This is the old United Nations University website. Visit the new site at http://unu.edu
Subjects and methods
Results
Discussion
References
Detty Siti Nurdiati, Mohammad Hakimi, Abdul Wahab, and Anna Winkvist
Detti Siti Nurdiati and Mohammad Hakimi are affiliated with the Department of Obstetrics and Gynaecology and the Community Health and Nutrition Research Laboratory in the Faculty of Medicine in Gadjah Mada University in Yogyakarta, Indonesia. Abdul Wahab is affiliated with the Community Health and Nutrition Research Laboratory in the Faculty of Medicine in Gadjah Mada University in Yogyakarta, and Anna Winkvist is affiliated with the Department of Epidemiology and Public Health in Umeå University, Umeå Sweden.
Mention of the names of firms and commercial products does not imply endorsement by the United Nations University.
Abstract
There are few studies on the nutritional status of non-pregnant women. A population-based, cross-sectional study of nutritional status in 5,817 non-pregnant women 15 to 49 years of age was conducted in Purworejo District, Indonesia, in 1996. Weight, height, mid-upper-arm circumference (MUAC), and triceps skinfold thickness were measured, and information on socio-economic, demographic, and reproductive factors was collected. Seventeen percent of the women had chronic energy deficiency and 11% were obese. Mean weight, MUAC, and triceps skinfold thickness corresponded to the 25th percentile of standards and mean height to the 5th percentile. Obesity was more common among older women and chronic energy deficiency among both the oldest and the youngest women. Women working in agriculture, not using contraceptives, and not owning a television, radio, or refrigerator were more likely to have chronic energy deficiency. In summary, both chronic energy deficiency and obesity existed in Purworejo, and risk factors were identified. Interventions are needed to improve the nutritional status of girls and women before and after pregnancy.
Introduction
The Safe Motherhood Initiative is a global effort to reduce maternal mortality and morbidity, particularly in the developing world. In 1994 it was noted that Indonesia had the highest maternal mortality among the ASEAN countries: 390 per 100,000 live births, with an inter-provincial variation of 130 to 750 [1]. In rural Java, maternal mortality ranged from 360 to 570. Recognizing the need for urgent measures, the Indonesian government set the target of reducing maternal mortality from 390 per 100,000 to 225 per 100,000 in its Sixth Five-Year Development Plan for 1994-1999 (“Repelita VI”). Factors shown to contribute to the high rate of maternal mortality were limited accessibility, efficiency, and coverage of health care during pregnancy and childbirth; poor nutritional status of the women; poor availability and use of family-planning services, especially among high-risk mothers; women’s heavy workload; and traditional beliefs relating to women’s status and fertility [2]. Thus, efforts to reduce maternal mortality ultimately need to address all of these factors.
Maternal nutritional status is important for a host of reasons-for the woman herself, for her capacity to reproduce, and for the development of her children, with implications for the health and reproductive capacity of the next generation’s mothers. However, for decades, issues in women’s nutrition have centred on nutrition during pregnancy and lactation. Much of the concern has thus been for the newborn’s health and well-being [3-11]. The nutritional issues of women themselves have rarely been investigated. In earlier nutritional research, only a few publications made women’s own health the main objective, and not many nutritional data are available from non-pregnant women. As a result, insufficient attention has been paid to the extent, causes, and consequences of malnutrition among women. The result has been inadequate resources, both public and private, allotted to the improvement of women’s nutrition for their own sake [12].
Only recently has attention been paid to the link between women’s own nutritional and health status and their multiple roles in family and society. Several reviews have emphasized the vulnerability of women throughout their life cycle [12-14]. The biologic and socio-economic differences between women and men sometimes place women at higher risk for malnutrition and mortality. In some countries, girls are treated differently in terms of access to health care, food, and education. Also, girls are at risk for early pregnancy, which may affect their own prepubertal growth. Women of reproductive age are subject to numerous stresses affecting their health and well-being. Finally, elderly women in many societies are deprived [15].
The objectives of this research were to estimate the nutritional status as reflected by anthropometry and to explore the associations between demographic and socio-economic factors and nutritional status among non-pregnant women of reproductive age in Central Java, Indonesia.
The research was conducted in Purworejo District, Central Java, which consists of 16 sub-districts and 494 villages with a total population of 729,825. The total area of Purworejo is 1,035 km2, including both lowlands and mountains. The infant mortality rate has declined over the past five years, and the current estimate is 52 per 1,000 live births. The total fertility rate is 3.13 per woman.
The study took advantage of a large surveillance programme that was initiated in Purworejo District in June 1994 by the Community Health and Nutrition Research Laboratory, Gadjah Mada University, Yogyakarta. A 10% sample of the total population was monitored here, i.e., approximately 14,500 households. A two-stage cluster sampling method with probability proportional to the estimated size of the cluster was used to select households representative of the district. The sampling frame for the first stage consisted of a 20% sample of the Central Bureau of Statistics enumeration areas used in the 1990 census. The frame for the second stage was the household listings of each enumeration area prepared for the 1993 Agricultural Census, from which an equal sample of households was systematically sampled [16].
Through this surveillance system, we had access to an updated list of women of reproductive age (15-49 years) consisting of 13,094 women. For cultural reasons, only married women defined as at risk of becoming pregnant (i.e., not currently pregnant or sterilized) were invited to participate. Women from this sample who later become pregnant were enrolled in a pregnancy study for which the measurements described in this article represent the baseline measures. Pregnancy was identified by the woman’s own report (last menstrual period). In case the pregnancy was not later diagnosed by the midwives (by fundal palpation), the fieldworkers carried out a b-human chorionic gonadotropin test for pregnancy. Only 15 women originally defined as non-pregnant at the time of measurement were later found to have been pregnant; these were removed from the analyses. Individual written informed consent was obtained. Ethical approval was provided by the research ethics committees of the Medical Faculty of Gadjah Mada University and Umeå University.
Measurements
The cross-sectional study of nutritional status among non-pregnant women took place between January and March 1996. Eligible women were invited to undergo anthropometric measurements at the health post in each village. Training and standardization in anthropometric measurements, consisting of weight, height, mid-upper-arm circumference (MUAC), and triceps skinfold thickness, were carried out in December 1995. Weights were measured within 0.1 kg with calibrated electronic scales (Seca Model 835, CMS, London) with clothes as light as possible. A portable stadiometer kit (CMS) was used to measure body height with an accuracy of 0.1 cm. MUAC was measured on the left arm to the nearest centimetre with the insertion-type arm circumference tapes obtained from UNICEF. The Harpenden skinfold caliper (CMS) was used for the skinfold thickness measurements, with an accuracy of 0.1 mm. As reference, the female Canadian standard for weight and height [17] and the US National Health and Nutrition Examination Survey (NHANES) standard for MUAC and triceps skinfold thickness [18] were used.
Women were classified as chronically energy deficient or obese as described by James et al. [19] and the World Health Organization [20]. Chronic energy deficiency grades I, II, and III correspond to body mass index (BMI) 17.0-18.4, 16.0-16.9, and < 16.0, respectively. Women with BMI 18.5-24.9 were classified as normal. Finally, obesity grades I, II, and III correspond to BMI 25.0-29.9, 30.0-39.9, and ³ 40, respectively.
Socio-economic and demographic information for the women were retrieved from the surveillance data collected between August and October 1995. These were categorized in dose agreement with the Indonesian Demographic and Health Survey categorization scheme [1]. A system of quality control of the information collected was instituted in order to ensure good validity of data.
Data analyses
Women were excluded from the analyses if their anthropometric data were incomplete or seemed to be affected by measurement errors. In those cases where only one of the indicators of nutritional status was below the 5th or above the 95th percentile, while others were closer to the median, this was classified as measurement or recording error and the subject was excluded.
Data analyses were performed with the Statistical Package for Social Science (SPSS version 7.1, 1997). The chi-square and t tests were used to compare the nutritional status and background factors of the selected sample with those for all women in the surveillance area. Univariate logistic regressions were used to examine the relationships between nutritional status and other individual characteristics. Variables significant in the univariate logistic regressions were included in multivariate logistic regressions, where possible associated factors were evaluated simultaneously. The logistic regression analyses were constructed with chronic energy deficiency and obesity as dichotomous variables (all chronic energy deficiency categories vs. normal, extreme chronic energy deficiency vs. normal, all obesity categories vs. normal, and extreme obesity vs. normal). Interactions among independent variables likely to show joint effects were evaluated by the construction of a new variable that expressed the combined effect of both. Only significant variables were kept in the final multivariate models (confidence interval not including 1).
Table 1. Mid-upper-arm circumference (MUAC) and background variables for the different subsamples of women
|
Background variable |
At risk of pregnancy (n = 8,154) |
Not at risk of pregnancy (n = 4,940) |
Total |
|||
|
Attended measurement sessions (n = 5,880) |
Did not attend measurement sessions |
|||||
|
Data complete (n = 5,817) |
Data incomplete (n = 63) |
|||||
|
MUAC (cm)-mean ± SDa |
25.8-12.9 |
27.4 4.9 |
25.3 ± 3.0 |
23.7 ± 2.5 |
24.8 2.8 |
|
|
Age (yr)-mean ± SD |
34.9 ± 7.3 |
36.4 6.3 |
35.3 ± 7.9 |
22.6 ± 7.9 |
30.4 9.74 |
|
|
Marital status (%) |
||||||
|
|
Unmarried |
0b |
0 |
0.1 |
65.5 |
24.7 |
|
|
Married |
94.3 |
95.2 |
91.1 |
32.7 |
70.5 |
|
|
Divorced |
1.5 |
1.6 |
2.8 |
0.6 |
1.4 |
|
|
Widowed |
3.3 |
1.6 |
4.6 |
0.7 |
2.6 |
|
|
Separated |
0.9 |
1.6 |
1.5 |
0.4 |
0.8 |
|
Parity |
||||||
|
|
0 |
3.2b |
3.2 |
6.9 |
73.4 |
30.3 |
|
|
1-2 |
40.1 |
33.3 |
39.6 |
16.9 |
31.2 |
|
|
3-4 |
37.2 |
39.7 |
34.3 |
7.1 |
25.3 |
|
|
³ 5 |
19.4 |
23.8 |
19.3 |
2.7 |
13.1 |
|
Contraceptive use |
||||||
|
|
Yes |
57.0b |
61.9 |
48.2 |
4.7 |
35.7 |
|
|
No |
36.7 |
31.7 |
40.3 |
70.5 |
50.1 |
|
|
No data |
6.4 |
6.3 |
11.4 |
24.8 |
14.2 |
|
Residence (%) |
||||||
|
|
Rural |
91.9b |
88.9 |
79.9 |
81.6 |
85.9 |
|
|
Urban |
8.1 |
11.1 |
20.1 |
18.4 |
14.1 |
|
Years of education (%) |
||||||
|
|
0 |
7.2 b |
7.9 |
7.8 |
4.2 |
6.2 |
|
|
1-6 |
69.9 |
65.1 |
57.4 |
32.8 |
53.7 |
|
|
7-9 |
13.1 |
17.5 |
14.0 |
27.4 |
18.7 |
|
|
³ 10 |
9.7 |
9.5 |
20.8 |
35.6 |
21.4 |
|
Occupation (%) |
||||||
|
|
Unemployed or housewife |
24.8b |
30.2 |
26.8 |
67.2 |
41.2 |
|
|
Agricultural |
55.3 |
49.2 |
40.2 |
17.7 |
38.5 |
|
|
Non- agricultural |
19.9 |
20.6 |
33.0 |
15.1 |
20.4 |
Representativeness of the study sample
All 8,154 married women defined as at risk of becoming pregnant were invited to the measurement sessions (fig. 1). Among 4,940 women defined as not at risk of becoming pregnant, 554 women had already become pregnant. Of those invited to participate, 2,274 (28%) did not participate, 282 because they were not at home and the remainder for unknown reasons. Thus, 72% of the women at risk of becoming pregnant attended the measurement sessions. Measurements on 63 women were incomplete or affected by measurement errors. Hence, 5,817 (71%) of the eligible women were included in these analyses, of whom 567 (10%) were breastfeeding.
FIG. 1. Sample selection for nutritional analyses
Because of the selection process, women not at risk of becoming pregnant were different in several ways from women at risk of becoming pregnant. Women in the former group were younger, were more likely to be unmarried, had fewer children, and were more educated, albeit still unemployed. To evaluate possible selection bias, nutritional status and background characteristics were compared for those who attended and those who did not attend the measurement sessions. Among those attending, the group with incomplete data was small (63 women) and was therefore combined with the group with complete data for statistical testing. MUAC had been measured on all 13,094 women between August and October 1995 and therefore could be used for comparative purposes.
The mean MUAC was not significantly different between those who did and did not attend the measurement sessions (t test, p =. 14). However, there were significant differences between these two groups in most background factors (p <.001). The women attending were somewhat less educated and more likely to use contraceptives, live in rural areas, and work in agriculture. However, the difference between the two groups in mean age was barely significant (p =.05).
Nutritional status among women in the study sample
The mean anthropometric values for the 5,817 women in the sample were 47.8 ± 7.9 kg weight, 149.1 ± 5.1 cm height, 25.8 ± 2.9 cm MUAC, 15.0 ± 6.3 mm triceps skinfold thickness, and 21.2 ± 3.1 BMI.
A comparison of the nutritional status of the women with reference data indicated that overall, 37% were below the 5th percentile for weight and 49% were below the 5th percentile for height. In total, 22% were below the 5th percentile for MUAC and 30% were below the 5th percentile for triceps skinfold thickness. The proportion of women falling below the 5th percentiles was significantly different among the different age strata (p <.001). The proportion was lowest for women in the middle of the age range for all four indicators (fig. 2).
The total prevalence of chronic energy deficiency among the women was 17% and the total prevalence of obesity was 11%. Further, chronic energy deficiency grades III, II, I, normal, obese I, and obese II were found among 1.2%, 3.0%, 12.8%, 71.7%, 10.0%, and 1.4% of the women, respectively. Obesity was most common among older women, and chronic energy deficiency was most common among the youngest and the oldest (fig. 3).
Relationship between nutritional status and background factors
Chronic energy deficiency
In univariate logistic regression analyses with normal versus chronic energy deficiency grades I, II, and III as binary dependent variables, chronic energy deficiency was less common among women in the middle range of age (table 2). Chronic energy deficiency was more prevalent among women who worked in agriculture or at home. Women who did not provide any data on the use of contraceptives were more likely to have chronic energy deficiency. Women who did not own a radio, television set, or refrigerator were also more likely to have chronic energy deficiency. However, chronic energy deficiency was less likely among women who did not own a bicycle and among women who had good water supplies.
FIG. 3. Distribution of chronic energy deficiency and obesity according to age group
The association between occupation and chronic energy deficiency remained significant after adjustment for other factors, and so did the association between chronic energy deficiency and the use of contraceptive methods. Women who worked in agriculture or at home had a 28% to 38% higher risk of chronic energy deficiency than non-agricultural workers. Women who did not answer questions about the use of contraceptive methods had almost twice the risk of chronic energy deficiency as women who used contraceptives. Not owning a television set or a radio was associated with chronic energy deficiency.
When normal and extreme chronic energy deficiency(grade III) were compared in univariate analyses, not using contraceptives and not owning a television remained significant risk factors (table 3). Chronic energy deficiency grade III was less common among women who lived in hilly and highland areas. Finally, chronic energy deficiency III was more common among women who lived in houses with wooden floors. In multivariate analyses, all these variables remained significant.
Table 2. Relationship between background variables and chronic energy deficiency: Odds ratios (OR) and 95% confidence intervals (CI) are given for the risk of having chronic energy deficiency I, II, and III versus being normal according to logistic regression analysis (n = 5,817)
|
Background variable |
n |
Univariate analysis |
Multivariate analysisa |
|||
|
OR |
95% CI |
OR |
95% CI |
|||
|
Age (yr) |
||||||
|
|
45-49 |
567 |
1.00 |
|
1.00 |
|
|
|
40-44 |
895 |
0.84 |
0.66-1.08 |
1.00 |
0.77-1.31 |
|
|
35-39 |
1,174 |
0.55 |
0.44-0.72 |
0.67 |
0.52-0.88 |
|
|
30-34 |
1,177 |
0.62 |
0.49-0.80 |
0.78 |
0.60-1.02 |
|
|
25-29 |
828 |
0.72 |
0.56-0.93 |
0.89 |
0.67-1.18 |
|
|
15-24 |
486 |
0.85 |
0.64-1.13 |
1.05 |
0.77-1.43 |
|
Contraceptive use |
||||||
|
|
Yes |
2,871 |
1.00 |
|
1.00 |
|
|
|
No |
1,921 |
1.35 |
1.16-1.56 |
1.30 |
1.12-1.51 |
|
|
No data |
335 |
2.25 |
1.75-2.89 |
1.94 |
1.46-2.57 |
|
Occupation |
||||||
|
|
Non-agricultural |
934 |
1.00 |
|
1.00 |
|
|
|
Unemployed or housewife |
1,233 |
1.44 |
1.15-1.80 |
1.37 |
1.08-1.73 |
|
|
Agricultural |
2,960 |
1.40 |
1.14-1.71 |
1.25 |
1.01-1.54 |
|
Water supply |
||||||
|
|
Private or public tap |
141 |
1.00 |
|
|
|
|
|
Private or public pump or well |
3,467 |
1.72 |
1.04-2.85 |
|
|
|
|
Spring, river, rain, or other |
1,519 |
1.44 |
0.86-2.40 |
|
|
|
Radio ownership |
||||||
|
|
Yes |
4,341 |
1.00 |
|
1.00 |
|
|
|
No |
786 |
1.28 |
1.06-1.54 |
1.23 |
1.02-1.49 |
|
Television ownership |
||||||
|
|
Yes |
1,664 |
1.00 |
|
1.00 |
|
|
|
No |
3,463 |
1.35 |
1.16-1.58 |
1.33 |
1.12-1.57 |
|
Refrigerator ownership |
||||||
|
|
Yes |
81 |
1.00 |
|
|
|
|
|
No |
5,046 |
2.17 |
1.04-4.51 |
|
|
|
Bicycle ownership |
||||||
|
|
Yes |
3,831 |
1.00 |
|
1.00 |
|
|
|
No |
1,296 |
0.77 |
0.65-0.91 |
0.66 |
0.56-0.79 |
None of the combined variables (interactions) were significantly associated with the risk of chronic energy deficiency grades I, II, and III or only chronic energy deficiency grade III.
Obesity
Obesity grades I and II was least common among young women according to both univariate and multivariate analyses (table 4). Women with high parity, those who used contraceptives, and those who were more educated were more likely to be obese, as were women who lived in urban and lowland areas. However, women who worked in agriculture were the least likely to be obese. Obesity was more common among women who had better sanitation, as indicated by good water sources, tile floors, and latrine facilities in their homes, and who owned a television set, bicycle, or motorcycle. However, the associations between obesity and parity, education, urban or rural residence, altitude, type of floor, type of latrine, and refrigerator ownership became non-significant in multivariate analyses.
Older age, high parity, non-agricultural work, and having electricity, a television set, a refrigerator, and a motorcycle were also significant risk factors for extreme obesity (grade II) (table 5). Women who lived in hilly or highland areas and those whose homes had bare earth floors, poor water supplies, and no latrine facilities were less likely to be classified as having obesity grade II. Multivariate analyses that included age, occupation, altitude, type of floor, and motorcycle ownership as independent variables indicated that all five were significantly associated with the risk of obesity grade II.
Table 3. Relationship between background variables and chronic energy deficiency: Odds ratios (OR) and 95% confidence intervals (CI) are given for the risk of having chronic energy deficiency III versus being normal according to logistic regression analysis (n = 5,817)
|
Background variable |
n |
Univariate analysis |
Multivariate analysisa |
|||
|
OR |
95% CI |
OR |
95% CI |
|||
|
Age (yr) |
||||||
|
|
45-49 |
433 |
1.00 |
|
|
|
|
|
40-44 |
715 |
0.53 |
0.26-1.10 |
|
|
|
|
35-39 |
1,007 |
0.38 |
0.18-0.78 |
|
|
|
|
30-34 |
990 |
0.38 |
0.49-0.79 |
|
|
|
|
25-29 |
677 |
0.32 |
0.13-0.75 |
|
|
|
|
15-24 |
383 |
0.28 |
0.09-0.84 |
|
|
|
Contraceptive use |
||||||
|
|
Yes |
2,420 |
1.00 |
|
1.00 |
|
|
|
No |
1,552 |
2.21 |
1.32-3.71 |
2.33 |
1.37-3.97 |
|
|
No data |
243 |
4.12 |
1.95-8.68 |
4.40 |
2.06-9.40 |
|
Education (yr) |
||||||
|
|
³ 7 |
320 |
1.00 |
|
|
|
|
|
1-6 |
2,981 |
1.54 |
0.78-3.05 |
|
|
|
|
None |
914 |
2.90 |
1.19-7.03 |
|
|
|
Altitude |
||||||
|
|
Lowland |
892 |
1.00 |
|
1.00 |
|
|
|
Coastal |
1,958 |
0.65 |
0.39-1.11 |
0.57 |
0.33-0.99 |
|
|
Hills or highland |
1,365 |
0.29 |
0.14-0.60 |
0.27 |
0.13-0.57 |
|
Type of floor |
||||||
|
|
Ceramic or tile |
2,230 |
1.00 |
|
1.00 |
|
|
|
Wood |
17 |
7.49 |
1.66-33.87 |
6.24 |
1.33-29.39 |
|
|
Soil |
1,968 |
0.81 |
0.50-1.32 |
0.68 |
0.39-1.19 |
|
Television ownership |
||||||
|
|
Yes |
1,411 |
1.00 |
|
1.00 |
|
|
|
No |
2,804 |
1.83 |
1.03-3.25 |
2.21 |
1.19-4.12 |
None of the combined variables (interactions) were significantly associated with the risk of being classified into obesity grades I and II or grade II alone.
The coverage of women at risk of becoming pregnant was 72%. Because some of the background factors differed between those attending and those not attending the measurement sessions, extrapolation of our results should be done with some caution. Relatively fewer highly educated women in the city were represented.
For cultural reasons, only married women were available to the study. However, most women in Purworejo are Moslem, and practically all women who become pregnant are married. Nevertheless, information on unmarried women would have been desirable. Most importantly, the mean MUAC did not differ between those attending and those not attending the measurement sessions, indicating that we did not have a selection bias as to anthropometry on our study. Thus, overall the study was performed on a large, population-based sample where representativeness was evaluated and deemed appropriate.
The mean weight of the women in Purworejo was 47.8 ± 7.9 kg. This was higher than the mean weight obtained in previous studies in Indonesia in East Java (data collected in 1982-1985 and 1987-1989) and West Java (data collected in 1991-1992), which were 42 and 46 kg, respectively [21, 22]. This could be due to improved economic conditions in Purworejo District. On the other hand, the mean height of the Purworejo women was slightly lower: 149.1 ± 5.1 cm, as compared with 150 cm in East Java [21] and 152 cm in West Java [22]. Several major achievements in Indonesian national development took place during the 1980s and 1990s: economic growth, self-sufficiency in food, a sharp decline in infant mortality, a reduction in population growth, and a significant reduction in the number of the poor. The breakdown by province of these achievements, particularly economic growth, also showed a similar pattern to the national figure [23]. These developments might have positively influenced the current nutritional status of the women, whereas their relatively short stature may reflect poorer conditions in their childhood. Environmental and socio-economic factors influence childhood growth in height and weight more than genetic factors [24]. Comparison of our sample with other Indonesian or foreign studies should be made with caution, because we have selected only. women at risk of becoming pregnant, whereas other studies may include pregnant as well as non-pregnant women.
Table 4. Relationship between background variables and obesity: Odds ratios (OR) and 95% confidence intervals (CI) are given for the risk of having obesity I and II versus being normal according to logistic regression analysis (n = 5,817)
|
Background variable |
n |
Univariate analysis |
Multivariate analysisa |
|||
|
OR |
95% CI |
OR |
95% CI |
|||
|
Age (yr) |
||||||
|
|
45-49 |
533 |
1.00 |
|
1.00 |
|
|
|
40-44 |
848 |
0.84 |
0.64-1.11 |
0.74 |
0.55-1.00 |
|
|
35-39 |
1,174 |
0.74 |
0.57-0.96 |
0.60 |
0.45-0.80 |
|
|
30-34 |
1,100 |
0.50 |
0.38-0.67 |
0.40 |
0.29-0.55 |
|
|
25-29 |
750 |
0.48 |
0.35-0.66 |
0.42 |
0.30-0.59 |
|
|
15-24 |
400 |
0.22 |
0.14-0.36 |
0.22 |
0.13-0.36 |
|
Parity |
||||||
|
|
0-2 |
2,077 |
1.00 |
|
|
|
|
|
3-4 |
1,792 |
1.14 |
0.94-1.37 |
|
|
|
|
³ 5 |
936 |
1.43 |
1.16-1.78 |
|
|
|
Contraceptive use |
||||||
|
|
Yes |
2,822 |
1.00 |
|
1.00 |
|
|
|
No |
1,718 |
0.75 |
0.63-0.89 |
0.82 |
0.68-0.99 |
|
|
No data |
936 |
0.80 |
0.54-1.16 |
0.63 |
0.41-0.96 |
|
Education (yr) |
||||||
|
|
³ 7 |
332 |
1.00 |
|
|
|
|
|
1-6 |
3,359 |
0.63 |
0.53-0.75 |
|
|
|
|
None |
1,114 |
0.30 |
0.19-0.48 |
|
|
|
Occupation |
||||||
|
|
Non-agricultural |
1,010 |
1.00 |
|
1.00 |
|
|
|
Unemployed or housewife |
1,187 |
0.77 |
0.62-0.95 |
0.90 |
0.72-1.13 |
|
|
Agricultural |
2,608 |
0.37 |
0.30-0.45 |
0.52 |
0.41-0.65 |
|
Residence |
||||||
|
|
Urban |
412 |
1.00 |
|
|
|
|
|
Rural |
4,393 |
0.35 |
0.28-0.44 |
|
|
|
Altitude |
||||||
|
|
Lowland |
1,064 |
1.00 |
|
|
|
|
|
Coastal |
2,254 |
0.75 |
0.62-0.91 |
|
|
|
|
Hills or highland |
1,487 |
0.43 |
0.34-0.55 |
|
|
|
Water supply |
||||||
|
|
Private or public tap |
191 |
1.00 |
|
1.00 |
|
|
|
Private or public pump or well |
3,265 |
0.32 |
0.24-0.44 |
0.43 |
0.31-0.60 |
|
|
Spring, river, rain, or other |
1,349 |
0.14 |
0.09-0.19 |
0.28 |
0.19-0.42 |
|
Type of floor |
||||||
|
|
Ceramic or tile |
2,671 |
1.00 |
|
|
|
|
|
Wood |
18 |
0.91 |
0.26-3.17 |
|
|
|
|
Soil |
2,116 |
0.41 |
0.12-0.50 |
|
|
|
Type of latrine |
||||||
|
|
Private septic tank |
1,350 |
1.00 |
|
|
|
|
|
Private, no septic tank |
426 |
0.69 |
0.52-0.93 |
|
|
|
|
Shared or public toilet |
282 |
0.49 |
0.33-0.73 |
|
|
|
|
River, pond, or yard |
2,747 |
0.44 |
0.37-0.53 |
|
|
|
Electricity |
||||||
|
|
Yes |
3,066 |
1.00 |
|
|
|
|
|
No |
1739 |
0.46 |
0.38-0.55 |
|
|
|
Television ownership |
||||||
|
|
Yes |
1,751 |
1.00 |
|
1.00 |
|
|
|
No |
3,054 |
0.43 |
0.37-0.51 |
0.80 |
0.65-0.97 |
|
Refrigerator ownership |
|
|
|
|
|
|
|
|
Yes |
115 |
1.00 |
|
|
|
|
|
No |
4,690 |
0.26 |
0.18-0.39 |
|
|
|
Bicycle ownership |
|
|
|
|
|
|
|
|
Yes |
3,629 |
1.00 |
|
1.00 |
|
|
|
No |
1,176 |
0.45 |
0.36-0.57 |
0.70 |
0.53-0.91 |
|
Motorcycle ownership |
|
|
|
|
|
|
|
|
Yes |
654 |
1.00 |
|
1.00 |
|
|
|
No |
4,151 |
0.38 |
0.31-0.46 |
0.67 |
0.54-0.85 |
The mean BMI was 21.2 ± 3. 1. However, the interpretation of BMI should be based on practical BMI cut-offs [25], since the prevalence of thinness and overweight varies widely from country to country. Still, the mean BMI of the Indonesian women was higher than the average BMI of women of reproductive age in Ethiopia (18.5 ± 1.8) and India (18.0 ± 2.1), and slightly lower than that for Zimbabwian (22.0 ± 3.3) and Thai (21.4 ± 2.5) women [26, 27]. When the data from Purworejo women were compared with standards for chronic energy deficiency, most women were classified as normal (71.7%). However, when the sample was compared with the Canadian standards for weight of women 15 to 49 years old [17], the mean weight was below the 25th percentile. Further, when the sample was compared with the Canadian standards for height, the mean height was below the 5th percentile. Thus, the relatively normal BMI of these women may be a reflection of a large height deficit masking any weight deficit, again, as the result of poorer conditions earlier in life.
For adult women over 15 years old, the MUAC cutoff commonly used for indicating risk is 22.5 cm. The mean MUAC for the study sample was 25.8 ± 2.9 cm, which was higher than this cut-off point, although still only in the 25th percentile of the NHANES standards. The mean triceps skinfold thickness was 15.0 ± 6.3 mm, which lies in the 25th percentile of the NHANES standards [18].
We have shown that 17% of the Purworejo women had chronic energy deficiency, with 1.2%, 3.0%, and 12.8% having chronic energy deficiency grades I, II, and III, respectively. Thus, the prevalence of chronic energy deficiency was lower than that in East Java (41%) and also lower than that in other developing countries such as India (61%) and Ethiopia (57%). Still, it is higher than that in Zimbabwe (11%) and Thailand (11.6%) [21,26,27]. The percentage of obesity among Purworejo women was 11.4%, much higher than the less than I% obesity among Indian and Ethiopian women but lower than the 13.6% obesity among Thai women. Obesity among women is associated with increased risk of non-insulin-dependent diabetes mellitus, coronary heart disease, stroke, hypertension, gall bladder disease, menstrual irregularities, and cancers of the breast, cervix, endometrium, ovary, and gall bladder [28]. Obesity affects many women in Western countries; 35% of adult women in the United States are obese. It is more common among women of lower socio-economic status in Western countries, whereas in developing countries the opposite is true. As there is an increasing trend in cardiovascular disease in Indonesia [29], further increase of obesity among women should be prevented.
Both chronic energy deficiency and obesity were more prevalent among older women, and chronic energy deficiency was also common among the youngest women. Even though the distribution of chronic energy deficiency was significantly different among age groups, this association disappeared in the multivariate analyses, probably because some of the covariates also were related to age. However, obesity was clearly linked to older age. It has also been shown that BMI correlates well with age [30-33]. In many non-Western countries, low BMI is found especially among older people. In contrast, in Western populations low BMI is found predominantly among younger people. Thus, Purworejo districts contain subpopulations similar to both non-Western and Western countries.
Table 5. Relationship between background variables and obesity: Odds ratios (OR) and 95% confidence intervals (CI) are given for the risk of having obesity II versus being normal according to logistic regression analysis (n = 5,817)
|
Background variable |
n |
Univariate analysis |
Multivariate analysisa |
|||
|
OR |
95% CI |
OR |
95% CI |
|||
|
Age (yr) |
||||||
|
|
45-49 |
447 |
1.00 |
|
1.00 |
|
|
|
40-44 |
721 |
0.58 |
0.31-1.08 |
0.51 |
0.27-0.98 |
|
|
35-39 |
1,013 |
0.41 |
0.22-0.76 |
0.35 |
0.18-0.67 |
|
|
30-34 |
989 |
0.25 |
0.12-0.51 |
0.20 |
0.09-0.41 |
|
|
25-29 |
675 |
0.18 |
0.07-0.45 |
0.15 |
0.06-0.38 |
|
|
15-24 |
380 |
0.05 |
0.01-0.40 |
0.05 |
0.01-0.37 |
|
Parity |
||||||
|
|
0-2 |
1,848 |
1.00 |
|
|
|
|
|
3-4 |
1,573 |
1.22 |
0.72-2.08 |
|
|
|
|
³ 5 |
804 |
2.17 |
1.25-3.77 |
|
|
|
Occupation |
||||||
|
|
Non-agricultural |
824 |
1.00 |
|
1.00 |
|
|
|
Unemployed or housewife |
1,015 |
0.89 |
0.54-1.48 |
1.13 |
0.67-1.91 |
|
|
Agricultural |
2,386 |
0.19 |
0.10-0.35 |
0.26 |
0.14-0.50 |
|
Residence |
||||||
|
|
Urban |
314 |
1.00 |
|
|
|
|
|
Rural |
3,911 |
0.25 |
0.14-0.42 |
|
|
|
Altitude |
||||||
|
|
Lowland |
899 |
1.00 |
|
1.00 |
|
|
|
Coastal |
1,960 |
0.52 |
0.32-0.85 |
0.56 |
0.34-0.93 |
|
|
Hills or highland |
1,366 |
0.27 |
0.14-0.51 |
0.49 |
0.24-1.00 |
|
Water supply |
||||||
|
|
Private or public tap |
127 |
1.00 |
|
|
|
|
|
Private or public pump or well |
2,835 |
0.74 |
0.27-2.07 |
|
|
|
|
Spring, river, rain, or other |
1,263 |
0.20 |
0.06-0.66 |
|
|
|
Type of floor |
||||||
|
|
Ceramic or tile |
2,256 |
1.00 |
|
1.00 |
|
|
|
Wood |
16 |
2.25 |
0.29-17.27 |
4.27 |
0.52-34.87 |
|
|
Soil |
1,953 |
0.23 |
0.12-0.41 |
0.45 |
6.23-0.87 |
|
Type of latrine |
||||||
|
|
Private septic tank |
1,104 |
1.00 |
|
|
|
|
|
Private, no septic tank |
372 |
1.02 |
0.51-2.05 |
|
|
|
|
Shared or public toilet |
253 |
0.40 |
0.12-1.32 |
|
|
|
|
River, pond, or yard |
2,496 |
0.45 |
0.27-0.73 |
|
|
|
Electricity |
||||||
|
|
Yes |
2,621 |
1.00 |
|
|
|
|
|
No |
1,604 |
0.26 |
0.14-0.49 |
|
|
|
Television ownership |
||||||
|
|
Yes |
1,448 |
1.00 |
|
|
|
|
|
No |
2,777 |
0.26 |
0.16-0.42 |
|
|
|
Refrigerator ownership |
||||||
|
|
Yes |
77 |
1.00 |
|
|
|
|
|
No |
4,148 |
0.34 |
0.12-0.94 |
|
|
|
Motorcycle ownership |
||||||
|
|
Yes |
512 |
1.00 |
|
1.00 |
|
|
|
No |
3,713 |
0.24 |
0.15-0.39 |
0.53 |
0.32-0.89 |
Many studies have found associations between nutritional status and reproductive and socio-economic status [27, 31, 33-37]. The Purworejo women who worked in agriculture had the highest risk of chronic energy deficiency compared with non- agricultural workers (when possible confounding factors were adjusted for), and they were less likely to be obese. Being a housewife or being unemployed increased the risk of chronic energy deficiency by 28% to 38%. These findings were similar to those in Cuba [38], where the highest proportions of underweight and chronic energy deficiency were found among agricultural workers and housewives. This phenomenon may be explained by findings in Ethiopia, where the physical activity levels for agricultural, domestic, and other productive work in relation to maintenance energy cost in women were 2.8, 2.0, and 1.3, respectively, based on actual measurements [39]. Unfortunately, many types of heavy physical activity are such that individuals of low BMI are clearly at a disadvantage, and those types of activity could be important in some types of agricultural work [40].
As expected, women with better economic status, as indicated by the availability of a radio, television, refrigerator, drinking water, and tile floors, had better nutritional status. These findings were similar to those of Achadi and co-workers in West Java, where women owning a motorcycle and a radio had better nutritional status [22]. However, in our study higher proportions of both chronic energy deficiency and obesity were found among women who owned a bicycle. Thus, appropriate socio-economic indicators for these two study sites may differ. In West Java a motorcycle and radio may indicate relative wealth, whereas in Purworejo a radio, television, and refrigerator may instead indicate a similar level of wealth. Ownership of a bicycle increased with age, and so did the prevalence of both chronic energy deficiency and obesity; this may explain a spurious relationship.
We found that women who used contraception had the best nutritional status. Women not using any contraception had a 31% increased risk, and women without any information on contraceptive methods had more than a 95% higher risk of chronic energy deficiency. Conversely, these women were less likely to be obese. Women without information on contraceptive methods were older, had more children, had less education, and worked in agriculture. Thus, this variable could also act as a proxy for these other background factors.
Only in univariate analyses did parity show a significant association with obesity (but not with chronic energy deficiency); women with parity greater than four had a higher risk of being obese. In multivariate analyses, parity showed no significant association with obesity. However, the East Java study also found the prevalence of chronic energy deficiency to be unrelated to parity [41]. It could be that the degree of undernutrition influences reproductive performance and fecundity, rather than reproduction leading to maternal depletion. However, evaluation of women’s nutritional status across age and parity is probably not the correct way to investigate maternal depletion [42, 43].
Conclusions
In Purworejo, Central Java, Indonesia, 17% percent of non-pregnant women of reproductive age had chronic energy deficiency, 71.7% were normal, and 11.4% were obese. The height deficit was more severe than the weight deficit, indicating past malnutrition combined with an improved situation in today’s society. The major causes of malnutrition-limited resources and poor socioeconomic status of the population-will need to be addressed through programmes that increase the purchasing power of the poor throughout the year. There is a need to improve the nutritional status of both girls and women before and after pregnancy. Intervention during pregnancy, a period with high nutritional demands, may be too late. Efforts to enhance work opportunity, expand access to primary and secondary education, improve dietary intake, and facilitate the use of health and nutrition services are needed to improve women’s nutritional status.
Acknowledgments
We thank all those involved with the conduct of the study. It was supported by grants from Sida/SAREC (Swedish International Development Authority/Swedish Agency for Research Cooperation in the Developing Countries), Sweden (SWE-94-149), and the World Bank through the Community Health and Nutrition Development Project of the Ministry of Health, Indonesia (IBRD Loan No. 3550-IND).
1. Central Bureau of Statistics (CBS) Indonesia, State Ministry of Population/National Family Planning Coordinating Board (NFPCB), Ministry of Health (MOH), Macro International (MI). Indonesia demographic and health survey 1994 (IDHS, 1994). Calverton, Md, USA: CBS and MI, 1995.
2. BAPPENAS, UNICEF. Summary: Situation analysis of children and women in Indonesia. Jakarta: UNICEF, 1995.
3. Kramer MS, McLean FH, Eason EL, Usher RH. Maternal nutrition and spontaneous preterm birth. Am J Epidemiol 1992;136:574-83.
4. Shepard MJ, Bakketeig LS, Jacobsen G, O’Connor T, Bracken MB. Maternal body mass, proportional weight gain, and fetal growth in parous women. Paediatr Peri-natal Epidemiol 1996;10:207-19.
5. Siega-Riz AM, Adair LS, Hobel CJ. Institute of Medicine maternal weight gain recommendations and pregnancy outcome in predominantly Hispanic population. Obstet Gynecol 1994;84:565-73.
6. Siega-Riz AM, Adair LS, Hobel CJ. Maternal underweight status and inadequate rate of weight gain during the third trimester of pregnancy increases the risk of preterm delivery. J Nutr 1996;126:146-53.
7. Snyder J, Gray-Donald K, Koski KG. Predictors of infant birth weight in gestational diabetes. Am J Clin Nutr 1994;59:1409-14.
8. Cogswell ME, Serdula MK, Hungerford DW, Yip R. Gestational weight gain among average-weight and overweight women-What is excessive? Am J Obstet Gynecol 1995;172:705-12.
9. Cogswell ME, Yip R. The influence of fetal and maternal factors on the distribution of birthweight. Semin Perinatol 1995;19:222-40.
10. World Health Organization. Maternal anthropometry and pregnancy outcomes. A WHO collaborative study. Bull WHO 1995;73 (suppl).
11. Anderson MA, Krasovec K. Maternal nutrition and pregnancy outcome. Scientific Publication No. 529. Washington, DC: Pan American Health Organization, 1991:1-14.
12. Leslie J. Women’s nutrition: the key to improving family health in developing countries? Health Pol Plan 1991; 6(11):1-19.
13. Tinker A, Daly P, Green C, Saxenian H, Lakshminarayanan R, Gill K. Women’s health and nutrition. World Bank Discussion Paper No. 256. Washington, DC: World Bank, 1995.
14. Merchant KM, Kurtz KM. Women’s nutrition through the life cycle: social and biological vulnerabilities. In: Koblinsky M, Timyan J, Gay J, eds. San Francisco, Calif, USA: Westview Press, 1993:63-90.
15. United Nations Administrative Committee on Coordination/Sub-Committee on Nutrition. Second report on the world nutrition situation. Vol I: Global and regional results. Geneva: ACC/SCN, 1992.
16. Wilopo SA, Community Health and Nutrition Research Laboratory Team. Key issues on the research design, data collection and management. Reprints of the Community Health and Nutrition Research Laboratory No. 2. Yogyakarta, Indonesia: Community Health and Nutrition Research Laboratory, Faculty of Medicine, Gadjah Mada University, 1997.
17. Nutrition Canada. Anthropometry report: height, weight and body dimensions. Ottawa: Bureau of Nutritional Sciences, Health Protection Branch, Health and Welfare, 1980.
18. Frisancho AR. New norms of upper limb fat and muscle areas for assessment of nutritional status. Am J Clin Nutr 1981;34:2540-5.
19. James WPT, Ferro-Luzzi A, Waterlow JC. Definition of chronic energy deficiency in adults. Eur J Clin Nutr 1988; 42:969-81.
20. World Health Organization. Physical status: the use and interpretation of anthropometry. WHO Technical Report Series No. 854. Geneva: WHO, 1995.
21. Kusin JA, Kardjati S, Renqvist U, Goei K. Reproduction and maternal nutrition in Madura, Indonesia. Trop Geogr Med 1992;44:248-55.
22. Achadi EL, Hansell MJ, Sloan NL, Anderson MA. Women’s nutritional status, iron consumption and weight gain during pregnancy in relation to neonatal weight and length in West Java, Indonesia. Int J Gynecol Obstet 1995;48 (suppl): S103-19.
23. Soekirman, Tarwotjo I, Jus’at I, Sumodiningrat G, Jalal F. Economic growth, equity and nutritional improvement in Indonesia. Geneva: United Nations Administrative Committee on Coordination/Sub-Committee on Nutrition, 1992.
24. Habicht JP, Martorell R, Yarbrough C, Malina RM, Klein RE. Height and weight standards for preschool children. How relevant are ethnic differences in growth potential? Lancet 1974;1:611-5.
25. Onis M, Habicht JP. Anthropometric reference data for international use: recommendations from a World Health Organization expert committee. Am J Clin Nutr 1996; 64:650-8.
26. Ferro-Luzzi A, Sette S, Franklin M, James WPT. A simplified approach of assessing adult chronic energy deficiency. Eur J Clin Nutr 1992;46:173-86.
27. Sanchaisuriya P, Pongpaew P, Saowakontha S, Supawan V, Migesena P, Schelp FP. Nutritional health and parasitic infections of rural Thai women of the child bearing age. J Med Assoc Thai 1993;76:139-44.
28. Wolinsky I, Klimis-Tavantzis D. Nutritional concerns of women. New York: CRC Press, 1996.
29. Boedhi-Darmojo R. The pattern of cardiovascular disease in Indonesia. World Health Stat Q 1993;46:119-24.
30 Strickland SS, Ulijaszek SJ. Body mass index and illness in rural Sarawak. For J Clin Nutr 1994;48(suppl 3): S98-109.
31. De Vasconcellos MTL. Body mass index: its relationship with food consumption and socioeconomic variables in Brazil. Eur J Clin Nutr 1994;48(suppl 3): S115-23.
32. Giay T, Khoi HH. Use of body mass index in the assessment of adult nutritional status in Vietnam. Eur J Clin Nutr 1994;48(suppl 3):S124-30.
33. Shetty PS, James WPT. Body mass index: a measure of chronic energy deficiency in adults. Food and Agriculture Organization Food and Nutrition Paper No. 56. Rome: FAO, 1994.
34. Delpeuch F, Cornu A, Massamba J-P, Traissac P, Maire B. Is body mass index sensitively related to socio-economic status and to economic adjustment? A case study from the Congo. Eur J Clin Nutr 1994;48(suppl 3):S141-7.
35. Huffman SL, Wolff M, Lowell S. Nutrition and fertility in Bangladesh: nutritional status of nonpregnant women. Am J Clin Nutr 1985;42:725-38.
36. Allen LH, Lung’aho MS, Shaheen M, Harrison GG, Neumann C, Kirksey A. Maternal body mass index and pregnancy outcome in the Nutrition Collaborative Research Support Program. Eur J Clin Nutr 1994;48(suppl 3):S68-77.
37. McGuire J, Popkin BM. Beating the zero-sum game: women and nutrition in the third world. Part 1. Food Nutr Bull 1989; 11:38-63.
38. Berdasco A. Body mass index values in the Cuban adult population. Eur J Clin Nutr 1994;48(suppl 3):S155-64.
39. Ferro Luzzi A, Scaccini C, Taffese C, Aberra B, Demeke T. Seasonal energy deficiency in Ethiopian rural women. Eur J Clin Nutr 1990;44:7-18.
40. Durnin JVGA. Low body mass index, physical work capacity and physical activity levels. Eur J Clin Nutr 1994;48(suppl 3):S39-44.
41. Kusin JA, Kardjati S, Renqvist UH. Chronic undernutrition in pregnancy and lactation. Proc Nutr Soc 1993; 52:19-28.
42. Winkvist A, Rasmussen KM, Habicht JP. A new definition of the maternal depletion syndrome. Am J Public Health 1994;82:691-4.
43. Leslie J, Pelto GH, Rasmussen KM. Nutrition of women in developing countries. Food Nutr Bull 1988;10:4-7.
