square method in order to remove the fluctuation. This was done by trial and error using first-degree to fourth-degree equations. The criterion of the best fitted curve was the highest R2. A fourth-degree polynomial fitted with R2 = .99. It can be seen in table 1 that expected values are close to observed values, suggesting best fit of the distribution. These expected values were then used to compute life table estimates of non-sterile women.

The probability of obtaining sterile values (qx) was calculated from the data in table 1. F(x) was the proportion of non-sterile women in the age group x to x+n. Then

q(x+n/2) = F(x=N)/F(x)

This is simply the numerical differentiation of the fitted curve at the point (x+n/2). With this qx value, the exact number of non-sterile women at the beginning of the interval Ix was computed by the usual life-table method, where

Ix =I(x-n) (I-qx-n), I17 =1,000

The expected non-sterile life of women aged 17 years was estimated to be 25.7 years (table 2). Similarly, the non-sterile life of women currently aged 32 was found to be 11.5 years, and a small proportion of women 47 years old were fertile for another 2.5 years. In terms of the life-table concept, women currently 17 years old will remain non-sterile for another 25.7 years (i.e., they will reach sterility at age 42.7 years). Again, women currently aged 32 and 47 years will reach sterility at the ages of 43.5 and 49.5 years, respectively. The overall median age at sterility of the currently married women was about 43.6 years. This was calculated from the Ix (number of non-sterile women) column by linear interpolation with class interval x, where Ix = 0.5. The extremely low median age at menopause recorded in Bangladesh suggests that poor nutrition may be associated with premature menopause.

TABLE 2. Construction of Non-sterile Life Table of the Study Women

Age Number of Non-sterile Women(a) Probability of Becoming Sterile during Interval, Exact Number of Non-sterile Women at Beginning of Interval, Expected Non-sterile Life at Beginning of Interval,
    qx Ix ex
17-21 501 0.010 1,000 25.65
22-26 388 0.000 990 20.88
27-31 342 0.042 990 15.88
32-36 400 0.152 948 11.47
37-41 224 0.269 804 8.08
42-46 113 0.474 588 5.13
47-51 23 1.000 309 2.50
52 -56 -      

Median age at sterility - x:( lx =0.5) = 43.6 years.
a. Total N= 1,991.

TABLE 3. Non-sterile Life of Married Women by Age at the Beginning of the Interval According to Nutritional Status

Age Expected Non-sterile Life
Height Weight Arm circumference
< 148cm >= 148cm <41 kg >= kg <21.9cm >=21.9cm
e(x) e(y) e(x) e(y) e(x) e(y)
17-21 25.60 26.01 25.29 26.89 25.41 26.57
22-26 20.60 21.24 20.52 21.89 20.64 21.57
27-31 15.78 16.24 15.91 17.09 15.82 16.76
32-36 11.67 11.54 11.82 12.24 11.57 11.92
37-41 8.22 7.99 8.15 8.38 8.09 8.04
42-46 5.22 5.04 5.13 5.31 5.18 5.00
47-51 2.50 2.50 2.50 2.50 2.50 2.50
Median age at sterility 43.6 43.7 43.6 44.8 43.5 44.3


Many studies indicate an association between socioeconomic characteristics and nutritional status in a population. In order to determine whether there is any relationship between these two variables, women in the study sample were classified according to height, weight, and arm circumference.

Height and Age at Secondary Sterility

The ages at secondary sterility among the study women according to nutritional status are presented in table 3. The age at secondary sterility of currently married women according to height was estimated by the life-table technique. The information provided here suggests that there were no significant differences at age of sterility between short and taller women. Although the proportions of non-sterility for both groups of women decline gradually at higher ages, there is little difference in the age pattern of secondary sterility. In terms of life-table estimates (table 3), we find that the median ages at sterility were 43.6 and 43.7 years, respectively, for women less than 148 cm and 148 cm or above. Apparently height, which is primarily determined by childhood and adolescent growth, has no impact on age at secondary sterility in women.

Weight and Age at Sterility

Weight was considered another important nutritional status variable to determine age at sterility. The variations in weight are even more conspicuous than those for height. Currently married women were classified by age at sterility according to their weights. The results suggest that weight makes little difference in the age pattern of secondary sterility, although some difference is evident in older age groups (table 3). The life-table estimate suggests some difference in non-sterile life for heavier vs. thinner women. The probability of reaching sterility for both groups was also estimated. The expected non-sterile life of women currently 17 years old was estimated to be 26.9 and 25.3 years, respectively, for the two groups. The median ages of secondary sterility were found to be 43.6 for thinner women and 44.8 years for heavier women, a difference of more than one year. This suggests that thinner women may experience an earlier menopause.

Arm Circumference and Age at Sterility

Table 3 a)so shows predicted age at sterility by arm circumference. The average arm circumference of the study women was 21.8 cm, but this figure is below the standard one. For simplicity of discussion, study women were classified into two groups- those having arm circumferences of less than 21.9 cm and those 21.9 cm and above. The purpose was to see whether arm circumference had any effect on age of the women at sterility.

The detailed pattern of age at sterility of currently married women by two nutritional groups can be seen in table 3. The expected non-sterile life was estimated according to current sterility schedule. The expected non-sterile life of women currently aged 17 years was estimated to be 25.4 and 26.6 years, respectively, according to arm circumference of less than 21.9 cm and 21.9 cm and above. The median ages at sterility calculated from the life table by linear interpolation were 43.4 and 44.3 years, respectively, with a difference of about one year. This figure implies that sterility occurs earlier in thinner women.


The purpose of this study was to estimate age at secondary sterility by nutritional status of currently married women from a sample population in rural Bangladesh. The detailed investigation of nutritional status and age at sterility gave evidence that nutritional status was an important influence on sterility status.

The reason for constructing life-table estimates of age at secondary sterility for currently married women is that, once women become sterile, there is little possibility of their becoming fertile again. The results presented here show that there appears to be some difference in age at secondary sterility between better nourished and poorly nourished women. The estimates do suggest that thinner women experience an earlier menopause. There appears to be an inverse relationship between the age at menopause and anthropometric status in poorly nourished populations.


1. R. H. Gray, "Biological Factors Other than Nutrition and Lactation Which May Influence Natural Fertility: A Review," paper presented at a Seminar on Natural Fertility (INED, Paris, 1977).

2. H. W. Mosley, "The Effects of Nutrition on Natural Fertility," paper presented at a Seminar on Natural Fertility (INED, Paris, 1 977).

3. L. Henry, Human Fertility: The Basic Components (Chicago University Press, Chicago, III., USA, 1977), p. 95.

4. R. E. Frisch, "Demographic Implications of the Biological Determinants of Female Fecundity," Soc. Biol., 22 (1): 17 (1 974).

5. R. E. Frisch, "Population, Food Intake and Fertility," Science, 199: 22 ( 1978).

6. R. E. Frisch and J. W. McArthur, "Menstrual Cycles: Fatness as a Determinant of Minimum Weight for Height Necessary for Their Maintenance and Onset," Science, 185: 949 (1974).

Contents - Previous - Next